Practical Methods for Biocatalysis and Biotransformations

Practical Methods forBiocatalysis and

Biotransformations

Editors

JOHN WHITTALL

Manchester Interdisciplinary Biocentre,University of Manchester, United Kingdom

PETER SUTTON

Synthetic Chemistry, GlaxoSmithKline R&D Ltd,United Kingdom

A John Wiley and Sons, Ltd., Publication

Practical Methods for Biocatalysis and

Biotransformations

Practical Methods forBiocatalysis and

Biotransformations

Editors

JOHN WHITTALL

Manchester Interdisciplinary Biocentre,University of Manchester, United Kingdom

PETER SUTTON

Synthetic Chemistry, GlaxoSmithKline R&D Ltd,United Kingdom

A John Wiley and Sons, Ltd., Publication

This edition first published 2010

� 2010 John Wiley & Sons Ltd

Registered office

John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, United Kingdom

For details of our global editorial offices, for customer services and for information about how to apply for

permission to reuse the copyright material in this book please see our website at www.wiley.com.

The right of the author to be identified as the author of this work has been asserted in accordance with the

Copyright, Designs and Patents Act 1988.

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in

any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by

the UK Copyright, Designs and Patents Act 1988, without the prior permission of the publisher.

Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be

available in electronic books.

Designations used by companies to distinguish their products are often claimed as trademarks. All brand names

and product names used in this book are trade names, service marks, trademarks or registered trademarks of their

respective owners. The publisher is not associated with any product or vendor mentioned in this book. This

publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It

is sold on the understanding that the publisher is not engaged in rendering professional services. If professional

advice or other expert assistance is required, the services of a competent professional should be sought.

The publisher and the author make no representations or warranties with respect to the accuracy or completeness

of the contents of this work and specifically disclaim all warranties, including without limitation any implied

warranties of fitness for a particular purpose. This work is sold with the understanding that the publisher is not

engaged in rendering professional services. The advice and strategies contained herein may not be suitable for

every situation. In view of ongoing research, equipment modifications, changes in governmental regulations, and

the constant flow of information relating to the use of experimental reagents, equipment, and devices, the reader is

urged to review and evaluate the information provided in the package insert or instructions for each chemical,

piece of equipment, reagent, or device for, among other things, any changes in the instructions or indication of

usage and for added warnings and precautions. The fact that an organization or Website is referred to in this work

as a citation and/or a potential source of further information does not mean that the author or the publisher

endorses the information the organization or Website may provide or recommendations it may make. Further,

readers should be aware that Internet Websites listed in this work may have changed or disappeared between when

this work was written and when it is read. No warranty may be created or extended by any promotional statements

for this work. Neither the publisher nor the author shall be liable for any damages arising herefrom.

Library of Congress Cataloging-in-Publication Data

Practical methods for biocatalysis and biotransformations / editors,

John Whittall, Peter Sutton.

p. ; cm.

Includes bibliographical references and index.

ISBN 978-0-470-51927-1

1. Enzymes—Biotechnology. 2. Biotransformation (Metabolism) 3. Organic compounds—

Synthesis. I. Whittall, John. II. Sutton, Peter (Peter W.)

[DNLM: 1. Biocatalysis. 2. Biotransformation. 3. Enzymes. QU 135 P895 2009]

TP248.65.E59P73 2009

660.6034—dc22

2009030811

A catalogue record for this book is available from the British Library.

ISBN 978-0-470-51927-1

Set in 10/12pt Times by Integra Software Services Pvt. Ltd, Pondicherry, India

Printed and bound in Great Britain by CPI Antony Rowe, Chippenham, Wiltshire

www.wiley.com

Contents

Preface xi

Abbreviations xiii

List of Contributors xix

1 Biotransformations in Small-molecule Pharmaceutical Development 1

Joseph P. Adams, Andrew J. Collis, Richard K. Henderson and Peter

W. Sutton

2 Biocatalyst Identification and Scale-up: Molecular Biology for

Chemists 83

Kathleen H. McClean

3 Kinetic Resolutions Using Biotransformations 117

3.1 Stereo- and Enantio-selective Hydrolysis of rac-2-Octylsulfate Using

Whole Resting Cells of Pseudomonas spp. 117

Petra Gadler and Kurt Faber

3.2 Protease-catalyzed Resolutions Using the 3-(3-Pyridine)propionyl

Anchor Group: p-Toluenesulfonamide 121

Christopher K. Savile and Romas J. Kazlauskas

3.3 Desymmetrization of Prochiral Ketones Using Enzymes 125

Andrew J. Carnell

3.4 Enzymatic Resolution of 1-Methyl-tetrahydroisoquinoline

using Candida rugosa Lipase 129

Gary Breen

4 Dynamic Kinetic Resolution for the Synthesis of Esters, Amides and

Acids Using Lipases 133

4.1 Dynamic Kinetic Resolution of 1-Phenylethanol by Immobilized

Lipase Coupled with In Situ Racemization over Zeolite Beta 133

Kam Loon Fow, Yongzhong Zhu, Gaik Khuan Chuah and Stephan

Jaenicke

4.2 Synthesis of the (R)-Butyrate Esters of Secondary Alcohols by

Dynamic Kinetic Resolution Employing a Bis(tetrafluorosuccinato)-

bridged Ru(II) Complex 137

S.F.G.M. van Nispen, J. van Buijtenen, J.A.J.M. Vekemans,

J. Meuldijk and L.A. Hulshof

4.3 Dynamic Kinetic Resolution 6,7-Dimethoxy-1-methyl-1,2,3,4-

tetrahydroisoquinoline 141

Michael Page, John Blacker and Matthew Stirling

4.4 Dynamic Kinetic Resolution of Primary Amines with a Recyclable

Palladium Nanocatalyst (Pd/AlO(OH)) for Racemization 148

Soo-Byung Ko, Mahn-Joo Kim and Jaiwook Park

4.5 Dynamic Kinetic Resolution of Amines Involving Biocatalysis and

In Situ Free-radical-mediated Racemization 153

Stephane Gastaldi, Gerard Gil and Michele P. Bertrand

4.6 Chemoenzymatic Dynamic Kinetic Resolution of (S)-Ibuprofen 157

A.H. Kamaruddin and F. Hamzah

4.7 Dynamic Kinetic Resolution Synthesis of a Fluorinated Amino Acid

Ester Amide by a Continuous Process Lipase-mediated Ethanolysis

of an Azalactone 162

Matthew Truppo, David Pollard, Jeffrey Moore and Paul Devine

5 Enzymatic Selectivity in Synthetic Methods 165

5.1 Alcalase-catalysed Syntheses of Hydrophilic Di- and Tri-peptides in

Organic Solvents 165

Xue-Zhong Zhang, Rui-Zhen Hou, Li Xu and Yi-Bing Huang

5.2 Selective Alkoxycarbonylation of 1�,25-Dihydroxyvitamin D3 Diol

Precursor with Candida antarctica Lipase B 170

Miguel Ferrero, Susana Fernandez and Vicente Gotor

5.3 The Use of Lipase Enzymes for the Synthesis of Polymers and

Polymer Intermediates 173

Alan Taylor

5.4 Bioconversion of 3-Cyanopyridine into Nicotinic Acid with Gordona

terrae NDB1165 182

Tek Chand Bhalla

5.5 Enzyme-promoted Desymmetrization of Prochiral Dinitriles 186

Marloes A. Wijdeven, Piotr Kiełbasinski and Floris P.J.T. Rutjes

5.6 Epoxide Hydrolase-catalyzed Synthesis of (R)-3-Benzyloxy-2-

methylpropane-1,2-diol 190

Takeshi Sugai, Aya Fujino, Hitomi Yamaguchi and Masaya Ikunaka

5.7 One-pot Biocatalytic Synthesis of Methyl (S)-4-Chloro-3-

hydroxybutanoate and Methyl (S)-4-Cyano-3-hydroxybutanoate 199

Maja Majeric Elenkov, Lixia Tang, Bernhard Hauer and Dick

B. Janssen

vi Contents

6 Aldolase Enzymes for Complex Synthesis 203

6.1 One-step Synthesis of L-Fructose Using Rhamnulose-1-phosphate

Aldolase in Borate Buffer 203

William A. Greenberg and Chi-Huey Wong

6.2 Straightforward Fructose-1,6-bisphosphate Aldolase-mediated

Synthesis of Aminocyclitols 206

Marielle Lemaire and Lahssen El Blidi

6.3 Synthesis of D-Fagomine by Aldol Addition of Dihydroxyacetone to

N-Cbz-3-AminopropanalCatalysedbyD-Fructose-6-phosphateAldolase 212

Jose A. Castillo, Teodor Parella, Tomoyuki Inoue, Georg

A. Sprenger, Jesus Joglar and Pere Clapes

6.4 Chemoenzymatic Synthesis of 5-Thio-D-xylopyranose 218

Franck Charmantray, Philippe Dellis, Virgil Helaine, Soth Samreth

and Laurence Hecquet

7 Enzymatic Synthesis of Glycosides and Glucuronides 227

7.1 Glycosynthase-assisted Oligosaccharide Synthesis 227

Adrian Scaffidi and Robert V. Stick

7.2 Glycosyl Azides: Novel Substrates for Enzymatic

Transglycosylations 232

Vladimır Kren and Pavla Bojarova

7.3 Facile Synthesis of Alkyl �-D-Glucopyranosides from D-Glucose and

the Corresponding Alcohols Using Fruit Seed Meals 236

Wen-Ya Lu, Guo-Qiang Lin, Hui-Lei Yu, Ai-Ming Tong and

Jian-He Xu

7.4 Laccase-mediated Oxidation of Natural Glycosides 240

Cosimo Chirivı, Francesca Sagui and Sergio Riva

7.5 Biocatalysed Synthesis of Monoglucuronides of Hydroxytyrosol,

Tyrosol, Homovanillic Alcohol and 3-(40-Hydroxyphenyl)propanol

Using Liver Cell Microsomal Fractions 245

Olha Khymenets, Pere Clapes, Teodor Parella, Marıa-Isabel Covas,

Rafael de la Torre, and Jesus Joglar

7.6 Synthesis of the Acyl Glucuronide of Mycophenolic Acid 251

Matthias Kittelmann, Lukas Oberer, Reiner Aichholz and Oreste

Ghisalba

8 Synthesis of Cyanohydrins Using Hydroxynitrile Lyases 255

8.1 Synthesis of (S)-2-Hydroxy-2-methylbutyric Acid by a

Chemoenzymatic Methodology 255

Manuela Avi and Herfried Griengl

8.2 (S)-Selective Cyanohydrin Formation from Aromatic Ketones Using

Hydroxynitrile Lyases 259

Chris Roberge, Fred Fleitz and Paul Devine

Contents vii

8.3 Hydroxynitrile-lyase-catalysed Synthesis of Enantiopure

(S)-Acetophenone Cyanohydrins 262

Jan von Langermann, Annett Mell, Eckhard Paetzold and Udo Kragl

8.4 (R)- and (S)-Cyanohydrin Formation from Pyridine-

3-carboxaldehyde Using CLEATM-immobilized Hydroxynitrile Lyases 266

Chris Roberge, Fred Fleitz and Paul Devine

8.5 A New (R)-Hydroxynitrile Lyase from Prunus mume for Asymmetric

Synthesis of Cyanohydrins 269

Yasuhisa Asano

9 Synthesis of Chiral sec-Alcohols by Ketone Reduction 273

9.1 Asymmetric Synthesis of (S)-Bis(trifluoromethyl)phenylethanol by

Biocatalytic Reduction of Bis(trifluoromethyl)acetophenone 273

David Pollard, Matthew Truppo and Jeffrey Moore

9.2 Enantioselective and Diastereoselective Enzyme-catalyzed Dynamic

Kinetic Resolution of an Unsaturated Ketone 276

Birgit Kosjek, David Tellers and Jeffrey Moore

9.3 Enzyme-catalysed Synthesis of �-Alkyl-�-hydroxy Ketones and

Esters by Isolated Ketoreductases 278

Ioulia Smonou and Dimitris Kalaitzakis

9.4 Asymmetric Reduction of Phenyl Ring-containing Ketones Using

Xerogel-encapsulated W110A Secondary Alcohol Dehydrogenase

from Thermoanaerobacter ethanolicus 284

Musa M. Musa, Karla I. Ziegelmann-Fjeld, Claire Vieille, J. Gregory

Zeikus and Robert S. Phillips

9.5 (R)- and (S)-Enantioselective Diaryl Methanol Synthesis Using

Enzymatic Reduction of Diaryl Ketones 288

Matthew Truppo, Krista Morley, David Pollard and Paul Devine

9.6 Highly Enantioselective and Efficient Synthesis of

Methyl (R)-o-Chloromandelate, Key Intermediate for Clopidogrel

Synthesis, with Recombinant Escherichia coli 291

Tadashi Ema, Nobuyasu Okita, Sayaka Ide and Takashi Sakai

10 Reduction of Functional Groups 295

10.1 Reduction of Carboxylic Acids by Carboxylic Acid Reductase

Heterologously Expressed in Escherichia coli 295

Andrew S. Lamm, Arshdeep Khare and John P.N. Rosazza

10.2 Light-driven Stereoselective Biocatalytic Oxidations and Reductions 299

Andreas Taglieber, Frank Schulz, Frank Hollmann, Monika Rusek

and Manfred T. Reetz

10.3 Unnatural Amino Acids by Enzymatic Transamination: Synthesis of

Glutamic Acid Analogues with Aspartate Aminotransferase 306

Thierry Gefflaut, Emmanuelle Sagot and Jean Bolte

viii Contents

10.4 Synthesis of L-Pipecolic Acid with �1-Piperidine-2-carboxylate

Reductase from Pseudomonas putida 310

Hisaaki Mihara and Nobuyoshi Esaki

10.5 Synthesis of Substituted Derivatives of L-Phenylalanine and of other

Non-natural L-Amino Acids Using Engineered Mutants of

Phenylalanine Dehydrogenase 314

Philip Conway, Francesca Paradisi and Paul Engel

11 Enzymatic Oxidation Chemistry 319

11.1 Monoamine Oxidase-catalysed Reactions: Application Towards the

Chemo-enzymatic Deracemization of the Alkaloid (–)-Crispine A 319

Andrew J. Ellis, Renate Reiss, Timothy J. Snape and Nicholas

J. Turner

11.2 Glucose Oxidase-catalysed Synthesis of Aldonic Acids 323

Fabio Pezzotti, Helene Therisod and Michel Therisod

11.3 Oxidation and Halo-hydroxylation of Monoterpenes with

Chloroperoxidase from Leptoxyphium fumago 327

Bjoern-Arne Kaup, Umberto Piantini, Matthias Wust and Jens

Schrader

11.4 Chloroperoxidase-catalyzed Oxidation of Phenyl Methylsulfide in

Ionic Liquids 330

Cinzia Chiappe

11.5 Stereoselective Synthesis of �-Hydroxy Sulfoxides Catalyzed by

Cyclohexanone Monooxygenase 332

Stefano Colonna, Nicoletta Gaggero, Sara Pellegrino and Francesca

Zambianchi

11.6 Enantioselective Kinetic Resolution of Racemic 3-Phenylbutan-2-

one Using a Baeyer–Villiger Monooxygenase 337

Anett Kirschner and Uwe T. Bornscheuer

11.7 Desymmetrization of 1-Methylbicyclo[3.3.0]octane-2,8-dione by the

Retro-claisenase 6-Oxo Camphor Hydrolase 341

Gideon Grogan and Cheryl Hill

11.8 Synthesis of Optically Pure Chiral Lactones by Cyclopentadecanone

Monooxygenase-catalyzed Baeyer–Villiger Oxidations 344

Shaozhao Wang, Jianzhong Yang and Peter C.K. Lau

12 Whole-cell Oxidations and Dehalogenations 351

12.1 Biotransformations of Naphthalene to 4-Hydroxy-1-tetralone by

Streptomyces griseus NRRL 8090 351

Arshdeep Khare, Andrew S. Lamm and John P.N. Rosazza

12.2 Hydroxylation of Imidacloprid for the Synthesis of Olefin

Imidacloprid by Stenotrophomonas maltophilia CGMCC 1.1788 355

Sheng Yuan and Yi-jun Dai

Contents ix

12.3 Biocatalytic Synthesis of 6-Hydroxy Fluvastatin using Motierella

rammaniana DSM 62752 in Shake Flask Culture and on Multi-gram

Scale using a Wave Bioreactor 359

Matthias Kittelmann, Maria Serrano Correia, Anton Kuhn, Serge

Parel, Jurgen Kuhnol, Reiner Aichholz, Monique Ponelle and Oreste

Ghisalba

12.4 Synthesis of 1-Adamantanol from Adamantane through

Regioselective Hydroxylation by Streptomyces griseoplanus Cells 367

Koichi Mitsukura, Yoshinori Kondo, Toyokazu Yoshida and Toru

Nagasawa

12.5 Enantioselective Benzylic Microbial Hydroxylation of Indan and

Tetralin 369

Renata P. Limberger, Cleber V. Ursini, Paulo J.S. Moran and

J. Augusto R. Rodrigues

12.6 Stereospecific Biotransformation of (R,S)-Linalool by Corynespora

cassiicola DSM 62475 into Linalool Oxides 376

Marco-Antonio Mirata and Jens Schrader

12.7 The Biocatalytic Synthesis of 4-Fluorocatechol from Fluorobenzene 379

Louise C. Nolan and Kevin E. O’Connor

12.8 Synthesis of Enantiopure (S)-Styrene Oxide by Selective Oxidation

of Styrene by Recombinant Escherichia coli JM101 (pSPZ10) 385

Katja Buehler and Andreas Schmid

12.9 Biotransformation of �-Bromo and �,�0-Dibromo Alkanone into

�-Hydroxyketone and �-Diketone by Spirulina platensis 391

Takamitsu Utsukihara and C. Akira Horiuchi

Index 397

x Contents

Preface

During the early to mid 1990s Professor Stan Roberts was chief editor of a series of loose-

leaf laboratory protocols detailing the use of biotransformations in synthetic organic

chemistry that were collected together and published in book form (Preparative

Biotransformations, Wiley, Chichester, 1999). This led to the publication of the series of

books Catalysts for Fine Chemical Synthesis, volumes 1–5, by the same publisher which

covered the application of chemo- and bio-catalytic procedures for the synthesis of fine

chemicals; for this series, Dr John Whittall became co-editor on the homogeneous cata-

lysis volumes. Following the format of this series, Practical Methods in Biocatalysis and

Biotransformations has been prepared. In keeping with these earlier formats, we aim to

provide the readership with enough information to understand when a biocatalytic or

biotransformation method would be a suitable practical method to carry out their synthetic

transformation.

In recent times, the employment of enzymes and whole cells to perform a range of

organic reactions has become much more commonplace, and biotransformation has

become accepted as a powerful method for application in synthetic organic chemistry.

However, for chemists developing synthetic methods for a particular target molecule, the

understanding of the advantages and limitations of biocatalysis and biotransformation is

not always clear. Therefore, this book intends to review the industrial background to when

biotransformations are used and introduce the nonmicrobiologist to the background of how

biocatalysts are discovered and developed and then give detailed experimental procedures

for a comprehensive range of useful biotransformation methods.

In order to place the later chapters in proper context, Chapter 1 offers a comprehensive

review of biotransformation from the perspective of a large pharmaceutical company

(GSK) and Chapter 2 gives an introduction that allows an appreciation of molecular

biology for scientists with no formal training in this area.

In the remaining chapters, key biotransformations have been identified from the recent

primary literature (learned journals) and the respective authors have amplified the dis-

closure of their methodologies in this volume. These disclosures often contain additional

equipment and experimental details to those found in the experimental section of most

journals, allowing the reader to decide whether these methods are suitable for addressing

their needs.

Chapter 3 describes the application of lipases, proteases and sulfatases for the kinetic

resolution of a range of interesting molecules. A selection of dynamic kinetic resolution

(DKR) procedures is disclosed in Chapter 4. DKRs are attracting a significant amount of

interest as they allow access to >50 % yields of single enantiopure products from

racemates. Other useful synthetic applications of hydrolase enzymes are covered in

Chapter 5, including desymmetrization and regio- and chemo-selective transformations.

Chapters 6 and 7 cover sugar-type chemistry, focusing on aldol and glycosylation

methods which can offer substantial advantages over traditional chemical approaches.

Chapter 8 describes the application of hydroxyl nitrile lyases to the synthesis of new

chiral cyanohydrins and �-hydroxy acids and includes new approaches to the transforma-

tion of ‘difficult’ aldehyde and ketone substrates using substrate engineering and immo-

bilization techniques.

The latter part of the book is dedicated to redox biotransformation application, with

Chapter 9 disclosing several methods for the synthesis of chiral secondary alcohols using a

range of commercially available ketoreductases (alcohol dehydrogenases) which are now

being applied regularly on a large scale.

Chapter 10 covers reductive enzymes with an emphasis on transaminase enzymes,

which are enjoying widespread application in the synthesis of nonnatural amino acids

which are key building blocks for several products of industrial importance.

The use of a range of oxidative enzymes in synthesis is covered in Chapter 11, whilst the

very powerful technique of regio- and stereo-specific biohydroxylation of even complex

molecules by fermenting whole-cell methods is covered in Chapter 12.

The Editors are most grateful to the authors who have submitted details of their

procedures in the prescribed format for inclusion in this book. We hope that this book

will increase the exposure of these methods to the chemical community and contribute to

the expanded employment of biocatalysis in organic synthesis.

John Whittall, Manchester

Peter Sutton, Stevenage

2009

xii Preface

Abbreviations

A adenine

ABTS 2,20-azino-bis-3-ethylbenzothiazoline-6-sulfonic acid

7-ACA 7-aminocephalosporanic acid

ACN acetonitrile

AcOH acetic acid

ACS GCI American Chemical Society Green Chemistry Institute

7-ADCA 7-aminodesacetoxycephalosporanic acid

ADH alcohol dehydrogenase (alternative name for a ketoreductases or KREDs)

ADH-RE alcohol dehydrogenase from Rhodococcus erythropolis

AIBN 2,20-azobis(2-methylpropionitrile)

6-APA 6-aminopenicillanic acid

API active pharmaceutical ingredient

Ara-G 9-b-D-arabinofuranosylguanidine

Ara-U 9-b-D-arabinofuranosyluridine

AspAT aspartate aminotransferase

AT aminotransferases

AZT 30-azido-20,30-dideoxythymidine (zidovudine)

BEHP bis(2-ethylhexyl)phthalate

BES N,N-bis(2-hydroxyethyl)-2-aminoethanesulfonic acid

BLAST basic local alignment search tool

BREP butanol-rinsed enzyme preparation

BSA bovine serum albumin

BSA N-bromosuccinimide

BVMO Baeyer–Villiger monooxygenase

C cytosine

CAL-A lipase A from Candida antarctica

CAL-B lipase B from Candida antarctica

Car carboxylic acid reductase

CASTing combinatorial active site saturation test

Cbz Benzyloxycarbonyl

CCL lipase from Candida cylindracea (now known as lipase from

Candida rugosa or CRL)

CDI 1,10-carbonyldiimidazole

CDW cell dry weight

cGMP current good manufacturing practice

CHMO cyclohexanone monooxygenase

CINV chemotherapy-induced nausea

CLEA cross-linked enzyme aggregate

CLEC cross-linked enzyme crystal

CNS Central nervous system

CPDMO cyclopentadecanone monooxygenase

CPO chloroperoxidase

CRL lipase from Candida rugosa

CSA cysteine sulfinic acid

CYP cytochrome P450

DBDMH N,N0-dibromodimethylhydantoin

DBE di-n-butylether

DCM dichloromethane

DCW dry cell weight

DDI drug–drug interaction

DERA 2-deoxyribose-5-phosphate aldolase

dGTP deoxyguanosine triphosphate

DHA dihydroxyacetone

DHAP dihydroxyacetone phosphate

DHF dihydrofolate

DIPE diisopropylether

DKR dynamic kinetic resolution

DMAP 4-dimethylaminopyridine

DMF dimethylformamide

DMSO dimethylsulfoxide

DNA deoxyribonucleic acid

DNAse deoxyribonuclease

dNTP deoxyribonucleotide triphosphate

DoE design of experiment

DOT dissolved oxygen tension

DSMZ Deutsche Sammlung von Mikroorganismen und Zellkulturen

dsDNA double-stranded DNA

D4T dideoxydidehydrothymidine

DTT dithiothreitol

dUDP 20-deoxyuridine-50-diphosphate

dUMP 20-deoxyuridine-50-monophosphate

E enantiomeric ratio

EDC 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride

EDTA ethylenediaminetetraacetic acid

xiv Abbreviations

EHS environmental health and safety

epPCR error-prone PCR

EtOAc Ethyl acetate

FACS fluorescence-activated cell sorting

FAD flavin adenine dinucleotide

FADH2 flavin adenine dinucleotide, reduced form

FASTA FAST ALL (a programme for fast protein comparison or fast

nucleotide sequence comparison)

FDA Food and Drug Administration (United States)

FDH formate dehydrogenase

FMN flavin mononucleotide (riboflavin-50-phosphate)

FPLC fast protein liquid chromatography

FruA fructose-1,6-bisphosphate aldolase

FSA D-fructose-6-phosphate aldolase

FTIR Fourier-transform infrared spectroscopy

GABA g-aminobutyric acid

G guanine

GC gas chromatography

GDH glucose dehydrogenase

GlcI glucose isomerase

GMO genetically modified organism

GMM genetically modified microorganism

G6P glucose-6-phosphate

G6PDH glucose-6-phosphate dehydrogenase

GPC gel permeation chromatography

GPO L-glycerol-3-phosphate oxidase

GR glucocorticoid receptor

GRAS generally recognized as safe

GSK GlaxoSmithKline

HBV hepatitis B virus

HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid

HIV human immunodeficiency virus

HMQC heteronuclear multiple quantum coherence

HNL hydroxynitrile lyase

HOBt 1-hydroxybenzotriazole

HOPhPr hydroxyphenylpropanol

HOTYR hydroxytyrosol

HPA hydroxypyruvate

HPI N-hydroxyphthalimide

HPLC high-performance liquid chromatography

HTS high-throughput screening

HVAlc Homovanillic alcohol

Abbreviations xv

IMI imidacloprid

Indels insertions and deletions

IP intellectual property

IPTG isopropyl-b-D-thiogalactopyranoside

ISPR in situ product removal

KPB potassium phosphate buffer

KR kinetic resolution

KRED ketoreductase (alternative name for an alcohol dehydrogenase or ADH)

LAS lovastatin ammonium salt

LB Luria–Bertani

LCA life cycle analysis

LovD acyltransferase from the lovastatin biosynthetic pathway

Mab monoclonal antibody

MAO-N monoamine oxidase

MEA malt extract agar

MES 2-morpholino ethansulfonic acid monohydrate

MGF minimum genome factories

MML lipase from Mucor sp.

MOPS 3-morpholino propane sulfonic acid

m.p. melting point

MPA mycophenolic acid

MPLC medium-pressure chromatography

mRNA messenger RNA

MS molecular sieves

MTBE tert-butylmethylether

MTQ methyl-tetrahydroisoquinoline

MYB malt yeast broth

NADþ b-nicotinamide adenine dinucleotide

NADH b-nicotinamide adenine dinucleotide, reduced form

NADPH b-nicotinamide adenine dinucleotide 20-phosphate, reduced form

NADPþ b-nicotinamide adenine dinucleotide 20-phosphate

NAG N-acetyl-D-glucosamine

NAM N-acetyl-D-mannosamine

NANA N-acetyl-D-neuraminic acid

NCE new chemical entity

NK-1 neurokinin-1

NME new molecular entity

NMR nuclear magnetic resonance

NP nucleoside phosphorylase

OCH 6-oxo camphor hydrolase

ORI origin of replication

P450 cytochrome P450

xvi Abbreviations

P450 BM-3 cytochrome P450 BM-3 from Bacillus megaterium

PAMO phenylacetone monooxygenase

Pase acid phosphatase

PAT process analytical technology

PCL lipase from Pseudomonas cepacia (now renamed to Burkholderia

cepacia)

PCR polymerase chain reaction

PDCB potato–dextrose–carrot broth

PEP phosphoenolpyruvic acid

PFL lipase from Pseudomonas fluorescens

PGA penicillin G acylase

Pip2C D1-piperideine-2-carboxylate reductase

PLE pig liver esterase

pNPG p-nitrophenyl-b-D-glucopyranoside

PNP purine nucleoside phosphorylase

PPL porcine pancreatic lipase

ProSAR protein sequence–activity relationship

QbD quality by design

QSAR quantitative structure–activity relationship

RAMA rabbit muscle aldolase (fructose-1,6-bis-phosphate aldolase)

R&D research and development

rDNA recombinant DNA

rRNA ribosomal RNA

Rf retention factor

RhaD rhamnulose-1-phosphate aldolase

RNA ribonucleic acid

ROH generic alcohol

Rt retention time

SAS simvastatin ammonium salt

SCR Saccharomyces cerevisiae carbonyl reductase

SIGEX substrate-induced gene-expression screening

SMB simulated moving bed chromatography

SOT Spirulina–Ogawa–Terui

ssDNA single-stranded DNA

T thymine

Taq a thermostable DNA polymerase from Thermus aquaticus

TBDMSCl tert-butyldimethylsilyl chloride

TBME tert-butylmethylether

TCA trichloroacetic acid

TDP thymidine 50-phosphate

TdR thymidine

TEMPO 2,2,6,6-tetramethyl-1-piperidinyloxy

Abbreviations xvii

TFA trifluoroacetic acid

THF tetrahydrofuran

THFo tetrahydrofolate

ThDP thiamine pyrophosphate

TK transketolase

TLC thin-layer chromatography

TMP thymidine 50-monophosphate

TMS tetramethyl silane

TMOS tetramethyl orthosilicate

TMSOTf trimethylsilyl triflate

TP thymidine-50-phosphorylase

tris-HCl tris(hydroxymethyl)aminomethane HCl

TTN total turnover number

TYR tyrosol

U unit of enzyme activity (mmol min�1)

U uracil

UdR 20-deoxyuridine

UDP uridine-50-diphosphate

UTP uridine-50-triphosphate

UDPGA uridine-50-diphosphoglucuronic acid

UDPGT uridine-50-diphosphoglucuronyl transferase

URDP uridine-50-phosphorylase

UV ultraviolet

VVM gas volume flow per unit of liquid volume per minute

WFCC World Federation for Culture Collections

YPG yeast extract–peptone–glucose

xviii Abbreviations

List of Contributors

Joseph P. Adams, GlaxoSmithKline, Synthetic Chemistry, Gunnels Wood Road,

Stevenage, Hertfordshire SG1 2NY, UK

Reiner Aichholz, Metabolism and Pharmacokinetics, NIBR, Novartis Pharma AG,

CH-4002 Basel, Switzerland

Yasuhisa Asano, Biotechnology Research Center and Department of Biotechnology,

Toyama Prefectural University, 5180 Kurokawa, Imizu, Toyama 939-0398, Japan

Manuela Avi, Institute of Organic Chemistry, Graz University of Technology,

Stremayrgasse 16, 8010 Graz, Austria

Michele P. Bertrand, Laboratoire de Chimie Moleculaire Organique, LCP UMR 6264,

Boite 562, Universite Paul Cezanne, Aix-Marseille III, Faculte des Sciences St Jerome,

Avenue Escadrille Normandie-Niemen, 13397 Marseille Cedex 20, France

Tek Chand Bhalla, Department of Biotechnology, Himachal Pradesh University, Shimla

171005, India

John Blacker, NPIL Pharma Ltd, Leeds Road, Huddersfield, HD1 9GA, UK

Lahssen El Blidi, Laboratoire SEESIB, UMR 6504 CNRS, Universite Blaise Pascal,

24 avenue des Landais 63177 Aubiere cedex, France

Pavla Bojarova, Institute of Microbiology, Center of Biocatalysis and

Biotransformations, Academy of Sciences of the Czech Republic, Vıdenska 1083,

CZ-142 20 Prague 4, Czech Republic

Jean Bolte, Department of Chemistry, Universite Blaise Pascal, Clermont-Ferrand, France

Uwe T. Bornscheuer, Department of Biotechnology and Enzyme Catalysis, Institute of

Biochemistry, Greifswald University, Felix-Hausdorff-Str. 4, 17487 Greifswald,

Germany

Gary Breen, GlaxoSmithKline, Synthetic Chemistry, Leigh, Tonbridge, Kent, TN11

9AN, UK

Katja Buehler, Laboratory of Chemical Biotechnology, Faculty of Biochemical and

Chemical Engineering, TU Dortmund, Emil-Figge-Strasse 66, 44221 Dortmund,

Germany

J. van Buijtenen, Eindhoven University of Technology, Laboratory of Macromolecular

and Organic Chemistry, PO Box 513, 5600 MB Eindhoven, The Netherlands

Andrew J. Carnell, Department of Chemistry, Robert Robinson Laboratories, University

of Liverpool, Liverpool, L69 7ZD, UK

Jose A. Castillo, Biotransformation and Bioactive Molecules Group, Instituto de Quimica

Avanzada de Cataluna, Consejo Superior de Investigaciones Cientificas, Jordi Girona

18-26, 08034 Barcelona, Spain

Franck Charmantray, Laboratoire SEESIB, UMR 6504 CNRS, Universite Blaise Pascal,

24 avenue des Landais, 63177 Aubiere, France

Cinzia Chiappe, Dipartimento di Chimica e Chimica Industriale, Universit di Pisa, 56126

Pisa, Italy

Cosimo Chirivı, Istituto di Chimica del Riconoscimento Molecolare, C.N.R., Via Mario

Bianco 9, 20131 Milano, Italy

Gaik Khuan Chuah, Department of Chemistry, National University of Singapore, Kent

Ridge, Singapore 119260, Republic of Singapore

Pere Clapes, Biotransformation and Bioactive Molecules Group, Instituto de Quimica



Andrew J. Collis, GlaxoSmithKline, Biotechnology and Environmental Shared Service,

North Lonsdale Road, Ulverston, Cumbria LA12 9DR, UK

Stefano Colonna, Dipartimento di Scienze Molecolari Applicate ai Biosistemi

(DISMAB), Facolta di Farmacia, Universita degli Studi di Milano, via Venezian 21,

20133 Milano, Italy

Philip Conway, School of Biomolecular and Biomedical Science, University College

Dublin, Belfield, Dublin 4, Ireland

Maria Serrano Correia, Rua Maria Auxiliadora, n�147, 6�andar porta 3, Bairro do

Rosario, P-2750-616 Cascais, Portugal

Marıa-Isabel Covas, Research Unit on Lipids and Cardiovascular Epidemiology, Institut

Municipal d’Investigacio Medica (IMIM). Universitat Pompeu Fabra (CEXS-UPF),

Barcelona, Spain

Yi-jun Dai, Nanjing Engineering Research Center for microbiology, Jiangsu Key

Laboratory for Biodiversity and Biotechnology, College of Life Science, Nanjing

Normal University, 1, Wenyuan Rd, Nanjing 210046, PR China

Philippe Dellis, Synkem, 47 rue de Longvic, 21300 Chenove, France

Paul Devine, Process Research, Merck Research Laboratories, Merck & Co. Inc. Rahway,

NJ, USA

xx List of Contributors

Andrew J. Ellis, School of Chemistry, Manchester Interdisciplinary Biocentre, University

of Manchester, 131 Princess Street, Manchester, M1 7DN, UK

Tadashi Ema, Division of Chemistry and Biochemistry, Graduate School of Natural

Science and Technology, Okayama University, Tsushima, Okayama 700-8530, Japan

Paul Engel, School of Biomolecular and Biomedical Science, University College Dublin,

Belfield, Dublin 4, Ireland

Nobuyoshi Esaki, Institute for Chemical Research, Kyoto University, Uji, Kyoto

611-0011, Japan

Kurt Faber, Department of Chemistry, Organic and Bioorganic Chemistry, University of

Graz, Heinrichstrasse 28, 8010 Graz, Austria

Susana Fernandez, Departamento de Quımica Organica e Inorganica and Instituto

Universitario de Biotecnologıa de Asturias, Universidad de Oviedo, 33006-Oviedo

(Asturias), Spain

Miguel Ferrero, Departamento de Quımica Organica e Inorganica and Instituto

Universitario de Biotecnologıa de Asturias, Universidad de Oviedo, 33006-Oviedo

(Asturias), Spain

Fred Fleitz, Process Research, Merck Research Laboratories, Merck & Co. Inc. Rahway,

NJ, USA

Kam Loon Fow, Department of Chemistry, National University of Singapore, Kent


Aya Fujino, Department of Chemistry, Faculty of Science and Technology, Keio

University, Hiyoshi, Kohoku-ku, Yokohama 223-8522, Japan

Petra Gadler, Department of Chemistry, Organic and Bioorganic Chemistry, University

of Graz, Heinrichstrasse 28, 8010 Graz, Austria

Nicoletta Gaggero, Dipartimento di Scienze Molecolari Applicate ai Biosistemi

(DISMAB), Facolta di Farmacia, Universita degli Studi di Milano, via Venezian 21,

20133 Milano, Italy

Stephane Gastaldi, Laboratoire de Chimie Moleculaire Organique, LCP UMR 6264,

Boite 562, Universite Paul Cezanne, Aix-Marseille III, Faculte des Sciences St Jerome,

Avenue Escadrille Normandie-Niemen, 13397 Marseille Cedex 20, France

Thierry Gefflaut, Department of Chemistry, Universite Blaise Pascal, Clermont-Ferrand,

France

Oreste Ghisalba, Ghisalba Life Sciences GmbH, Habshagstrasse 8c, CH-4153 Reinach,

Switzerland

Gerard Gil, Laboratoire de Stereochimie Dynamique et Chiralite, ISM2, UMR 6263,

Universite Paul Cezanne, Aix-Marseille III, Faculte des Sciences St Jerome, Avenue

Escadrille Normandie-Niemen, 13397 Marseille Cedex 20, France

List of Contributors xxi

Vicente Gotor, Departamento de Quımica Organica e Inorganica and Instituto Universitario

de Biotecnologıa de Asturias, Universidad de Oviedo, 33006-Oviedo (Asturias), Spain

William A. Greenberg, Department of Chemistry, The Scripps Research Institute, 10550

North Torrey Pines Rd., La Jolla, CA 92307, USA

Herfried Griengl, Research Centre Applied Biocatalysis, Petersgasse 14, 8010 Graz, Austria

Gideon Grogan, York Structural Biology Laboratory, Department of Chemistry,

University of York, Heslington, York, YO10 5YW, UK

F. Hamzah, School of Chemical Engineering, Engineering Campus, Universiti Sains

Malaysia, Seri Ampangan, 14300, Nibong Tebal, Penang, Malaysia

Bernhard Hauer, Institute of Technical Biochemistry, University of Stuttgart,

Allmandring 31, 70569 Stuttgart, Germany

Laurence Hecquet, Laboratoire SEESIB, UMR 6504 CNRS, Universite Blaise Pascal,

24 avenue des Landais, 63177 Aubiere, France

Virgil Helaine, Laboratoire SEESIB, UMR 6504 CNRS, Universite Blaise Pascal, 24 ave-

nue des Landais, 63177 Aubiere, France

Richard K. Henderson, GlaxoSmithKline, Centre of Excellence for Sustainability and

Environment, Park Road, Ware, Hertfordshire SG12 0DP, UK

Cheryl Hill, York Structural Biology Laboratory, Department of Chemistry, University of

York, Heslington, York, YO10 5YW, UK

Frank Hollmann, Max-Planck-Institut fur Kohlenforschung, Kaiser-Wilhelm-Platz 1,

45470 Mulheim/Ruhr, Germany

C. Akira Horiuchi, Department of Chemistry, Rikkyo (St.Paul’s) University, Nishi-

Ikebukuro, Toshima-Ku, Tokyo 171-8501, Japan

Rui-Zhen Hou, Key Laboratory for Molecular Enzymology and Engineering of Ministry

of Education, Jilin University, Changchun, 130021, PR China

Yi-Bing Huang, Key Laboratory for Molecular Enzymology and Engineering of Ministry

of Education, Jilin University, Changchun, 130021, PR China

L.A. Hulshof, Eindhoven University of Technology, Laboratory of Macromolecular and

Organic Chemistry, PO Box 513, 5600 MB Eindhoven, The Netherlands

Sayaka Ide, Division of Chemistry and Biochemistry, Graduate School of Natural Science

and Technology, Okayama University, Tsushima, Okayama 700-8530, Japan

Masaya Ikunaka, Fine Chemicals Department, Nagase & Co., Ltd., 5-1, Nihonbashi-

Kobunacho, Chuo-ku, Tokyo 103-8355, Japan

Tomoyuki Inoue, Institute of Microbiology, University of Stuttgart, Allmandring

31, 70569 Stuttgart, Germany

Stephan Jaenicke, Department of Chemistry, National University of Singapore, Kent


xxii List of Contributors

Dick B. Janssen, Biochemical Laboratory, Groningen Biomolecular Sciences and

Biotechnology Institute, University of Groningen, Nijenborgh 4, 9747 AG, Groningen,

The Netherlands

Jesus Joglar, Biotransformation and Bioactive Molecules Group, Instituto de Quimica



Dimitris Kalaitzakis, Department of Chemistry, University of Crete, Iraklion-Voutes,

71003 Crete, Greece

Azlina Kamaruddin, School of Chemical Engineering, Engineering Campus,

Universiti Sains Malaysia, Seri Ampangan, 14300, Nibong Tebal, Penang,

Malaysia

Bjoern-Arne Kaup, DECHEMA e.V., Karl-Winnacker-Institut, Biochemical

Engineering Group, Theodor-Heuss-Allee 25, 60486 Frankfurt, Germany

Romas J. Kazlauskas, Department of Biochemistry, Molecular Biology & Biophysics

and The Biotechnology Institute, University of Minnesota, 1479 Gortner Avenue, Saint

Paul, MN 55108, USA

Arshdeep Khare, Center for Biocatalysis and Bioprocessing, 2501 Crosspark Road, Suite

C100 MTF, University of Iowa, Iowa City, Iowa, IA 52242-5000, USA

Olha Khymenets, Pharmacology Research Unit, Institut Municipal d’Investigacio

Medica (IMIM), Barcelona, Spain

Piotr Kiełbasinski, Institute for Molecules and Materials, Radboud University Nijmegen,

Toernooiveld 1, NL-6525 ED Nijmegen, The Netherlands

Mahn-Joo Kim, Department of Chemistry, Pohang University of Science and Technology

(POSTECH), San-31, Hyojadong, Pohang 790-784, Korea

Anett Kirschner, Department of Biotechnology and Enzyme Catalysis, Institute of

Biochemistry, Greifswald University, Felix-Hausdorff-Str. 4, 17487 Greifswald,

Germany

Matthias Kittelmann, GDC/PSB/Bioreactions, Novartis Institutes of Biomedical

Research (NIBR), Novartis Pharma AG, CH-4002 Basel, Switzerland

Soo-Byung Ko, Department of Chemistry, Pohang University of Science and Technology


Yoshinori Kondo, Department of Biomolecular Science, Gifu University, Yanagido 1-1,

Gifu 501-1193, Japan

Birgit Kosjek, Process Research, Merck Research Laboratories, Merck & Co. Inc.

Rahway, NJ, USA

Udo Kragl, Institut fur Chemie, Universitat Rostock, Albert-Einstein-Str. 3a, 18059

Rostock, Germany

List of Contributors xxiii

Vladimır Kren, Institute of Microbiology, Center of Biocatalysis and Biotransformations,

Academy of Sciences of the Czech Republic, Vıdenska 1083, CZ-142 20 Prague 4, Czech

Republic

Anton Kuhn, GDC/PSB/Bioreactions, Novartis Institutes of Biomedical Research

(NIBR), Novartis Pharma AG, CH-4002 Basel, Switzerland

Jurgen Kuhnol, GDC/PSB/Separations, NIBR, Novartis Pharma AG, CH-4002 Basel,

Switzerland

Andrew S. Lamm, Center for Biocatalysis and Bioprocessing, 2501 Crosspark Road,

Suite C100 MTF, University of Iowa, Iowa City, Iowa, IA 52242-5000, USA

Jan von Langermann, Max-Planck-Institut fur Dynamik komplexer technischer

Systeme, Physikalisch-Chemische Grundlagen der Prozesstechnik, Sandtorstr.1 D-39106

Magdeburg, Germany

Peter C.K. Lau, Biotechnology Research Institute, National Research Council Canada,

Montreal, Quebec H4P 2R2, Canada

Marielle Lemaire, Laboratoire SEESIB, UMR 6504 CNRS, Universite Blaise Pascal, 24

avenue des Landais 63177 Aubiere cedex, France

Renata P. Limberger, State University of Campinas, Institute of Chemistry, CP 6154,

13084-971, Campinas-SP, Brazil

Guo-Qiang Lin, Laboratory of Biocatalysis and Bioprocessing, State Key Laboratory of

Bioreactor Engineering, East China University of Science and Technology, Shanghai

200237, PR China

Wen-Ya Lu, Laboratory of Biocatalysis and Bioprocessing, State Key Laboratory of


200237, PR China

Maja Majeric Elenkov, Laboratory for Stereoselective Catalysis and Biocatalysis, Ru�der

Boskovic Institute, Bijenicka c. 54, 10002 Zagreb, Croatia

Kathleen H. McClean, C-Tech Innovation Ltd, Capenhurst Technology Park,

Capenhurst, Chester, CH1 6EH, UK

Annett Mell, Institut fur Chemie, Universitat Rostock, Albert-Einstein-Str. 3a, 18059

Rostock, Germany

J. Meuldijk, Eindhoven University of Technology, Laboratory of Macromolecular and

Organic Chemistry, PO Box 513, 5600 MB Eindhoven, The Netherlands

Hisaaki Mihara, Department of Biotechnology, Institute of Science and Engineering,

College of Life Sciences, Ritsumeikan University, Kusatsu, Shiga 525-8577, Japan

Marco-Antonio Mirata, DECHEMA e.V., Karl-Winnacker-Institut, Biochemical

Engineering Group, Theodor-Heuss-Allee 25, 60486 Frankfurt, Germany

xxiv List of Contributors

Koichi Mitsukura, Department of Biomolecular Science, Gifu University, Yanagido 1-1,


Jeffrey Moore, Process Research, Merck Research Laboratories, Merck & Co. Inc.

Rahway, NJ, USA

Paulo J. S. Moran, State University of Campinas, Institute of Chemistry, CP 6154,


Krista Morley, Process Research, Merck Research Laboratories, Merck & Co. Inc.

Rahway, NJ, USA

Musa M. Musa, Department of Chemistry and of Biochemistry and Molecular Biology,

University of Georgia, Athens, GA 30602, USA

Toru Nagasawa, Department of Biomolecular Science, Gifu University, Yanagido 1-1,


S.F.G.M. van Nispen, Eindhoven University of Technology, Laboratory of Macromolecular

and Organic Chemistry, PO Box 513, 5600 MB Eindhoven, The Netherlands

Louise C. Nolan, School of Biomolecular and Biomedical Science, Conway Institute for

Biomolecular and Biomedical Research, National University of Ireland, University

College Dublin, Ardmore House, Belfield, Dublin 4, Republic of Ireland

Kevin E. O’Connor, School of Biomolecular and Biomedical Science, Conway Institute

for Biomolecular and Biomedical Research, National University of Ireland, University

College Dublin, Ardmore House, Belfield, Dublin 4, Republic of Ireland

Lukas Oberer, Analytical and Imaging Sciences, Novartis Institutes of Biomedical

Research, Novartis Pharma AG, CH-4002 Basel, Switzerland

Nobuyasu Okita, Division of Chemistry and Biochemistry, Graduate School of Natural


Eckhard Paetzold, Leibniz-Institut fur Katalyse, A.-Einstein-Str. 29a,18059 Rostock

Germany

Michael Page, Department of Chemical and Biological Sciences, The University of

Huddersfield, Huddersfield, HD1 3DH, UK

Francesca Paradisi, School of Chemistry and Chemical Biology, University College

Dublin, Belfield, Dublin 4, Ireland

Serge Parel, Biofocus DPI AG, Gewerbestrasse 16, CH-4123 Allschwil, Switzerland

Teodor Parella, Servei de Ressonancia Magnetica Nuclear, Universitat Autonoma de

Barcelona, 08193 Bellaterra, Barcelona, Spain

Jaiwook Park, Department of Chemistry, Pohang University of Science and Technology


Sara Pellegrino, Dipartimento di Scienze Molecolari Applicate ai Biosistemi (DISMAB),

Facolta di Farmacia, Universita degli Studi di Milano, via Venezian 21, 20133 Milano, Italy

List of Contributors xxv

Fabio Pezzotti, ECBB, ICMMO, Univ Paris-Sud, UMR 8182, F-91405 Orsay, France

Robert S. Phillips, Department of Chemistry and of Biochemistry and Molecular Biology,

University of Georgia, Athens, GA 30602, USA

Umberto Piantini, Institute of Life Technologies, University of Applied Sciences Valais,

Route du Rawyl 47, 1950 Sion, Switzerland

David Pollard, Process Research, Merck Research Laboratories, Merck & Co. Inc.

Rahway, NJ, USA

Monique Ponelle, Analytical and Imaging Sciences, NIBR, Novartis Pharma AG,

CH-4002 Basel, Switzerland

Manfred T. Reetz, Max-Planck-Institut fur Kohlenforschung, Kaiser-Wilhelm-Platz 1,


Renate Reiss, School of Chemistry, Manchester Interdisciplinary Biocentre, University of

Manchester, 131 Princess Street, Manchester, M1 7DN, UK

Sergio Riva, Istituto di Chimica del Riconoscimento Molecolare, C.N.R., Via Mario


Chris Roberge, Process Research, Merck Research Laboratories, Merck & Co. Inc.

Rahway, NJ, USA

J. Augusto R. Rodrigues, State University of Campinas, Institute of Chemistry, CP 6154,


John P. N. Rosazza, Center for Biocatalysis and Bioprocessing, 2501 Crosspark Road,

Suite C100 MTF, University of Iowa, Iowa City, Iowa, IA 52242-5000, USA

Monika Rusek, Max-Planck-Institut fur Kohlenforschung, Kaiser-Wilhelm-Platz 1,


Floris P. J. T. Rutjes, Institute for Molecules and Materials, Radboud University

Nijmegen, Toernooiveld 1, NL-6525 ED Nijmegen, The Netherlands

Emmanuelle Sagot, Department of Chemistry, Universite Blaise Pascal, Clermont-

Ferrand, France

Francesca Sagui, Istituto di Chimica del Riconoscimento Molecolare, C.N.R., Via Mario


Takashi Sakai, Division of Chemistry and Biochemistry, Graduate School of Natural


Soth Samreth, Fournier Pharma, 50 rue de Dijon, 21121 Daix, France

Christopher K. Savile, Department of Biochemistry, Molecular Biology & Biophysics

and The Biotechnology Institute, University of Minnesota, 1479 Gortner Avenue, Saint

Paul, MN 55108 USA

Adrian Scaffidi, Chemistry M313, School of Biomedical, Biomolecular and Chemical

Sciences, University of Western Australia, Crawley, WA 6009, Australia

xxvi List of Contributors

Andreas Schmid, Laboratory of Chemical Biotechnology, Faculty of Biochemical and

Chemical Engineering, TU Dortmund, Emil-Figge-Strasse 66, 44221 Dortmund, Germany

Jens Schrader, DECHEMA e.V., Karl-Winnacker-Institut, Biochemical Engineering

Group, Theodor-Heuss-Allee 25, 60486 Frankfurt, Germany

Frank Schulz, Max-Planck-Institut fur Kohlenforschung, Kaiser-Wilhelm-Platz 1, 45470

Mulheim/Ruhr, Germany

Ioulia Smonou, Department of Chemistry, University of Crete, Iraklion-Voutes, 71003

Crete, Greece

Timothy Snape, School of Chemistry, Manchester Interdisciplinary Biocentre, University

of Manchester, 131 Princess Street, Manchester, M1 7DN, UK

Georg A. Sprenger, Institute of Microbiology, University of Stuttgart, Allmandring 31,

70569 Stuttgart, Germany

Robert V Stick, Chemistry M313, School of Biomedical, Biomolecular and Chemical

Sciences, University of Western Australia, Crawley, WA 6009, Australia

Matthew Stirling, Department of Chemical and Biological Sciences, The University of

Huddersfield, Huddersfield, HD1 3DH, UK

Takeshi Sugai, Faculty of Pharmacy, Keio University, 1-5-30, Shibakoen, Minato-ku,

Tokyo 105-8512, Japan

Peter Sutton, GlaxoSmithKline, Synthetic Chemistry, Gunnels Wood Road, Stevenage,

Hertfordshire SG1 2NY, UK

Andreas Taglieber, Max-Planck-Institut fur Kohlenforschung, Kaiser-Wilhelm-Platz 1,


Lixia Tang, Biochemical Laboratory, Groningen Biomolecular Sciences and

Biotechnology Institute, University of Groningen, Nijenborgh 4, 9747 AG, Groningen,

The Netherlands

Alan Taylor, Centre for Material Science, University of Central Lancashire, Preston

Lancashire, UK

David Tellers, Process Research, Merck Research Laboratories, Merck & Co. Inc.

Rahway, NJ, USA

Helene Therisod, ECBB, ICMMO, Univ Paris-Sud, UMR 8182, F-91405 Orsay,

France

Michel Therisod, ECBB, ICMMO, Univ Paris-Sud, UMR 8182, F-91405 Orsay, France

Ai-Ming Tong, Laboratory of Biocatalysis and Bioprocessing, State Key Laboratory of


200237, PR China

Rafael de la Torre, Pharmacology Research Unit, Institut Municipal d’Investigacio

Medica (IMIM), Barcelona, Spain

List of Contributors xxvii

Matthew Truppo, Process Research, Merck Research Laboratories, Merck & Co. Inc.

Rahway, NJ, USA

Nicholas J. Turner, School of Chemistry, Manchester Interdisciplinary Biocentre,

University of Manchester, 131 Princess Street, Manchester, M1 7DN, UK

Cleber V. Ursini, State University of Campinas, Institute of Chemistry, CP 6154, 13084-

971, Campinas-SP, Brazil

Takamitsu Utsukihara, Department of Chemistry, Rikkyo (St.Paul’s) University, Nishi-

Ikebukuro, Toshima-Ku, Tokyo 171-8501, Japan

J.A.J.M. Vekemans, Eindhoven University of Technology, Laboratory of

Macromolecular and Organic Chemistry, PO Box 513, 5600 MB Eindhoven, The

Netherlands

Claire Vieille, Biochemistry and Molecular Biology, Michigan State University, East

Lansing, MI 48824, USA

Shaozhao Wang, Biotechnology Research Institute, National Research Council Canada,


Marloes A. Wijdeven, Institute for Molecules and Materials, Radboud University

Nijmegen, Toernooiveld 1, NL-6525 ED Nijmegen, The Netherlands

Chi-Huey Wong, Department of Chemistry, The Scripps Research Institute, 10550 North

Torrey Pines Rd., La Jolla, CA 92307, USA

Matthias Wust, Institute of Life Technologies, University of Applied Sciences Valais,

Route du Rawyl 47, 1950 Sion, Switzerland

Li Xu, Key Laboratory for Molecular Enzymology and Engineering of Ministry of

Education, Jilin University, Changchun, 130021, PR China

Jian-He Xu, Laboratory of Biocatalysis and Bioprocessing, State Key Laboratory of


200237, PR China

Hitomi Yamaguchi, Research & Development Center, Nagase & Co., Ltd., 2-2-3,

Murotani, Nishi-ku, Kobe 651-2241, Japan

Jianzhong Yang, Biotechnology Research Institute, National Research Council Canada,


Toyokazu Yoshida, Department of Biomolecular Science, Gifu University, Yanagido

1-1, Gifu 501-1193, Japan

Hui-Lei Yu, Laboratory of Biocatalysis and Bioprocessing, State Key Laboratory of


200237, PR China

Sheng Yuan, Nanjing Engineering Research Center for microbiology, Jiangsu Key

Laboratory for Biodiversity and Biotechnology, College of Life Science, Nanjing

Normal University, 1, Wenyuan Rd, Nanjing 210046, PR China

xxviii List of Contributors

Francesca Zambianchi, Istituto di Chimica del Riconoscimento Molecolare, CNR, via

Mario Bianco 9, 20131 Milano, Italy

J. Gregory Zeikus, Biochemistry and Molecular Biology, Michigan State University,

East Lansing, MI 48824, USA

Karla I. Ziegelmann-Fjeld, Biochemistry and Molecular Biology, Michigan State

University, East Lansing, MI 48824, USA

Xue-Zhong Zhang, Key Laboratory for Molecular Enzymology and Engineering of

Ministry of Education, Jilin University, Changchun, 130021, PR China

Yongzhong Zhu, Department of Chemistry, National University of Singapore, Kent


List of Contributors xxix

1

Biotransformations in Small-moleculePharmaceutical Development

Joseph P. Adams, Andrew J. Collis, Richard K. Henderson and Peter W. Sutton

1.1 Introduction

The demand for medicines that treat illnesses formerly associated with the developed

world is expanding at a time when some countries are becoming increasingly affluent. As a

result, the global pharmaceutical market is predicted to grow to $800 billion by the year

2020.1 However, as demand increases for products, the pharmaceutical industry is facing

increasing pressures that can primarily be attributed to three factors:

1. As the global population ages and lifestyles become more sedentary, the cost of

healthcare is becoming increasingly unsustainable. This is no more so than in the

USA, where, although prescription products contribute only 10 % of healthcare costs,

they are perceived to be much higher by the consumer and so represent an easy political

target for cost cuts through price controls.

2. Erosion of product lifetimes as a result of greater generic competition means that a pro-

duct can expect to lose the majority of its market in as little as 3 months after patent expiry.

3. Spiralling R&D costs. Typically, it takes 10 years at a cost of $500 million to bring a

drug to market.2 Fewer new molecular entities (NMEs) and biologics are reaching the

market as a result of a shift of research focus away from already established and

crowded therapeutic areas into new, unproven biological areas (Figure 1.1).3

Practical Methods for Biocatalysis and Biotransformations Edited by John Whittall and Peter Sutton

� 2009 John Wiley & Sons, Ltd

Whereas new drugs reaching the market do not necessarily look any more complex or

contain any more stereocentres than in the past (Figure 1.2),3 the complexity of drug

candidates under development has increased on average. In addition, following the FDA’s

1992 policy statement on stereoisomers, it is now clearly more economical to progress an

active pharmaceutical ingredient (API) in enantiopure form, as can be seen from the trend

towards the launch of single-enantiomer new chemical entities (NCEs) (Figure 1.3).

0

5000

10000

15000

20000

25000

30000

35000

40000

45000

50000

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

R&

D S

pen

din

g (

US

$ M

illio

ns)

0

10

20

30

40

50

60

No

. of

NM

Es

and

Bio

log

ics

Ap

pro

ved

R&D Spending NMEs and New Biologics Approved

Figure 1.1 R&D spending versus the number of NMEs and biologics approved by the US Foodand Drug Administration (FDA). (Reprinted with permission from Pharma 2020: The vision:Which path will you take?, PricewaterhouseCoopers, 2007.)

02

468

10

121416

1820

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year of Launch

NC

Es

achiral chiral

Figure 1.2 Number of chiral and achiral marketed NCEs. (Reprinted with permission fromFarina, V., Reeves, J.T., Senanayake, C.H. and Song, J.J. Asymmetric synthesis of activepharmaceutical ingredients. Chem. Rev. 2006, 106, 2734–2793. Copyright 2006, AmericanChemical Society)

2 Biotransformations in Small-molecule Pharmaceutical Development

This trend seems likely to continue, with over 50 % of current drug candidates being

developed as single enantiomers.4 An obvious consequence has been an explosion of

research activity into asymmetric synthetic methods.

These combined issues, on average, make pharmaceutical companies as much as 50 %

riskier than other big industries.5 Led by the FDA’s Current Good Manufacturing Practices

for the 21st Century initiative,6 the pharmaceutical industry has begun to apply a risk

management and quality systems approach, practiced in some other industries for decades,

to products throughout their lifetimes.7 The guidance from the FDA’s Process Analytical

Technologies regulatory framework, developed over the last decade, aims to build quality

by design into pharmaceutical products through better process understanding and

increased innovation. The ultimate goal is to minimize risk to the patient whilst encoura-

ging the industry to cut operating costs.8

Encouraged by this more flexible regulatory approach, there is an increased willingness

within the industry to adopt ‘new’ technologies.9

1.2 Current Status of Biocatalysis

A biotransformation, as defined by Straathof et al.,10 is ‘a process that describes a reaction

or a set of simultaneous reactions in which a pre-formed precursor molecule is converted

using enzymes and/or whole cells, or combinations thereof, either free or immobilised’.

Fermentation processes, with de novo product formation from a carbon and energy source,

such as glucose via primary metabolism, are outside the scope of this chapter and book

unless employed in conjunction with a biotransformation.

Biocatalysis has long been known as a green technology, capable of delivering

highly stereo-, chemo- and regioselective transformations that can sometimes allow

0

2

4

6

8

10

12

14

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year of Launch

NC

Es

single enantiomer racemate

Figure 1.3 Number of single enantiomer versus racemic NCEs. (Reprinted with permissionfrom Farina, V., Reeves, J.T., Senanayake, C.H. and Song, J.J. Asymmetric synthesis of activepharmaceutical ingredients. Chem. Rev. 2006, 106, 2734–2793. Copyright 2006, AmericanChemical Society)

1.2 Current Status of Biocatalysis 3

the number of steps in a synthetic route to be reduced. Numerous industrial biotransfor-

mations (announced to be commercialized at a scale of >100 kg per annum) are in

operation worldwide, many of which have been described by Liese et al.11 Most of these

known biotransformations are used to produce building blocks that are subsequently

supplied to the pharmaceutical industry (Figure 1.4).10

Biocatalysis is still an emerging field; hence, some transformations are more established

than others.12 Panke et al.13 have performed a survey of patent applications in the area of

biocatalysis granted between the years 2000 and 2004. They found that although hydrolases,

which perform hydrolyses and esterifications, still command widespread attention and

remain the most utilized class of enzyme (Figure 1.5), significant focus has turned towards

the use of biocatalysts with different activities and in particular alcohol dehydrogenases

(ADHs) – also known as ketoreductases (KREDs) – used for asymmetric ketone reduction.

Whereas the number of industrial biotransformations ‘known’ to be operating in 2002

was 134, the number of chiral drug candidates is much greater. Farina et al.3 have estimated

between 500 and 1000 single-enantiomer APIs to be in development each year in the

global pipeline. This implies that biotransformations might supply only a small percentage

of chiral centres. This might be partially attributable to the reluctance of the pharmaceutical

industry to innovate in the absence of the recently established regulatory directives, or to

the lack of commercial enzymes available on a large scale. However, the main factor lies in

the strategy used to incorporate chirality into drug candidates.

Polymers

Cosmetics

Food

Animal Feed

Agro

Other sectors

Pharma

Figure 1.4 Number of biotransformations used catagorised by industrial sector (based on 134processes). (Reprinted from Straathof, A.J.J., Panke, S. and Schmid, A. The production of finechemicals by biotransformations. Curr. Opin. Biotechnol. 2002, 13, 548–556 with permissionfrom Elsevier.)

Oxidoreductases

Oxidising cells

Reducing cells

Isomerases

Lyases

Hydrolases

Transferases

Figure 1.5 Enzyme Types Used in Industrial Biotransformations (based on 134 processes).(Reprinted from Straathof, A.J.J., Panke, S. and Schmid, A. The production of fine chemicals bybiotransformations. Curr. Opin. Biotechnol. 2002, 13, 548–556 with permission from Elsevier.)


In line with the construction of a target molecule from smaller, complex fragments, it is

generally preferred to introduce chirality into a synthetic route at an early stage through the

purchase of simple chiral starting materials from the fine chemical industry. This has been

demonstrated by Carey et al.,14 who performed a survey of 128 drug candidate syntheses,

many of which were in an early phase of development. They found that, of the 69 chiral

drug candidates considered, 55 % of the 135 chiral centres present were bought in from the

fine chemicals industry. In cases where it was necessary to generate chirality in-house, the

favoured method was racemate resolution (28 % of chiral centres – with classical salt

formation employed in two-thirds of cases and dynamic kinetic resolution, chromatogra-

phy and biocatalytic methods evenly distributed in the remainder) followed by chemical

asymmetrization (10 % of chiral centres – see Section 1.3.4 for definition) and diaster-

eoselective induction (7 % of chiral centres). Another important source of chirality (which

was not exemplified in that article) is fermentation technology, which provides access to

many of the important chiral scaffolds that have been employed by the industry in well-

known classes of drug, such as b-lactam antibiotics and, more recently, first-generation

statins.

Some 35 % of the chiral building blocks that are bought in from the fine chemical

industry, such as both proteinogenic and non-proteinogenic amino acids, carboxylic acids,

amines, alcohols and epoxides, are produced using generic biocatalytic technologies, and

this is expected to increase to 70 % by 2010.12 Far more chiral centres present in APIs are

derived from industrial biotransformations than would be expected by counting the

number of known processes.15 This can also be noted from the procedures given in this

book, the vast majority of which provide biocatalytic routes to chiral building blocks.

These chiral building blocks, in turn, will be dictated by current drug candidates within the

pharmaceutical industry’s pipeline.

Given the wide utility of biocatalysis in the fine chemical industry, why is there such an

in-house reliance on classical methods of enantioseparation? In fact, why is biocatalysis

not applied more generally as a replacement for atom-inefficient or hazardous reactions

that are intensively used in the pharmaceutical industry, such as amidation, reduction and

oxidation?16

The sparse incorporation of biocatalysts into the process chemist’s toolbox is at least in

part due to a number of long-standing issues that differ depending on the drug development

phase. At an early stage of development, where little resource is available for new route

development, biocatalysis options are often neglected due to a lack of sufficient commer-

cially available biocatalysts.17 In contrast, classical salt formation regularly provides

access to chiral material in >99.5 % enantiomeric purity; hence its widespread adoption.

At a later phase biocatalysis may be considered, but the longer development times often

needed and the more advanced state of competing chemical routes put it at a disadvantage.

Many of these issues have been or are being addressed. For example, with the continued

expansion in the number of microorganisms whose genomes have been sequenced, the

application of bioinformatics techniques is leading to a rapid expansion in the number of

commercially available enzymes such as ADHs (ketone reduction), nitrilases (nitrile hydro-

lysis), enoate reductases (�,b-unsaturated olefin reduction) and transaminases (reductive

amination).18 Having identified a putative enzyme gene by sequence similarity, it can now

be quickly and cheaply generated by using oligonucleotide synthesis services that are

provided by a number of companies. However, it is predicted that about 99 % of

1.2 Current Status of Biocatalysis 5

microorganisms are ‘non-cultivable’, and metagenomics – the extraction of environmental

DNA – is proving highly successful in accessing novel biocatalysts from this untapped

resource.19

Further expansion in the number of commercially available enzymes and the modifica-

tion of hits to suit process requirements is being fuelled by advances in enzyme engineer-

ing20 and high-throughput screening (HTS) technologies.21 A particularly elegant

approach is the protein sequence activity relationship (ProSAR) technology developed

by Codexis.22 By using a multivariate analysis approach, libraries of enzyme variants

containing programmed mutations generated from different sources of diversity are

screened against a given substrate and sequenced. Positive mutations and interactions

between different mutations can then be ascertained, allowing more active variants to be

predicted in silico, thus reducing the bottleneck often caused by screening. As the impact

of individual mutations is understood, the possibility of missing important ones or carrying

false hits through to the next round is reduced compared with traditional hit-based

directed-evolution strategies. Similar multivariate techniques are used on a daily basis

within the pharmaceutical industry (quantitative structure–activity relationships in med-

icinal chemistry and design of experiments to understand and optimize chemical reactions

during process development) to interpret complex data sets. So powerful is this technique

at producing ADHs suitable for process applications that one pharmaceutical company

now considers biocatalysis as their first option for asymmetric ketone reduction.

There is also a greater appreciation within the biotech community that alternative routes

to a drug candidate are always available and, although they are sometimes technologically

inferior, will always be favoured if freedom to operate is at stake.

1.3 Application of Biocatalysis in the Pharmaceutical Industry

This section primarily focuses on examples of biotransformations that have been devel-

oped for the preparation of small-molecule APIs (molecular weight <1000), metabolites

and late-stage intermediates. Particular emphasis will be given to the incorporation of

biotransformation steps into synthetic routes and their advantages over competing tech-

nologies. Later sections will then expand on the key topics of ‘enzymes in organic

solvents,’ ‘enzyme immobilization’ and ‘green chemistry’ that are introduced in earlier

sections.

No attempt has been made to cover all drug classes or enzyme classes; instead, a flavour of

the potential benefits that can be achieved by the adoption of biocatalytic methods as a

compliment to chemical approaches is given. Biocatalytic methods of accessing chiral

building blocks will only occasionally be discussed here and the reader is referred to a

number of comprehensive reviews that have been published elsewhere.15,23

1.3.1 Drug Metabolites and Metabolic Transformations

Metabolites are generated by the body’s own biochemical processes as a way to facilitate

excretion of xenobiotics. The enzymes catalysing in vivo modification of drugs and drug-

like molecules have a fundamental significance for the pharmaceutical industry. This was

once primarily the field of the pharmacologist, but interest in metabolic reactions


increasingly extends to the synthetic chemist arising from the requirement to synthesize

specific drug metabolites, as well as the realization that some of these enzymes could be

exploited as general synthetic tools.

A thorough understanding of the metabolic fate of a drug candidate is essential in the

assessment of its efficacy and toxicity and to safeguard patient welfare through the

identification of potential drug–drug interactions (DDIs).24 It is also an integral part of

the drug discovery process, allowing molecular redesign based on the identification of

active metabolites and an appreciation of how they arise. In fact, metabolism can lead to

new structures that need to be covered in patent claims.

Metabolic reactions (also known as biotransformations) can be divided into two cate-

gories: functionalization, where a functional group is created or modified, and conjuga-

tion, where another molecule is transferred to the substrate.25 These are also known as

phase I and phase II metabolic reactions respectively (Scheme 1.1). Functionalization

encompasses redox reactions, hydrolyses and hydrations, whereas conjugation can involve

a wide variety of transformations, such as glucuronidation, methylation, sulfoxidation and

phosphorylation (Table 1.1). Most of these drug-metabolizing enzymes are expressed

intracellularly in the liver at comparatively high levels relative to the rest of the body.

The most prominent metabolic transformations are catalysed by oxidoreductases,

hydrolases and transferases (glutathione transferases and glucuronyl transferases), oxida-

tive transformations being quantitatively of greatest importance. Metabolic enzymes are

often highly promiscuous, transforming a wide variety of xenobiotics. Given the number

of different metabolic enzymes, their promiscuity and the multiple activities of some, it is

not surprising that the metabolism of a xenobiotic often results in a soup of different

compounds, all present in low abundance.

Direct isolation of sufficient quantities of each metabolite for structural characteriza-

tion, assay validation and pharmacological or toxicological testing from in vivo studies

using biological specimens is, therefore, often impossible, particularly from drugs with a

low therapeutic index. Furthermore, many metabolites have structural modifications

which are difficult to replicate by traditional chemical methods. A number of synthetic

steps may be required to prepare such metabolites from the API, or, in the worst case, a

completely new synthetic route may need to be developed.

It seems logical that drug metabolites should be prepared using the specific enzymes

involved in their formation as biocatalysts for in vitro synthesis. This would overcome the

problems inherent in the use of crude mammalian tissue extracts with their cocktails of

Drug

Phase Ireactions

Metabolites

Phase IIreaction

Conjugated metabolitesPhase II

reaction

Scheme 1.1 Categories of metabolite

1.3 Application of Biocatalysis in the Pharmaceutical Industry 7

Table 1.1 The main classes of mammalian metabolic transformations.26 (Reprinted with kindpermission of Springer ScienceþBusiness Media.)

Reaction Type Substrate Enzyme Product

Phase I Metabolism

Hydrolysis Esters Esterases AlcoholsAmides Amidases AminesEpoxides Epoxide hydrolases Diols

Reduction Ketones Alcohol reductases AlcoholsAlkenes Hydrogenases AlkanesNitro and azo Nitro and azo

reductasesNitroso, oximes,amines

Hydroxylation Aromatic, allylic, benzylicor saturated carboncontaining

CYP450 or flavinmonooxygenases

PhenolsAlcohols

Epoxidation Alkenes Epoxides

N-, O-, S-Dealkylation

N-, O-, S-Alkyl Amines, alcohols,thiols

C-Oxidation Alcohols, aldehydes,ketones

Aldehydes, ketones,carboxylic acids

N-, S-Oxidation Secondary and tertiaryamines

N-Oxides

S-Alkyl Sulfoxides, sulfones

N-Hydroxylation

Secondary and tertiaryamines

Oximes

Deamination Primary amines Monoamine oxidases Aldehydes

Phase II Metabolism

Glucuronidation Alcohols/phenols Glucuronyltransferases a- or �-glucuronidesCarboxylic acidsAminesThiols

Glycosylation Alcohols/phenols Glycosyltransferases a- or �-glycosidesCarboxylic acidsAminesThiols

Thiolconjugation

Epoxides Glutathione-S-transferases

Glutathione orN-acetyl cysteinethioethers

Glycineconjugation

Carboxylic acids N-Transferases Glycinamideconjugates

Carbamoylation Alcohols O-Carbamoylderivatives

Acetylation Primary amines Acetyltransferases AcetamidesHydrazines

O-Methylation Phenols Methyltransferases Methyl aryl ethers

Sulfation Alcohols/phenols Sulfotransferases Sulfate estersAmines Sulfonamides


metabolic enzymes. However, such a system is rarely possible, as these enzymes present

particular technical obstacles, being usually membrane bound rather than soluble and

requiring a range of biochemical cofactors. Some examples are given in later sections.

The classes of reaction observed in drug metabolism are not exclusive to mammalian

systems; hence, microorganisms can often be used to produce metabolites. Compared with

mammalian enzyme preparations such as liver homogenates and other tissue preparations,

microbial cultures provide low-cost maintenance, long-term stability and easier scale-up to

prepare purified metabolites. Microbial systems tend to show higher tolerance of xenobio-

tics, allowing higher concentrations of metabolite to be produced, and greater regio- and

stereo-specificity, limiting the number of metabolites to be separated for purification.

The ideal would be a collection of microorganisms, each mimicking a single metabolic

reaction, together covering the full range of mammalian drug metabolism. Numerous

different panels of microorganisms that allow the biotransformation of a wide variety of

different substrates have been reported since the initial studies on these so-called ‘micro-

bial models of mammalian metabolism’ by Smith and Rosazza in the 1970s.27 The best of

these systems will provide both phase I and phase II metabolites in milligram yields.

For example, the antihypertensive drug Irbesartan is known to give at least eight urinary

metabolites in mammals and humans, which include hydroxylated, ring-opened and

N-glucuronylated products. To provide sufficient quantities of these metabolites for

further structural and stereochemical characterization, Azerad and co-workers28 screened

10 fungal strains and 28 bacterial strains that are regularly used for drug hydroxylation

within their laboratory for activity towards irbesartan. The hydrolysis product 1 was

produced by three-quarters of the strains tested, whereas the hydroxylated products 2–5

were produced equally by about one-quarter of the strains (Scheme 1.2). The metabolite 6,

tentatively assigned as an N-glycosidic conjugate similar to the mammalian N-glucuronide

metabolite, proved to be the least accessible metabolite, produced by only four strains.

Although small amounts of metabolites were detected in fungal incubations, actinomy-

cetes or filamentous bacteria were found to be more productive both quantitatively and

qualitatively. Thus, Streptomyces strains produced the highest levels of metabolites 2–5

and were the only strains capable of producing metabolite 6. Some of these strains were

then used to access 20–100 mg quantities of each metabolite. This mirrors the general

findings that fungi and acinomycetes (including Streptomyces, Nocardia, Actinoplanes,

Mycobacteria and Corynebacteria) are most useful for biotransformation. Other bacterial

strains tend to consume the xenobiotic as a carbon or nitrogen source, making metabolite

isolation problematic.29

1.3.1.1 Phase I Metabolic Transformations

One reaction characteristic of phase I metabolism is monooxygenase-catalysed hydroxy-

lation at specific C�H bonds without chemical activation. Different enzymes show varying

degrees of regio-, chemo-, and enantio-specificity, so the reaction is usually challenging

for the synthetic chemist to reproduce once preparation of such a metabolite is required. It

should be evident that biocatalysts with such capabilities would also be highly desirable as

tools for chemical synthesis in general.

The vast majority of hydroxylations in mammalian systems result from the action of

cytochrome P450s (CYPs – also known as P450s), which are a superfamily of


monooxygenases catalysing an array of different reactions.30 It is estimated that about

90 % of all marketed drugs and drug candidates are substrates for CYPs. Of more than 60

known human CYPs, CYP1A2, CYP2B6, CYP2C9, CYP2C19, CYP2D6, CYP2E1 and

CYP3A4 are of particular significance in the metabolism of xenobiotics.31

These CYPs are isoenzymes (or isozymes), catalysing essentially the same reaction, but

for different substrate ranges with specificity determined by their different amino acid

sequences. CYPs (and other metabolic enzymes) often react with individual substrates in a

highly regio-, chemo- or stereo-selective manner, each isozyme displaying its own unique

selectivity. Some examples of selective CYP-catalysed transformations are shown in

Scheme 1.3.

Although mammalian CYPs are attractive candidates for use as commercial biocatalysts,

many functional characteristics limit the opportunities to exploit such a system. Association

of the enzymes with membranes prevents easy extraction and purification and limits the

opportunities to produce useful recombinant enzymes by cloning the relevant genes for

expression in microbial systems. All P450s have a porphyrin-haem active site that requires a

second protein to reduce the iron component, often cytochrome P450 reductase or

N

N

NN

N NH

OBu

N

N

NN

N NH

OBu

OH OH

N

N

NN

N NH

OBu

O

O OH

OH

OH

OH

NH

NN

N NH

O

NH

O

Bu

NH

NN

N NH

O

NH

O

Bu

Irbesartan

1

2 and 3 4 and 5

6

Scheme 1.2 Metabolism of irbesartan


cytochrome b3. In addition, a reduced nucleotide cofactor (reduced-form nicotinamide

adenine dinucleotide (NADH) or reduced-form nicotinamide adenine dinucleotide phos-

phate (NADPH)) must be provided in stoichiometric quantities. Despite these difficulties, a

range of commercially available solutions exists for synthesis of small quantities of meta-

bolites. The effort put into development of these systems reflects both their importance to the

pharmaceutical industry and the limited availability of off-the-shelf microbial oxidations.

Liver microsomal preparations contain a spectrum of P450 isozymes with competing

activities. Depending upon the source, these may or may not correspond to the specificities

of the human CYPs. Such systems serve to evaluate the spectrum of potential metabolites

from a given substrate molecule, but they have more limited value for synthesis of single

metabolites in useful yield. To this end, considerable effort has been directed towards

development of genetically engineered cell lines that express single specific CYPs.32

These systems essentially provide a microsomal preparation of P450s expressed together

with an appropriate reductase component in insect cells, yeasts or bacteria.

N

NH

NH2 N

NH

NH2

OH

N

NN

N

O

ON

NN

NH

O

O N

NN

N

O

O N

NH

N

N

O

O

NH

NH

OO

NH

NH

OO

O

S

H2N NH2

OO S

H2N NHOH

OO

NH

NH

OO

Debrisoquine

CYP2D6

CYP1A2 + +

80% 11% 4%Caffeine

CYP2C

Dapsone

CYP2C9

CYP2E1

CYP3A4

Phenytoin(Racemic)

+

Scheme 1.3 Selected examples of CYP-catalysed oxidations31


More recent developments include the modification of the genes encoding several

human CYP genes and the corresponding reductase to allow expression as soluble proteins

in Escherichia coli. This provides a water-soluble enzyme system for a limited number of

pharmacologically relevant P450s, but retains the basic disadvantages of slow reaction

rate, sensitivity to substrate concentration and the need for added reductase and reduced

NAD(P) cofactors. All these systems remain very much a tool for synthesis of limited

quantities of drug metabolites rather than scaleable synthetic tools.

Although few of the P450s characterized from microbial sources have substrate speci-

ficity corresponding to that of the human liver CYPs, these enzymes may tolerate higher

substrate concentrations, promising higher yields of metabolites. Of particular interest are

enzymes such as cytochrome P450 BM-3 from Bacillus megaterium (CYP102), which is a

natural fusion of monooxygenase with a reductase, and offers a system for biocatalysis

with fewer components. Very recently, a kit of variants of P450 BM-3, developed by the

Arnold group,33 has become commercially available in a 96-well plate format. This

collection of enzyme variants, generated by directed-evolution techniques, is claimed to

accept a broader substrate range and offer greater potential for use at scale than

human CYPs.

Currently, microbial whole cells must be considered the main option to produce larger

quantities of phase I metabolites through biotransformation. These processes are generally

regarded as scaleable through standard fermentation technology, although careful screen-

ing of microorganisms is required to achieve the reaction required in the absence of

competing side reactions or catabolism of the substrate and product. Some microorganisms

are particularly effective at mimicking phase I metabolic reactions such as hydroxylation,

and a recent review on how these can be employed in the study of drug metabolism has

been published by Ghisalba and Kittelmann.34 The oxidation of fluvastatin by different

microorganisms provides a good example of how these systems can be beneficial in

selective metabolite preparation (Scheme 1.4).35

N

F

OHOH

CO2Na

N

F

OHOH

CO2NaOH

N

F

OHOH

CO2Na

OH

Mortierella rammaniana

Fluvastatin-Na

Streptomyces violascensATCC 31560

Scheme 1.4 Selective microbial hydroxylation of fluvastatin


For a manufacturing-scale process the requirement to maximize overall yield and

eliminate side reactions is greater still, yet whole-cell biocatalysis remains the sole option

for oxidative biotransformation. For example, a whole-cell monooxygenase-based oxida-

tion of 2-methylquinoxaline has been reported by Wong et al.36 that uses a strain of

Pseudomonas putida possessing an aryl ADH and a benzaldehyde dehydrogenase that

together generate 2-quinoxaline carboxylic acid in three steps at a reported 86 % overall

yield (Scheme 1.5).

This example shows the fortuitous interaction of multiple enzymes in a single microbial

system. However, competing activities are more likely using a natural system, and there

will be an ongoing desire to develop a scalable process using either an isolated stable and

active P450 monooxygenase, or a recombinant whole-cell system based on a cloned and

overexpressed P450. At the time of writing, neither system yet exists, although the recently

characterized microbial enzymes with fused reductase clearly offer some potential for

development. Until recently, the challenges of providing the necessary NADH or NADPH

cofactors to permit the use of any oxidoreductase outside of a whole-cell system would

have been considered a barrier to the development of a viable process, but these problems

have been solved in response to the increasing application of enzymes for asymmetric

ketone reduction (see Section 1.3.4.4).

1.3.1.2 Phase II Metabolic Transformations

In phase II biotransformations, the conjugating functional group is generally transferred

to the target molecule from an activated cofactor or ‘coenzyme’. Most such reactions use

transferase mechanisms found throughout biology; for example, acetyltransferases requir-

ing acetylcoenzyme A or methyltransferases dependent on S-adenosylmethionine.

Amongst phase II reactions, glucuronidation is a special case, as it is seldom observed

outside of mammalian metabolism of xenobiotics.

Phase II drug metabolites are often the final form in which a xenobiotic is solubilized for

release from the body, and many display significant biological activity. Synthesis of

purified phase II metabolites, therefore, is a requirement of the drug development process.

Most of the enzymes involved in mammalian phase II metabolism are, like P450s, poor

candidates for in vitro biocatalysis, being membrane associated and requiring activated

N

N

N

NOH

N

N CHO

N

N COOH

monooxygenase

aryl alcoholdehydrogenase

benzaldehydedehydrogenase

Scheme 1.5 Biotransformation of 2-methylquinoxaline by Pseudomonas putida ATCC 33015


coenzymes that may be both costly and unstable. For example, glucuronidation using

uridine diphosphate (UDP) cofactor is catalysed by UDP-glucuronyltransferases using the

catalytic cycle shown in Scheme 1.6.

Currently, glucuronides that cannot be easily synthesized chemically are prepared using

liver microsomes,25 which can allow access to hundreds-of-milligram quantities of the

desired metabolite after purification. Gene cloning for single mammalian UDP-glucuro-

nosyltransferases has been less successful than for the human CYPs, or perhaps less

rigorously attempted. Between mammalian species, the spectrum of glucuronyltransferase

isoenzymes can vary significantly, particularly in the case of N-glucuronidation.38 These

differences may be exploited by the chemist seeking to prepare a particular glucuronide for

pharmacological studies.

For example, Kittelmann et al.39 were interested in the pharmacologically active acylglu-

curonide of mycophenolic acid, an immunosuppressant. In humans, this molecule is glucur-

onidated to afford a 1:80 mixture of acylglucuronide and inactive 7-O-glucuronide

respectively (Scheme 1.7). By screening a range of liver homogenates from different

mammalian species, the group was able to produce the two metabolites in a 1:1 mixture

that allowed the preparation of multi-hundred-milligram quantities of the desired metabolite.

OOH

OUDPOH

OH

OH

OOH

OUDP

HO2C

OHOH

OOH

OOH

OH

OH

P O–

O–

O

OOHHO2C

OHOH OR

UDP-glucuronic acid

β-D-glucuronide

glucose-1-phosphate

UDP-glucose

UDP

ROH

phosphoenolpyruvate

pyruvate

UTP

pyrophosphate

2 NAD+

2 NADH

enzymes: A pyruvate kinase; B UDP-glucose pyrophosphorylase; C UDP-glucose dehydrogenase; D UDP-glucuronosyl transferase

A

B

C

D

Scheme 1.6 Uridine diphosphate glucuronide transferase cycle.37 (Reproduced by permis-sion of the Royal Society of Chemistry.)


Glucuronidation specifically is rarely observed in microorganisms; however, enzymatic

glycosylation as a general reaction is common to most living systems, the most common

case being simple O- or N-glucosylation. Glucuronic acid is derived from glucose by

oxidation at the 6-carbon; hence, it is worth considering glucosides as targets potentially

accessible using microbial biotransformations and subsequently converting these to the

corresponding glucuronides. Recent work by Baratto et al.40 has shown that glucuronides

can be readily accessed from glycosides by a mild laccase/2,2,6,6-tetramethyl-1-piperidi-

nyloxy (TEMPO) oxidation (Scheme 1.8).

Laccases are oxidoreductases, primarily secreted by fungi, available in industrial

quantities for use in the fabrics industry. Their natural role is in the breakdown of

O

O OH

CO2H

OMe

O

O

O O

CO2H

OH

OH

OH

HO2C

OMe

O OH

OH

OH

HO2C

O

O

O OH

OOMe

Liver homogenate

Mycophenolic acid

7-O-glucuronide

Acylglucuronide

+

Scheme 1.7 Glucuronidation of mycophenolic acid with liver homogenate.

O

O

O

SMe

OHOH

OH

OH

NHAc

OMeMeO

O

O

O

SMe

OHOH

OH

NHAc

OMeMeO

CO2H

Laccase, O2, TEMPO

Glucuronide

Scheme 1.8 Selective laccase/mediator oxidation of the natural glycoside thiocolchicoside


polyphenols (present in plants as lignin) using molecular oxygen as the oxidant. More

recently, their substrate range has been dramatically increased to allow allylic and benzylic

alcohol oxidation by the use of catalytic quantities of mediators such as 2,20-azino-bis-3-

ethylbenzothiazoline-6-sulfonic acid (ABTS), hydroxybenzotriazole (HOBt), N-hydro-

xyphthalimide (HPI) and TEMPO.41 The mediator undergoes oxidation by the laccase,

which in turn oxidizes the substrate of interest before returning to its original state. Any

selectivity towards the substrate, therefore, is based on the chemical interaction between

substrate and activated mediator and should be relatively broad. The feasibility of prepar-

ing a range of glucuronides using laccase/mediator oxidation may now be limited only by

access to the required glycoside.

Of all the biotransformations involved in phase II metabolism it is glycosylation that

potentially has the greatest significance as a tool in the synthesis of pharmaceutical

molecules, extending well beyond the synthesis of glucuronides. The glycosylation reac-

tion is involved in the synthesis of a range of molecules that is probably the most abundant

in biology, from macromolecules such as polysaccharides and glycoproteins down to

small-molecule glycosides, with functions ranging from structure and storage to signalling

and detoxification.42 Glycosylated molecules have increasing application both as active

ingredients and in drug delivery. There are many opportunities for application of bioca-

talysis in an area where the chemistry is increasingly complex.

Many glycosylated natural products display potent pharmacological activity, including

large molecules that are strictly beyond the scope of this review, although of immense

significance to the pharmaceutical industry. Biopharmaceuticals represent a broad array of

macromolecular natural products or natural product analogues whose development has

been rapidly expanding since the introduction of recombinant insulin 20 years ago.43 There

are over 400 biopharmaceuticals currently under development, the majority being vac-

cines and monoclonal antibodies primarily targeting cancer, infectious and autoimmune

diseases. Owing to their highly complex structures, biopharmaceutical drugs are generally

produced by recombinant cell culture. The majority of such processes currently use

mammalian cell culture rather than microbial fermentation as a system for protein expres-

sion owing to the requirement for biochemical modification of the protein, which may not

be feasible in microbial cells. Post-translational modifications of proteins include glyco-

sylation and other relatively complex conjugations that are often essential for biological

activity.44

Many pharmacologically active glycosides of low or intermediate molecular weight are

also produced as natural products, extracted from plants (digoxin), animals (heparin

fractions) or microbial cultures (macrolide and aminoglycoside antibiotics). Currently,

biocatalytic glycosylation is confined to relatively simple operations rather than in vitro

synthesis of these more complex molecules, and natural products may be used as starting

materials for semisynthetic molecules. Compared with proteins and nucleic acids, which

are produced by template-driven biosynthesis, there is no equivalent to ensure the struc-

tural fidelity of carbohydrate macromolecules and other glycosylated products.

In the case of glycoprotein biopharmaceuticals, the consequence of the multiple glyco-

sylated products is the need for strict control of the manufacturing process to maintain a

reproducible spectrum of products with consistent therapeutic profile. That biopharma-

ceuticals often contain mixtures of related products can also be advantageous to the

pharmaceutical industry in warding off generic competition, due to the challenge of


replicating an exact process to generate an equivalent product. The challenge of multiple

glycosylation products, however, extends to much smaller molecules and is common to

both chemical and biological methods of glycosylation.

The anticoagulant fondaparinux, a synthetic analogue of the terminal fragment of

heparin, is synthesized using multiple protection/deprotection steps that result in a route

of up to 50 steps. There is, as yet, no enzymatic system that approaches the capability to

make such a molecule.45 As this modified pentasaccharide is a natural product, it should,

in theory, be accessible through a series of biotransformations, but we currently lack the

biocatalytic tools to achieve more than a few steps and would still need to use some

protection steps to avoid multiple products. Enzymatic synthesis in vivo depends largely

on the levels and selectivities of glycosylating enzymes to achieve multistep reactions, a

situation that has been mimicked in vitro for simpler systems.46

Most of the enzymes involved in the biosynthesis of glycosides are, like the UDP-

glucuronyltransferases of phase II drug metabolism, members of the Leloir glycosyltrans-

ferase superfamily. Enzymes in this category catalyse transfer of the saccharide from a

sugar nucleotide, usually a UDP or thymidine diphosphate glycoside, to an acceptor

nucleophile such as an alcohol or amine (Scheme 1.9). The most abundant class of

enzymes forming glycosidic bonds in nature, they are usually highly regioselective and

enantioselective and have been widely applied in organic synthesis.47,48

Many of these enzymes are membrane associated, like the mammalian UDP-glucur-

osyltransferases, but there are also many soluble enzymes, such as the UDP-glucosyl-

transferases that serve to solubilize xenobiotics in plants by forming their glucosides.

Plants appear to be particularly rich sources of glycosyltransferases: less than 30 such

enzymes have been identified in the human genome,49 yet there are 117 putative glycosyl-

transferases in the genome of Arabidopsis thaliana, a species used as a model system by

molecular biologists due to the small size of its genome relative to that of other plants.50

Lim et al.51 successfully produced a panel of transgenic organisms where glycosyltrans-

ferases from A. thaliana are expressed in bacterial cells, potentially facilitating their use in

chemical synthesis. Glycosyltransferases are generally employed in whole-cell biotrans-

formations so that the required sugar nucleotide may be generated in situ by a mechanism

similar to that outlined in Scheme 1.6.

‘Non-Leloir’ glycosyltransferases, using non-nucleotide donors such as simple sugar

phosphates, are also found throughout biology and may be applied in biocatalysis; for

example, in the N-transglycosylation reactions described in Section 1.3.3. There are, in

addition, a large number of glycosidases which hydrolyse glycosidic bonds. Specificity for

ORO(HO)n (HO)nNN

O

OHOH

O O PO

OHO P

O

OHO

O

O

β-Glycosyl transferase

ROH+ UDP

Scheme 1.9 General scheme for a UDP-�-glycosyltransferase-catalysed transformation(where ROH can be another sugar or any alcohol)


these enzymes tends to be significantly lower than for the glycosyltransferases, potentially

allowing for the development of a more general biocatalyst for glycosylation using the

reverse reaction. Glycosides may be accessed through glycosidase-catalysed reverse

hydrolysis under thermodynamic or kinetic control by using free or activated sugar donors

respectively. However, this approach is hampered by low conversions.52

This limitation was overcome by Mackenzie et al.,53 who developed glycosidase

variants containing an amine in place of the active-site carboxylate nucleophile that is

responsible for glycoside cleavage. This excellent example of rational enzyme modifica-

tion resulted in a new class of artificial enzyme known as glycosynthases, which are

capable of selectively producing oligosaccharides in high yield from glycosyl fluoride

donors of opposite anomeric configuration to that of the desired products. This pioneering

work led to the development of other artificial enzymes, such as retaining glycosynthases,

where the anomeric configuration of the glycosyl donor is retained, and thioglycoligases

and thioglycosynthases, which form S-glycosidic bonds.52 Glycosynthases are now avail-

able for the formation of a diverse range of b-(1!3)-, b-(1!4)-, b-(1!6)- and �-(1!4)-

linked oligosaccharides.48

The increasing availability of biocatalytic tools for glycosylation is applicable to the

drug discovery process as well as to metabolite preparation and synthesis of established

glycosidic active ingredients. There is growing interest in the discovery of new polysac-

charide-containing drugs. Glycorandomization, a powerful biocatalytic approach to the

generation of libraries of unnatural polysaccharides through the use of enzyme variants

with relaxed substrate specificity, represents one important approach towards this end.48

In summary, biocatalysis offers a number of alternatives to chemical synthesis for the

selective preparation of metabolites. The use of recombinant human CYPs is an attractive

method, as little screening is required and the enzymes catalyse a broad range of metabolic

reactions. Unfortunately, owing to their instability and high cost, they are unlikely to prove

suitable for the preparation of large quantities of a desired metabolite. In contrast, micro-

organisms offer a cheap and scalable alternative and their diversity can often allow the

identification of a suitable system. However, microorganisms often contain competing

activities, and so the availability of kits of microbial P450s and glycosylating enzymes,

together with the development of new methodologies such as the laccase/mediator-cata-

lysed oxidation of glycosides, offer distinct advantages.

1.3.2 Regioselective/Chemoselective Biotransformations

Selective reaction at only one position in a molecule that contains two or more of the same

functionality, or different functionalities that react in a similar manner, can be difficult to

achieve chemically without lengthy protection/deprotection strategies. In contrast, such

regio- and chemo-selective transformations can frequently be realized surprisingly easily

with a biocatalyst, as demonstrated by the following examples.

The b-lactams, mainly penicillins and cephalosporins, are by production volume the

most important class of antibiotics worldwide, enjoying wide applicability towards a range

of infectious bacteria. Most of the key molecules are semi-synthetic products produced by

chemical modification of fermentation products. Production of these molecules has con-

tributed significantly to the development of large-scale microbial fermentation technol-

ogy, and also of large-scale biocatalytic processing.


Semi-synthetic penicillins are accessed from 6-aminopenicillanic acid, (6-APA),

derived from fermented penicillin G. Starting materials for semi-synthetic cephalosporins

are either 7-aminodesacetoxycephalosporanic acid (7-ADCA), which is also derived from

penicillin G or 7-aminocephalosporanic acid (7-ACA), derived from fermented cephalos-

porin C (Scheme 1.10). These three key building blocks are produced in thousands of

tonnes annually worldwide. The relatively labile nature of these molecules has encouraged

the development of mild biocatalytic methods for selective hydrolysis and attachment of

side chains.

Penicillin acylases or amidohydrolases, which cleave the amide side chain of penicillin

G, have been known for almost 50 years.54 As one of the first enzymes to be developed for

use at scale in the pharmaceutical industry, penicillin G acylase (PGA) has often been used

as a model system for academic studies from molecular biology to biochemical engineer-

ing. Despite extensive screening, however, for decades there was no equivalent enzyme to

generate 7-ACA by cleaving the polar D-�-aminoadipoyl side chain from cephalosporin C.

The traditional chemical approach to 7-ACA requires the protection of the amine and

carboxylic acid groups of cephalosporin C by treatment with an acid chloride.55 The

resulting mixed anhydride is then converted to the imodyl chloride using phosphorus

pentachloride, which is subsequently broken down to 7-ACA with methanol and water via

a transient imodyl ether (Scheme 1.11). The use of subzero reaction temperatures and

numerous hazardous reagents, required in order to avoid hydrolysis of the acetate and

highly labile b-lactam moieties, have a significant cost and environmental impact on this

high-tonnage product.

N

SNH2

O

OO

H

CO2H

N

S

O

NH2

CO2H

N

S

O

NH2

CO2H

N

SNH

O

OO

H

CO2H

ONH2

HO2C

N

S

O

NH

O

CO2H

7-ACA

7-ADCA

6-APA

Cephalosporin C

Penicillin G

Scheme 1.10 Key intermediates for the production of semi-synthetic cephalosporins


An alternative two-step biocatalytic route, first developed at Glaxo in the 1970s, utilized

a D-amino acid oxidase and an amidase to provide 7-ACA under physiological conditions

(Scheme 1.12).56 This process has since been established in several companies, with minor

modifications. In fact, 7-ACA was manufactured by GSK at Ulverston (Cumbria, UK)

using both the chemical and biocatalytic processes in parallel for a period of 2 years

during which time the environmental benefits of the biocatalytic process were assessed

(see Section 1.6).

A single-step process using one biocatalyst might be expected to provide even greater

environmental and cost benefits. In addition to providing a simplified process, the single-

enzyme process affords D-�-aminoadipic acid as a co-product. Being an optically pure

chiral product, �-aminoadipic acid is of potential commercial value in contrast to the

ammonia and glutamic acid co-products resulting from the two-enzyme process. However,

S

N OAcO

NH

ONH2

HO2C

CO2H

CO2H

S

N

CO2COCH2Cl

OAc

O

NH

O

ClCH2CO2CO

NH

OCl

S

N OAcO

H2N

S

N

CO2COCH2Cl

OAc

O

N

Cl

ClCH2CO2CO

N

ClCl

S

N

CO2COCH2Cl

OAc

O

NClCH2CO2CO

NCl

OMe

OMe

Cephalosporin C K salt

7-ACA

Chloroacetyl chloride,

Dimethyl aniline, PCl3, PCl5 MeOH

H2O, NH3, MeOH

Dimethyl aniline, DCM

Scheme 1.11 Chemical route to 7-ACA; DCM: dichloromethane

N

SH2N

O

OO

H

CO2H

N

SNH

O

OO

H

CO2H

ONH2

HO2C

N

SNH

O

OO

H

CO2H

OO

HO2C

7-ACA

Cephalosporin C

D-amino acid oxidase,

H2O, pH 7.3, 25 oC

Glutaryl-7-ACA acylase,

H2O, pH 8.3, 30 °C

Scheme 1.12 Two-enzyme route to 7-ACA


this option proved to be somewhat elusive until the discovery of a cephalosporin C amidase

from Pseudomonas sp. SE-83 (Scheme 1.13).57 Cephalosporin C acylases have subse-

quently been found in other bacterial and fungal strains.58

There has also been extensive activity towards the replacement of the entire chemical

route to 7-ADCA (Scheme 1.14) with a biocatalytic one. This is somewhat more complex

than the above example, as the penicillin fermentation product requires ring expansion as

well as side-chain hydrolysis in order to arrive at the desired nucleus. The penicillin

nucleus can be converted to the cephalosporin nucleus using expandase enzymes, a process

that occurs naturally during the biosynthesis of cephalosporin C by Acremonium chryso-

genum and cephamycin C by Streptomyces clavuligerus from isopenicillin N (6-APA

containing a 6-L-�-aminoadipoyl side chain).59

The expandase of cephalosporin C biosynthesis is fused with a hydroxylase acting on

the 3-methyl group of the cephalosporin, these being two separate enzymes in cephamycin

C biosynthesis. For this reason the S. clavuligerus expandase is used where 7-ADCA is

the desired end product. Both expandases are highly specific for the 6-position

amide: during the biosynthesis of cephalosporin C, isopenicillin N must be isomerized

to the D-�-aminoadipoyl analogue, penicillin N, before ring expansion can be catalysed.

Not surprisingly, cheap, commercially available penicillins, such as penicillin G or

6-APA, are not substrates for the expandase.

N

S

O

NH

O

CO2H

N

S

O

NH

O

CO2H

N

S

O

NH2

CO2H

i) Sulfoxidation

ii) Rearrangement

Deacylation

Penicillin G

7-ADCA

Cephalosporin G

Scheme 1.14 Chemical route to 7-ADCA

N

SH2N

O

OO

H

CO2H

N

SNH

O

OO

H

CO2H

ONH2

HO2C

7-ACACephalosporin C

Cephalosporin C amidase

Scheme 1.13 Single-enzyme route to 7-ACA


An elegant solution to production of 7-ADCA was achieved by Conder et al.60 by

combining modifications to the fermentation process, strain and downstream biocatalytic

treatment. The expandase of S. clavuligerus was found to be active with adipoyl-6-APA as

substrate. It was recognized that adipoyl-6-APA could be generated as a fermentation

product by feeding adipic acid, in the same way that phenylacetic acid is fed to generate

penicillin G. Feeding adipic acid to a Penicillium chrysogenum recombinant strain carry-

ing a cloned expandase gene, adipoyl-7-ADCA can be directly obtained (Scheme 1.15).

The side chain of adipoyl-7-ADCA can then be removed in a subsequent step by treatment

with an acylase closely related to that used to remove the glutaryl side chain in the two-

enzyme process for 7-ACA.

The extensive literature of b-lactam antibiotics biotechnology will show many further

examples of genetic manipulation towards the formation of the three nuclei for semi-

synthetics production; however, enzymatic methods have also been sought towards the

synthesis of the final active antibiotics themselves. Further elaboration of 6-APA, 7-ACA

or 7-ADCA requires the acylation of the 6- or 7-amino groups without affecting other

sensitive functionality present in the molecules. Traditional approaches employ bulky

coupling reagents, chlorinated organic solvents (such as dichloromethane) and atom-

inefficient protection/deprotection strategies to achieve this goal. For example, in the

production of cephalexin, 7-ADCA is esterified to protect the carboxylic acid function-

ality, prior to 7-aminoacylation using a heavily functionalized mixed anhydride derivative

of (R)-phenylglycine (Scheme 1.16).61

Given that hydrolysis is a reversible reaction, the principle of microscopic reversibility

implies that biocatalytic aminoacylation should also be applicable as a mild and efficient

alternative method of introducing the side chain of both penicillin- and cephalosporin-

based antibiotics. This is the case, with PGAs proving to be particularly effective bioca-

talysts towards the aminoacylation of both penicillin and cephalosporin nuclei with a

variety of carboxylic acids.62 Amoxicillin and cephalexin, two of the most important

b-lactam antibiotics, contain an (R)-phenylglycine side chain which cannot be directly

introduced as the amino acid due to its zwitterionic nature at the moderate pH values at

N

SNH H

OO

CO2H

CO2H

S

NO

NH

O

CO2H

CO2H N

S

O

H2N

CO2H

Expandase

Adipyl-6-APA

Acylase

Sucrose+

Adipic acid

Penicillium

chrysogenum

7-ADCA

Scheme 1.15 Biocatalytic route to 7-ADCA


which PGAs operate. Amino acid esters and primary amides are not zwitterionic, and so

(R)-phenylglycine, or other amino acids, can instead be chemoselectively introduced by a

kinetically controlled PGA-catalysed reaction. PGAs are not stable in organic solvent, and

so the aminoacylation reactions are performed in an aqueous environment at high substrate

concentrations to minimize competing hydrolytic reactions.

During the some 40 years of development that have been devoted towards achieving

current levels of efficiency in the production of b-lactam antibiotics, many contributions

have been made towards our knowledge of biocatalytic processes, particularly enzyme

immobilization techniques (see Section 1.5).63 Even so, the biosynthesis of semi-synthetic

antibiotics still holds further challenges. One limitation of the current cephalexin biopro-

cess is the inhibition of PGA by phenylacetic acid, which prevents the adoption of a single-

pot side-chain exchange and, ultimately, a single-stage biosynthetic route. Schroen et al.64

overcame this limitation by employing adipoyl-7-ADCA instead of penicillin G as starting

material in the cephalexin process. PGA is not inhibited by adipic acid and so cephalexin

can be accessed using an efficient tandem adipoyl-acylase-catalysed hydrolysis/PGA-

catalysed aminoacylation procedure (Scheme 1.17).

Prodrugs provide a vehicle by which the bioavailablility of a drug displaying poor water

solubility can be enhanced or a method of targeting diseased areas of the body. Following

uptake, the drug is frequently released by the action of metabolic enzymes. For example,

the human enzyme believed to be primarily responsible for the rapid in vivo hydrolysis of

valaciclovir to aciclovir has recently been isolated and characterized (Scheme 1.18).65

NH2

N

S

O

NH

O

CO2H

S

NO

NH

O

CO2H

CO2HAdipoyl acylase,

PGA, phenylglycine ester or

phenylglycine amideCephalexin

Scheme 1.17 Preparation of cephalexin using a tandem hydrolysis/amidation approach

NH

O

O O

EtO2C

N

S

O

H2N

CO2H

N

S

O

H2N

CO2R

NH

N

S

O

NH

O

CO2R

EtO2CNH2

N

S

O

NH

O

CO2H

Protection Couple

Deprotect

7-ADCA

Cephalexin

Scheme 1.16 Chemical route to cephalexin


Given that many prodrugs are broken down by enzymic action, their enzymatic synthesis

should also be feasible.

Unlike aciclovir, many other nucleoside analogues contain a number of hydroxyl

groups, and so chemical synthesis of the desired ester prodrug with adequate regioselectivity

can be challenging. For example, attempts to prepare the L-alanine prodrug of ribavirin, a

powerful antiviral agent used to treat hepatitis C, by direct chemical esterification resulted

in a mixture of products.66 This could only be overcome by the use of a three-step

procedure involving acetonide protection/deprotection of the secondary hydroxyl moieties.

At first sight, it appears that it should be feasible to prepare such esters regioselectively

using a similar biocatalytic approach to that employed for the 6- and 7-amino acylation of

6-APA and 7-ADCA shown above. Unfortunately, owing to the poor nucleophilicity of

alcohols, biocatalytic esterification in aqueous media is far more challenging than amida-

tion. Therefore, it was not until the pioneering work of Klibanov and co-workers,67 who

first demonstrated the use of enzymes in neat organic solvents, that this option became

viable (see Section 1.4).

Employing methodology developed by the Gotor group,68 Zaks and co-workers69 were

able to produce 50-N-CBz-(S)-alaninyl ribavarin with complete selectivity using the widely

utilized lipase B from Candida antarctica (CALB) and the oxime ester of N-CBz-pro-

tected L-alanine, an irreversible acyl donor used to shift the reaction equilibrium towards

product formation (Scheme 1.19). After optimization, about 80 kg of the CBz-protected

prodrug was produced in>80 % isolated yield by the treatment of ribavirin with 0.8 weight

equivalents of CALB in tetrahydrofuran (THF) at 60 �C for 24 h. This approach has also

been used to produce ester prodrugs of other nucleoside antivirals, such as nelarabine.70

The preparation of N-CBz-(S)-valinyl lobucavir provides a particularly challenging

example, where only one of two primary alcohols is acylated with excellent regioselec-

tivity (Scheme 1.20).71

N

NN

NH

O

H2NO

O

O

NH2

N

NN

NH

O

H2NO

OH

Valacyclovirase

Scheme 1.18 Hydrolysis of valaciclovir by a human hydrolase enzyme

CbzNH CO2H

N

N

N

O

OHOH

OH

H2NOC

OH OH

N

N

N

OO

H2NOC

OCbzHN

CALB, acetone oxime,

Ribavirin

Scheme 1.19 Regioselective preparation of 50-N-CBz-(S)-alaninyl ribavarin


Simvastatin is a semi-synthetic statin that is produced from the natural statin lovasta-

tin.72 Both are potent antihypercholesterolemic agents with simvastatin differing from

lovastatin by just one additional methyl substituent residing on the 2-(S)-methylbutyrate

side chain (Figure 1.6).

Lovastatin is produced by fermentation from the filamentous fungus Aspergillus terreus

and can be converted to simvastatin by a single-step chemical methylation.73 However,

this transformation is hampered by low yields, which result in downstream processing

issues resulting from difficulties in the separation of starting material and product.

Simvastatin is instead produced using a lengthier protection/deprotection strategy.74

To overcome these separation issues, Schimmel et al.75 sought a hydrolase enzyme

capable of selectively hydrolysing the 2-(S)-methylbutyrate ester of lovastatin ammonium

salt (LAS) whilst leaving the more hindered 2-dimethylbutyrate ester of the simvastatin

ammonium salt (SAS) unchanged. After screening 150 microorganisms, the fungus

Clonostachys compactiuscula was found to produce a suitable esterase. By applying this

esterase to inseparable LAS/SAS mixtures resulting from the single-step chemical methy-

lation, they were able to hydrolyse LAS selectively to the more polar, readily separable

monacolin J ammonium salt, thus providing a two-step conversion of lovastatin to

simvastatin (Scheme 1.21).

Regioselective esterification of the 8-hydroxyl group of accumulated monacolin J,

produced using a truncated lovastatin biosynthetic pathway, could provide a viable

biocatalytic route to simvastatin. With this aim, Xie and Tang76 cloned and expressed

the acyl transferase LovD from the lovastatin biosynthetic pathway into E. coli; they found

O

OO

OH O

HO

OO

OH O

H

SimvastatinLovastatin

Figure 1.6 Structures of lovastatin and simvastatin.

N

NH

N

N

OH

OH

O

NH2 N

NH

N

N

O

OH

O

NH2

O

NHCbzO

O

NHCbz

O2N

Immobilized PCL,

Scheme 1.20 Regioselective esterification of lobucavir (PCL: Pseudomonas cepacia lipase,now known as Burkholderia cepacia)


that LovD was not only active towards lovastatin synthesis, but also capable of producing

simvastatin using simple �-dimethylbutyrate thioesters (Scheme 1.22). Whereas a bio-

transformation using partially purified LovD in aqueous solution gave a conversion of only

60 % due to competing hydrolysis, whole-cell reactions went to completion to afford 4–6 g

L�1 product concentrations. The authors speculated that the superior results obtained from

the whole-cell reactions might result from active transport of simvastatin out of the cells

which are subsequently impermeable to re-entry, whereas the more polar monacolin J can

diffuse in both directions. The efficiency of the transformation was later improved by

knocking out the gene expressing the BioH enzyme which is responsible for competing

thioester hydrolysis.77

Using molecular biology techniques to redirect primary metabolic pathways, microorgan-

isms may be engineered to overproduce a wide range of biochemical intermediates, such as

amino acids and vitamins.78 This principle can be extended by introducing novel enzymes

and, thereby, novel biotransformation steps into microbial hosts in order to generate

unnatural products from natural precursors. Such a modification may be lethal for the host

cell, requiring the application of techniques developed for controlled, conditional gene

expression in the production of recombinant proteins.79 This is illustrated by the engineered

microbial production of the nucleoside thymidine (TdR), an important starting material for

synthesis of the antiretrovirals zidovudine and stavudine (Scheme 1.23).

Although thymidine-50-triphosphate is an almost universal component of DNA, it is

exclusively derived from thymidine-50-monophosphate (TMP). In contrast, TdR does not

occur naturally and so it is impossible to manufacture TdR by manipulation of existing

metabolic pathways as for most biochemical intermediates. This problem was addressed

OH

O

OH O

HO

OO

OH O

H

O

S

SimvastatinMonacolin J

LovD

Scheme 1.22 Biocatalytic regioselective acylation of monacolin J

O

O

CO2NH4OH

OH

OH

CO2NH4OH

OH

O

O

CO2NH4OH

OH

LAS SAS

+ + SASEsterase from

C. compactiuscula

Monacolin J ammonium salt

Scheme 1.21 Selective enzymatic hydrolysis of LAS/SAS mixtures


by McCandliss and Anderson80 using a gene encoding TMPase, a phosphohydrolase

acting on TMP to generate TdR. Such enzymes are found only in rare bacterial viruses

with DNA incorporating uracil in place of thymine. This potentially lethal gene, capable of

knocking out normal DNA synthesis, was coupled to an inducible promoter allowing strict

control of its expression. Typical genetic refinements used in metabolic engineering were

introduced, knocking out enzymes that would degrade TdR and enhancing pathways

leading into TMP synthesis.

Deoxyribonucleotides are derived metabolically by reduction of the corresponding

ribonucleotide, an arrangement that reflects the greater abundance of RNA compared

with DNA. This reduction occurs at the level of the nucleoside diphosphates. TMP is

derived by methylation of 20-deoxyuridine-50-monophosphate (dUMP), itself derived from

the corresponding 20-deoxyuridine-50-diphosphate, which is generated by reduction of the

analogous ribonucleotide uridine-50-diphosphate. In order to enhance the process to

commercial levels of productivity in an engineered strain of E. coli, Anderson et al.81

added a number of recombinant genes encoding both a ribonucleotide reductase and the

thioredoxin required to regenerate its reduced and active form.

TMPase acts to dephosphorylate both TMP and its precursor dUMP, forming a mixture

of TdR and 20-deoxyuridine (UdR). As a starting material for zidovudine synthesis, TdR

must be essentially free of this impurity, which would pass through the manufacturing

process to form a demethylated analogue of zidovudine. Separation of TdR and UdR

requires difficult and costly downstream processing; hence, the key to a commercial

process is metabolic engineering to minimize biosynthetic UdR.

Anderson et al.82 investigated a range of solutions to this problem, each based on the

principle of efficient methylation of dUMP to TMP to avoid the accumulation of a

significant pool of free dUMP that could be converted to UdR. Various techniques were

used to increase the efficiency of the methylation reaction itself using alternative forms of

thymidylate synthase with enhanced catalytic activity and altered regulation. However, the

most significant improvement was by enhancing activity of the enzymes recycling and

replenishing the methyl donor 5,10-methylenetetrahydrofolate (Scheme 1.24).

N

NH

OO

O

OH

OPOHO

OH

N

NH

OO

O

OH

OH

N

NH

OO

O

OH

N3

N

NH

OO

O

OH

Phosphohydrolase

Thymidine-5'-monophosphate (TMP) Thymidine (TdR)

Zidovudine

Stavudine

Scheme 1.23 Enzyme-catalysed hydrolysis of thymidine-50-monophosphate


Biosynthetic production of thymidine is overall a complex process combining the

controlled introduction of a novel biotransformation step into a biological system with

selective enhancement or knock-out of a series of existing metabolic steps. Metabolic

engineering to enhance cofactor recycling at both ribonucleotide reduction and dUMP

methylation steps has important parallels in other systems, as whole-cell biotransforma-

tions are frequently employed as a means to supply, in situ, high-cost and usually labile

cofactors.

Atorvastatin, an antihypercholesterolemic agent, is a synthetic drug that was initially

produced in kilogram quantities using an 11-step chemical route. The syn-1,3-diol-con-

taining side chain was produced from the chiral starting material, isoascorbic acid

(Scheme 1.25).83

Numerous biocatalytic routes to this challenging intermediate have been reported.84

For example, Fox et al.85 have recently developed an efficient regioselective cyanation

starting from low-cost epichlorohydrin (Scheme 1.26). Initial experiments found that

halohydrin dehydrogenase from Agrobacterium radiobacter expressed in E. coli pro-

duced the desired product, but inefficiently. To meet the projected cost requirements for

economic viability, the product needed to be produced at 100 g L�1 with complete

conversion and a 4000-fold increase in volumetric productivity. The biocatalyst needed

to function under neutral conditions to avoid by-product formation, which causes down-

stream processing issues.

Using ProSAR technology (see Section 1.2), the group identified a variant halohydrin

dehalogenase containing 37 mutations that gave the required volumetric productivity

increase. This methodology is also applicable to other antihypercholesterolemic drugs,

such as rosuvastatin and fluvastatin (Figure 1.7).

N

NH

OO

O

OH

OPOH

O

OH

N

NH

OO

O

OH

OPOH

O

OH

TMPdUMP

DHFoCH2-THFo

THFo

CH2-THFo THFo

NADPH + H+

NADP+

(regeneration)

(regeneration)

thymidylate synthase(EC 2.1.1.145)

thymidylate synthase(EC 2.1.1.148)

dihydrofolate reductase(EC 1.5.1.3)

FADH2 FAD

CH2-THFo 5,10-methylenetetrahydrofolate

THFo tetrahydrofolateDHFo dihydrofolateNADPH reduced nicotinamide adenine dinucleotide phosphateNADP

+ nicotinamide adenine dinucleotide phosphate

FADH2 reduced flavin adenine dinucleotide

FAD flavin adenine dinucleotide

Scheme 1.24 Methylation in TMP biosynthesis


In conclusion, regioselective biocatalysis has been extensively employed to access both

semi-synthetic and synthetic pharmaceuticals. The methodology is particularly attractive

for the streamlining of processes through the elimination of protecting-group strategies

and to avoid the use of hazardous reagents.

Cl

OHCO2Et

OHCO2EtNC

OCO2Et

Halohydrin dehalogenase

NaCN

Scheme 1.26 Halohydrin-catalysed cyanation of epichlorohydrin

NOH OH

F

CO2H

N

N

OH OH

F

NS

CO2HOO

FluvastatinRosuvastatin

Figure 1.7 Some other statins containing a 1,3-syn-diol side chain.

OO

OHOH

OHOH

OHCO2MeBr

OO

NH2

CO2tBu

OONC CO2tBu

OOHNC CO2tBu

OTBDMS

NC CO2H

NO

NH

OH OHCO2H

F

i) NaBH4, Et2BOMe, MeOH, –90

o C

ii) Me2C(OMe)2, MeSO3H

H2, Raney Ni, MeOH, 50 psi

Atorvastatin

i) TBDMSCl, Im, DMAP

ii) NaCN, DMSO

i) CDI, Mg(O2CCH2CO2tBu)2

ii) Bu4NF, AcOH, THF

iii) NaOH

Scheme 1.25 Chemical synthesis of the atorvastatin side chain


1.3.3 Diastereoselective Biotransformations

Diastereoselective reactions, where one or more chiral centres are generated in a selective

manner within a molecule that already contains chirality, to produce single diastereoi-

somers (epimers) are very common in nature. Some examples of chemical processes which

harness the properties of biocatalysts are shown below.

Highly diastereoselective enzyme-catalysed glycosylation reactions allow access to

functionalized sugars and highly complex polysaccharides and provide an important

pathway by which xenobiotics are metabolized (see Section 1.3.1). A similar transforma-

tion is the nucleoside phosphorylase (NP)-catalysed reversible cleavage of the N-glycosi-

dic linkage of a nucleoside in the presence of phosphate to generate the corresponding

pentose sugar phosphate and free nucleobase. NPs are ubiquitous in biology, and substrate

ranges include deoxyribonucleosides and/or ribonucleosides with purine or pyrimidine

nucleobases. N-Transglycosylation can be achieved by coupling cleavage of the sugar

from a donor nucleoside to synthesis of a new nucleoside using a second, acceptor base.

This reaction, which is completely regioselective towards the base and diastereoselective

towards formation of the b-anomer at the sugar is a useful strategy for synthesis of

nucleoside analogues, including many antiviral and anticancer agents, such as ribavirin

or, indirectly via thymidine, zidovudine and stavudine (Scheme 1.27).86

Using guanosine or 20-deoxyguanosine as starting material for the synthesis of ribonu-

cleosides or deoxyribonucleosides respectively, the reaction can be driven towards com-

pletion by precipitation of the highly insoluble guanine co-product. This approach has

NH

N N

N

O

OHOH

NH2

O

OH

N

NH

O

OH

OH

O

O

OH

N

NH

OOH

O

O

N3

N

N

N

O

OHOH

OH

H2NOC

N

NH

OOH

O

O

N

NH

O

OH

OH

O

O

Methyluridine

ZidovudineRibavirin Stavudine

Guanosine Thymidine

Scheme 1.27 Antiretroviral nucleosides accessible by NP catalysis


been used for direct synthesis of the antiviral ribavirin in approximately 75 % yield using

bacterial cells (Brevibacterium) (Scheme 1.28).87

Like many reported N-transglycosylations, this reaction uses uncharacterized nucleo-

side phosphorylases from whole cells held at 50–60 �C, a temperature well above the range

for viability of the parent microorganism. Remarkable temperature stability has been

reported for three well-known NPs of E. coli: purine nucleoside phosphorylase (PNP),

uridine phosphorylase (URDP) and thymidine phosphorylase.88

Certain NPs can use pentoses other than ribose or deoxyribose as substrates, enabling

the synthesis of nucleoside analogues with unnatural sugar moieties: for example, in the

synthesis of purine arabinonucleosides.89 A convenient donor for transarabinosylation is

9-b-D-arabinofuranosyluridine (Ara-U), which can be accessed from uridine using a two-

step chemical process to invert the 20-hydroxyl group.90 A general protocol for preparation

of purine analogues using Ara-U with a mixture of purified URDP and PNP from E. coli is

described by Averett et al.95 The enzymes are used in varying proportions, depending upon

the reaction rate for the required purine nucleoside synthesis, and are sufficiently robust for

addition of water-miscible solvents to aid substrate solubility.

The URDP/PNP/Ara-U process is used to manufacture nelarabine, a water-soluble

prodrug of 9-b-D-arabinofuranosylguanidine produced as a treatment for acute lympho-

blastic leukaemia (Scheme 1.29).70,92 The two-enzyme process is run at 200 g L�1

NH

N N

N

O

OHOH

NH2

O

OH

N

N

N

O

OHOH

OH

H2NOC

NH

NN

CONH2

NH

N NH

N

NH2

O

RibavirinGuanosine

+H3PO4, bacterial cells, 60 °C

+

Guanine(precipitated)

Scheme 1.28 Enzymatic direct synthesis of ribavirin

NH

NH

O

O

N

N

N

NH

NH2

OCH3N

N N

N

O

OHOH

NH2

OH

OMe

N

NH

OOH

O

O

O

OOH

O

OPO3H2

OOH

O

H OHH OH

H OH

OPO3H2

URDP+

α-D-Arabinose-1-phosphate

+PNP

H3PO4+

H3PO4+

Ara-U

Nelarabine6-Methoxyguanine

Scheme 1.29 Preparation of nelarabine from Ara-U


substrate concentration and can be driven to 90 % conversion over 50 h by using the correct

ratio of the two enzymes. As with other NP-catalysed transformations, the process is run at

50 �C. To improve thermostability and facilitate reuse, the enzymes are co-immobilized

onto the same support.

For design of a simple manufacturing process, the thermostability of the NP enzymes is

a very useful feature. Although heat treatment can be used as part of a purification protocol

to isolate the enzymes from contaminating materials, the high temperature of operation

itself excludes undesired enzyme-catalysed side reactions. For example, in the synthesis of

9-b-D-arabinofuranosyladenine from Ara-U and adenine, using a wet cell paste of

Enterobacter aerogenes, adenine and Ara-U mainly underwent deamination at lower

temperatures to form hypoxanthine and uracil respectively.93 At elevated temperature,

deamination was completely eliminated and the rate of transarabinosylation increased.

One drawback of biocatalysis is that enzymes are not available in both enantiomeric

forms. Particularly where a class of enzymes whose natural substrates are optically active,

such as nucleosides, it can be difficult if not impossible to find an alternative enzyme that

will accept the unnatural substrate enantiomer. This is not insurmountable if directed-

evolution approaches are used, but it can be prohibitively expensive, especially when the

desired product is in an early stage of development or required for use only as an analytical

reference or standard.

During the development of nelarabine, the opposite enantiomer (ent-nelarabine) was

required as an analytical marker.94 The lengthy chemical route to ent-nelarabine and

the fact that this chemical route is necessary illustrate both the advantages and disadvantages

of biocatalytic approaches. The chemical synthesis of ent-nelarabine is not straightforward,

commencing with a three-step global acetylation of the sugar (Scheme 1.30). As chemical

glycosylation using arabinose results predominantly in formation of the undesired

�-anomer, ribose is instead used as the starting sugar. The enhanced diastereoselectivity

N

N N

N

NH2

Cl

O

AcO OAc

OAc

O

OH OH

OHOH O

OH OH

OHMeO O

AcO OAc

OAcMeO

N

NNH

N

NH2

Cl

N

N N

N

NH2

OMe

O

OH OH

OH

N

N N

N

NH2

OMe

O

OH OH

OH

O

AcO OAc

OAcAcO

5 steps

MeOH, HCl Ac2O, pyridine

BSA, TMSOTf, MeCN, 75 °C

NaOMe, MeOH

AcOH, Ac2O, H2SO4

Scheme 1.30 Chemical synthesis of ent-nelarabine


gained by the use of ribose in the glycosylation reaction also has a penalty, in that five

additional steps are required in order to invert the 20-alcohol of the resultant b-riboside.

Furthermore, the unnatural 6-methoxyguanine base reacts chemically at N-7, as opposed to

N-9 selectivity of the biocatalytic approach. This could be rectified by instead using

6-chloroguanine (to deactivate N-7) which could later be converted to the methoxide with

concomitant acetate deprotection. Thus, ent-nelarabine was produced using an overall

10-step procedure.

Carbon–carbon bond lyases, used in the reverse, synthetic direction have also enjoyed

significant application in the pharmaceutical industry. For example N-acetyl-D-neuraminic

acid (NANA), an intermediate in the chemoenzymatic synthesis of the influenza virus

sialidase inhibitor zanamavir, may be synthesized using NANA aldolase.

In nature, NANA arises through condensation of phosphoenolpyruvic acid with

N-acetyl-D-mannosamine (NAM) catalysed by the biosynthetic enzyme NANA

synthase.95 Owing to the labile nature of phosphoenolpyruvate, the use of this reaction

in the synthesis of NANA has been limited to whole-cell systems where this substance can

be generated biosynthetically in situ.96 Most recently, the NANA synthase reaction forms

the basis of fermentation processes for total biosynthesis of NANA.97

Catabolic enzyme NANA aldolase catalyses cleavage of NANA to form NAM and

pyruvic acid, the latter being a more attractive material for a chemoenzymatic process. It

has long been known that the reverse reaction may be used for NANA synthesis.98

However, this approach to a manufacturing process also has complications.

NAM is produced by base-catalysed epimerization of N-acetyl-D-glucosamine

(NAG), generating an unfavourable 1:4 mixture of NAM:NAG. NAG, although not a

substrate for the aldolase, inhibits the reaction. In addition, excess pyruvate is required to

push the equilibrium in favour of product formation (Scheme 1.31). Although 90 %

yields can be obtained at laboratory scale using E. coli NANA aldolase using a

NAG:NAM mixture, the NANA product is difficult to separate from the excess pyruvate

required to achieve this.

OH

NHAc

OHOH

OOH

OH

NHAc

OHOH

OOH

OH

AcHN

OOH

OH

OHOH

CO2HO

CO2H NH

AcHN

OOH

OHOH

CO2H

NH

NH2

NAMNAG

epimerization

NANA

immobilized NANA-aldolase

Zanamavir

(excess)

(base or enzyme catalysed)

Scheme 1.31 Aldolase-catalysed preparation of NANA


Cipolletti et al.99 recently described a crystallization procedure to isolate NAM in>98 %

purity from a 4:1 NAG:NAM epimerate, potentially enabling the use of a NAG-free process.

However, Mahmoudian et al.100 developed a scalable process using selective precipitation of

NAG from aqueous solutions of NAG/NAM epimerate with isopropanol to generate a

NAM-enriched solution as substrate for the enzymatic synthesis. Precipitated NAG could

be recycled by base-catalysed epimerization. The NAM-enriched starting material allowed

NANA product concentrations of 155 g L�1 to be attained by using just two equivalents of

pyruvate. Because of the lower pyruvate content, NANA could be purified by a simple

crystallization following removal of the Eupergit C-immobilized aldolase by filtration.

As an alternative to chemical epimerization, NAG epimerase may be used to maintain

a constant NAM:NAG ratio in a one-pot reaction with pyruvate and NANA aldolase.101

The epimerase is itself inhibited by pyruvate, which must, therefore, be added continu-

ously or via aliquots to the reaction. In a refined version of this reaction at laboratory scale,

Kragl et al.102 produced NANA by a continuous process, using a membrane reactor to

contain both enzymes in solution.

1.3.4 Asymmetric Biocatalysis

Asymmetric synthesis can refer to any process which accesses homochiral products. We

will focus on asymmetric synthesis from racemic or prochiral starting materials in the

presence of an enantioselective catalyst (enzyme). There are four general methodologies

commonly applied: kinetic resolution, dynamic kinetic resolution, deracemization and

asymmetrization.

The process of obtaining homochiral product from a racemate is known as kinetic

resolution. Kinetic resolution functions by the transformation of two enantiomers of a

racemic mixture at different rates. The objective is to effect a change in the physical

properties of one enantiomer to such an extent that the resulting product is readily

separable from the other. The technique suffers from the inherent inability to access

>50 % of the desired enantiomer unless the unwanted enantiomer can be racemized and

recycled or inverted.

Dynamic kinetic resolution (DKR) is an extension to the kinetic resolution process, in

which an enantioselective catalyst is usually used in tandem with a chemoselective

catalyst. The chemoselective catalyst is used to racemize the starting material of the

kinetic resolution process whilst leaving the product unchanged. As a consequence, the

enantioselective catalyst is constantly supplied with fresh fast-reacting enantiomer so that

the process can be driven to theoretical yields of up to 100 %. There are special cases where

the starting material spontaneously racemizes under the reaction conditions and so a

second catalyst is not required.

An alternative method of obtaining theoretical yields of up to 100 % of homochiral

product from racemic mixtures is known as deracemization. This process again employs

two catalysts in tandem and so bears much similarity to the DKR process. However, here

an enantioselective catalyst preferentially transforms one enantiomer of starting material

into a prochiral product. The prochiral product is then converted back into racemic starting

material using an achiral catalyst, resulting in an overall enrichment towards one enantio-

mer of starting material. Further enrichment results by allowing the process to run over

multiple cycles, until only one enantiomer remains.


The process of obtaining homochiral product from a prochiral starting material is known

as asymmetrization. This encompasses reactions where a faster rate of attack of a reactive

species occurs on one enantiotopic face of a prochiral trigonal biplanar system, or at one

enantiotopic substituent of a C2 symmetrical system, resulting in the preferential formation

of one product enantiomer. The latter is also frequently referred to as the ‘meso-trick’ or

‘desymmetrization’. These transformations can be more easily defined in pictorial form

(Figure 1.8).

Unlike kinetic resolution, catalytic desymmetrization and asymmetrization can afford

enantiopure products in theoretical yields of 100 % and are more generally applicable than

DKR or deracemization techniques.

This section will only discuss examples of catalytic kinetic resolution, DKR, desymme-

trization and asymmetrization. Deracemization will not be considered because, although

an important developing technology, examples of its application to the production of chiral

late-stage intermediates in API production have yet to appear.

1.3.4.1 Kinetic Resolution

This technique can allow the rapid development of processes for the separation of large

quantities of enantiomers and can be ideal for early-stage ‘fit for purpose’ campaigns

(where little resource is allocated to process development) in spite of the limitation in

attainable yield. This can be useful in providing sufficient homochiral product for

biological evaluation and the preparation of analytical standards of both enantiomeric

forms.

Most kinetic resolutions of pharmaceutical intermediates that have been reported

involve the use of hydrolases, particularly lipases and proteases. This is because many

hydrolases are commercially available (in bulk and kit form),104 do not require cofactors

and are active in many organic solvents (see Section 1.4). Processes can therefore, often be

developed rapidly, using high substrate concentrations and without specialist knowledge.

CX

B

DD

XC B

D

XB

D

C

A

CX

B

AD

(R )-product

(S)-product

Pro-R

Pro-S

Re

Si

Where X = carbon or heteroatomA,B,C and D are any substituent in decreasing CIP priority

Desymmetrisation

Asymmetric Transformation

Figure 1.8 Schematic representation of asymmetrization reactions.103 (Reprinted with per-mission from the American Chemical Society Copyright (2005))


A second-generation manufacturing process involving a highly enantio- and diastereo-

selective lipase-catalysed kinetic resolution step has recently been reported for the produc-

tion of pregabalin, a lipophilic g-aminobutyric acid analogue that was developed for the

treatment of several central nervous system disorders (Scheme 1.32).105

Following a screen of hydrolase enzymes, the lipase from Thermomyces lanuginosus

was selected based on its high activity and enantioselectivity. This enzyme is commer-

cially available in industrial quantities as Lipolase, a cheap catalyst of importance to the

detergents industry due to its high thermal stability and broad pH tolerance. Product

inhibition was observed at concentrations over 1 M, and so divalent ion species were

added as complexation agents. In the presence of calcium acetate, the reaction proceeded

to completion at substrate concentrations up to 3 M, although only substoichiometric

quantities were required, implying that the additive plays a more complex role than

envisaged from the original rationale. A high concentration resulted in the added benefit

of dramatically improved phase splitting during work-up, which facilitated product isola-

tion and catalyst removal. The optimized biotransformation was successfully demon-

strated in a manufacturing trial at 3.5 t scale in an 8000 L reactor.

(3R,3aS,6aR)-Hexahydrofuro[2,3-b]furan-3-ol (bisfuran alcohol), a key building block

in the synthesis of human immunodeficiency virus (HIV) protease inhibitors such as

brecanavir, can be accessed using a number of asymmetric approaches which include

lipase resolution.106 At first glance, lipase-catalysed acylation appears to be an attractive

possibility for resolution, as there is the potential to remove the undesired alcohol through

derivatization whilst leaving the desired enantiomer unchanged for subsequent chemical

transformation. However, the desired alcohol is extremely water soluble, which eliminates

the possibility of a simple extractive work-up.107 In contrast, highly enantioselective

hydrolytic resolution of the racemic acetate, using either PCL or CALB, affords the

unwanted enantiomer as an alcohol that can be removed from the desired (R)-acetate on

partition between dichloromethane and water.108 During the separation process, a thick

emulsion is formed if free enzyme is present. Emulsion formation can be avoided if an

immobilized enzyme is used, but enzyme immobilization generally dilutes catalyst activ-

ity due to the large quantity of inert support that is required. Thus, high loadings of

Novozym 435 (a commercially available form of CALB specifically designed for use in

organic solvents) were required to perform the reaction at a reasonable rate, and this led to

additional problems such as product absorption and catalyst swelling. By instead

CO2Et

CO2Et

CNCO2Et

CN

CO2H

CO2Et

CO2Et

CN

CO2H

NH2

Lipolase

pH 7, 24 h+

Pregabalin

Scheme 1.32 Kinetic resolution of a key intermediate to pregabalin


employing commercially available ChiroCLEC-PC (a cross-linked crystalline form of

PCL), the reaction proceeded rapidly at low loadings (0.05 wt %) comparable to those of

the free enzyme, whilst facilitating catalyst recovery and avoiding emulsion formation

(Scheme 1.33) (T.C. Lovelace, personal communication).109

An example where enzyme-catalysed acylation has been used to good effect was

reported by Vaidyanathan et al.110 for the preparation of an androgen receptor antagonist

that was being developed as a treatment for alopecia and oily skin. The group were

concerned that chromatographic separation of a racemic hydroxynitrile intermediate

would afford ultrapure material with an impurity profile that would not be representative

of a future commercial process. Enzymatic resolution could provide a practical solution,

but enantioselective acylation with commonly used acyl donors like vinyl acetate would

afford a neutral product that might be difficult to separate from the starting material and,

therefore, also require chromatographic purification. The authors rationalized that, by

employing succinic anhydride, previously demonstrated to be an effective acyl donor

when used with some lipases,111 an acidic product would result that could then be easily

separated from the remaining alcohol by extraction with aqueous base.

By screening a variety of lipases in organic solvent for their ability to acylate the

racemic hydroxynitrile with succinic anhydride, Novozym 435 was found to yield the

best results, affording product in 94–95 % ee at conversions of 47–49 % (Scheme 1.34).

After optimization, the reaction was successfully run at 22 kg scale. The immobilized

catalyst could be easily isolated by filtration and reused.

Given that resolution can only achieve a maximum yield of 50 %, the approach is

inherently inefficient. Additionally, classical resolution and simulated moving bed

OO

OAcH

H

OO

OAcH

H

OO

OHH

H

O

O

SN

OO

OH

H

O

NH

OH

NS

O O+ChiroCLEC-PC

Brecanavir

(Racemic)

Scheme 1.33 Kinetic resolution of a bisfuran intermediate of brecanavir

O

CN CN

CF3OH

CN

OO O

O

CNO

CO2HOH

CN

Novozym 435,

TBME, 40 – 50 °C +

(Racemic)

Scheme 1.34 Lipase resolution of a key intermediate in the synthetic route to an androgenreceptor antagonist. TBME: tert-butyl methyl ether


chromatography can provide attractive alternatives, as development times are frequently

shorter, there are no intellectual property or sourcing issues and the techniques are often

more accessible to the process chemist. Some examples of where biotransformations have

ultimately been replaced by alternative technologies are discussed below.

Lotrafiban, a nonpeptidic glycoprotein IIb/IIIa receptor anagonist that was under devel-

opment as a treatment for the prevention of platelet aggregation and thrombus formation,

was initially prepared using an 11-step linear sequence starting from methyl Cbz-L-

aspartate (Scheme 1.35).112 An overall yield of 9 % and issues with obtaining the product

in sufficient enantiopurity led the group to look for an alternative route via the enzymatic

resolution of a racemic ester intermediate.

The ester was screened against a panel of enzymes for hydrolysis activity from which

only Novozym 435 efficiently hydrolysed the desired (S)-enantiomer.113 After significant

optimization studies using Novozym 435, a process was established where a 100 g L�1

slurry of racemic ester in commercial tert-butanol (which is supplied as a mixture contain-

ing 12 % water – anhydrous tert-butanol could not be used due to its higher melting point),

furnished the desired acid in 43 % yield and >99 % ee (Scheme 1.36). The reaction

was performed at 50 �C as a compromise that gave satisfactory substrate concentration

NHCbz

CO2MeHO2C

F N O

NH2

CO2tBu

CO2Me

NNH

OMeO2C

CO2tBuN

NNH

OMeO2C

O

HCl.

Lotrafiban

NH

Scheme 1.35 Medicinal chemistry approach to lotrafiban

NNH

OMeO2C

NH

N

NN

O

O

HO2C

NNH

OMeO2C

NNH

OHO2C

HCl.

Lotrafiban+Novozym 435

Scheme 1.36 Kinetic resolution of an ester intermediate in the synthetic route to lotrafiban


(2.4 g L�1) whilst allowing the catalyst to be reused up to 10 times (running at 60 �C, a

fivefold reduction in catalyst activity was observed after a single cycle). The undesired

enantiomer, remaining as the ester, was separated from the acidic product by selective

crystallization and was subsequently racemized and recycled. This route was ultimately

run on scale at the site of primary manufacture.

In an attempt to find an improved reaction solvent, Roberts et al.113 investigated a

number of ionic liquids. Using [BMIM][PF6], an eightfold increase in substrate concen-

tration was observed compared with 88 % v/v tert-butanol, which resulted in a threefold

increase in reaction rate and allowed the isolation of acid in comparable yield and

enantiopurity to that obtained using the developed process.

Ultimately, an alternative route based on asymmetric hydrogenation using a rhodium

catalyst employing the Josiphos ligand was identified, but only demonstrated on a 10 g

scale before the project was terminated (Scheme 1.37).114

Lamivudine (also known as Epivir and 3TC) is a potent antiviral drug used in the

treatment of HIV and hepatitis B virus (HBV) infections. Although both enantiomers are

equipotent antiviral agents, the unnatural enantiomer (with respect to natural nucleosides)

is far less cytotoxic, and so a method of selectively accessing the single enantiomer was

required.

Asymmetric routes to lamivudine have recently been reviewed.115 A number of these

are biocatalytic, the most elegant of which is a highly enantioselective kinetic resolution

process based on the use of cytidine deaminase from E. coli.116 The process is particularly

impressive given that the reaction site is five atoms away from the nearest chiral centre

(Scheme 1.38).

NNH

OMeO2C

NNH

OMeO2C

P(iPr)2

P(iBu)2

Rh(COD)2BF4/L*

FeL* =

Scheme 1.37 Asymmetric synthesis of an ester intermediate in the synthetic route tolotrafiban

N

N

S

OOH

NH2

O N

NH

S

OOH

O

O N

N

S

OOH

NH2

O+

Cytidine deaminase on Eupergit C,

Lamivudine

pH 7 buffer (166 vols), 32 oC, 35 h

(Racemic)

Scheme 1.38 Cytidine-catalysed kinetic resolution of racemic lamivudine


Cytidine deaminase was not commercially available, but it is produced by numerous

microorganisms and can be induced at high levels in enteric bacteria, such as E. coli, in the

presence of cytidine. To overcome the need to add cytidine, mutant strains that express the

deaminase constitutively were sought through ultraviolet irradiation of the native micro-

organism. A selection process was developed to detect strains of interest that took

advantage of the fact that both cytidine deaminase and uridine phosphorylase are induced

by cytidine as they share the same repressor.117 Thus, any mutant that grows well on

uridine in the absence of cytidine is likely to have a defective repressor and express both

enzymes constitutively. Using this procedure, the mutant E. coli strain 3732E was devel-

oped that gave high deaminase expression independent of cytidine concentration.

However, a higher level of expression was required for pilot studies, and so a recombinant

strain, overexpressing the deaminase gene from strain 3732E, was developed that pro-

duced 80 times more deaminase than the mutant strain. Crude cell extracts of the cytidine

deaminase variant, immobilized on Eupergit C, were used on a multikilogram scale. The

desired enantiopure product could be selectively extracted by adsorption onto an anion-

exchange column and isolated in 40 % yield after subsequent recrystallization. The

biocatalytic approach was ultimately replaced by the classical resolution of an early-

stage intermediate in the final production route. Even so, the deaminase had proven

valuable for achieving preclinical supplies.118

Other examples of efficient enzymatic resolutions by reaction at a remote position from

stereocentres have been reported, such as the lipase-catalysed resolution of a synthetic

intermediate of escitalopram.119 This property of enzymes has also been effectively used

to resolve sterically hindered compounds by the introduction of a tether so that the

enzyme-catalysed reaction can be performed at an artificially created, but less hindered,

remote location. An example is the resolution of tertiary alcohols by the introduction of a

glyoxylate ester.120

Most of the examples encountered so far have employed cheap, commercially available

enzymes or enzymes that can be readily produced in-house. When a proprietary enzyme,

developed by a third party, is used, additional factors such as royalty payments, freedom to

operate and single-source supply require consideration. An example is the production of

the key (1R,4S)-azabicyclo[2.2.1]hept-5-en-3-one intermediate used in the manufacture of

abacavir, another potent reverse transcriptase inhibitor used for the treatment of HIV and

HBV infection. Enantiocomplimentary microorganisms (Rhodococcus equi NCIB 40213

and Pseudomonas solanacearum NCIB 40249) were first isolated from the environment

under conditions to select for growth on N-acyl compounds as the sole source of carbon

and energy.121 Mutant strains of Pseudomonas solanacearum NCIB 40249, hyperexpres-

sing g-lactamase, resulted in a highly enantioselective kinetic resolution process using

substrate concentrations of >100 g L�1 (Scheme 1.39 where R¼H). The process was

initially run using whole cells, as the g-lactamase was too unstable to isolate, but this

resulted in complex downstream processing. Through further microbial screening, a new

lactamase that was sufficiently stable to isolate was identified122 and subsequently cloned

(internal presentation from Dow). Using this recombinant lactamase, a highly efficient

process was developed that uses 500 g L�1 substrate concentrations and a significantly

improved workup.

In a bid to find a process that employs a commercially available biocatalyst,

Mahmoudian et al. rationalized that Boc-protection of the racemic lactam should activate


the amide bond towards nucleophilic attack. After screening a variety of commercially

available hydrolases towards hydrolysis of this substrate in 1:1 THF/buffer mixtures (to

eliminate background hydrolysis), a number of hits were obtained. Of these hits, savinase

(protease from Bacillus lentus) proved to be highly enantioselective towards hydrolysis of

the undesired enantiomer, leaving the (1R,4S)-Boc-lactam in>99 % ee at 50 % conversion

(Scheme 1.39 where R¼ tBuOC(O)O).123 Savinase and other alkaline proteases are

produced in industrial quantities for use in the detergent industry.104b,c

Carnell and co-workers have recently applied lipase-catalysed resolution to formally

desymmetrize prochiral ketones that would not normally be considered as candidates for

enzyme resolution, through enantioselective hydrolysis of the chemically prepared race-

mic enol acetate.124 For example, an NK-2 antagonist was formally desymmetrized by this

approach using Pseudomonas fluorescens lipase (PFL) (Scheme 1.40).125 By recycling the

prochiral ketone product, up to 82 % yields of the desired (S)-enol acetate (99 % ee) could

be realized.126 This method offers a mild alternative to methodologies such as base-

catalysed asymmetric deprotonation, which requires low temperature, and biocatalytic

Baeyer–Villiger oxidation, which is difficult to scale up.

NRO

NHO

NH2 CO2H

NN

NN

NH

NH2

OH

NOBoc

NHBocHO2C

+

Abacavir

R = H R =

tBuOC(O)O (Racemic)

+

γ-Lactamase

Savinase

Scheme 1.39 Enzymatic kinetic resolution approaches to abacavir

O

Ar CN

OAc

Ar CN

OAc

Ar CN

R'N

O

ArNHR''

O

Ar CN

+NK-2 antagonists

PFL, n-BuOH

THF

Scheme 1.40 Access to NK-2 antagonists by the lipase-catalysed resolution of enol acetates


1.3.4.2 Dynamic Kinetic Resolution

As seen in Section 1.3.4.1 (synthesis of lotrafiban), the recycling of an unwanted enantio-

mer resulting from a kinetic resolution allows theoretical yields of up to 100 % to be

achieved, but it can also create a bottleneck in a production process. DKR, where a starting

material undergoes racemization in situ, either spontaneously or through the action of a

second catalyst, offers a more efficient approach. This technique has been applied,

particularly in academia, to the preparation of a broad range of chiral building blocks,

and a number of recent reviews are available.127

Odanacatib is currently under clinical development for the treatment of post-menopau-

sal osteoporosis.128 The medicinal chemistry route to the (S)-fluoroleucine moiety, requir-

ing six synthetic steps from an expensive protected aspartic acid derivative and the use of

numerous hazardous reagents, was not suitable for scale-up. A more efficient chemoenzy-

matic approach was instead sought, based on the enzyme-catalysed DKR of racemic

2-phenyl-4-substituted-5(4H)-oxazolones developed by Sih and co-workers.129 The

desired racemic azalactone, efficiently produced in a high-yielding, two-pot, four-step

process underwent Novozym 435-catalysed ethanolysis in EtOH/TBME in the presence of

20 mol % of triethylamine to furnish ethyl N-Bz-(S)-g-fluoroleucinate in 80 % isolated

yield and 95 % ee (Scheme 1.41).130 Unfortunately, benzoyl deprotection of the resultant

product could not be effected without significant formation of the desfluoro compound. By

instead using the 2-(3-butenyl)-oxazolone, the amino acid derivative was produced in

comparable yields, but moderate enantioselectivity (78 % ee). However, deprotection of

the 4-pentenamide by hydroxybromination using N,N0-dibromodimethylhydantoin

and trifluoroacetic acid in water/MeCN afforded the desired product in high yield.131

Recrystallization from TBME or isopropyl acetate with H2SO4 afforded the product

as the hydrogen sulfate salt in 80 % yield and 97 % ee. This procedure was used to

produce >250 kg of API.

In addition to the moderate enantioselectivity, the DKR required one weight equivalent

of catalyst to compensate for the background ethanolysis reaction. Furthermore, a sig-

nificant quantity of hydrolysis product was produced, resulting from the water content of

CNNH

F

MeO2S

CF3

O

NH

NH

F

O

O

ROEtO

NF

O

R

ON

F

R

OH

ON

F

O

R

NH2

F

O

OEt

Odanacatib

Novozym 435, EtOHBase Base

Deprotect

H2SO4.

Scheme 1.41 Preparation of a g-fluoroleucinate intermediate of odanacatib by enzyme-catalysed DKR


the catalyst that is required for enzyme activity (see Section 1.4). By using a continuous

flow format, the biotransformation was greatly improved.132 Not only could the catalyst

loading be substantially reduced to 0.05 weight equivalents, but catalyst lifetime was also

increased 20-fold due to the absence of shear forces. Product was thus obtained in 90 %

yields and 86 % ee in kilogram quantities. The yield of hydrolysis product was reduced,

possibly as a result of the catalyst operating at suboptimal water activity due to stripping by

solvent.

To provide a more efficient route to roxifiban, a drug candidate for the treatment of a

range of cardiovascular disorders, Pesti et al. wanted to convert the hydrolytic kinetic

resolution of an isoxazoline ester intermediate, using Amano PS30 (PCL), into a

DKR.133 Attempts to effect a DKR through adjustment of the reaction pH were

unsuccessful even though the ester was prone to base-catalysed racemization via an

intramolecular Michael/retro-Michael mechanism. Based on literature precedent for

the DKR of �-thiophenyl esters, an efficient DKR process was finally established

through Amano PS30-catalysed hydrolysis of the n-propyl thioester in triethylamine

and aqueous pH 9 buffer solution to furnish the (R)-acid in 80 % yield and >99.9 % ee

(Scheme 1.42).

Clopidogrel is a potent antithrombotic agent, the chiral portion of which can be

accessed from (R)-2-chloromandelic acid. Mandelic acid derivatives are an important

class of compound in their own right owing to their use as chiral resolving agents and

as building blocks for pharmaceuticals. They can be accessed in enantiomerically

pure form by a number of biocatalytic routes, such as nitrile hydrolysis, asymmetric

cyanohydrin formation (see Section 1.3.4.5), ketoester reduction (see Scheme 1.53),

ester hydrolysis/transesterification,134 O-acetyl hydrolysis135 or hydroxyacid oxidation

(Scheme 1.43).136

One of the most attractive biocatalytic options is the nitrilase-catalysed enantioselective

hydrolysis of the racemic cyanohydrin. The hydroxyacid is produced directly without need

for protection/deprotection steps and cyanohydrins racemize spontaneously at neutral or

ON

COSPr

CN

OHN

COSPr

CN

ON

CN

CO2H

NHCO2Bu

CO2Me

NO

NH2

NHO

NH

Lipase from Ps. cepacia,

phosphate buffer

AcOH.

Roxifiban

Trimethylamine

Scheme 1.42 Enzymatic DKR of a thioester intermediate of odanacatib


high pH through the reversible loss of HCN. Another attractive aspect is that, like other

hydrolases, nitrilase enzymes require no cofactor.

DeSantis et al.137 have reported the discovery of new nitrilases through the screening of

genomic libraries created by the extraction of DNA from various environments (metage-

nomics). In preliminary experiments, using 25 mM mandelonitrile in pH 8 buffer contain-

ing 10 % methanol and 0.12 g mL�1 of one of these nitrilases, the acid was produced

quantitatively with 98 % ee within 10 min. The product was subsequently shown to be

(R)-mandelic acid after isolation in 86 % yield. In a parallel reaction, (R)-2-chloromande-

lic acid was produced at a seventeenth of the rate (Scheme 1.44).

OH

HO2C

OH

NC

OH

NC

OR

OAc

HO2C

OH

MeO2C

OH

HO2C

O

HO2CR R

R

R

R R

R

HCN

Hydroxynitrilase

Nitrilase

Monooxygenase Esterase

Esterase

Alcohol dehydrogenase

Scheme 1.43 Some potential biocatalytic approaches to optically pure mandelate derivatives

SN

OMeOCl

OH Cl

HO2C

OH Cl

NC

Cl

O

Clopidogrel

HCN

pH 8

Nitrilase

Scheme 1.44 Nitrilase-catalysed preparation of a cyanohyrin intermediate to clopidogrel


1.3.4.3 Desymmetrization

The initial synthetic route to the antifungal agent posaconazole employed an asymmetric

Sharpless–Katsuki epoxidation to afford an (R)-epoxide intermediate in high yield and

88–92 % ee (Scheme 1.45).138 The optical purity could satisfactorily be improved to

>98 % ee after one recrystallization of the diol product obtained after ring opening of the

epoxide, with retention of stereochemistry, by sodium triazole. Unfortunately, ditosyla-

tion and subsequent base-catalysed ring closure of a later triol intermediate gave an

almost equimolar mixture of cis- and trans-THF products that required chromatographic

separation.

This was overcome by acetylation of the same triol intermediate, using Novozym 435

(immobilized CALB) in vinyl acetate and acetonitrile, to afford the monoacetate in 95 %

yield and 97 % diastereoselectivity (Scheme 1.46).139 The monoacetate was then readily

converted to the desired cis-THF derivative by alcohol activation and cyclization as

described above.

By performing the desymmetrization on a prochiral diol, a far more efficient asym-

metric biocatalytic route was subsequently developed. Enzyme screening found that

NN

N

OH

F

F

OH

OH O

NN

N

F

F

OTs

O

NN

N

F

F

OTs

N

NN

N

NO

NN

N

F

F

O

O

OH

OHF

F NN

N

OH

F

F

OH

NN

N

F

F

OO

NN

N

F

F

O

CO2Et

O

F

F

OH

+

40 : 60 Cis/Trans

i. TsCl, Et3N, DMAP, CH2Cl2–THF

Posaconazole

ii. NaH, PhCH3, 100 oC

Sharpless-Katsuki

L (+)-tartrate

ii. NaH, DMF

i. MsCl, Et3N, CH2Cl2, 0-5 °C, Na diethyl malonate,

Sodium triazole, DMF

NaBH4, LiCl, EtOH

DMF, 50–55 oC

Scheme 1.45 Chemical synthesis of posaconazole


CALB was again the favoured catalyst, selectively acetylating the pro-S alcohol

(Scheme 1.47). To obtain the desired (S)-monoacetate in sufficient enantiopurity, the

reaction was not terminated when all starting material had been consumed, but allowed

to run a little further to transform a small portion of monoacetate to diacetate. This resulted

in enantioenrichment of the desired (S)-monoacetate by the preferential acetylation of the

unwanted (R)-monoacetate to prochiral diacetate.

This apparent swap of selectivity is a result of the predictable steric interactions of most

commercially available lipases with primary and secondary alcohols and carboxylic acids.

In fact, a simple predictive tool, known as the ‘Kazlauskas rules’, has been developed

where attack is favoured towards substrates of configuration shown in Figure 1.9.140 These

rules are highly predictive for secondary alcohols and less reliable for primary alcohols and

carboxylic acids.

In the case of the primary alcohols of Scheme 1.47, CALB operates in an anti-

Kazlauskas fashion, resulting in anti-Kazlauskas diol acetylation to produce the (S)-

monoacetate and anti-Kazlauskas acetylation of the (R)-monoacetate to produce diol

(Figure 1.10). In contrast, CALB is observed to act in a Kazlauskas fashion toward the

secondary alcohol shown in Scheme 1.34 and the ester shown in Scheme 1.36.

F

F

OH

OH

O

I

F

F

OAcF

F

OH

OAc

F

F

OAc

OAc

CALB

vinyl acetate, MeCN

I2, NaHCO3,

MeCN, 0 °C +

Scheme 1.47 Lipase-catalysed desymmetrization of a posaconazole intermediate

NN

N

OH

F

F

OH

OH

NN

N

OH

F

F

OAc

OHCALB

vinyl acetate, MeCN

Scheme 1.46 Lipase-catalysed diastereoselective acetylation of a posaconazole intermediate

M L

OH

M L

HO

M L

CO2H

Figure 1.9 Kazlauskas rules: preferential action of a lipase on alcohols and carboxylic acids(M and L indicate medium- and large-sized substituents respectively)


The desired S-monoacetate could thus be obtained in 81 % yield and 97 % ee at pilot-plant

scale. The tertiary centre could then be constructed by diastereoselective iodocyclization of

the resultant monoester, thus removing the need for the Sharpless–Katsuki epoxidation.

Diacetate remained unchanged during this step and could be removed at a later stage.

Moderate yields of monoacylated product (74–81 %) were initially obtained using vinyl

acetate as acylating agent as significant diacetylated by-product formation was necessary

to achieve sufficiently high monoacetate enantiopurity. The ultimate route developed for

the manufacture of multi-ton quantities of posaconazole used isobutyric anhydride as the

acylating agent (Scheme 1.48).141 This more bulky acylating agent proved to be superior,

affording>90 % yields of the desired product at low temperature (�14 �C) in the presence

of NaHCO3 to suppress background reaction and acyl migration respectively.

Desymmetrization is not restricted to a single class of enzyme. For example, Madrell

et al.142 reported the gram-scale preparation of a key intermediate of the lovastatin lactone

through the desymmetrization of 3-(benzyloxy)glutaronitrile using whole cells from

Brevibacterium R312. The transformation occurs via a dual nitrile hydratase/amidase-

catalysed hydrolysis to afford acid in 65 % yield and 88 % ee (Scheme 1.49).

F

F

HO

HOF

F

AcO

HO

Figure 1.10 Anti-Kazlauskas action of CALB on the primary alcohol intermediates ofposaconazole

O

OO

OH O

H

OBn

CNCN

OBn

CN CO2H

Lovastatin

Brevibacterium R312

Scheme 1.49 Synthesis of a key hydroxyacid intermediate of lovastatin

OH

OH

FF

OH

O

FFO

O

O

O

CALB, NaHCO3

Scheme 1.48 Industrial-scale desymmetrization of a posaconazole intermediate


Using a similar approach, Bergeron et al.143 prepared the side chain of atorvastatin via a

nitrilase catalysed desymmetrization of 3-hydroxyglutaronitrile. The dinitrile was pre-

pared in two steps from epichlorohydrin, albeit in moderate yield. A highly enantioselec-

tive desymmetrization was then performed using the nitrilase BD9570, developed by Burk

and co-workers,144 expressed in a strain of Pseudomonas fluorescens (Scheme 1.50). The

enzyme was obtained solely as a soluble, active multimer in excess of 25 g L�1 by

fermentation, a quantity that represented >50 % of the total cell protein. An advantage

of high-level protein expression is greatly simplified downstream processing of the

enzyme, a contributing factor to the enzyme cost. In addition, if the enzyme is inexpensive

there is no need to recycle, therefore potentially obviating the need for catalyst immobi-

lization. However, reaction workup was problematic due to the high water solubility of the

product and the presence of cell debris resulting from the use of crude catalyst.

1.3.4.4 Asymmetric Ketone Reduction

Microbial reduction has been recognized for decades as a laboratory method of preparing

alcohols from ketones with exquisite enantioselectivity. The baker’s yeast system repre-

sents one of the better known examples of biocatalysis, taught on many undergraduate

chemistry courses. Numerous other microorganisms also produce the ADH enzymes

(KREDs) responsible for asymmetric ketone reduction, and so suitable biocatalysts have

traditionally been identified by extensive microbial screening. Homann et al.145 have

recently reported the identification of a subset of 60 ADH-producing microbial cultures

that cut microbial screening time from weeks to days.

The advantage of using living microorganisms for bioreduction is that they can be

readily sourced from the environment and the cofactors (necessary to regenerate the

reduced form of the ADH enzyme and, thus, allowing catalyst turnover) are constantly

generated by the intact cellular metabolic machinery. However, reduction using native

microorganisms does have several drawbacks. Microorganisms often contain a number of

ADHs that can display different or opposite enantioselectivities towards a given substrate.

Also, enzymes displaying competing activities might be present or the desired enzyme

might not be sufficiently active towards a chosen substrate or poorly expressed by the

native organism. Furthermore, most living cells only tolerate low substrate and organic

solvent concentrations. For example, Barbieri et al.146 used whole cells from Geotrichum

candidum to produce 2 g L�1 titres of (S)-chlorohydrin in 90 % yield and 93 % ee. The

chlorohydrin can be used as a chiral building block in the synthesis of sertraline, an

antidepressant and anorectic agent (Scheme 1.51). To overcome product inhibition, two

OHCNCl

OH

2H

OCl

OHCNNC

OHNC CO NC CO2Et

Nitrilase

i. HCN

ii. Base

NaCN

pH 7.5, 16 h EtOH

H2SO4

Scheme 1.50 Nitrilase desymmetrization approach to the atorvastatin statin side chain


weight equivalents of the nonionic macroreticular resin Amberlite XAD-1180 was used for

in situ product removal. This resulted in a twofold increase in product titre from an

unoptimized reaction and both yield and enantioselectivity also increased.

Product extraction from large volumes of fermentation broth can be complex, requiring

large volumes of organic solvent or solid-phase extraction techniques, which can some-

times greatly reduce or even cancel out the benefits of the biotransformation itself, such as

shorter route and environmentally benign conditions.

Given the large capital investment required for specialist equipment, the fermentation

needs to display considerable production cost benefits over the chemical process to be

considered seriously as a route to API manufacture.

Partially purified or isolated ADHs offer several advantages:

• higher substrate concentrations

• higher solvent tolerance

• simplified downstream processing.

Unlike the whole-cell system, enzymatic reductions require the addition of a hydride

donating cofactor to regenerate the reduced form of the enzyme. Depending on the chosen

ADH, the cofactor is usually NADH or NADPH, both of which are prohibitively expensive

for use in stoichiometric quantities at scale. Given the criticality of cofactor cost, numerous

methods of in situ cofactor regeneration, both chemical and biocatalytic, have been

investigated. However, only biocatalytic regeneration has so far proven to be sufficiently

selective to provide the cofactor total turnover numbers of at least 105 required in

production.147

Biocatalytic approaches to cofactor regeneration can be divided into coupled-enzyme

methods and coupled-substrate methods.148 In the coupled-enzyme method, the oxidized

cofactors (NADþ and NADPþ) are recycled in situ by performing an oxidation reaction

using a second enzyme and an inexpensive auxiliary substrate. This second enzyme must

employ the same cofactor, but neither enzyme should be able to accept the same substrate.

ClO

ClCl

ClOH

ClCl

O

Cl

Cl

NHMe

ClCl

O

Cl

Cl

O

G. candidum,

XAD-1180 resin

NaOH

Diethyl malonate

Na, dioxane

Sertraline

Scheme 1.51 Chemoenzymatic approach to sertraline


Furthermore, the oxidation reaction needs to be irreversible so as to drive the reduction

reaction to completion. NADþ and NADPþ are most frequently recycled using formate

dehydrogenase (FDH) and glucose dehydrogenase (GDH) enzymes respectively as the

second enzyme. By the introduction of formate and glucose as co-substrates, the oxidized

forms of FDH and GDH irreversibly generate carbon dioxide and D-glucono-1,5-lactone

respectively, thereby driving the reduction to completion. Alternatively, another ADH can

be employed as the second enzyme in the presence of an inexpensive ketone so long as the

resultant alcohol can be removed from the reaction mixture in some way as it forms.

Davis et al.149 adopted the coupled-enzyme method to access the (S)-hydroxyester

(Scheme 1.52) that is subsequently fed into the halohydrin-dehydrogenase-catalysed

cyanation process shown in Scheme 1.26. Reaction workup using wild-type enzymes

gave an emulsion that settled slowly, thus wasting valuable plant time. Modification of

both ADH and GDH enzymes allowed improved separation as well as increased reaction

rate and catalyst stability.

Recombinant cells expressing a cloned ADH have also been used in a coupled enzyme

method to efficiently produce the (R)-2-chloromandelate intermediate in the synthetic

route to clopidogrel in 90 % yield and >99 % ee at 200 gL�1 substrate concentration

(Scheme 1.53).150 This procedure does not use hydrogen cyanide and, therefore, represents

a less hazardous alternative to the nitrilase- and hydroxynitrilase (HnL)-catalysed

approaches shown in Scheme 1.44 and Scheme 1.56 respectively.

The coupled substrate method is perhaps the simplest approach to asymmetric ketone

reduction, using a single recombinant ADH to perform the oxidation of a cheap auxiliary

OH Cl

MeO2C

O Cl

MeO2C ADH

NADPH NADP+

D-glucoseD-glucono-1,5-lactoneGDH

Scheme 1.53 ADH approach to the (R)-2-chloromandelate intermediate to clopidogrel

Cl

OCO2Et Cl

OHCO2Et

ADH

NADPH NADP+

D-glucoseD-glucono-1,5-lactoneGDH

Scheme 1.52 ADH reduction approach to the atorvastatin side chain


substrate (such as a low molecular weight alcohol) in addition to the desired reduction. By

using a large excess of sacrificial alcohol, the reaction can be driven towards formation of

the desired reduced product.

Montelukast, a leukotriene antagonist used for the treatment of asthma, is produced as

a single enantiomer. Asymmetric reduction of the ketone with most hydrogenations

and metal hydrides is precluded due to the presence of other sensitive functionality.

Using (�)-b-chlorodiisopinocamphenylborane ((�)-DIP-Cl) as the reducing agent at

�20 �C, the desired alcohol can be produced in 80 % isolated yield and 99.5 % ee,151

but 1.8 equivalents of this moisture-sensitive and corrosive reagent are required (Scheme

1.54). In light of the need to use stoichiometric quantities of reagent, the development of

more efficient catalytic methods has been the subject of extensive research.

Using a microbial screening strategy, Shafiee et al.152 found that the chiral hydroxyester

can be generated from Microbacterium campoquemadoensis in >95 % ee. The whole-cell

reaction was optimized to produce 500 mg mL�1 product concentrations after 280 h. The

ADH responsible was purified and found to be NADPH dependent and active in hexane or

DMSO/aqueous mixtures, but no attempt to clone this enzyme has been reported.

Ulijn et al. identified an enzyme, capable of enantioselectively reducing the ketone, from

their extensive collection of ADH variants; further modification of the hit resulted in a

biocatalyst that produces the desired (S)-alcohol in >99.9 % ee at concentrations of

100 gL�1 in a solid-to-solid biotransformation,153 where both starting material and product

display only sparing solubility in the reaction medium.154 High conversions (>99 %) are

achieved by the substrate-coupled method, using 50 % v/v isopropyl alcohol concentrations

to drive the reaction by continuous acetone removal (Scheme 1.55). The product can be

easily isolated by filtration and washing.

NCl

OHS

CO2Na

NCl

O CO2Me

NCl

OH CO2Me

Montelukast

(–)-DIP-Cl (1.8 equivs), THF,

–20 °C, 4 h

Scheme 1.54 Preparation of a montelukast intermediate using a chemical asymmetric catalyst

NCl

O CO2Me

NCl

OH CO2MeADH, NADPH

IPA,toluene, water, 45 oC

Scheme 1.55 Alcohol dehydrogenase preparation of a montelukast intermediate


Both enantiomers of 1-[3,5-bis(trifluoromethyl)phenyl]ethan-2-ol are of importance

in the pharmaceutical industry, and so considerable effort has been expended in

their asymmetric synthesis. The (R)-enantiomer is a building block for aprepitant, a

neurokinin-1 (NK-1) antagonist used for the treatment of chemotherapy-induced nausea

(Figure 1.11).155

Gelo-Pujic et al.156 recently reported the results of a comparison between enzymatic,

microbial and chemocatalytic asymmetric reduction of 1-[3,5-bis(trifluoromethyl)phe-

nyl]ethanone. Whereas both biocatalytic methods gave high product ees, both systems

only functioned at low substrate concentrations and the enzymatic method gave inferior

conversions to the whole-cell system. The chemocatalytic method gave moderate pro-

duct ees but could be performed at high substrate concentrations and gave high yields.

However, the enzymatic approach was only tested using the substrate-coupled method.

In sharp contrast, Pollard et al.157 efficiently prepared both alcohol enantiomers with

different isolated ADHs using the enzyme-coupled method. For example, using the

commercially available ADH from Rhodococcus erythropolis and a GDH cofactor

recycling system they produced (R)-alcohol in >98 % yield and >99 % ee at

200 g L�1 concentrations on a 25 kg scale. Caution clearly needs to be taken in the

proper choice of reaction conditions.

1.3.4.5 Asymmetrization Using Other Biocatalysts

Another class of biocatalyst of great potential for the preparation of chiral intermediates

through asymmetric carbon–carbon bond formation is the HnLs. A range of HnLs are

commercially available which are enjoying increasing interest in the pharmaceutical

industry. In addition to the nitrilase and ADH approach to the (R)-2-chloromandelate

intermediate to clopidogrel discussed earlier (Schemes 1.44 and 1.53), asymmetric cyana-

tion of 2-chlorobenzaldehyde using the crude HnL from Prunus amygdalus (almond meal)

has also been reported.158 The reaction is run at low pH (to slow the background reaction),

to afford the cyanohydrin in 90 % ee (Scheme 1.56).

Several approaches to statin side-chain intermediates have so far been discussed. Whereas

these chemoenzymatic approaches provide clear benefits over the chemical processes, they

do not harness the true potential of biocatalysis as the biotransformations have simply been

inserted into the existing chemical route. Wong and co-workers have developed a more

biosynthetic-like approach by using a mutant 2-deoxyribose-5-phosphate aldolase (DERA)

NHN

NN

O

O

F

O

CF3

CF3

HOCF3

CF3

Aprepitant

H

Figure 1.11 Aprepitant and an (R)-1[3,5-bis(trifluoromethyl)phenyl]ethan-2-ol intermediate


(Scheme 1.57).159 Although the natural donor aldehyde is D-2-deoxyribose-5-phosphate,

non-phosphorylated donor aldehydes are also tolerated and the enzyme displays some

flexibility towards both donor and acceptor. Importantly, as both donor and acceptor

substrates are aldehydes, the enzyme can perform sequential aldol reactions allowing the

preparation of a key lactol intermediate to the atorvastatin side chain in a single step.

Following substantial modification, this approach is now operated on an industrial scale to

produce this intermediate in >100 gL�1 concentrations.84

In a recent patent, Hu et al. reported a similar procedure where the acceptor aldehyde

contains aminoalkyl substituents in place of chloride.160 Subsequent to lactol oxidation

and amine deprotection, these intermediates can directly undergo Paal–Knorr cyclization

with the appropriate diketone to produce atorvastatin, thus avoiding the use of cyanide

chemistry.

The flexibility of DERA enzymes makes them a valuable synthetic tool for the

quick access to a range of polyoxgenated products, such as the cytotoxic agent epothilone

A (Scheme 1.58).161

Of the known classes of aldolase, DERA (statin side chain) and pyruvate aldolases

(sialic acids) have been shown to be of particular value in API production as they use

readily accessible substrates.162 Glycine-dependent aldolases are another valuable class

that allow access to b-hydroxy amino acid derivatives. In contrast, dihydroxyacetone

phosphate (DHAP) aldolases, which also access two stereogenic centres simultaneously,

O

ClO

OH

OH

Cl

O

OO O

OCl

O+

DERA+

Scheme 1.57 DERA approach to the atorvastatin side chain

SN

OMeOCl

OH Cl

HO2C

OH Cl

NC

Cl

O

Clopidogrel

Low pH

ChemicalHnL, HCN,

Hydrolysis

Scheme 1.56 Preparation of a clopidogrel hydroxyacid intermediates with HnL


have only been of academic interest as they require expensive phosphorylated aldehyde

donors and produce phosphorylated products that require subsequent deprotection. This is

beginning to change with the discoveries of fructose-6-phosphate aldolase (FSA) that

accepts dihydroxyacetone (DHA) as substrate163 and that DHAP aldolase can accept DHA

when used in borate buffer due to the transient formation of a borate ester that mimics

phosphate.164

As more enzyme kits become commercially available, the screening for a suitable

catalyst can now be performed in a matter of hours rather than days or weeks.

Furthermore, both the screening and biotransformation can be performed by nonspecia-

lists. This increases the likelihood of uptake of a biocatalytic process, as a proof of concept

can be more readily obtained without the commitment of considerable resource. For these

reasons, the use of ADHs by pharmaceutical companies has increased considerably in

recent years.

1.4 Enzymes in Organic Solvent

Biocatalysis has traditionally been performed in aqueous environments, but this is of

limited value for the vast majority of nonpolar reactants used in chemical synthesis. For

a long time it was assumed that all organic solvents act as denaturants, primarily based on

the flawed extrapolation of data obtained from the exposure of aqueous solutions of

enzyme to a few water-miscible solvents, such as alcohols and acetone, to that of all

organic solvents.165

This assumption has since been swept aside and it is now recognized that a broad range

of enzymes retain their activity on exposure to organic solvents or organic solvent–water

mixtures. The addition of organic solvent allows the coupling of the exquisite selectivities

observed from traditional approaches with numerous other advantages, such as:

• increased concentrations of nonpolar reactants;

• enablement of reactions that have unfavourable thermodynamic equilibria in water;

• enhanced biocatalyst stability towards heat and autolysis;

• compartmentalization of substrate/product from enzyme (reduced substrate/product

inhibition);

• modification of enzyme selectivity;

• selective inhibition of competing enzymes;

N

S

O

O

O O

OH

OH

O

OH

OH

OH O O OtBu

O

PMPO

OTBSO

OHOMe

OMe

O O

OH

OHN

SI

OAc

Epothilone A+

DERA

+ii. DERA

i. Dowex (H+)

Scheme 1.58 DERA approach to the synthesis of epothilone A


• reduced background reaction;

• improved workup;

• better integration into synthetic routes;

• greater potential for tandem chemoenzymatic processes.

The field of biocatalysis in organic media is now of considerable industrial importance,

enjoying widespread application, particularly in the preparation of enantiopure

intermediates.166

Enzyme catalysis in nonconventional media can be divided into a number of different

categories depending on whether the aqueous and organic phases are miscible or immis-

cible and whether the biocatalyst is dissolved or not. In this section, only ‘free’ enzymes

will be considered. Thus, the field can be simplified to just two categories, depending on

whether the solvent is water miscible or immiscible (systems employing water-immiscible

solvents, where water is present in quantities that are below its solubility limit, have been

considered as monophasic):

1. monophasic biocatalysis

2. biphasic biocatalysis.

The state of the catalyst (homogeneous or heterogeneous) is dictated by the relative

quantities of solvent and water used.

1.4.1 Monophasic Biocatalysis

The structural integrity of enzymes in aqueous solution is often compromised by the

addition of small quantities of water-miscible organic solvents.167 However, there are

numerous examples, particularly using extremophiles,168 where enzymes have been suc-

cessfully employed in organic solvent–aqueous mixtures.166b A good example is the

savinase-catalysed kinetic resolution of an activated racemic lactam precursor to abacavir

in 1:1 THF/water (Scheme 1.39). The organic solvent is beneficial as it retards the rate of

the unselective background hydrolysis.

The use of water-miscible organic solvent–water mixtures is a particularly attractive

method for use with cofactor-dependent enzymes due to its simplicity. The high water

content can allow dissolution of both enzyme and cofactor, whilst the water-miscible

solvent can provide a dual role in both substrate dissolution and as a cosubstrate for

cofactor recycling (substrate-coupled cofactor recycling).148 The asymmetric reduction

of a ketone intermediate of montelukast using an engineered ADH in the presence of 50 %

v/v isopropanol offers a powerful demonstration of this methodology (Scheme 1.55).

It might be expected that in miscible organic solvent–water mixtures of increasing

organic solvent content, the structural integrity of many enzymes will progressively

diminish due to loss of essential hydrogen bonding. In fact, this is not the case, as

demonstrated by Griebenow and Klibanov,165 who used Fourier-transform infrared spec-

troscopy to assess the effect of acetonitrile–water mixtures (0–100 %) on the secondary

structure of lysozyme. Rather than a gradual loss in secondary structure with increasing

organic solvent content, they observed an inverse bell-shaped relationship, with maximum

�-helicity occurring at both high water and high organic solvent content. Reduced enzyme

solubility at high organic solvent content might have provided an attractive rationale, but

this was not supported by the data. A similar trend was observed using Bacillus subtilisin

1.4 Enzymes in Organic Solvent 55

protease (also known as subtilisin Carlsberg) and other water-miscible organic solvents.

The authors concluded that enzyme denaturation increases as the organic solvent content

increases. At the same time, a decline in water content reduces conformational mobility so

that the enzyme becomes kinetically trapped in an active conformation.

In addition to the retention of structural integrity in neat organic solvents, Klibanov and

co-workers demonstrated that a diverse range of enzymes, from hydrolases and perox-

idases to cofactor-dependent alcohol oxidases and ADHs, also retain activity.67 This

pioneering work single-handedly led to the popularization of biocatalysis in neat organic

solvent.

Recent literature has shown that nonaqueous biocatalysis is not limited to traditional

organic solvents, with examples that employ ionic liquids169 and supercritical fluids170

now widespread. Reaction in organic solvent has also led to the discovery that some

enzymes display promiscuity towards reaction type as well as substrate type,171 with the

HnL-catalysed asymmetric Henry reaction,172 and the lipase-catalysed Michael-type

addition of thiols to �,b-unsaturated enones providing some recent examples.173

Enhanced rigidity of enzymes in nonaqueous media also imparts greater thermostability,

allowing reactions to be run at temperatures of up to 100 �C over prolonged time

periods.174 For example, the kinetic resolution of a key intermediate in the synthesis of

lotrafiban using Novozym 435 as catalyst (Scheme 1.36) can be performed at temperatures

of 70 �C over prolonged reaction times without enzyme degradation. However, a lower

temperature of 50 �C was employed in the final production route due to limitations of the

immobilization technique used rather than the enzyme. In the 88 % tert-butanol–12 %

water solvent mixture, required to provide sufficient substrate solubility, substantial

enzyme desorption from the support at higher temperatures limited reuse of this expensive

catalyst.

Efficient biocatalysis in neat organic solvent depends on the careful choice of the

method of ‘dehydrated’ enzyme preparation and solvent used. Optimization of these

factors towards a given transformation is often known as ‘catalyst formulation’ and

‘solvent, or medium, engineering’ respectively, both of which will be briefly discussed

below. ‘Catalyst engineering’ which also provides a powerful method of improving

activity and stability, is discussed in Chapter 2.

1.4.1.1 Catalyst Formulation

A requirement of biocatalysis in neat organic solvent is the use of a dehydrated form of an

enzyme that displays the desired activity. A number of techniques are available for the

preparation of dehydrated enzymes, some of which are discussed in a recent review by

Griebenow and Barletta.175 The techniques that have been most commonly used are:

• lyophilization

• precipitation

• immobilization (see Section 1.5).

The resultant dehydrated enzyme preparations often display comparable activity to

untreated enzyme when reconstituted in aqueous buffer. However, in the case of many

enzymes, activity in a suitable neat organic solvent can be three to five orders of

magnitude lower than in water. This was recognized by Klibanov early on in the


development of the field, and so many of the basic principles leading to reduced

efficiency have been elucidated. These have been extensively reviewed and will only

be briefly discussed here.176,177

A major cause of suboptimal activity in organic solvent results from the removal of

‘essential water’ during enzyme dehydration. All enzymes require some water in order to

retain activity through the provision of conformational flexibility.178 Particularly in the

case of lipases, the amount of water can be so low that it appears that none is required. For

example, following the development of suitable techniques to analyse low water concen-

trations,179 it has been reported that the lipase from Rhizomucor miehei retains 30 % of its

optimum activity with as little as two or three water molecules per molecule of

enzyme.180,181 Owing to the apparent absence of water in some exceptional cases, the

term ‘biocatalysis in anhydrous solvent’ is commonly used, although in the vast majority

of cases a monolayer of water is required for optimal activity (although this is often still

well below its solubility limit in water-immiscible solvent).67

Numerous ‘tricks’ have been developed to retain activity of the dehydrated enzyme

preparation. Activity can be dramatically enhanced by adding a small quantity of water to

the enzyme prior to use,182 but this can be detrimental in transformations where it can

participate as a reactant, particularly where the reagents are expensive. Retention of

activity without the need to partially rehydrate has, therefore, been the focus of intensive

investigation. Some effective strategies, such as co-lyophilization in the presence of

lyoprotectants (sugars or hydrophilic polymers) and the use of additives such as crown

ethers, substrate or transition-state analogues (molecular imprinting) or inorganic salts,

have recently been reviewed by Serdakowski and Dordick.177 Some of these techniques

can lead to dramatic changes in enantioselectivity and activity.183

The ionization state of polar (ionogenic) residues of the dehydrated enzyme preparation

can also have a substantial impact on conformation and, hence, on activity in organic

solvent. The ionization state can be optimized through pH control of the aqueous solution

from which the enzyme was last in contact. Commonly referred to as the ‘pH memory’

effect, optimum activity in organic solvent is usually attained by preparing the dehydrated

enzyme from an aqueous solution of optimal pH for enzyme activity in conventional

media. In many cases, charged species are generated during the course of a transformation

that can affect the enzyme ionization state. This can be controlled through the addition of

solid-state buffers to the reaction mixture.184

Because enzymes are insoluble in organic solvent, mass-transfer limitations apply as

with any heterogeneous catalyst. Water-soluble enzymes (which represent the majority of

enzymes currently used in biocatalysis) have hydrophilic surfaces and so tend to form

aggregates or stick to reaction vessel walls rather than form the fine dispersions that are

required for optimum efficiency. This can be overcome by enzyme immobilization, as

discussed in Section 1.5.

1.4.1.2 Solvent Engineering

Enzyme activity varies greatly depending on solvent choice, as illustrated by Zaks and

Klibanov185 for the transesterification of tributyrin and heptanol by three different lipases.

Using these data, Laane et al.186 found that enzyme activity correlates closely with solvent

hydrophobicity (log P) for the lipases from Mucor sp. (MML) and Candida cylindracea


(CCL – now known as lipase from Candida rugosa (CRL)) but not porcine pancreatic

lipase (PPL) (Figure 1.12).

It was postulated that the differences in enzyme activity observed primarily result from

interactions between enzyme-bound water and solvent, rather than enzyme and solvent. As

enzyme-associated water is noncovalently attached, with some molecules more tightly

bound than others, enzyme hydration is a dynamic process for which there will be

competition between enzyme and solvent. Solvents of greater hydrophilicity will strip

more water from the enzyme, decreasing enzyme mobility and ultimately resulting in

reversible enzyme deactivation. Each enzyme, having a unique sequence (and in some

cases covalently or noncovalently attached cofactors and/or carbohydrates), will also have

different affinities for water, so that in the case of PPL the enzyme is sufficiently

hydrophilic to retain water in all but the most hydrophilic solvents.

The impact of water on enzyme activity is powerfully demonstrated by the chymotrypsin-

catalysed transesterification of ethyl N-acetyl-L-phenylalaninate with propanol. In dry acet-

one, the reaction is over 7000 times slower than in dry octane. However, by adding 1.5 % v/v

water to acetone, the reaction rate dramatically increases to two-thirds the rate of that in dry

octane.67a Zaks and Klibanov also demonstrated the effect of water stripping on enzyme

activity by incubating chymotrypsin in various organic solvents and then assessing the

resulting enzyme water content. Activity in the different organic solvents was found to

correlate well with water retained by the enzyme. Halling was able to rationalize such

findings by realizing that a given enzyme requires a defined quantity of water to attain

optimal activity. This can be expressed in terms of thermodynamic water activity, which

essentially describes the amount of water bound to the enzyme.187 Thus, optimum chymo-

trypsin activity in acetone is realized at the same thermodynamic water activity as that in

Figure 1.12 Transesterification activity of PPL, CCL and MML in various organic solvents


octane even though the total water content of each system is very different. However, at

comparable water activity, variations in optimum enzyme activity observed in each solvent

show that the direct effect of solvent on the enzyme is also an important factor which may

account for the activity deviations from the activity/log P relationship seen in Figure 1.12.

The choice of organic solvent can also have a dramatic effect on selectivity.166a In contrast

to enzyme activity, in the majority of examples reported there appears to be no correlation

between solvent physical properties and enantioselectivity. In fact, investigating the effect

of various solvents towards a number of lipases, Secundo et al.188 also found that the

optimal solvent differed with both enzyme and substrate. A number of theories have been

postulated in order to explain these effects in individual cases, but none has any general

predictive value.183b This is somewhat intriguing given that differences in enantioselectivity

simply relate to a change in the relative rate of conversion of each enantiomer.

Reaction in organic solvent can sometimes provide superior selectivity to that observed

in aqueous solution. For example, Keeling et al.189 recently produced enantioenriched

�-trifluoromethyl-�-tosyloxymethyl epoxide, a key intermediate in the synthetic route to a

series of nonsteroidal glucocorticoid receptor agonist drug candidates, through the enan-

tioselective acylation of a prochiral triol using the lipase from Burkholderia cepacia in

vinyl butyrate and TBME (Scheme 1.59). In contrast, attempts to access the opposite

enantiomer by desymmetrization of the 1,3-diester by lipase-catalysed hydrolysis resulted

in rapid hydrolysis to triol under a variety of conditions.

1.4.2 Biphasic Biocatalysis

Biocatalysis in biphasic mixtures of water-immiscible organic solvent and water involves

the transfer of low concentrations of substrate from the organic to aqueous phase during

agitation. The substrate then undergoes transformation before returning to the organic

phase. The partition of substrate/product between the two phases is independent of their

ratio and so the volume of the organic phase can be much greater than the aqueous phase,

allowing high-intensity transformations to be achieved whilst simultaneously minimizing

exposure of the enzyme to organic species. The technique is particularly valuable for

transformations in which the enzyme is sensitive to inhibition by high concentrations of

substrate or product and transformations where cofactor recycling is required.

Biphasic conditions can also be used to suppress background reaction. HnL-catalysed

asymmetric addition of cyanide to aldehydes and ketones provides an important example,

OHOHOHF3C

OOHOHF3C

O

Pr O

O

PrO CF3

NN

N

O

R2

OHR1NH

O NH2

F3C

R

Lipase, vinyl butyrate

TBME

85% yield92% ee

Scheme 1.59 Synthesis of nonsteroidal GR agonists


allowing chiral intermediates to APIs such as clopidogrel to be accessed in excellent enantio-

purity (Scheme 1.56). However, whereas the biphasic method of controlling background

reaction works well with nonpolar substrates, it is less effective with polar, water-soluble

substrates such as 3-pyridinecarboxaldehyde. Such substrates require transformation under

nearly anhydrous conditions where, unfortunately, HnLs rapidly deactivate. Faced with this

issue, Roberge et al.190 have recently reported that HnLs, immobilized as cross-linked enzyme

aggregates (CLEAs), retain their activity in nearly anhydrous conditions (see Section 1.5.2 for

further details of CLEAs). Using two different commercially available HnL CLEAs they were

able to produce either of the enantiomers of 3-pyridinecarboxaldehyde cyanohydrin in

moderate to high yield and >90 % ee in dichloromethane containing just 0.18 % water.

The solvent present in biphasic reactions can still have an effect on the enzyme even

though the enzyme functions primarily in an aqueous microenvironment. A particularly

dramatic example is the lipase AH (lipase from Burkholderia cepacia)-catalysed desym-

metrization of prochiral 1,4-dihydropyridine dicarboxylic esters, where either enantiomer

can be accessed in high enantioselectivity by using either water-saturated cyclohexane or

diisopropyl ether (DIPE) respectively (Scheme 1.60).191 The acyl group used in acylation

and deacylation can also have a dramatic effect on enantioselectivity.134

In conclusion, by using organic solvents, biotransformations can achieve productivities

suitable for pharmaceutical manufacture. Biocatalysis under organic solvent–aqueous con-

ditions can be applied to a broad range of enzymes as the methodology is compatible with

cofactor recycling, whereas biocatalysis in nearly anhydrous solvent facilitates numerous

transformations that are thermodynamically disfavoured in the presence of water, although

limited to use with enzymes that do not require cofactors, particularly hydrolases. In

selecting an appropriate solvent, it is necessary to screen each new biotransformation on a

case-by-case basis to ensure that optimum enzyme activity, stability and selectivity are

NH

OO

O OOO

NO2

OO

NH

OO

O OOH

NO2

O

NH

OO

OHOO

NO2

O

lipase AH,cyclohexane,water

lipase AH,DIPE,water

R-ent 87% yield, 89% ee

S-ent 88% yield, >99% ee

Scheme 1.60 Resolution of a prochiral 1,4-dihydropyridine dicarboxylic ester with lipase AHin the presence of cyclohexane or DIPE


achieved. For optimal activity under nearly anhydrous conditions, attention should also be

paid to water activity and the dehydrated enzyme formulation used. Water stripping is

particularly important to consider when setting up a continuous process.

1.5 Enzyme Immobilization

Ballesteros et al.192 defined immobilized biocatalysts as ‘enzymes, cells or organelles

(or combinations of these) which are in a state that permits their reuse’. Enzyme immo-

bilization represents only a small part of this field, but is the most commonly employed in

pharmaceutical production.

Immobilized enzymes are frequently used in biocatalysis to overcome limitations such as:

• insufficient stability towards temperature, pH, shear stress or autolysis;

• necessity to recycle the enzyme for economical reasons;

• biological contamination of the product causing complex downstream processing;

• emulsion formation during product extraction;

• poor catalyst dispersion in the reaction mixture;

• insufficient activity;

• inappropriate form if required for a continuous process.

Where immobilization is necessary, any resulting biocatalyst should be:

• toxicologically safe;

• low cost;

• sufficiently active and selective;

• chemically and thermally stable under process and storage conditions;

• insoluble towards the reaction solvent;

• mechanically strong;

• of uniform particle size;

• resistant to microbial attack;

• reusable.

Numerous different immobilization methods have been reported that take advantage of

various enzyme properties such as size, chemically reactive functionality, ionic groups or

hydrophobic domains.193 Based on these properties, enzyme immobilization can be split

into three main classes (which are also applicable to the immobilization of cell cultures):

• noncovalent attachment;

• covalent attachment and cross-linking;

• entrapment.

In spite of the immense quantity of available literature, it can still be a challenge to

determine which immobilization technique is suitable for a particular application, and so it

is usually necessary to test a number of options on a case-by-case basis.

1.5.1 Noncovalent Attachment

Noncovalent attachment is a popular method of immobilization, and numerous different

support materials have been employed, ranging from organic supports, like cellulose,

1.5 Enzyme Immobilization 61

chitin, ion-exchange resins and polyacrylamide, to inorganic supports, such as celite, salts,

zeolites or even iron particles. However, the technique is disfavoured for industrial

applications as the enzyme is weakly bound and, therefore, prone to leaching, potentially

leading to product contamination and inefficient recycling.

Many lipases are commercially available in a noncovalently immobilized form either

adsorbed onto celite, which aids dispersion in organic solvent, or onto a hydrophobic

support such as accurel. As a result, noncovalently immobilized lipases are frequently

employed, in spite of the above limitations, owing to their availability. Lipase immobiliza-

tion on hydrophobic supports is particularly useful, as it takes advantage of the unique

property of this enzyme class towards interfacial activation at the surface of oil droplets.194

Unlike other enzymes, most lipases contain what is often referred to as a lid or flap that

masks the active site. This lid is hydrophilic on the external surface and hydrophobic on the

internal surface, so that in aqueous solution the lipase exists in an equilibrium lying

primarily towards the inactive or closed form. On adsorption to an oil droplet, the flap

undergoes a conformational change to the ‘open form,’ resulting in activation. As dis-

cussed in Section 1.4, enzymes are more rigid in organic solvent and so the lipase can be

trapped in the form that was predominant in the aqueous solution from which it was last in

contact.195 On immobilization, the hydrophobic support itself can mimic an oil droplet,

resulting in hyperactivation of the lipase. It is not uncommon for an immobilized lipase to

display greatly enhanced activity over that of the free enzyme.

1.5.2 Covalent Attachment and Cross-linking

Immobilization of an enzyme through covalent attachment is a widely used technique, as the

catalyst can be used in either aqueous or organic media without leaching and provides a

suitable catalyst form for use in multipurpose apparatus or more specialized equipment such

as a continuous reactor. Covalent attachment is usually achieved via attack from nucleo-

philic groups of the enzyme onto electrophilic moieties on the support (although the reverse

has also been reported). Given that most enzymes have numerous reactive substituents

(Table 1.2), multipoint attachment to the support can occur, which can have a significant

stabilizing effect. A drawback of this technique can result from the formation of covalent

linkages in or near to the enzyme active site, causing deactivation. However, this outcome

can usually be circumvented by using another of the many alternative supports available.

Table 1.2 Reactive functionality of amino acid residues frequently present in proteins

Functional group Amino acid

Primary amine L-Lysine and N-terminusThiol L-CysteineCarboxylic acid L-Aspartate, L-glutamate and C-terminusPhenol L-TyrosineGuanidine L-ArgenineImidazole L-HistidineDisulfide L-CystineIndole L-TryptophanThioether L-MethionineAlcohol L-Serine, L-threonine


Eupergit C and, more recently, Sepabeads EC-EP are mesoporous supports that have

proven to be of particular importance in pharmaceutical production. Both are highly

hydrophilic macroporous resins, containing high densities of epoxide groups on the sur-

face. Available as beads of 100–250 mm in diameter and 20–40 nm pore diameter, these

resins display high chemical and mechanical stability, tolerating a wide range of pH and

solvents.

About 60 mg of purified enzyme per gram of resin can generally be immobilized onto

Sepabeads EC-EP, under extremely mild conditions, using enzyme dissolved in buffers of

high salt concentration. An initial rapid adsorption takes place followed by slower covalent

bond formation, after which the remaining epoxides (as much as 99 % of the original

groups) can be opened or ‘capped’ using a nucleophilic species. Crude enzyme prepara-

tions can also be used, as other cell debris will either irreversibly bind to the support along

with the enzyme or can be easily washed away after immobilization is complete. To

exemplify the mildness and robustness of this technique, 85–90 % of PGA active sites have

been reported to remain competent following immobilization to Eupergit C.196

Furthermore, the immobilized catalyst lost only 40 % of its activity over >800 cycles.

Covalent enzyme attachment to an inert support is inherently inefficient, as enzyme

activity is diluted and additional material costs are incurred. An attractive alternative that

circumvents both of these issues is to cross-link enzyme molecules together using a

bifunctional linker. This technique gained huge popularity with the emergence of cross-

linked enzyme crystals (CLECs).197 CLECs are produced by crystallization of purified

enzyme and subsequent cross-linking, usually with glutaraldehyde, which is an FDA-

approved fixing agent for the immobilization of glucose isomerase used in high-fructose

corn syrup production.198 CLECs proved to be excellent biocatalysts, displaying high

activity, stability and separation properties, as demonstrated by their use in the resolution

of the bisfuran alcohol intermediate of brecanavir (Scheme 1.33). Unfortunately, enzyme

purification and crystallization can be labour intensive to develop and inefficient, resulting

in an extremely active but highly expensive catalyst. This led to poor uptake of the

technology and withdrawal of CLECs from the marketplace.

More recently, CLEAs have been introduced. They provide many of the positive

attributes of CLECs but can be rapidly prepared from partially purified enzyme prepara-

tions with minimal technical expertise.199 Essentially, their preparation involves enzyme

precipitation (see Section 1.4.1.1) with in situ cross-linking, or vice versa. Glutaraldehyde

is usually employed as the cross-linking agent, although bulkier linkers, such as dextran

polyaldehyde, have been successfully used where cross-linking with the smaller reagent

results in activity loss through interaction with the enzyme active site.200

1.5.3 Entrapment

Entrapment involves the physical confinement of an enzyme in a semipermeable matrix,

in much the same manner as nature handles soluble enzymes.201 This should represent an

extremely mild method of immobilization, as the enzyme remains free, albeit confined to

a small space. Two techniques, which at first sight appear unrelated, have been well

utilized:

• entrapment in a polymer matrix;

• entrapment behind a membrane.

1.5 Enzyme Immobilization 63

Entrapment in polymeric matrices is a variation of noncovalent attachment where the

support is instead generated in the presence of the enzyme. A particularly popular entrap-

ment technique is sol–gel encapsulation, where the enzyme is trapped within an SiO2

matrix formed by acid- or base-catalysed hydrolysis of tetraalkoxysilanes in the presence

of enzyme.202 The technique can be tuned to provide the appropriate microenvironment for

each enzyme in much the same way as can be done with other immobilization methods.203

Pharmaceutical production generally uses multipurpose equipment, and so entrapment

behind a membrane would require significant capital expenditure on specialized equip-

ment. In spite of this, the use of membrane reactors in biocatalysis represents an efficient

method of enzyme immobilization, given the large molecular weight difference between

enzymes (10–150 kDa) and most substrates (300–500 Da). The reader is referred to some

recent reviews on the topic.204

In summary, enzyme immobilization is extremely important in the scale-up of many

biocatalytic processes. The preferred method for pharmaceutical production involves

covalent binding through cross-linking or attachment to a support. Noncovalent attach-

ment is less attractive, but it is heavily utilized owing to the commercial availability of

industrial quantities of some enzymes immobilized using this technique.

1.6 Green Chemistry

The use of biocatalysis in the manufacture of APIs can address some of the 12 principles of

green chemistry set out by Anastas and Warner.205 For example, biocatalytic processes can:

• increase atom efficiency;

• operate under mild conditions;

• reduce protection/deprotection steps;

• avoid the use of stoichiometric reagents;

• avoid the use of toxic/hazardous chemistry.

However, these statements are generalizations, and it is not necessarily true to say that

all biotransformations will be greener than the chemical alternative. Therefore, it is

important to analyse each comparison objectively on a case-by-case basis using a multi-

variate process to take into account the complexity of the analysis. Designing greener

processes involves, for example:

• designing efficient processes that minimize the resources (mass and energy) needed to

produce the desired product;

• considering the environmental, health and safety profile of the materials used in the

process;

• considering the environmental life cycle of the process;

• considering the economic viability of the process;

• considering the waste generated in the process, both in nature and quantity, whether it is

hazardous, benign, can be recycled or recovered and used in this or another process.

It is not easy or straightforward to determine how green a process is, and there have been

a number of different approaches taken. Sheldon’s E-factor was one of the first measures

of greenness proposed in the 1980s, to highlight the amount of waste generated in order to


produce 1 kg of chemical product across different branches of the chemical industry.206

Simply put, the higher the E number, the more waste is generated to produce 1 kg of

product. Within the pharmaceutical industry there have been other variations of measuring

the mass efficiency, such as the mass intensity proposed by Constable et al.207 and the

process mass intensity proposed by the ACS GCI Pharmaceutical Roundtable.208

Measuring greenness is not just about determining the quantity of waste; one must also

consider the efficiency of the chemistry or biochemistry (atom efficiency, reaction mass

efficiency) and the nature of the materials involved as reagents, solvents and as waste.209

One should also consider the process conditions used, all within the context of the 12

principles of green chemistry. The next factor to take into account when trying to evaluate

the greenness is the environmental life cycle impact of the materials used in the process.

Determining the life cycle for every material used in a pharmaceutical synthetic process is

a complex task, as often the life cycle data for every material is just not available.

However, GSK have developed a methodology and a tool to enable good estimations of

the life cycle impacts so that comparisons between different development options can be

made.210 Data have recently been added to the tool to enable life cycle comparisons for

routes using enzymes as catalysts or involving a fermentation step. GSK have also

developed a framework for analysing and comparing two processes based upon the suite

of metrics discussed above.211

This framework was used as the basis for a comparison of the environmental, health,

safety and life cycle (EHS and LCA) impacts of the chemical (Scheme 1.11) and two

enzyme biocatalytic (Scheme 1.12) 7-ACA processes, recently reported by Henderson

et al.55 The measures used accounted for the chemical and process efficiencies, the nature

of the materials used and waste generated, as well as determining the overall life cycle

environmental impacts from ‘cradle to gate’ of each process. This analysis showed that the

bioprocess could be classed as ‘greener’ when compared with the purely chemical process.

The chemical process uses more hazardous materials and solvents, and requires about 25 %

more process energy than the enzymatic process. When accounting for the cradle-to-gate

environmental life cycle, the chemical process has a larger overall environmental impact,

mainly derived from the production of raw materials. In comparison with the enzyme-

catalysed process, the chemical process uses approximately 60 % more energy, about 16 %

more mass (excluding water), has double the greenhouse gas impact and about 30 % higher

photochemical ozone creation potential and acidification impact. Only the yield of the

chemical process was higher, showing that yield is not a good measure of greenness, which

reinforces the message that it is important to take a more holistic view, since assessing

greenness is a multivariate and complex process. One of the aims of the analysis was to

develop a methodology and framework for objective comparisons of two very different

types of synthetic process, which could then be applied to other different systems.

A secondary aim was to test the hypothesis that biotransformations are greener than

chemical transformations. By the application of such rigorous and academic analyses

one can test this hypothesis for a number of different systems, including once-through

fermentations and enzyme-catalysed systems, where the amounts of waste generated will

be significantly different.

To celebrate the fifteenth anniversary of his E-factor, Sheldon compared different mea-

sures of greenness212 with the E-factor and reminds us of the value of the headline number,

which challenges those in the pharmaceutical industry to improve the efficiency of

1.6 Green Chemistry 65

pharmaceutical processes by moving away from continually using stoichiometric reagents

towards catalytic reagents. While it is true to say that the absolute volumes of waste are low

compared with fine chemicals or petrochemicals, the challenge remains valid today that the

pharmaceutical industry has the opportunity to embrace catalytic technology as one way to

improve the mass efficiency of processes. The application of biocatalytic technology in the

pharmaceutical industry is one way of addressing that challenge.

1.7 Future Perspectives

Biocatalysis contributes significantly to the generation of APIs through the supply of chiral

building blocks from the fine chemical industry. In contrast, there is a clear underutiliza-

tion within the pharmaceutical industry, where it could provide more efficient and greener

methods of late-stage intermediate and API production.

The ACS GCI Pharmaceutical Roundtable recently set out to prioritize the areas of

chemical synthesis where improved methodology would realize the greatest beneficial

impact on pharmaceutical production. This resulted in the publication of a ‘wish list’ of

currently utilized transformations that require better reagents and aspirational transforma-

tions that would provide shorter routes were they available (Table 1.3).16

A recent categorization of biotransformations by Pollard and Woodley12 (Figure 1.13),

based on the availability of commercial enzymes, together with the examples given in this

book demonstrate that biocatalysis can meet many of these pharmaceutical needs as shown

by the highlighted entries in Table 1.3.

Table 1.3 List of key areas of green chemistry of importance to the pharmaceutical industry (inascending order); areas where biocatalytic precedent exists are given in bold.

Reactions currently used but better reagentspreferred

More aspirational reactions

Amide formation avoiding poor atom economyreagents

C�H activation of aromatics (crosscoupling reactions avoiding thepreparation of haloaromatics)

OH activation for nucleophilic substitution Aldehyde or ketoneþNH3þ ‘X’ to givechiral amine

Reduction of amides without hydride reagents Asymmetric hydrogenation ofunfunctionalizedolefins/enamines/imines

Oxidation/epoxidation methods withoutthe use of chlorinated solvents

New greener fluorination methods

Safer and more environmentally friendlyMitsunobu reactions

N-Centred chemistry avoiding azides,hydrazine etc.

Friedel–Crafts reaction on unactivatedsystems

Asymmetric hydramination

Nitrations Green sources of electophilic nitrogen(not TsN3, nitroso, or diimide)Asymmetric hydrocyanation


Routes to APIs are predominantly designed by synthetic organic chemists who are well

versed in the adoption of new technologies. To maximize uptake of biocatalytic techni-

ques, the most efficient approach is to provide them with reasonably priced kits of enzymes

that can easily be used without specialist knowledge. Greater availability of comprehen-

sive commercial kits with diverse applications and better tools to predict improved

biocatalyst properties in silico should diminish the current perception by many chemists

that enzymes are exotic catalysts only to be used as a last resort. However, this expansion

requires significant investment from specialist enzyme producers, many of whom subse-

quently base their business models on the generation of royalties from the use of their

proprietary biocatalysts or biocatalytic processes. The use of proprietary enzymes in

pharmaceutical production can be cost effective where a biocatalyst is involved in an

asymmetric or regioselective transformation if traditional chemical approaches generate

substantial waste or require additional steps, but is probably precluded for achiral trans-

formations such as the replacement of an atom-inefficient coupling reagent for amide bond

R R′

O

R R′

NH2

O R′′

R′′′

O R′′

R′′′

R R′O

R R′

OH

OH

R R′O

R R′R R′

OH

RR′

O

R

O O

R

NH

R″R

R′

NR″

R

R′

R

O O

R′R″ R

OH O

R″R′

R

XH

R′ R″

O

OH R

X

R′

O

R″

R

O

R′ R

OH

R′

R R'

CN

R R′

CO2H

R

O

R′ R

OH

R′CN

R

O O

R′ R

OHR′

O

R = Alkyl or arylR′ = Alkyl, aryl or CO2R

Transaminase

Enoate reductase

Emerging Chemistries

Epoxide hydrolase

Monooxygenase

Monooxygenase

Monooxygenase

N-oxidase

+Aldolase

+

R, R′, R″ = Alkyl or arylX = O, N or S

Lipase/Protease

ADH

Established Chemistries

R, R′, R″ = Alkyl or aryl

Nitrilase

Hydroxynitrilase

Expanding Chemistries

+ HCN

+Decarboxylase

Figure 1.13 Status of various biotransformations (not exhaustive). (Reprinted from Pollard,D.J. and Woodley, J.M. Biocatalysis for pharmaceutical intermediates: the future is now. TrendsBiotechnol. 2007, 25, 66–73 with permission from Elsevier.)

1.7 Future Perspectives 67

formation. However, as the field matures and these enzymes become cheaper, such

applications should become competitive.

The above applications consider biocatalysis from the perspective of the synthetic

organic chemist rather than the biochemist. Slotting single-step biotransformations into

chemical syntheses is unlikely to use biocatalysis to its full potential. Undoubtedly,

isolated enzymes offer an attractive solution to rapid biocatalyst identification, and

advances in molecular biology and biotransformation technology have provided a number

of techniques by which hits can be modified to fit a required process, or vice versa.

However, there is also a significant cost associated with the isolation of enzymes at

scale. It is far more attractive to use crude lysates or whole cells; but, as shown in previous

sections, these have their own disadvantages. The use of crude lysates can increase

downstream processing complexity, and alleviation of this issue by immobilization adds

extra costs associated with production time and additional materials. Whole-cell biocata-

lysis can also require complex downstream processing and is generally hampered by low

substrate concentrations.

To realize the full potential of biocatalysis, a long-term approach might instead harness

nature’s tandem biocatalytic approach to the construction of complex secondary metabo-

lites for the production of synthetic molecules.213 Whilst product concentration is gen-

erally lower than that of a chemical process, this is offset by the ability to generate

molecules of high complexity in a single step and to eliminate costly isolation steps.

Fermentation scientists have been harnessing natural, highly selective biosynthetic path-

ways to produce complex pharmaceutical intermediates from cheap raw materials for

decades. Some of the most important pharmaceutical core molecules, such as penicillins

and cephalosporins, are economically produced in this way. The wide differences between

biosynthetic and chemical approaches to a target API can be gleaned by comparison of the

alternative routes that have been reported for the synthesis of orlistat ((�)-tetrahydrolip-

statin), a potent gastrointestinal lipase inhibitor used in the treatment of obesity

(Figure 1.14). Orlistat can be prepared by hydrogenation of the highly lipophilic secondary

metabolite lipstatin. Lipstatin itself is produced by fermentation (or, more correctly, a

tandem biotransformation) from linoleic acid via a key enzyme-catalysed Claisen con-

densation using Streptomyces toxytricini under aerobic conditions.214 In contrast, the

chemical approach to orlistat, based on the classical resolution or asymmetric synthesis

of a highly functionalized six-membered ring lactone,215,216 is considered to be one of the

most complex in the pharmaceutical industry, requiring four isolation steps and a number

of protection/deprotections.217

However, molecules currently produced by fermentation are usually natural products,

whereas most current drug candidates are synthetic. If lipstatin was not known to be a

OO

H23C11

OO

NHCHO

OO

NHCHO

OO

OrlistatLipstatin

Figure 1.14 Structures of lipstatin and orlistat


natural product, would a biosynthetic approach have been developed? Most likely, bioca-

talytic approaches would be limited to the insertion of individual transformations into the

current chemical route. In fact, lipase resolution of the six-membered lactone intermediate

produced from the chemical approach to orlistat has been reported.218 Metabolic engineer-

ing of unnatural biosynthetic pathways, by the insertion of non-native genes into a host

organism, offers great hope in this respect but is currently still in its infancy.214 The

production of thymidine represents the first example of its successful implementation in

pharmaceutical production (Scheme 1.23).

Thymidine, although a synthetic molecule, bears considerable resemblance to other

natural products, whereas many drug candidates have no counterpart in nature and will

likely require transformations for which there is no biocatalytic precedent. To build an

entirely artificial biosynthetic pathway using genetically modified organisms would

require a monumental screening effort, given that the vast majority of enzymes involved

in biosynthetic pathways have not yet been characterized and their specificities remain

unevaluated. Furthermore, should it be necessary to insert a chemocatalytic step into the

middle of a biosynthesis, transport across cell membranes would also require considera-

tion. An alternative approach might instead be to express the required enzymes together in

a genetically modified microorganism and use partially purified isolates, perhaps in

tandem with chemocatalysts. The one-pot synthesis of corrin, a biosynthetic intermediate

of vitamin B12, with 20 % unoptimized yield by 12 isolated enzymes demonstrates that

complex tandem processes are feasible using isolated enzymes (Scheme 1.61),219 and the

numerous chemoenzymatic processes available in the literature (some of which appear

later in the book) demonstrate that chemocatalysts can be efficiently inserted into bioca-

talytic processes.

1.8 Concluding Remarks

Biocatalysis has enjoyed widespread application in the preparation of chiral building

blocks but has generally been employed on a limited basis for the production of more

complex, late-stage pharmaceutical intermediates. Owing to pressure on the industry to

develop more efficient and greener processes, along with rapid advances in the field of

biocatalysis, this is beginning to change.220

O

NH2 CO2H

NNH

N NH

HO2C

CO2H

HO2C

HO2C

CO2H

CO2HHO2C

12 EnzymesVitamin B12

Corrin

Scheme 1.61 Tandem biocatalytic synthesis of corrin

1.8 Concluding Remarks 69

The recent commercialization of more diverse ranges of enzymes, combined with a

plethora of successful applications originating from both academia and the fine chemical

industry, is placing biocatalysis in the mainstream as an addition to the chemist’s toolbox

rather than an exotic curiosity. It is likely that, as the field matures, a greater diversity of

non-natural molecules of greater complexity will become accessible through the tandem

use of biocatalysts and genetically modified microorganisms. Together with advances in

chemocatalysis, this will significantly impact on pharmaceutical production by improving

efficiency and reducing waste.

Acknowledgements

I would like to thank John Gray and Shiping Xie for their help in proof reading this chapter.

References

1. PricewaterhouseCoopers, Pharma 2020: The vision, 2007. http://www.pwc.com/Extweb/onli-neforms.nsf/docid/DA9F87D21EE8BCAA852575A5005B3484?opendocument&doc¼vision.

2. McAndrews, P., Lilly sings a new tune: chorus unit brings high efficiency note to early R&D.The Pink Sheet, 2007, 69, 26.

3. Farina, V., Reeves, J.T., Senanayake, C.H. and Song, J.J., Asymmetric synthesis of activepharmaceutical ingredients. Chem. Rev., 2006, 106, 2734–2793.

4. Agrawal, Y.K., Bhatt, H.G., Raval, H.G., Oza, P.M. and Gogoi, P.J., Chirality – a new era oftherapeutics. Mini-Rev Med. Chem., 2007, 7, 451–460.

5. KPMG, Pressure points: risk management in the pharmaceuticals industry, 2006.6. US Food and Drug Administration, Pharmaceutical current good manufacturing practices

(cGMPs) for the 21st century – a risk-based approach: final report, 2004.7. McKenzie, P., Kiang, S., Tom, J., Rubin, A.E. and Futran, M., Can pharmaceutical process

development become high tech? Am. Inst. Chem. Eng., 2006, 52, 3990–3994.8. US Food and Drug Administration, Guidance for industry PAT – a framework for innovative

pharmaceutical development, manufacturing, and quality assurance, 2004.9. Rubin, A.E., Tummala, S., Both, D.A., Wang, C. and Delaney, E.J., Emerging technologies

supporting chemical process R&D and their increasing impact on productivity in the pharma-ceutical industry. Chem. Rev., 2006, 106, 2794–2810.

10. Straathof, A.J.J., Panke, S. and Schmid, A., The production of fine chemicals by biotransforma-tions. Curr. Opin. Biotechnol., 2002, 13, 548–556.

11. Liese, A., Seelbach K. and Wandrey C. (eds), Industrial Biotransformations. Wiley–VCH, 2006.12. Pollard, D.J. and Woodley, J.M., Biocatalysis for pharmaceutical intermediates: the future is

now. Trends Biotechnol., 2007, 25, 66–73.13. Panke, S., Held, M. and Wubbolts, M., Trends and innovations in industrial biocatalysis for the

production of fine chemicals. Curr. Opin. Biotechnol., 2004, 15, 272–279.14. Carey, J.S., Laffan, D., Thomson, C. and Williams, M.T., Analysis of the reactions used for the

preparation of drug candidate molecules. Org. Biomol. Chem., 2006, 4, 2337–2347.15. Breuer, M., Ditrich, K., Habicher, T., Hauer, B., Keßeler,M., Sturmer,R. and Zelinski, T.,

Industrial methods for the production of optically active intermediates. Angew. Chem. Int. Ed.,2004, 43, 788 – 824.

16. Constable, D.J.C., Dunn, P.J., Hayler, J.D., Humphrey, G.R., Leazer Jr, J.L., Linderman, R.J.,Lorenz, K., Manley, J., Pearlman, B.A., Wells, A., Zaks, A. and Zhang, T.Y., Key greenchemistry research areas – a perspective from pharmaceutical manufacturers. Green Chem.,2007, 9, 411–420.


17. Schoemaker, H.E., Mink, D., Wubbolt, M.G., Dispelling the myths – biocatalysis in industrialsynthesis. Science, 2003, 299, 1694–1697.

18. Vink, M.K.S. and Rozzell, J.D., Expanded opportunities for biocatalysis. PharmaChem, May2006, 17–18.

19. (a) Lefevre, F., Jarrin, C., Ginolhac, A., Auriol, D. and Nalin, R., Environmental metagenomics:an innovative resource for industrial biocatalysis. Biocatal. Biotrans., 2007, 25, 242–250. (b)Ferrer, M., Martınez-Abarca,F. and Golyshin, P.N., Mining genomes and ‘metagenomes’ fornovel catalysts. Curr. Opin. Biotechnol., 2005, 16, 588–593. (c) Cowan, D.A., Arslanoglu, A.,Burton, S.G., Baker, G.C., Cameron, R.A., Smith, J.J. and Meyer, Q., Metagenomics, genediscovery and the ideal biocatalyst. Biochem. Soc. Trans., 2004, 32, 298–302.

20. For some recent reviews on enzyme modification see: (a) Reetz, M.T., Evolution in the test-tubeas a means to creat selective biocatalysts. Chimia, 2007, 61, 100–103; (b) Sylvestre, J., Chautard,H., Cedrone, F. and Delcourt, M., Directed evolution of biocatalysts. Org. Proc. Res. Dev., 2006,10, 562–571; (c) Hibbert, E.G. and Dalby, P.A., Directed evolution strategies for improvedenzymatic performance. Microb. Cell Fact., 2004, 4, 29; (d) Otten, L.G. and Quax, W.J.,Directed evolution: selecting today’s biocatalysts. Biomol. Eng., 2005, 22, 1–9.

21. For some recent reviews on HTS, see: (a) Aharoni, A., Griffiths, A.D. and Tawfik, D.S., High-throughput screens and selections of enzyme-encoding genes. Curr. Opin. Chem. Biol., 2005, 9,210–216; (b) Reymond, J. and Babiak, P., Screening systems. Adv. Biochem. Eng Biotechnol.,2007, 105, 31–58.

22. Fox, R.J., Davis, S.C., Mundorff, E.C., Newman, L.M., Gavrilovic, V., Ma, S.K., Chung, L.M.,Ching, C., Tam, S., Muley, S., Grate, J., Gruber, J., Whitman, J.C., Sheldon, R.A. and Huisman,G.W., Improving catalytic function by ProSAR-driven enzyme evolution. Nat. Biotechnol.,2007, 25, 338–344.

23. Hilterhaus, L. and Liese, A., Building blocks. Adv. Biochem. Eng. Biotechnol., 2007, 105,133–173.

24. Grossman, S.J., Overview: drug metabolism in the modern pharmaceutical industry. In DrugMetabolism in Drug Design and Development, Zhang, D., Zhu, M. and Humphreys, W.G. (eds).J. Wiley & Sons, Inc., 2008, pp. 3–13.

25. Testa, B. and Kramer, S. D., The biochemistry of drug metabolism – an introduction part 1.Principles and overview. Chem. Biodiv., 2006, 3, 1053–1101 and references cited therein.

26. Azerad, R., Microbial models for drug metabolism. Adv. Biochem. Eng. Biotechnol., 1999, 63,169–218.

27. Smith, R.V. and Rosazza, J.P., Microbial models of mammalian metabolism. J. Pharm. Sci.,1975, 64, 1737–1759.

28. Alexandre, V., Ladril, S., Maurs, M. and Azerad, R., Microbial models of animal drug metabo-lism Part 5. Microbial preparation of human hydroxylated metabolites of irbesartan. J. Mol.Catal. B: Enzymatic, 2004, 29, 173–179.

29. For a recent review on microbial preparation of metabolites, see Venisetty, R.K. and Ciddi, V.,Application of microbial biotransformation for the new drug discovery using natural drugs assubstrates. Curr. Pharm. Biotechnol., 2003, 4, 153–167.

30. Testa, B. and Kramer, S.D., The biochemistry of drug metabolism – an introduction part 2.Redox reactions and their enzymes. Chem. Biodiv., 2007, 4, 257–404.

31. Smith, D.A., Ackland, M.J. and Jones, B.C., Properties of cytochrome P450 isoenzymes andtheir substrates. Part 2: properties of cytochrome P450 substrates. Drug Discov. Today, 1997, 2,479–486 and references cited therein.

32. (a) Hanlon, S.P., Friedburg, T., Wolf, C.R., Ghisalba, O. and Kittelmann, M., Recombinant yeastand bacteria that express human P450s: bioreactors for drug discovery, development, andbiotechnology. In Modern Biooxidation. Enzymes, Reactions and Applications. Schmid, R.D.and Urlacher, V.B. (eds). Wiley–VCH: Weinheim, 2007, pp. 233–252; (b) Crespi, C.L. andMiller, V.P., The use of heterologously expressed drug metabolizing enzymes – state of the artand prospects for the future. Pharmacol. Therap., 1999, 84, 121–131.

33. Schwaneberg, U., Otey, C., Cirino, P.C., Farinos, E. and Arnold, F.H., Cost-effective whole-cellassay for laboratory evolution of hydroxylases in Escherichia coli. J. Biomol. Screen., 2001, 6,111–117.

References 71

34. Ghisalba, O. and Kittelmann, M., Preparation of drug metabolites using fungal and bacterialstrains. In Modern Biooxidation, Schmid, R.D. and Urlacher, V. (eds). Wiley–VCH: Weinheim,2007, pp. 211–232.

35. This volume, Section 12.3, Kittelmann et al.36. Wong, J.W., Watson Jr, H.A., Bouressa, J.F., Burns, M.P., Cawley, J.J., Doro, A.E., Guzek, D.B.,

Hintz, M.A., McCormick, E.L., Scully, D.A., Siderewicz, J.M., Taylor, W.J., Truesdell, S.J. andWax, R.G. Biocatalytic oxidation of 2-methylquinoxaline to 2-quinoxalinecarboxylic acid. Org.Proc. Res. Dev., 2002, 6, 477–481.

37. Stachulski, A.V. and Jenkins, G.N., The synthesis of O-glucuronides. Nat. Prod. Rep., 1998, 15,173–186.

38. Chiu, S.H. and Huskey, S.W., Species differences in N-glucuronidation. Drug Metab. Dispos.,1998, 26, 838–847.

39. Kittelmann, M., Rheinegger, U., Espigat, A., Oberer, L., Aichholz, R., Francotte, E. andGhisalba, O., Preparative enzymatic synthesis of the acylglucuronide of mycophenolic acid.Adv. Synth. Catal., 2003, 345, 825–829.

40. Baratto, L., Candido, A., Marzorati, M., Sagui, F., Riva, S. and Danieli, B., Laccase-mediatedoxidation of natural glycosides. J. Mol. Catal. B: Enzymatic, 2006, 39, 3–8.

41. (a) Fabbrini, M., Galli, C., Gentili, P. and Macchitella, D., An oxidation of alcohols by oxygen withthe enzyme laccase and mediation by TEMPO. Tetrahedron Lett., 2001, 42, 7551–7553; (b) Niku-Paavola, M.-L. and Viikari, L., Enzymatic oxidation of alkenes. J. Mol. Catal. B: Enzymatic, 2000,10, 435–444; (c) Potthast, A., Rosenau, T., Chen,C.-L. and Gratzl, J.S., Selective enzymaticoxidation of aromatic methyl groups to aldehydes. J. Org. Chem. 1995, 60, 4320–4321.

42. Wong, C.-H. and Whitesides, G.M., Enzymes in synthetic organic chemistry. In TetrahedronOrganic Chemistry Series, Baldwin, J.E. and Magnus, P.D. (eds), Pergamon, 1994.

43. Newman, D.J., Cragg, G.M. and O’Keefe, B.R., Biopharmaceutical drugs from natural sources.In Modern Biopharmaceuticals, Knablein, J. (ed). Wiley–VCH: Weinheim, 2005, pp. 451–496.

44. Walsh, G. and Jefferis, R., Post-translational modifications in the context of therapeutic proteins.Nat. Biotechnol., 2006, 24, 1241–1252.

45. Meyer, A., Pellaux, R. and Panke, S. Bioengineering novel in vitro metabolic pathways usingsynthetic biology. Curr. Opin. Microbiol., 2007, 10, 246–253.

46. (a) Koeller, K.M. and Wong, C.-H., Synthesis of complex carbohydrates and glycoconjugates:enzyme-based and programmable one-pot strategies. Chem. Rev. 2000, 100, 4465–4493;(b) Khmelnitsky, Y.L., Current strategies for in vitro protein glycosylation. J. Mol. Catal. B,2004, 31, 73–81.

47. Ohrlein, R., Glycosyltransferase-catalysed synthesis of non-natural oligosaccharides. TopicsCurr. Chem., 1999, 200, 227–254.

48. Rowan, A.S. and Hamilton, C.J., Recent developments in preparative enzymatic syntheses ofcarbohydrates. Nat. Prod. Rep., 2006, 23, 412–443 and references cited therein.

49. Ullman, C.G. and Perkins, S.J., A classification of nucleotide diphospho-sugar glycosyltrans-ferases based on amino acid sequence similarities. Biochem. J., 1997, 326, 929–942.

50. Shao, H., He, X., Achnine, L., Blount, J.W., Dixon, R.A. and Wang, X., Crystal structures of amultifunctional triterpene/flavonoid glycosyltransferase from Medicago truncatula. The PlantCell, 2005, 17, 3141–3154.

51. Lim, E.-K., Ashford, D.A., Hou, B., Jackson, R.G. and Bowles, D.J., Biotechnol. Bioeng., 2004,87, 623–631.

52. Perusino, G., Cobucci-Ponzano, B., Rossi, M. and Moracci, M., Recent advances in the oligo-saccharide synthesis promoted by catalytically engineered glycosidases. Adv. Synth. Catal.,2005, 347, 941–950.

53. Mackenzie, L.F., Wang, Q., Warren, R.A.J. and Withers, S.G., Glycosynthases: mutant glyco-sidases for oligosaccharide synthesis. J. Am. Chem. Soc., 1998, 120, 5583–5584.

54. Spence, D.W. and Ramsden, M., Penicillin acylases. In Industrial Enzymes, Polaina, J. andMacCabe, A.P. (eds). Springer: Dordrecht, 2007, pp. 583–597.

55. Henderson R.K., Jimenez-Gonzalez C., Preston C., Constable D.J.C. and Woodley, J.M., EHS &LCA assessment for 7-ACA synthesis. A case study for comparing biocatalytic & chemicalsynthesis. Ind. Biotechnol., 2008, 4, 180–192.


56. Fildes, R.A., Potts, J.R. and Farthing, J.E., Process for preparing cephalosporin derivatives. PCTAppl., 1974, US 3,801,458.

57. Matsuda, A., Matsuyama, K., Yamamoto, K., Ichikawa, S. and Komatsu, K.-I., Cloning andcharacterization of the genes for two distinct cephalosporin acylases from a Pseudomonas strain.J. Bacteriol., 1987, 169, 5815–5820.

58. Matsumoto, K., Production of 6-APA, 7-ACA and 7-ADCA by immobilised penicillin andcephalosporin amidases. Bioproc. Technol. 1993, 16, 67–88.

59. Brakhage, A.A., Molecular regulation of b-lactam biosynthesis in filamentous fungi. Microbiol.Mol. Biol. Rev., 1998, 62, 547–585.

60. Conder, M.J., Crawford, L., McAda, P.C. and Rambosek, J.A., Novel bioprocess for preparing7-ADCA. PCT Appl., 1992, EP 0532341 A1.

61. Sheldon, R.A. and van Rantwijk, F., Biocatalysis for sustainable organic synthesis. Aust. J.Chem., 2004, 57, 281–289.

62. Fernandez-Lafuente,R., Rosell, C.M., Piatkowska, B. and Guisan, J.M., Synthesis of antibiotics(cephaloglycin) catalyzed by penicillin G acylase: evaluation and optimization of differentsynthetic approaches. Enzyme Microb. Technol., 1996, 19, 9–14.

63. Kallenberg, A.I., van Rantwijk, F. and Sheldon, R.A., Immobilization of penicillin G acylase: thekey to optimum performance. Adv. Synth. Catal., 2005, 347, 905–926.

64. Schroen, C.G.P.H., Nierstrasz, V.A., Bosma, R., Kroon, P.J., Tjeerdsma, P.S., DeVroom, E.,VanderLaan, J.M., Moody, H.M., Beeftink, H.H., Janssen, A.E.M. and Tramper, J., Integratedreactor concepts for the enzymatic kinetic synthesis of cephalexin. Biotech. Bioeng., 2002, 80,144–155.

65. Kim, I., Song, X., Vig, B.S., Mittal, S., Shin, H.-C., Lorenzi, P.J. and Amidon, G.L., A novelnucleoside prodrug-activating enzyme: substrate specificity of biphenyl hydrolase-like protein.Mol. Pharm., 2004, 1, 117–127.

66. Tamarez, M., Morgan, B., Wong, G.S.K., Tong, W., Bennett, F., Lovey, R., McCormick, J.L. andZaks, A., Pilot-scale lipase-catalyzed regioselective acylation of ribavirin in anhydrous media inthe synthesis of a novel prodrug intermediate. Org. Proc. Res. Dev., 2003, 7, 951–953.

67. (a) Zaks, A. and Klibanov, A.M., Enzymatic catalysis in nonaqueous solvent. J. Biol. Chem.,1988, 263, 3194–3201; (b) Klibanov, A.M., Asymmetric enzymatic oxidoreductions in organicsolvents. Curr. Opin. Biotechnol., 2003, 14, 427–431.

68. Ferrero, M. and Gotor, V., Biocatalytic selective modifications of conventional nucleosides,carbocyclic nucleosides, and C-nucleosides. Chem. Rev., 2000, 100, 4319–4347 and referencescited therein.

69. Tamarez, M., Morgan, B., Wong, G.S.K., Tong, W., Bennett, F., Lovey, R., McCormick, J.L.and Zaks, A., Pilot-scale lipase-catalyzed regioselective acylation of ribavirin in anhydrousmedia in the synthesis of a novel prodrug intermediate. Org. Proc. Res. Dev., 2003, 7,951–953.

70. Mahmoudian, M., Eaddy, J. and Dawson, M., Enzymic acylation of 506U78 (2-amino-9-�-D-arabinofuranosyl-6-methoxy-9H-purine), a powerful new anti-leukaemic agent. Biotechnol.Appl. Biochem., 1999, 29, 229–233.

71. Hanson, R.L., Shi, Z., Brzozowski, D.B., Banerjee, A., Kissick, T.P., Singh, J., Pullockaran, A.J.,North, J.T., Fan, J., Howell, J., Durand, S.C., Montana, M.A., Kronenthal, D.R., Mueller, R.H.and Patel, R.N., Regioselective enzymatic aminoacylation of lobucavir to give an intermediatefor lobucavir prodrug. Bioorg. Med. Chem., 2000, 8, 2681–2687.

72. Manzoni, M. and Rollini, M., Biosynthesis and biotechnological production of statins byfilamentous fungi and application of these cholesterol-lowering drugs. Appl. Microbiol.Biotechnol., 2002, 58, 555–564.

73. Sleteinger, M., Verhoeven, T.R. and Volante, R.P., Process for C-methylation of 2-methylbuty-rates. US PCT Appl., 1986, US 4,582,915.

74. Askin, D., Verhoeven, T.R., Liu, T.M.-H. and Shinkai, I., Synthesis of synvinolin: extremelyhigh conversion alkylation of an ester enolate. J. Org. Chem., 1991, 56, 4929–4932.

75. Schimmel, T.G., Borneman,W.S. and Conder, M.J., Purification and characterization of alovastatin esterase from Clonostachys compactiuscula. Appl. Environ. Microbiol., 1997, 63,1307–1311.

References 73

76. Xie, X. and Tang, Y., Efficient synthesis of simvastatin by use of whole-cell biocatalysis. Appl.Environ. Microbiol., 2007, 73, 2054–2060.

77. Xie, X., Wong, W.W. and Tang, Y., Improving simvastatin bioconversion in Escherichia coli bydeletion of bioH. Metabol. Eng., 2007, 9, 379–386.

78. Kern, A., Tilley, E., Hunter, I.S., Legısa, M. and Glieder, A., Engineering primary metabolicpathways of industrial micro-organisms. J. Biotechnol., 2007, 129, 6–29.

79. Sørensen, H.P. and Mortensen, K.K. Advanced genetic strategies for recombinant proteinexpression in Escherichia coli. J. Biotechnol., 2005, 115, 113–128.

80. McCandliss, R.J. and Anderson, D.M., Fermentation process for the production of pyrimidinedeoxyribonucleosides. PCT Int. Appl., 1991, WO91/09130.

81. Anderson, D. M., Liu, L., Podkovyrov, S. and Wang, B., PCT Int. Appl., 2001,WO2001002580 A1.

82. Anderson, D.M., Collis, A.J., Liu, L., Podkovyrov, S. and Preston, C., Thymidine production invarious strains/constructs. PCT Int. Appl., 2007, WO2007/090810 A1.

83. Roth, B.D., The discovery and development of atorvastatin, a potent novel hypolipidemic agent.Prog. Med. Chem., 2002, 40, 1–22 and references cited therein.

84. Muller, M., Chemoenzymatic synthesis of building blocks for statin side chains. Angew. Chem.Int. Ed., 2005, 44, 362–365.

85. Fox, R.J., Davis, S.C., Mundorff, E.C., Newman, L.M., Gavrilovic, V., Ma, S.K., Chung, L.M.,Ching, C., Tam, S., Muley, S., Grate, J., Gruber, J., Whitman, J.C., Sheldon, R.A. and Huisman,G.W., Improving catalytic function by ProSAR-driven enzyme evolution. Nat. Biotechnol.,2007, 25, 338–344.

86. Lewkowicz, E.S. and Iribarren, A.M., Nucleoside phosphorylases. Curr. Org. Chem., 2006, 10,1197–1215; Utagawa, T., Enzymatic preparation of nucleoside antibiotics. J. Mol. Catal. B,1999, 6, 215–222.

87. Shirae, H., Yokozeki, K., Uchiyama, M. and Kubota, K., Enzymatic production of ribavirinfrom purine nucleosides by Brevibacterium acetylicum. Agric. Biol. Chem., 1988, 52, 1777–1783.

88. Krenitsky, T.A., Koszalka, G.W. and Tuttle, J.V., Purine nucleoside synthesis, an efficientmethod employing nucleoside phosphorylases. Biochemistry, 1981, 20, 3615–3621.

89. Krenitsky, T.A., Elion, G.B. and Rideout, J.L., Enzymatic synthesis of arabinonucleosides, Eur.PCT Appl., 1979, EP 0002192; Krenitsky, T.A., Koszalka, G.W., Tuttle, J.V., Rideout, J.L. andElion, G.B., An enzymatic synthesis of purine D-arabinonucleosides. Carbohydrate Res., 1981,97, 139–146.

90. Komura, H., Yoshino, T. and Ishido, Y., Synthetic studies by the use of carbonates. II. Easymethod of preparing cyclic carbonates of polyhydroxy compounds by transesterification withethylene carbonate. Bull. Chem. Soc., 1973, 46, 550–553.

91. Averett, D.R., Koszalka, G.W., Fyfe, J.A., Roberts, G.B., Purifoy, D.J.M. and Krenitsky, T.A.,6-Methoxypurine arabinoside as a selective and potent inhibitor of varicella-zoster virus.Antimicrob. Agents Chemother. 1991, 35, 851–857.

92. Mahmoudian, M., Development of bioprocesses for the generation of anti-inflammatory, anti-viral and anti-leukaemic agents. Focus Biotechnol. 2001, 1, 249–265.

93. Utagawa, T., Morisawa, H., Yoshinaga, F., Yamazaki, A., Misugi, K. and Hirose, Y.,Microbiological synthesis of adenine arabinoside. Agric. Biol. Chem., 1985, 49, 1053–1058.

94. Herbal, K., Kitteringham, J., Voyle, M. and Whitehead, A.J., Synthesis of the enantiomer ofnelarabine. Tetrahedron Lett., 2005, 46, 2961–2964.

95. Warren, L. and Felsenfeld, H., Biosynthesis of N-acetylneuraminic acid. Biochem. Biophys.Res. Commun., 1961, 4, 232–235.

96. Tabata, K., Koizumi, S., Endo, T. and Ozaki, A., Production of N-acetyl-D-neuraminic acid bycoupling bacteria expressing N-acetyl-D-glucosamine 2-epimerase and N-acetyl-D-neuraminicacid synthetase. Enzyme Microb. Technol. 2002, 30, 327–333.

97. Samain, E., High yield production of sialic acid (Neu5Ac) by fermentation. PCT Int. Appl.,2008, WO 2008040717 A2; Lundgren, B.R. and Boddy, C.N., Sialic acid and N-acyl sialic acidanalog production by fermentation of metabolically and genetically engineered Escherichiacoli. Org. Biomol. Chem., 2007, 5, 1903–1909.


98. Comb, D.G. and Roseman, S., Composition and enzymic synthesis of N-acetylneuraminic acid(sialic acid). J. Am. Chem. Soc., 1958, 80, 497–499.

99. Cipolletti, G., Tamerlani, G., Lombardi, I. and Bartalucci, D., Process for the preparation ofN-acetyl-D-mannosamine monohydrate. PCT Int. Appl., 2007, WO 2007135086 A1.

100. Mahmoudian, M., Noble, D., Drake, C.S., Middleton, R.F., Montgomery, D.S., Piercey, J.E.,Ramlakhan, D., Todd, M. and Dawson, M.J., An efficient process for production ofN-acetylneuraminic acid using N-acetylneuraminic acid aldolase. Enzyme Microb. Technol.1997, 20, 393–400.

101. Kragl, U., Wandrey, C., Ghisalba, O. and Gygax, D., Enzymic preparation of N-acetyl neur-aminic acid. Ger. Offen. 1991, DE 3937891 A1; Maru, I., Ohnishi, J., Ohta, Y. and Tsukada, Y.,Why is sialic acid attracting interest now? Complete enzymatic synthesis of sialic acid withN-acylglucosamine 2-epimerase. J. Biosci. Bioeng., 2002, 93, 258–265; Maru, I., Ohnishi, J.,Ohta, Y. and Tsukada, Y., Simple and large-scale production of N-acetylneuraminic acid fromN-acetyl-D-glucosamine and pyruvate using N-acetyl-D-glucosamine 2-epimerase and N-acet-ylneuraminate lyase. Carbohydrate Res., 1998, 306, 575–578.

102. Kragl, U., Kittelmann, M., Ghisalba, O. and Wandrey, C., N-Acetylneuramic acid: from a rarechemical from natural sources to a multikilogram enzymatic synthesis for industrial applica-tion. Ann. N. Y. Acad. Sci., 1995, 750, 300–305.

103. Garcıa-Urdiales, E., Alfonso, I. and Gotor, V., Enantioselective enzymatic desymmetrizationsin organic synthesis. Chem. Rev., 2005, 105, 313–354.

104. (a) Hasan, F., Shah, A.A. and Hameed, A. Industrial applications of microbial lipases. EnzymeMicrob. Technol., 2006, 39, 235–251; (b) Maurer, K.-H., Detergent proteases. Curr. Opin.Biotechnol., 2004, 15, 330–334; (c) Gupta, R., Beg, Q.K. and Lorenz, P., Bacterial alkalineproteases: molecular approaches and industrial applications. Appl. Microbiol. Biotechnol.,2002, 59, 15–32.

105. Martinez, C.A., Hu, S., Dumond, Y., Tao, J., Kelleher, P. and Tully, L., Development of achemoenzymatic manufacturing process for pregabalin. Org. Proc. Res. Dev., 2008, 12, 392–398.

106. Canoy, W.L., Cooley, B.E., Corona, J.A., Lovelace, T.C., Millar, A., Weber, A.M., Xie, S. andZhang, Y., Efficient synthesis of (3R,3aS,6aR)-hexahydrofuro[2,3-b]furan-3-ol from glycolal-dehyde. Org. Lett., 2008, 10, 1103–1106 and references cited therein.

107. Yu, R.H., Polniaszek, R.P., Becker, M.W., Cook, C.M. and Yu, L.H.L., Research and devel-opment of an efficient synthesis of hexahydrofuro[2,3-b]furan-3-ol moiety – a key componentof the HIV protease inhibitor candidates. Org. Proc. Res. Dev., 2007, 11, 972–980.

108. Doan, B.D., Davis, R.D. and Lovelace, T.C., Process for preparing intermediates for HIVaspartyl protease inhibitors, particularly (3R,3aS,6aR)-hexahydrofuro[2,3-b]furan-3-ol and its(3R,3aS,6aR)-enantiomer. PCT Int. Appl., 2003, WO 2003024974 A2.

109. Roberts, J., A commercially viable route and process to HIV PI GW640385X. Abstracts ofPapers, 226th ACS National Meeting, New York, NY, United States, 2003, September 7–11.

110. Vaidyanathan, R., Hesmondhalgh, L. and Hu, S., A chemoenzymatic synthesis of an androgenreceptor antagonist. Org. Proc. Res. Dev., 2007, 11, 903–906.

111. Terao, Y., Tsuji, K., Murata, M., Achiwa, K., Nishio, T., Watanabe, N. and Seto, K., Facileprocess for enzymatic resolution of racemic alcohols. Chem. Pharm. Bull., 1989, 37, 1653–1655.

112. Andrews, I.P., Atkins, R.J., Breen, G.F., Carey, J.S., Forth, M.A., Morgan, D.O., Shamji, A.,Share, A.C., Smith, S.A.C., Walsgrove, T.C. and Wells, A.S., The development of a manufac-turing route for the GPIIb/IIIa receptor antagonist SB-214857-A. Part 1: synthesis of the keyintermediate 2,3,4,5-tetrahydro-4-methyl-3-oxo-1H-1,4-benzodiazepine-2-acetic acid methylester, SB-235349. Org. Proc. Res. Dev., 2003, 7, 655–662; Atkins, R.J., Banks, A., Bellingham,R.K., Breen, G.F., Carey, J.S., Etridge, S.K., Hayes, J.F., Hussain, N., Morgan, D.O., Oxley, P.,Passey, S.C., Walsgrove, T.C. and Wells, A.S., The development of a manufacturing route forthe GPIIb/IIIa receptor antagonist SB-214857-A. Part 2: conversion of the key intermediateSB-235349 to SB-214857-A. Org. Proc. Res. Dev., 2003, 7, 663–675.

113. Roberts, N.J., Seago, A., Carey, J.S., Freer, R., Preston, C. and Lye, G.J., Lipase catalysedresolution of the lotrafiban intermediate 2,3,4,5-tetrahydro-4-methyl-3-oxo-1H-1,4-benzodia-zepine-2-acetic acid methyl ester in ionic liquids: comparison to the industrial t-butanolprocess. Green Chem., 2004, 6, 475–482.

References 75

114. Butters, M., Catterick, D., Craig, A., Curzons, A., Dale, D., Gillmore, A., Green, S.P.,Marziano, I., Sherlock, J.-P. and White, W., Critical assessment of pharmaceutical processess –a rationale for changing the synthetic route. Chem. Rev., 2006, 106, 3002–3027.

115. Casu, F., Chiacchio, M.A., Romeo, R. and Gumina, G., Chiral synthesis of heterosubstitutednucleoside analogs from non-carbohydrate precursors. Curr. Org. Chem., 2007, 11, 1017–1032.

116. Mahmoudian, M., Baines, B.S., Drake, C.S., Hale, R.S., Jones, P., Piercey, J.E., Montgomery,D.S., Purvis, I.J., Storer, R., Dawson, M.J. and Lawrence, G.C., Enzymatic production ofoptically pure (20R-cis)-20-deoxy-30-thiacytidine (3TC, lamivudine): a potent anti-HIV agent.Enzyme Microb. Technol., 1993, 15, 749–755.

117. Munch-Petersen, A., Nygaard, P., Hammer-Jespersen, K. and Fiil, N., Mutants constitutivefor nucleoside-catabolizing enzymes in Escherichia coli K 12. Eur. J. Biochem., 1972, 27,208–215.

118. Mahmoudian, M. and Dawson, M.J., Chemoenzymic production of the antiviral agentEpivirTM. In Biotechnology of Antibiotics, Strohl, W.R. (ed.). Dekker: New York, 1997, pp.753–777.

119. Solares, L.F., Brieva, R., Quiros, M., Llorente, I., Bayod, M. and Gotor, V., Enzymaticresolution of a quaternary stereogenic centre as the key step in the synthesis of (S)-(þ)-citalopram. Tetrahedron Asymm., 2004, 15, 341–345.

120. Mateja Pogorevc, M. and Faber, K., Biocatalytic resolution of sterically hindered alcohols,carboxylic acids and esters containing fully substituted chiral centers by hydrolytic enzymes. J.Mol. Catal. B, 2000, 10, 357–376 and references cited therein.

121. Taylor, S.J.C., Sutherland, A.G., Lee, C., Wisdom, R., Thomas, S., Roberts, S.M. and Evans,C., Chemoenzymatic synthesis of (�)-carbovir utilizing a whole cell catalysed resolution of2-azabicyclo[2.2.1 ]hept-5-en-3-one. J. Chem. Soc. Chem. Commun., 1990, 1120–1121; Evans,C.T., Roberts, S.M., Shoberu, K.A. and Sutherland, A.G., Potential use of carbocyclic nucleo-sides for the treatment of AIDS: chemo-enzymatic syntheses of the enantiomers of carbovir. J.Chem. Soc. Perkin Trans. 1, 1992, 589–592.

122. Taylor, S.J.C, McCague, R., Wisdom, R., Lee, C., Dickson, K., Ruecroft, G., O’Brien, F.,Littlechild, J., Bevan, J., Roberts, S.M. and Evans, C.T., Development of the biocatalyticresolution of 2-azabicyclo [2,2,1] hept-5-en-3-one as an entry to single-enantiomer carbocyclicnucleosides. Tetrahedron Asymm., 1993, 4, 1117–1128.

123. Mahmoudian, M., Lowdon, A., Jones, M., Dawson, M. and Wallis, C., A practical enzymaticprocedure for the resolution of N-substituted 2-azabicyclo[2.2.1]hept-5-en-3-one. TetrahedronAsymm. 1999, 10, 1201–1206.

124. Carnell, A.J., Desymmetrisation of prochiral ketones using lipases. J. Mol. Catal. B, 2002, 19–20, 83–92.

125. Allan, G., Carnell, A.J., Escudero Hernandez, M.L. and Pettman, A., Chemoenzymatic synth-esis of a tachykinin NK-2 antagonist. Tetrahedron, 2001, 57, 8193–8202.

126. Carnell, A.J., Barkely, J. and Singh, A., Desymmetrisation of prochiral ketones by catalyticenantioselective hydrolysis of their enol esters using enzymes. Tetrahedron Lett., 1997, 38,7781–7784; Allan, G.R., Carnell, A.J. and Kroutil, W., One-pot deracemisation of an enolacetate derived from a prochiral cyclohexanone. Tetrahedron Lett., 2001, 42, 5959–5962.

127. Pellissier, H., Recent developments in dynamic kinetic resolution. Tetrahedron, 2008, 64,1563–1601; Turner, N.J., Enzyme catalysed deracemisation and dynamic kinetic resolutionreactions. Curr. Opin. Chem. Biol., 2004, 8, 114–119; Gruber, C.C., Lavandera, I., Faber, K.and Kroutil, W., From a racemate to a single enantiomer: deracemisation by stereoinversion.Adv. Synth. Catal., 2006, 348, 1789–1805; Pellissier, H., Dynamic kinetic resolution.Tetrahedron, 2003, 59, 8291–8327; Pamies, O. and Backvall, J.-E., Combination of enzymesand metal catalysts. A powerful approach in asymmetric catalysis. Chem. Rev., 2003, 103,3247–3261.

128. Hughes, G., O’Shea, P.D., Devine, P.N., Foster, B., Gauthier, D., Limanto, J., Truppo, M.,Pollard, D., Naber, J., McKay, D.J. and Volante, R.P., The discovery and development ofodanacatib: a selective inhibitor of cathepsin K for the treatment of osteoporosis. In 25th SCIProcess Development Symposium, 2007.


129. Gu, R.-L., Lee, I.S. and Sih, C.J., Chemo-enzymatic asymmetric synthesis of amino acids.Enantioselective hydrolyses of 2-phenyl-oxazolin-5-ones. Tetrahedron Lett., 1992, 33, 1953–1956; Crich, J., Brieva, R., Marquart, P., Gu, R.-L., Flemming, S. and Sih, C.J., Enzymicasymmetric synthesis of �-amino acids. Enantioselective cleavage of 4-substituted oxazolin-5-ones and thiazolin-5-ones. J. Org. Chem., 1993, 58, 3252–3258.

130. Limanto, J., Shaifee, A., Devine, P.N., Upadhyay, V., Desmond, R.A., Foster, B.R., Gauthier Jr,D.R., Reamer, R.A. and Volante, R.P., An efficient chemoenzymatic approach to (S)-�-fluoroleucine ethyl ester. J. Org. Chem., 2005, 70, 2372–2375.

131. Madsen, R., Roberts, C. and Fraser-Reid, B., The pent-4-enoyl group: a novel amine-protectinggroup that is readily cleaved under mild conditions. J. Org. Chem., 1995, 60, 7920–7926.

132. Truppo, M.D. and Moore, J.C., Process for making fluoroleucine ethyl esters. US PCT Appl.,2007, US 2007/0059812 A1.

133. Pesti, J.A., Yin, J., Zhang, L.-H. and Anzalone, L., Reversible Michael reaction–enzymatichydrolysis: a new variant of dynamic resolution. J. Am. Chem. Soc., 2001, 123, 11075–11076;Pesti, J.A., Yin, J., Zhang, L.-H., Anzalone, L., Waltermire, R.E., Ma, P., Gorko, E., Confalone,P.N., Fortunak, J., Silverman, C., Blackwell, J., Chung, J.C., Hrytsak, M.D., Cooke, M., Powell,L. and Ray, C., Efficient preparation of a key intermediate in the synthesis of roxifiban byenzymatic dynamic kinetic resolution on large scale. Org. Proc. Res. Dev., 2004, 8, 22–27.

134. Miyazawa, T., Kurita, S., Ueji, S., Yamada, T. and Shigeru, K., Resolution of mandelic acids bylipase-catalyzed transesterifications in organic media: inversion of enantioselectivity mediatedby the acyl donor. J. Chem. Soc. Perkin Trans. 1, 1992, 18, 2253–2255.

135. Groger, H., Enzymatic routes to enantiomerically pure aromatic �-hydroxy carboxylic acids: afurther example for the diversity of biocatalysis. Adv. Synth. Catal., 2001, 343, 547–558.

136. Huang, H.-R., Xu, J.-H., Xu, Y., Pan, J. and Liu, X., Preparation of (S)-mandelic acids byenantioselective degradation of racemates with a new isolate Pseudomonas putida ECU1009.Tetrahedron Asymm., 2005, 16, 2113–2117.

137. DeSantis, G., Zhu, Z., Greenberg, W.A., Wong, K., Chaplin, J., Hanson, S.R., Farwell, B.,Nicholson, L.W., Rand, C.L., Weiner, D.P., Robertson, D.E. and Burk, M.J., An enzyme libraryapproach to biocatalysis: development of nitrilases for enantioselective production of car-boxylic acid derivatives. J. Am. Chem. Soc., 2002, 124, 9024–9025.

138. Saksena, A.K., Girijavallabhan, V.M., Lovey, R.G., Pike, R.E., Desai, J.A., Ganguly, A.K.,Hare, R.S., Loebenberg, D., Cacciapuoti, A. and Parmegiani, R.M., Enantioselective synthesisof the optical isomers of broad-spectrum orally active antifungal azoles, SCH 42538 and SCH45012. Biorg. Med. Chem. Lett., 1994, 4, 2023–2028.

139. Morgan, B., Dodds, D.R., Zaks, A., Andrews, D.R. and Klesse, R., Enzymatic desymmetrisa-tion of prochiral 2-substituted-1,3-propanediols: a practical chemoenzymatic synthesis of a keyprecursor of SCH51048, a broad-spectrum orally active antifungal agent. J. Org. Chem., 1997,62, 7736–7743.

140. Schmid, R.D. and Verger, R., Lipases: interfacial enzymes with attractive applications. Angew.Chem. Int. Ed., 1998, 37, 1608–1633 and references cited therein.

141. Homann M.J., Suen W.-C., Zhang, N. and Zaks, A., Comparative analysis of chemical andbiocatalytic syntheses of drug intermediates. In Biocatalysis in the Pharmaceutical andBiotechnology Industries, Patel, R.N. (ed.), CRC Press, 2007, pp. 645–659 and referencescited therein.

142. Maddrell, S.J., Turner, N.J., Kerridge, A., Willetts, A.J. and Crosby, J., Nitrile hydrataseenzymes in organic synthesis: enantioselective synthesis of the lactone moiety of the mevinicacids. Tetrahedron Lett., 1996, 37, 6001–6004.

143. Bergeron, S., Chaplin, D.A., Edwards, J.H., Ellis, B.S.W., Hill, C.L., Holt-Tiffin, K., Knight,J.R., Mahoney, T., Osborne, A.P. and Ruecroft, G., Nitrilase-catalysed desymmetrisation of 3-hydroxyglutaronitrile: preparation of a statin side-chain intermediate. Org. Proc. Res. Dev.,2006, 10, 661–665.

144. DeSantis, G., Wong, K., Farwell, B., Chatman, K., Zhu, Z., Tomlinson, G., Huang, H., Tan, X.,Bibbs, L., Chen, P., Kretz, K. and Burk, M.J., Creation of a productive, highly enantioselectivenitrilase through gene site saturation mutagenesis (GSSM). J. Am. Chem. Soc., 2003, 125,11476–11477.

References 77

145. Homann, M.J., Vail, R.B., Previte, E., Tamarez, M., Morgan, B., Dodds, D.R. and Zaks, A.,Rapid identification of enantioselective ketone reductions using targeted microbial libraries.Tetrahedron, 2004, 60, 789–797.

146. Barbieri, C., Caruso, E., D’Arrigo, P., Fantoni, G.P. and Servi, S., Chemo-enzymatic synthesisof (R)- and (S)-3,4-dichlorophenylbutanolide intermediate in the synthesis of sertraline.Tetrahedron Asymm. 1999, 10, 3931–3937.

147. Leonida, M.D., Redox enzymes used in chiral syntheses coupled to coenzyme regeneration.Curr. Med. Chem., 2001, 8, 345–369.

148. Eckstein, M., Daubmann, T. and Kragl, U., Recent developments in NAD(P)H regeneration forenzymatic reductions in one- and two-phase systems. Biocatal. Biotrans., 2004, 22, 89–96.

149. Davis, S.C., Grate, J.H., Gray, D.R., Gruber, J.M., Huisman, G.W., Ma, S.K., Newman, L.M.,Sheldon, R. and Wang, L.A., Enzymatic processes for the production of 4-substituted3-hydroxybutyric acid derivatives. PCT Int. Appl., 2004, WO 2004015132 A2.

150. Ema, T., Okita, N., Ide, S. and Sakai, T., Highly enantioselective and efficient synthesis ofmethyl (R)-o-chloromandelate with recombinant E. coli: toward practical and green access toclopidogrel. Org. Biomol. Chem., 2007, 5, 1175–1176.

151. King, A.O., Corely, E.G., Anderson, R.K., Larsen, R.D., Verhoeven, T.R., Reider, P.J., Xiang,Y.B., Belley, M., Leblane, Y., Labelle, M., Prasit, P. and Zamboni, R.J., An efficient synthesisof LTD4 antagonist L-699,392. J. Org. Chem., 1993, 58, 3731–3735.

152. Shafiee, A., Motamedi, H. and King, A., Purification, characterization and immobilization ofan NADPH-dependent enzyme involved in the chiral specific reduction of the keto ester M, anintermediate in the synthesis of an anti-asthma drug, montelukast, from Microbacteriumcampoquemadoensis (MB5614). Appl. Microbiol. Biotechnol., 1998, 49, 709–717.

153. Ulijn, R,V., De Martin, L., Gardossi L. and Halling, P.J., Biocatalysis in reaction mixtures withundissolved solid substrates and products. Curr. Org. Chem., 2003, 7, 1333–1346.

154. Rozzell, D.,Enzymatic production of the key montelukast intermediate. Spec. Chem.Mag.,2008, April, 36–38.

155. Reddy, G.K., Gralla, R.J. and Hesketh, P.J., Novel neurokinin-1 antagonists as antiemetics forthe treatment of chemotherapy-induced emesis. Support. Cancer Ther., 2006, 3, 140–142.

156. Gelo-Pujic, M., Le Guyader, F. and Schlama, T., Microbial and homogenous asymmetriccatalysis in the reduction of 1-[3,5-bis(trifluoromethyl)phenyl]ethanone. TetrahedronAsymm. 2006, 17, 2000–2005.

157. Pollard, D., Truppo, M., Pollard, J., Chen, C.-Y. and Moore, J., Effective synthesis of (S)-3,5-bistrifluoromethylphenyl ethanol by asymmetric enzymatic reduction. Tetrahedron Asymm.2006, 17, 554–559.

158. Van Langen, L.M., van Rantwijk, F. and Sheldon, R.A., Enzymatic hydrocyanation of asterically hindered aldehyde. Optimization of a chemoenzymatic procedure for (R)-2-chloro-mandelic acid. Org. Proc. Res. Dev., 2003, 7, 828–831.

159. (a) Gijsen, H.J.M., Wong, C.-H., Unprecedented asymmetric aldol reactions with three aldehydesubstrates catalyzed by 2-deoxyribose-5-phosphate aldolase. J. Am. Chem. Soc., 1994, 116, 8422–8423; (b) Wong, C.-H., Garcia Junceda, E., Chen, L., Blanco, O., Gijsen, H.J.M. and Steensma,D.H., Recombinant 2-deoxyribose-5-phosphate aldolase in organic synthesis: use of sequentialtwo-substrate and three-substrate aldol reactions. J. Am. Chem. Soc., 1995, 117, 3333–3339;(c) DeSantis, G., Liu, J., Clark, D.P., Heine, A., Wilson, I.A. and Wong, C.-H., Structure-basedmutagenesis approaches toward expanding the substrate specificity of D-2-deoxyribose-5-phosphate aldolase. Bioorg. Med. Chem., 2003, 11, 43–52; (d) Liu, J., Hsu, C.-C. and Wong,C.-H., Sequential aldol condensation catalyzed by DERA mutant Ser238Asp and a formal totalsynthesis of atorvastatin. Tetrahedron Lett., 2004, 45, 2439–2441.

160. Hu, S., Tao, J. and Xie, Z., Process for producing atorvastatin, pharmaceutically acceptablesalts thereof and intermediates thereof. PTC Int. Appl., 2006, WO 2006/134482 A1.

161. Liu, J. and Wong C.-H., Aldolase-catalyzed asymmetric synthesis of novel pyranose syn-thons as a new entry to heterocycles and epothilones. Angew. Chem. Int. Ed., 2002, 41, 1404–1407.

162. Samland, A.K. and Sprenger, G.A., Microbial aldolases as C–C bonding enzymes – unknowntreasures and new developments. Appl. Microbiol. Biotechnol., 2006, 71, 253–264.


163. Schurmann, M. and Sprenger, G.A., Fructose-6-phosphate aldolase is a novel Class I aldolasefrom Escherichia coli and is related to a novel group of bacterial transaldolases. J. Biol. Chem.,2001, 276, 11055–11061.

164. Sugiyama, M., Hong, Z., Whalen, L.J., Greenberg, W.A. and Wong, C.-H., Borate as aphosphate ester mimic in aldolase-catalyzed reactions: practical synthesis of L-fructose andL-iminocyclitols. Adv. Synth. Catal. 2006, 348, 2555 – 2559.

165. Griebenow, K. and Klibanov, A.M., On protein denaturation in aqueous–organic mixtures butnot in pure organic solvents. J. Am. Chem. Soc., 1996, 118, 11695–11699.

166. For some recent reviews on the use of enzymes in nonconventional media, see: (a) Dreyer, S.,Lembrecht, J., Schumacher, J. and Kragl, U., Enzyme catalysis in nonaqueous media: past,present, and future in biocatalysis in the pharmaceutical and biotechnology industries, 2007,CRC Press, pp. 791–827; . (b) Torres, S. and Castro, G.R., Non-aqueous biocatalysis inhomogeneous solvent systems. Food Technol. Biotechnol., 2004, 42, 271–277; (c) Carrea, G.and Riva, S., Properties and synthetic applications of enzymes in organic solvent. Angew.Chem. Int. Ed., 2000, 39, 2226–2254.

167. Butler, L.G., Enzymes in non-aqueous solvents. Enzyme Microb. Technol., 1979, 1, 253–259.168. Gerard A. Sellek, G.A. and Chaudhuri, J.B., Biocatalysis in organic media using enzymes from

extremophiles. Enzyme Microb. Technol., 1999, 25, 471–482.169. Van Rantwijk, F. and Sheldon, R.A., Biocatalysis in ionic liquids. Chem. Rev., 2007, 107,

2757–2785.170. Hobbs, H.R. and Thomas, N.R., Biocatalysis in supercritical fluids, in fluorous solvents, and

under solvent-free conditions. Chem. Rev., 2007, 107, 2786–2820.171. Kazlauskas, R.J., Enhancing catalytic promiscuity for biocatalysis. Curr. Opin. Chem. Biol.,

2005, 9, 1–7.172. Gruber-Khadjawi, M., Purkarthofer, T., Skranc, W. and Griengl, H., hydroxynitrile lyase

catalyzed enzymatic nitroaldol (Henry) reaction. Adv. Synth. Catal. 2007, 349, 1445–1450.173. (a) Torre, O., Alfonso, I. and Gotor, V., Lipase catalysed Michael addition of secondary amines

to acrylonitrile. Chem. Commun., 2004, 1724–1725; (b) Cai, Y., Yao, S.-P., Wu, Q. and Lin, X.-F., Michael addition of imidazole with acrylates catalyzed by alkaline protease from Bacillussubtilis in organic media. Biotechnol. Lett., 2004, 26, 525–528.

174. Zaks, A. and Klibanov, A.M., Enzymatic catalysis in organic media at 100 �C. Science, 1984,224, 1249–1251.

175. Griebenow, K. and Barletta, G., Dehydrated protein powders as biocatalysts in nonaqueoussolvents. In Lyophilization of Biopharmaceuticals (Biotechnology: Pharmaceutical Aspects),Costantino, H.R. and Pikal, M.J. (eds). AAPS Press: Arlington, VA, 2004, pp. 643–668.

176. (a) Klibanov, A.M., Why are enzymes less active in organic solvents than in water? TrendsBiotechnol. 1997, 15, 97–101; (b) Klibanov, A.M., Improving enzymes by using them inorganic solvents. Nature, 2001, 409, 241–246.

177. Serdakowski, A.L. and Dordick, J.S., Enzyme activation for organic solvents made easy.Trends Biotechnol. 2008, 26, 48–54.

178. Partridge, J., Dennison, P.R., Moore, B.D. and Halling, P.J., Activity and mobility of subtilisinin low water organic media: hydration is more important than solvent dielectric. Biochim.Biophys. Acta, 1998, 1386, 79–89.

179. Bell, G., Halling, P.J., May, L., Moore, B.D., Robb, D.A., Ulijn, R. and Valivety, R.H., Methodsfor measurement and control of water in nonaqueous biocatalysis. In Methods inBiotechnology, Vol. 15: Enzymes in Nonaqueous Solvents: Methods and Protocols,Vulfson, E.N., Halling, P.J. and Holland, H.L. (eds). Humana Press: Totowa, NJ, 2001,pp. 105–126.

180. Valivety, R.H., Halling, P.J., Peilow, A.D. and Macrae, A.R., Lipases from different sourcesvary widely in dependence of catalytic activity on water activity. Biochim. Biophys. Acta, 1992,1122, 143–146.

181. Halling, P.J., What can we learn by studying enzymes in non-aqueous media? Phil. Trans. R.Soc. Lond. Ser. B, 2004, 359, 1287–1297.

182. Bell, G., Halling, P.J., Moore, B.D., Partridge, J. and Rees, D.G., Biocatalyst behaviour in low-water systems. Trends Biotechnol., 1995, 13, 468–473.

References 79

183. (a) Rich, J.O., Mozhaev, V.V., Dordick, J.S., Clark, D.S. and Khmelnitsky, Y.L., Molecularimprinting of enzymes with water-insoluble ligands for nonaqueous biocatalysis. J. Am. Chem.Soc., 2002, 124, 5254–5255 and references cited therein; . (b) Carrea, G., Ottolina, G. and Riva,S., Role of solvents in the control of enzyme selectivity in organic media. Trends Biotechnol.1995, 13, 63–70 and references cited therein.

184. Zacharis, E., Moore, B.D. and Halling, P.J., Control of enzyme activity in organic media bysolid-state acid–base buffers. J. Am. Chem. Soc., 1997, 119, 12396–12397.

185. Zaks, A. and Klibanov, A.M., Enzyme-catalyzed processes in organic solvents. Proc. Natl.Acad. Sci., 1985, 42, 3192–3196.

186. Laane, C., Boeren, S., Vos, K. and Veeger, C., Rules for optimization of biocatalysis in organicsolvents. Biotechnol. Bioeng., 1987, 30, 81–87.

187. Halling, P.J., Thermodynamic predictions for biocatalysis in nonconventional media: theory,tests, and recommendations for experimental design and analysis. Enzyme Microb. Technol.,1994, 16, 178–206.

188. Secundo, F., Riva, S. and Carrea, G., Effects of medium and of reaction conditions on theenantioselectivity of lipases in organic solvents and possible rationales, Tetrahedron Asymm.1992, 3, 267–280.

189. Keeling, S.P., Campbell, I.B., Coe, D.M., Cooper, T.W.J., Hardy, G.W., Jack, T.I., Jones, H.T.,Needham, D., Shipley, T.J., Skone, P.A., Sutton, P.W., Weingarten G.A. and Macdonald, S.J.F.,Efficient synthesis of an �-trifluoromethyl-�-tosyloxymethyl epoxide enabling stepwise doublefunctionalisation to afford CF3-substituted tertiary alcohols. Tetrahedron Lett. 2008, 49, 5101.

190. Roberge, C., Fleitz, F., Pollard, D. and Devine, P., Asymmetric synthesis of cyanohydrinderived from pyridine aldehyde with cross-linked aggregates of hydroxynitrile lyases.Tetrahedron Lett., 2007, 48, 1473–1477.

191. Hirose, Y., Kariya, K., Sasaki, I., Kurono, Y., Ebiike, H. and Achiwa, K., Drastic solvent effecton lipase-catalysed enantioselective hydroloysis of prochiral 1,4-dihydropyridines.Tetrahedron Lett., 1992, 33, 7157.

192. Ballesteros, A., van Beynum, G., Borud, O. and Buchholz, K., Guidelines for the characteriza-tion of immobilized biocatalysts. Enzyme Microb. Technol., 1983, 5, 304–307.

193. For some recent reviews, see: Prenosil, J.E., Kut, O.M., Dunn, I.J. and Heinzle, E., Immobilizedbiocatalysts. In Ullman’s Biotechnology and Biochemical Engineering, vol. 2. Wiley–VCH,Weinheim, 2007, pp. 683–734; Sheldon, R.A., Enzyme immobilization: the quest for optimumperformance. Adv. Synth. Catal., 2007, 349, 1289–1307; End, N. and Schoning, K.-U.,Immobilized biocatalysts in industrial research and production. Topics Curr. Chem., 2004,242, 273–317; Bornscheuer, U.T., Immobilizing enzymes: how to create more suitable bioca-talysts. Angew. Chem. Int. Ed., 2003, 42, 3336–3337; Cao, L. Immobilised enzymes: science orart? Curr. Opin. Chem. Biol., 2005, 9, 217–226.

194. Schmid, R.D. and Verger, R., Lipases: interfacial enzymes with attractive applications. Angew.Chem. Int. Ed., 1998, 37, 1608–1633; Hasan, F., Shah, A.A. and Hameed, A., Industrialapplications of microbial lipases. Enzyme. Microb. Technol., 2006, 39, 235–251.

195. Gonzalez-Navarro, H., Carmen Bano, M. and Abad, C., The closed/open model for lipaseactivation. Addressing intermediate active forms of fungal enzymes by trapping of conformersin water-restricted environments. Biochemistry, 2001, 40, 3174–3183.

196. Katchalski-Katzir, E. and Kraemer, D.M., Eupergit� C, a carrier for immobilization of enzymesof industrial potential. J. Mol. Catal. B, 2000, 10, 157–176.

197. Roy, J.J. and Abraham, T.E., Strategies in making cross-linked enzyme crystals. Chem. Rev.,2004, 104, 3705–3721.

198. Weetall, H.H. and Pitcher Jr, W.H., Scaling up an immobilized enzyme system. Science, 1986,232, 1396–1403.

199. Schoevaart, R., Wolbers, M.W., Golubovic, M., Ottens, M., Kieboom, A.P.G., van Rantwijk,F., van der Wielen, L.A.M. and Sheldon, R.A., Preparation, optimization, and structures ofcross-linked enzyme aggregates (CLEAs). Biotechnol. Bioeng., 2004, 87, 754–762.

200. Mateo, C., Palomo, J.M., van Lancen, L.M., van Rantwijk, F. and Sheldon, R.A., A new, mildcross-linking methodology to prepare cross-linked enzyme aggregates. Biotech. Bioeng., 2004,86, 273–276.


201. Dunker, A.K. and Fernandez, A., Engineering productive enzyme confinement. TrendsBiotechnol., 2007, 25, 189–190.

202. Pierre, A.C., The sol–gel encapsulation of enzymes. Biocatal. Biotrans., 2004, 22, 145–170 andreferences cited therein.

203. Reetz, M.T., Tielmann, P., Wiesenhofer, W., Konen, W. and Zonta, A., Second generation sol–gel encapsulated lipases: robust heterogeneous biocatalysts. Adv. Synth. Catal., 2003, 345,717–728.

204. Lutz, S., Rao, N.N. and Wandrey, C., Membranes in biotechnology. Chem. Eng. Technol.,2006, 29, 1404–1415; Gekas, V.C., Artificial membranes as carriers for the immobilization ofbiocatalysts. Enzyme Microb. Technol., 1986, 8, 450–460.

205. Anastas, P.T. and Warner, J.C., Green Chemistry: Theory and Practice, Oxford UniversityPress, 2000.

206. Sheldon, R.A., Organic synthesis - past, present and future. Chem. Ind., 1992, 23, 903–906.207. Constable, D.J.C., Curzons, A.D. and Cunningham, V.L., Metrics to ‘green chemistry’ – which

are the best? Green Chem. 2002, 4, 521–527.208. ACS GCI Pharmaceutical Roundtable, http://portal.acs.org/portal/acs/corg/content?_

nfpb¼true&_pageLabel¼PP_TRANSITIONMAIN&node_id¼1422&use_sec¼false&sec_url_var¼region1, (last access 24 June 2008).

209. Constable, D.J.C., Curzons, A.D., Freitas dos Santos, L.M., Geen, G.R., Kitteringham, J.,Smith, P., Hannah, R.E., McGuire, M.A., Webb, R.L., Yu, M., Hayler, J.D. and Richardson,J.E., Green chemistry measures for process research and development. Green Chem., 2001, 3,7–9; Curzons, A.D., Constable, D.J.C., Mortimera, D.N. and Cunningham, V.L., So you thinkyour process is green, how do you know? Using principles of sustainability to determine what isgreen – a corporate perspective. Green Chem., 2001, 3, 1–6.

210. Jimenez Gonzalez, C., Curzons, A.D., Constable, D.J.C. and Cunningham, V.L., Cradle-to-gatelife cycle inventory and assessment of pharmaceutical compounds. Int. J. Life Cycle Assess.,2004, 9, 115–121; Curzons, A.D., Jimenez Gonzalez, C., Duncan, A.L., Constable, D.J.C. andCunningham, V.L., Fast life cycle assessment of synthetic chemistry (FLASC) tool. Int. J. LifeCycle Assess., 2007, 12, 272–280; Jimenez Gonzalez, C., Overcash, M.R. and Curzons, A.,Waste treatment modules – a partial life cycle inventory. J. Chem. Technol. Biotechnol., 2001,76, 707–716.

211. Jimenez Gonzalez, C., Constable, D.J.C., Curzons, A.D. and Cunningham, V.L., DevelopingGSK’s green technology guidance: methodology for case-scenario comparison of technologies.Clean Technol. Environ. Pollut., 2002, 4, 44–53.

212. Sheldon, R.A., The E factor: fifteen years on. Green Chem., 2007, 9, 1273–1283.213. Menzel, A., Werner, H., Altenbuchner, J. and Groeger, H., From enzymes to ‘designer bugs’ in

reductive amination: a new process for the synthesis of L-tert-leucine using a whole cell-catalyst. Eng. Life Sci., 2004, 4, 573–576; Bruggink, A., Schoevaart, R. and Kieboom, T.,Concepts of nature in organic synthesis: cascade catalysis and multistep conversions in concert.Org. Proc. Res. Dev., 2003, 7, 622–640; Buckland, B.C., Robinson, D.K. and Chartrain, M.,Biocatalysis for pharmaceuticals – status and prospects for a key technology. Metab. Eng.,2000, 2, 42–48.

214. Bacher, A., Stohler, P. and Weber, W., Process for the production of lipstatin and tetrahydro-lipstatin. Eur. Pat. Appl., 2002, EP803576 A2.

215. (a) Fleming, M.P., Han, Y.-K., Hodges, L.M., Johnston, D.A., Micheli, R.P., Puentener, K.,Roberts, C.R., Scalone, M., Schwindt, M.A. and Topping, R.J., Process for the (enantiose-lective) synthesis of 3,6-dialkyl-5,6-dihydro-4-hydroxy-pyran-2-ones via intramolecularcyclization of (homochiral) �-halo esters. PCT Int. Appl., 2001, WO 2001057014 A2; (b)Birk, R., Karpf, M., Puntener, K., Scalone, M., Schwindt, M. and Zutter, U., With asymmetrichydrogenation towards a new, enantioselective synthesis of orlistat. Chimia, 2006, 60, 561–565; (c) Schwindt, M.A., Fleming, M.P., Han, Y.-K., Hodges, L.M., Johnston, D.A., Micheli,R.P., Roberts, C.R., Snyder, R., Topping, R.J., Puntener, K. and Scalone, M., enantioselectivesynthesis of a key intermediate in a new process for orlistat using asymmetric hydrogenationand a Grignard reagent promoted lactone cyclization. Org. Proc. Res. Dev., 2007, 11,524–533.

References 81

216. Karpf, M. and Zutter, U., Process for the preparation of oxetanones. Eur. Pat. Appl., 1991, EP443449 A2.

217. The chemical route to orlistat was ultimately favoured over the biosynthetic route as the latterrequired complex downstream processing that eliminated the benefits gained from the bio-transformation itself. More recent advances in metabolic engineering and drownstream proces-sing might have resulted in the development of a more competitive process.

218. Poechlauer, P. and Wagner, M., Enzymatic process to separate racemic mixtures of deltavalerolactones. US PCT Appl., 1995, US 5412110.

219. Scott, A.I., Discovering nature’s diverse pathways to vitamin B12: a 35-year odyssey. J. Org.Chem., 2003, 68, 2529–2539.

220. Bornscheuer, U.T. and Buchholz, K., Highlights in biocatalysis – historical landmarks andcurrent trends. Eng. Life Sci., 2005, 5, 309.


2

Biocatalyst Identification and Scale-up:Molecular Biology for Chemists

Kathleen H. McClean

2.1 History of Biotechnology

Biotechnology is a modern discipline with ancient origins. Fermentation has been

exploited by humans for more than 6000 years for the manufacture and preservation of

foodstuffs. The modern-day manufacture of beer, bread, wine, dairy products, and fer-

mented bean products are the descendants of these early experiments in food processing.

Over the past 200–300 years many of these processes have made the transition from

traditional or cottage industries to large (industrial)-scale, controlled manufacturing pro-

cesses. This was facilitated by the development of the discipline of microbiology, which

brought a better understanding of the identity and behaviour of the microorganisms

involved. Several of the bacteria, yeasts and filamentous fungi associated with these

traditional processes are still work-horses of modern biotechnology – organisms such as

Saccharomyces cerevisiae, lactic acid bacteria and Aspergillus oryzae. Many have been

used in industrial fermentations to produce a diverse range of products, such as vitamins,

amino acids, organic acids and solvents (Figure. 2.1). ‘White biotechnology’ had begun to

emerge as a means to produce platform chemicals, biofuels and chemical intermediates.

Early examples of biotransformation using microbes and defined chemical substrates

began to become established in the mid-nineteenth century. Pasteur noted in 1858 that

when a solution of an ammonium salt of (–)-tartaric acid was fed to a culture of the mould

Penicillium glaucum the (þ)-tartaric acid was consumed, leaving the (�)-tartaric acid,



thus providing a very early example of a whole-cell biotransformation. The pioneering

work of Eduard Buchner (1897) demonstrated the fermentation of sugar by cell-free

extracts of yeast, establishing the principle that fermentation occurred as a result of soluble

microbial catalysts (enzymes). This established the important principle that not all biolo-

gical transformations required living cells.

One of the first major pharmaceutical biotransformations was the development of the

synthesis of hydrocortisone in the late 1940s by whole-cell hydroxylation1 (Figure 2.2). Up

until then a 40-step synthetic route developed by the Noble Prize winning chemist

R.B.Woodward was the only source of this important drug substance and intermediate.2

Nowadays, a biocatalyst exists for the selective hydroxylation of every position on the

steroid nucleus.3

Biotransformations using native organisms (whole-cell or cell extracts) can be limited

by poor expression of the specific enzyme, by the presence of more than one enzyme

activities competing for the substrate or by the presence of isozymes displaying a range of

specificities. However, by 30–40 years ago, significant advances in microbial genetics had

paved the way for recombinant DNA (rDNA) technology which was to result in radical and

CO2H

CO2H

SOLVENTS AND BIOFUELS3CH2 HCdicAcitecAHO 3CO2HEthanol CH

Butanol CH3CH2CH2CH2OH 1,3-Propanediol HOCH2CH2CH2OH

Itaconic Acid

HO OH

O OHO

OH

H

2BRiboflavin VitaminVitamin C

N

S

HN

O

Me

Me

CO2H

O

Ph

Penicillin G

O

O

Me

n

Poly-Lactic Acid

(S)

(S)(Z)

(S)

(Z)

NMe2Me OH

OH

OH

CONH2

OO OHOH

Tetracycline

(R)

(S)(R)

(E)HO

HO

NH2

CO2H

Aminoshikimic Acid

FINE CHEMICALS, POLYMERS AND PHARMACEUTICAL INTERMEDIATES

VITAMINS

ANTIBIOTICS

N

N (E)

N

NH

(E)

O

O

(R)(R)

(S)

Me

Me

OH

OH

OH

OH

Figure 2.1 Structures of various industrial fermentation products

84 Biocatalyst Identification and Scale-up: Molecular Biology for Chemists

rapid advances in biotechnology. New techniques for manipulating DNA emerged, such as

the development of vectors and cloning strategies, which allowed rapid generation of

novel recombinant strains of microbes which could produce new enzymes. Many of the

first commercial products generated using these technologies were mass-produced extra-

cellular hydrolase enzymes (such as proteases: subtilisin, thermolysin and chymotrypsin).

Often, these ‘bulk enzymes’ were originally developed for consumer products or process

industry applications and were also exploited by chemists in synthetic reactions and for the

resolution of racemates.4 Rapid methods of random and specific site-directed mutagenesis

coupled with screening became routine in the enzyme manufacturing sector, enlarging the

portfolio of ‘process’ enzymes which could also be exploited for their biocatalytic

(synthetic) potential. The development of rDNA technology also accelerated research in

the regulation of genetic processes, and the development of technologies such as poly-

merase chain reaction (PCR) and automated gene sequencing have in turn helped to drive

advances in our understanding of how genetic information relates to function (bioinfor-

matics). As a result, the tremendous rate of expansion of information and the accessibility

of rDNA technologies (including robotics and advanced bioinformatics) has meant that

screening and genetic modification procedures which a decade ago could have taken

months or years can now often be completed in weeks or even days. This led to the

increased production of enzymes specifically for biotransformation applications. The

number of novel enzymes or enzyme genes identified has risen exponentially in the past

10 years, and a greater number and diversity of biocatalysts are available or accessible for

synthetic chemistry (in the same decade, a 35-fold increase in the number of articles with

‘biocatalysis’ in the title or abstract has been recorded by bibliographic resources such as

PubMed).

2.2 Identifying Potential Biocatalysts for Chemical Synthesis

Biological systems can offer many attractive features as catalysts for the synthetic chemist,

such as high substrate specificity, precise stereo- and regio-selectivity and mild reaction

conditions. The diversity of biochemical reactions is an indicator of the potential of

enzymes in synthetic applications. Therefore, when designing a synthetic route to a target

molecule, a number of steps might best be performed using a biocatalyst, particularly

where the generation of chirality is involved, and so biocatalysis should be routinely

considered. This section briefly outlines the options available to the researcher when

contemplating using a biocatalyst.

Me

Me

O

O

OHOHHO

Me

Me

O

O

OHOH

Whole Cells

Figure 2.2 Regio- and stereo-selective steroid hydroxylation

2.2 Identifying Potential Biocatalysts for Chemical Synthesis 85

2.2.1 Literature Precedents

As with chemical synthesis, the first step when prospecting for a particular biotransforma-

tion is to perform a literature search to check whether a suitable precedent has been

described. Extensive technical literature resources in the public domain provide both

examples of specific enzyme-catalysed reactions and descriptions of transformations

where enzyme activity is inferred if not explicitly described. Currently, searches of online

databases such as PubMed reveal over 2000 new publications per annum in the subject of

enzyme catalysis (excluding reviews).

In some fortunate instances, an exact match for the desired biotransformation might be

available in the literature that uses a commercial enzyme or microbial strain that

is accessible through culture collections and provides sufficient activity for use in

preparative-scale reactions.

2.2.2 Commercial Enzyme Sources

When the reaction of interest is catalysed by a stable enzyme and a single biotransforma-

tion step is required, it is often possible to find a suitable commercial enzyme for this step.

Some commonly used enzymes can be sourced from general supply houses such as Sigma–

Aldrich, but there are now several specialist suppliers who provide enzymes for biocata-

lysis applications. At the time of writing, the most comprehensive catalogue of enzymes is

probably that supplied by Codexis (www.codexis.com). Other major suppliers include

Meito Sangyo (www5.mediagalaxy.co.jp/meito/), Amano Enzyme (www.amano-enzyme.

co.jp/eng/company), ChiralVision (www.chiralvision.com) and Enzysource (www.

enzysource.com ). There are many other smaller suppliers, some of which also offer a

more ‘bespoke’ service providing a particular class of enzyme, or supplying additional

services such as enzyme formulation, immobilization or enzyme kits. Several academic

groups also maintain up-to-date pages of links to enzyme suppliers; these include the

CoEBio3 site at the University of Manchester (www.coebio3.org) and the biocatalysis

group pages at the University of Graz (http://borgc185.kfunigraz.ac.at/).

Many enzymes require the presence of small organic non-protein groups (cofactors) in

order to catalyse reactions. Dependence on cofactors may limit the usefulness of the

enzyme if the cofactors are expensive, unstable or difficult to recycle. This becomes an

issue when using redox enzymes such as dehydrogenases, which often require the presence

of the reduced form of nicotinamide adenine dinucleotide (NADH) or its phosphorylated

analogue (NADPH). Enzyme and cofactor recycling itself can be avoided by using whole-

cell systems, or the cofactor can be recycled in vitro by the action of a second enzyme and

the inclusion of a suitable substrate which is oxidized. An example is the use of formate

dehydrogenase (commercially available) for the oxidation of formic acid to CO2 for the

recycling of NADH from NADþ.

It should be noted that some commercial enzyme preparations may contain several

enzyme isomers (enzymes derived from one source which belong to the same enzyme class

but differ in specificity, stability or other properties). This is most often the case when the

commercial preparation was developed for a process industry application rather than a

specific chemical biotransformation application. Some fungal enzymes, such as laccase,

are sometimes supplied as crude enzyme mixtures. Fungal laccases are manufactured on a

huge scale (multitonne per annum) and are principally used in bulk processes such as wood


pulp processing and textile dye bleaching. If a very specific activity is required then the

chemist may need to request additional information from the supplier or manufacturer, or

be prepared to screen several commercial preparations. However, the growing trend to

produce enzymes using rDNA technology is reducing the frequency of these problems.

2.2.3 Culture Collections as Sources of Microorganisms

If the precise original biological source of the desired enzyme for a precedented transfor-

mation is known (say, from a microbe with a stock centre code or number), then it can be

easy to obtain that microbe and grow it according to the defined conditions in the literature

reference or stock centre recommendations. There are more than 500 registered microbial

stock centres worldwide where microbial cultures and cell lines are curated. Most of these

are accessible to the public, and samples of cultures or cell lines can be obtained for

relatively modest fees. Further information on stock centres, including contact details, is

available from the World Federation for Culture Collections (www.wfcc.info), which

provides links to culture collections. Although most developed countries have national

(and sometimes specialized) collections, there are a few major collections which are most

frequently used for the deposit of microbes of industrial interest (Table 2.1). These

collections are a vital resource for biotechnology, as they provide access to certified

pure microbial cultures as well as being valuable technical resources for microbial

identification with information on culture and characteristics of the microbes. In addition,

they provide safe repositories for materials covered by patent protection. Alongside the

large and often diverse culture collections there are many specialized collections which

Table 2.1 Some major microbial culture collections. Most of the larger collections haveonline searchable catalogues and provide other important information on pathogenicity, cellculture and maintenance, as well as bibliographic information relating to individual strains

Culture collection Abbreviation Web address Major collections

Belgian CoordinatedCollections ofMicroorganisms

BCCM www.bccm.belspo.be Bacteria, Fungi Yeasts DNA

Deutsche Sammlungvon Mikroorganismenund Zellkulturen

DSMZ www.dsmz.de Bacteria, fungi, yeasts, celllines, DNA, viruses,Archaea

American TypeCulture Collection

ATCC www.atcc.org Bacteria, fungi, yeasts, celllines, DNA, viruses,Archaea

Centraalbureau voorSchimmelcultures

CBS www.cbs.knaw.nl Fungi, yeasts, CBS also hostNCCB (NetherlandsBacterial CultureCollection)

National Collectionof Industrial Bacteria

NCIMB www.ncimb.com Bacteria (industrial food,environmental and marine),DNA resources


provide resources (materials and information) associated with one type or strain of

microbe (such as the Escherichia coli genetic stock centre: www.cgsc.biology.yale.edu).

Many of the larger commercial collections also have excellent online catalogues which

can also be searched using relevant keywords for recorded enzyme activity, metabolic

pathways, substrates or products, or their environmental origin. As the interest in using

culture collections as biocatalyst resources has increased, so the annotation of the catalo-

gues has improved, and most larger collections actively seek and collate information

documenting their strain’s biocatalytic capabilities.

What can be done if there is not an exact literature precedent for the required enzymatic

transformation? In these cases some sort of screening procedure may be necessary, either

by direct screening of commercial enzymes, culture collections and environmental iso-

lates, or by using a bioinformatics-based approach to identify potential enzymes based on

sequence information.

2.2.4 Enzyme and Gene Databases, Bioinformatics and the Search for

New Enzymes

Searching for information on enzymes can be made a lot easier by surveying the collated

information available at online databases dedicated to genetic information (sequence

based) and to enzymes (usually based on analysis of observed activity). The genetic

(sequence) information can refer to the gene sequences of cloned enzymes where activity

has been demonstrated, and also to the predicted enzyme genes ‘mined’ from the huge

resource of genetic information accrued from genome sequencing projects and metabo-

lomics experiments (the direct recovery of DNA from the environment). Originally,

information on genetic resources and enzyme activity was often collected independently,

but nowadays these resources are often integrated or cross-referenced. The correlation

between information relating to proteins (enzymes) and nucleic acid sequences

illustrates the increasing importance of bioinformatics – the application of mathematical

and computing techniques to interpretation of sequence information.

A good practical starting point is to look at databases which primarily deal with demon-

strated enzyme activity, since this may identify enzymes which could be relatively easy to

obtain. Most of these databases have their origins in sectors of biological research, although

increasingly the role of enzymes in biocatalysis is referenced in an accessible form.

Unfortunately, not all known enzymes are referenced in every database, and the occurrence

in a database may not mean the enzyme has been purified or that it comes from an accessible

microbial source. However, by knowing that the databases differ in their focus, this can be

used to collect complementary sets of information about individual enzymes. BRENDA

(www.brenda-enzymes.org) offers a comprehensive database of enzymes in the academic

literature. Much of the information presented focuses on functional information, so that

information on substrates, stability, pH range, etc. is particularly easy to obtain. The database

covers enzymes from all types of organisms. Searches can be made by many methods: by

enzyme name, by class, substrate, product, molecule structure (either by name or by using a

graphical interface), and by organism. It is also possible to search amino acid sequence

databases for families of sequence-related enzymes, or for sequence motifs, or even

to check if a gene is likely to encode an enzyme.5 The University of Minnesota

Biocatalysis/Biodegradation Database (http://umbbd.msi.umn.edu/index.html) provides a


comprehensive overview of microbial biocatalytic reactions and biodegradation pathways.6

This can be particularly useful if the substrate is a known xenobiotic. In many instances the

reactions described by this resource are referenced to original publications and named

microbial isolates. It is important to remember that the content of this database focuses on

the enzymatic activities of microbes in the environment and the mechanisms they use to

degrade organic compounds, including pollutants. The database can be searched by several

methods, including by pathway, by chemical compound (including graphical structure

search), by organism and by enzyme name. This site has organized biocatalytic reactions

into pathways, noting that in natural environments the steps of the biodegradative pathway

could exist in a single organism or in a range of microbes (a microbial consortium). Features

of this website also include a pathway prediction tool which attempts to generate plausible

biodegradation (biotransformation) routes for a given organic compound based on ‘rules’

generated by extensive review of the academic literature. The information is of particular use

if one needs to check whether a particular transformation is likely to occur (and may identify

an actual candidate enzyme); it is also important when carrying out whole-cell biotransfor-

mations, as it can be used to predict the likelihood of unwanted side reactions or downstream

modifications of the product. The site also hosts an excellent page of links to other online

resources in microbial biotechnology.

There are several other online searchable sites which can be used to search for enzymes,

and the most significant of these are linked to or embedded in bioinformatics resources.

These types of resource are probably more useful when considering obtaining a new enzyme

by cloning experiments or when planning mutagenesis experiments to generate enzymes

with altered properties. At its most basic, bioinformatics involves using basic sequence

analysis tools in the identification of features in nucleic acids or amino acid sequences.

Widely accessible programs are routinely used to identify gene coding sequences, features

such as regulatory sequences and introns (interrupting noncoding DNA sequences which

do not encode polypeptide) and prediction of proteolysis sites in polypeptide sequences.

However, the accumulation of large quantities of sequence information, combined with

advances in computational methods, has made it possible to perform many more complex

analyses. The principle activities in bioinformatics include mapping and analysing DNA

and protein sequences, aligning different DNA and protein sequences to compare them

(identifying homology) and creating and viewing three-dimensional models of protein

structures. All these activities can be applied in the search for new or improved enzymes.

One basic approach is to search databases of sequence information for predicted enzyme

sequences (sequence search service). Usually, when a DNA or RNA sequence is lodged in

a searchable database, the depositor will have performed a basic annotation which may

include predictions of enzyme amino acid sequences. Individual gene sequences can be

compared with all the other sequences in a database or library in searches for overall

sequence homology (e.g. ‘The Basic Local Alignment Search Tool’ (BLAST) finds

regions of local similarity between sequences7). These types of analysis can be used to

infer functional or evolutionary relationships between sequences. Protein sequence fea-

tures have been systematically analysed to help identify the specific features or motifs

characteristic of protein function (including enzyme activity) using facilities such as

PROSITE, a database of entries describing the domains, families and functional sites of

proteins, as well as their associated amino acid patterns, signatures and profiles.8 Further

information on these and other methods of ‘data mining’ and analyses are given at the


ExPASy (Expert Protein Analysis System, www.expasy.ch) or the National Centre for

Biotechnology Information (NCBI, www.ncbi.nlm.gov) websites. Both websites provide

bioinformatics tools, links to sequence databases and extensive bibliographic resources.

As an example of the wealth of information available on individual enzymes, at the time of

writing a search based on ‘nitrilase’ in the ‘Entrez protein’ section of NCBI will recover

more than 10 000 references to nitrilase enzyme amino acid sequences. These can be

rapidly screened online by organism, and the individual entries will have links to amino

acid and gene sequence, relevant literature and information on protein features (such as

conserved domains).

2.2.5 Metagenomics: Sampling DNA Directly from the Environment

Another approach now available when searching for new enzymes is to obtain the gene for

the enzyme from libraries of DNA recovered directly from the environment. This avoids

the need to know much about the original microorganism and also eliminates the need to

grow it in the laboratory. Much of the interest in metagenomics comes from the discovery

that the vast majority of microorganisms had gone unnoticed until relatively recently.9

Traditional microbiological methods rely upon laboratory cultivation of organisms. For

quite some time before the development of molecular biology techniques it had already

been recognized that many microbes eluded description or characterization because they

could not be cultured by standard microbiological techniques. Surveys of ribosomal RNA

genes taken directly from the environment revealed that cultivation-based methods find

less than 1% of the bacterial and archaeal species in a sample (Archaea are a group of

organisms which often resemble bacteria in morphology but have some features which

distinguish them from both prokaryotes an eukaryotes). This illustrated the extent of our

ignorance about the range of metabolic and species diversity in the microbial world.

In the recent past there has been a trend to sample microbes and microbial DNA from

‘mega diversity ecosystems’ typically found in locations such as Mexico, Central America

or South East Asia, or extreme environments found in regions of volcanic activity (Hawaii,

Iceland), deep ocean thermal vents and permafrost. Sampling DNA from multiple sites

probably increases the species diversity represented in the pooled samples. Samples from

particular environments might help to recover DNA enriched in genes for enzymes with

certain properties – thermostable enzymes could be expected to be found associated with

high-temperature environments; carbohydratases might be produced in the digestive tracts

of herbivores.10 Similarly, polluted or contaminated sites might be useful locations to look

for the DNA-encoding enzymes which act on the contaminants or related substances. For

several commercial organizations specializing in the development of novel enzymes for

industrial processes, these metagenomic pools are a valuable resource to be ‘mined’ using

high-throughput technologies to discover new enzymes (see Figure 2.3).11

These methods are represented in a rather simplistic manner, as with any set of

techniques there are some limitations: the DNA may be extensively damaged; it can be

difficult to recover entire intact genes on smaller fragments of DNA; redundancy could be

an issue where some species predominate; and specialized DNA cloning techniques may

be needed to maintain long fragments of genomic DNA. Recently, an additional screening

method has been proposed in which catabolic genes induced by various substrates are

identified from metagenomic DNA libraries by using automated cell-sorting screening


techniques (such as fluorescence activated cell sorting).12 This method was applied

successfully to isolate aromatic hydrocarbon-induced genes from a metagenomic library.

The challenges are to identify a subset of relevant clones from very large libraries of

random DNA clones. Nevertheless, metagenomics offers a methodology to sample true

enzymatic biodiversity which is several orders of magnitude greater than that which could

be found using ‘conventional’ microbiology.

In some senses, metagenomic screening is a reflection on, and expansion of, the historic

extensive screening of soil samples for culturable microbes which could produce interest-

ing secondary metabolites. The techniques now also sample the unculturable organisms

and can search for potential enzymes using sequence-based methodologies as well as by

demonstration of activity by expression cloning. There remain the difficulties which are

Sequence driven analysis Function driven analysis

Transformation

DNA fragments Digested vector

Ligation

Genomic DNA extraction

Figure 2.3 Metagenomic cloning experiments. Isolation of genomic DNA directly fromenvironments (soil, plants, mixed environments or thermal-vent worms are the examplesillustrated here) can recover DNA fragments which could encode for enzymes. The DNAfragments can be ligated to plasmids or DNA linkers, and then subjected to functional screen-ing (expression cloning) and/or sequence analysis. Amplification by PCR can sometimes beused to yield libraries enriched with clones containing selected sequence motifs relating tofamilies of enzymes


sometimes encountered when trying to express an active enzyme (some of these will be

discussed in later sections of this chapter – protein truncation, presence of introns,

misfolding, post-translational modifications, codon matching, inappropriate expression

host, etc.). The more fruitful approach may be first to recover and analyse the sequences

from the environment, identify the promising clones (e.g. those which have sequence

homology to known enzymes), then develop strategies for expression cloning (to produce

the enzyme in a host cell such as E. coli) and functional testing of the enzymes. This has

recently been used by several groups as a method to discover new enzymes involved in

polyketide synthesis.13 However, where there is access to existing high-throughput func-

tional screening methods (such as growth on single substrate), the direct functional screen

can also be useful.14 High-throughput functional screening systems typically employ

colorimetric or fluorescence-based assays to demonstrate activity, and often use robotic

systems to generate and analyse the many thousands of tests required in screening

experiments.

Bacteria-like microbes found in extreme environments (‘extremophiles’) were some-

times presumed to belong exclusively to the specialized ‘domain’ known as Archaea.

However, it has transpired that the arrangement of microbes in ecological niches was not as

simple as this – many Archaea have been found in temperate zones and in ecosystems

similar to those occupied by bacteria, and some ‘conventional’ microbes occupy extreme

environments.

Nevertheless enzymes isolated directly or indirectly from organisms found in extreme

environments are likely to harbour useful adaptations which could be exploited in indus-

trial processes (such as solvent tolerance and thermostability). As an example, the readily

available and versatile lipase B from Candida antarctica (CAL-B) enzyme, isolated from

an extremely cold environment (a lake on the Antarctic continent), surprisingly shows

remarkable thermal tolerance under certain conditions, notably in organic solvents, where

it can often be used at�60 �C. Structural studies of these enzymes can yield insights on the

features which confer these advantages, information which might be used in the future to

inform targeted evolution or modification of other enzymes, as well as contributing to the

general pool of known enzymes.15 Novel enzymes obtained via metagenomics may still

need to undergo performance optimization by molecular biology or protein engineering to

make them suited for industrial processes.

2.3 Molecular Biology for Improved Biocatalysts

Many reported biotransformations are initially only demonstrated on a very small scale,

the substrates or products may be subject to competing reactions if other enzymes are

present (this can be a serious issue in whole-cell biocatalysis), or the desired enzyme is

insufficiently active or produced in low levels. For many biotransformations a little care

and attention is needed in the growth of the microbe to achieve the desired results.

Production of a specific enzyme from a microbe can often be increased by growing the

cells in the presence of a very small concentration (typically micromolar) of an inducer.

The inducer could be a ‘natural’ enzyme substrate, a substrate mimic or a molecule which

is in some way associated with a substrate’s availability or role in metabolism. This

process is called induction and represents a genetic ‘switch’ which cells use to respond


to environmental changes, and can be used to control enzyme production both by wild-type

cells and when applied in rDNA technology. A typical example would be the induction of

lipase production by the presence of fats,16 whilst providing a protein substrate or starving

the cell of nitrogen could be a method to stimulate protease production.17 In contrast,

enzymes which are involved in essential tasks, such as central metabolic functions, are

available all the time to the cell – their genes are often referred to as housekeeping genes

and their constant level of enzyme production is known as constitutive expression. These

systems controlling levels of enzyme production are exploited in rDNA technology.

Controllable (inducible) enzyme production systems are generally used to produce

maximal yields of functional biocatalyst.

For laboratory-scale production of enzymes using rDNA technology in E. coli, many

commonly used expression systems use isopropyl-�-D-thiogalactopyranoside (IPTG) as an

inducer. Although convenient for small-scale production, IPTG is rather expensive and

toxic, often making it unsuitable for industrial-scale manufacture of enzymes. Luckily,

several alternative inducible control systems are available. A typical example of an

alternative system is the use of the relatively cheap and nontoxic sugar arabinose as an

inducer of gene expression in E. coli.18 Another strategy is to link the induction of enzyme

production to predictable changes in cell physiology, such as the change from the expo-

nential growth phase to the stationary phase. This has been demonstrated in industrial

strains of Streptomyces lividans, where the gene of interest has been linked to a promoter (a

genetic regulatory unit) which is only active in the stationary phase.19

Difficulties sometimes arise if the enzyme of interest is derived from an organism which

is difficult or impossible to grow using conventional laboratory facilities – organisms such

as extremophiles (requiring extreme temperatures, pressure or salt levels for culture), or

fastidious organisms which have complex (expensive) growth requirements, or from a

multicellular organisms which cannot readily be obtained or maintained in a conventional

microbiology laboratory. Where the enzyme is from a mammalian source there can be

additional safety, ethical or regulatory problems if the biotransformation is destined for

pharmaceutical manufacture (since the catalyst will probably be derived from human

tissue or slaughterhouse waste), or if consistent quality criteria of the supplies are required

which are derived from unregulated animal by-products. An interesting example of this

type of material is pig liver esterase (PLE), a versatile biocatalyst which fell into disuse at

least in part due to concerns over the safety of animal-derived products. Recently,

recombinant PLE enzymes (including commercially produced enzyme) have become

available, making its broader use in industrial applications possible once more.20

Enzymes from other plant or animal sources may also be difficult to obtain in quantity,

either due to limited access to the source material and/or difficulty in obtaining enough

purified active enzyme to perform the reaction. In other cases, a less-than-suitable enzyme

is available and one would like to change its properties such as its substrate selectivity or

process stability, or it might be thought necessary to find a new enzyme – either by

searching for one from nature and/or by engineering of existing enzymes. It is in these

instances that rDNA technology becomes an essential tool for producing quantities of

usable enzymes manufactured in a controlled manner by a microbial host. This technology

aims to extract the DNA from a gene pool and tries to isolate and clone the desired

gene(s) for a particular application. For example, the bacterium Rhodococcus erythropolis

NCIMB 11540 (from the collection of National Collections of Industrial Food and Marine

2.3 Molecular Biology for Improved Biocatalysts 93

Bacteria, Aberdeen, UK), was found to have a highly active nitrile hydratase/amidase

enzyme system, based on whole-cell biotransformation experiments.21 Subsequently,

individual enzymes (nitrile hydratase and amidase) from this strain were cloned and

expressed separately in E. coli.22 However, distribution of some strains or other materials

from these public collections may be limited, usually as a result of the restrictions on their

commercial use imposed by intellectual property rights.

How does an understanding of the principles of molecular biology assist in (re)design of

biocatalysts? Once the gene specifying an enzyme has been sequenced, the sequence

information can be used by the molecular biologist to ‘tailor’ the enzyme using a

combination of synthetic and genetic techniques at the DNA sequence level so as to

modify the enzyme’s catalytic activity, improve its process compatibility and in some

instances improve the actual yield of enzyme manufactured in the fermentation process.

It is also important to note that molecular biology, while it is a very powerful tool, is

probably most effective in industrial process development when used in conjunction with

other techniques such as enzyme formulation, immobilization and appropriate process

design engineering.23

rDNA technology has many applications in speciality chemical and pharmaceutical

manufacturing beyond the current topic of biocatalysis for small-molecule manufacturing.

More advanced related topics include metabolic engineering, advanced fermentation

processes and production of biopharmaceuticals, to name but a few. This brief overview

of the field will concentrate on the basic principles of rDNA technology for enzyme

production, new enzyme discovery and enzyme modification.

2.3.1 Introduction to rDNA Technology (Applied Molecular Biology)

As is the case for many scientific disciplines, molecular biology has developed its own

terminology which can appear complex and sometimes bewildering to other scientists.

Many acronyms have been developed which help to produce a ‘shorthand’ notation for

describing the manipulation of large biomolecules with correspondingly large formal

names. For those not familiar with these terminologies, a brief primer of some of the

basic principles may be useful. More comprehensive descriptions of the principles of

molecular biology are available in several textbooks; for example, see Ref. 24.

2.3.1.1 The Central Dogma of Molecular Biology

The ‘central dogma’ relates to the transfer of sequence-based information in biological

systems. DNA – usually the primary store of genetic information and maintained in the cell

as double-stranded DNA (dsDNA) molecules – can be faithfully copied (replicated) as new

DNA molecules (DNA replication) by an enzyme called DNA-dependent DNA polymer-

ase (usually simply called ‘DNA polymerase’). The sequence-based DNA information can

also be copied into messenger RNA (mRNA) by a process called transcription and the

resulting mRNA molecules are often referred to as transcripts. Proteins can then be

synthesized using the information in mRNA as a template (translation). When inheritable

information from a gene, such as the DNA sequence, is made into a functional gene

product (such as protein or RNA) the process is known as gene expression and an

expression system is required to express it. The basic principles of the central dogma are

illustrated in Figure 2.4.


The objective of gene cloning is frequently to express protein products from the cloned

gene sequences often in a foreign host (i.e. in a cell from a different species). Some

special instances of reverse transcription occur in certain viruses, where DNA can be

synthesized using RNA as the primary template. The reverse transcriptase system is

important when trying to clone and express proteins from eukaryotes (yeasts, fungi,

plants and animals), as the organization of their genomic DNA has some important

differences from that of prokaryotes (Bacteria and Archaea – see Section 2.3.1.5).

Genes from these organisms are frequently cloned as functional units in bacteria by

means of in vitro manipulation of the mRNA to recover a DNA sequence which can be

used to express proteins in bacteria.

2.3.1.2 Enzyme Tools in Molecular Biology

Several of the enzymes involved in the processes of replicating, transcription and reverse

transcription are available commercially and are used by molecular biologists in the

manipulation of nucleic acids. One of the most important of these is Taq polymerase

(Taq), which is a thermostable DNA polymerase named after the thermophilic bacterium

Thermus aquaticus from which it was originally isolated. This enzyme is especially

important, as it is central to the technique known as PCR, which allows sophisticated,

targeted in vitro amplification and manipulation of sections of DNA or RNA. DNA

DNA

RNA

Protein

ReplicationReverse

Transcription Transcription

Translation

Used by retrovirusesUsed by all cells

Figure 2.4 The central dogma of molecular biology – transcription of DNA to RNA to protein.This concept (often called the ‘dogma’) forms the backbone of molecular biology and isrepresented by four major stages. (1) The DNA replicates its information by conservativereplication (by means of DNA polymerases). (2) The DNA codes for the production of mRNAduring transcription. In some viruses the primary repository of genetic information is RNA, andthe equivalent DNA molecules can be generated by a process known as reverse transcription.(3) In eukaryotic cells, an additional step occurs: the mRNA is processed (essentially by splicingoff noncoding regions called introns) and the mature mRNA migrates from the nucleus to thecytoplasm. (4) mRNA carries coded information to specialized complexes of protein andribonucleic acids called ribosomes. The ribosomes ‘read’ this information and use it for proteinsynthesis. This process is called translation


polymerases have also been central to the development of rapid modern methods of gene

sequencing, which is one of the enabling technologies for the discipline of bioinformatics

(interpretation of sequence information). Amongst the other enzymes which are important

in molecular biology are those used to cleave DNA at specific sites (restriction endonu-

cleases) and to join fragments of DNA (DNA ligases). Various other DNA-modifying

enzymes are also used in vitro to help generate rDNA molecules (where DNA sequences

from two or more sources which would not normally occur together are incorporated into a

single ‘recombinant’ molecule).

2.3.1.3 The Genetic Code: Nucleotides, Codons and Amino Acids

In DNA molecules, the genetic code is represented by sequences of the four nucleotide

bases adenine (A), cytosine (C), guanine (G) and thymine (T). On transcription, each

template DNA base is represented in the equivalent mRNA by its complementary base;

thus:

DNA ! RNA

Adenine ! UracilThymine ! AdenineGuanine ! CytosineCytosine ! Guanine

On the basis of this equivalence, if the DNA sequence is known, then the corresponding

RNA sequence can be inferred, and vice versa.

One of the basic units of genetic information in the genetic code is the codon, which is a

specific tri-nucleotide sequence (triplet). There are four nucleotide bases (‘letters’) which

can be arranged in three-letter combinations, making 64 possible codons (43 combinations)

(see Figure 2.5).

If the individual nucleotides can be thought of as the letters of the alphabet, then the

codons resemble ‘words’, most of which ‘represent’ a corresponding amino acid. The

exceptions are some codons called stop codons (UAG, UAA and UGA in RNA) which

identify were a polypeptide ends, and the single start codon (AUG in RNA) which marks

the start of a polypeptide-coding region in a sequence, and also corresponds to the amino

acid methionine. Thus, identifying the codon sequence in coding DNA or mature mRNA

can predict the polypeptide’s amino acid sequence. Only one grouping of the nucleotides

into codons in a gene results in a correct amino acid sequence in the corresponding protein

(this is referred to as the open reading frame). The loss or addition of a single nucleotide in

a sequence causes a frameshift mutation, which results in a different sequence of codons

(and, hence, a changed amino acid sequence). Frameshift mutations result in expression of

altered polypeptide sequences, which are often truncated due to the premature occurrence

of stop codons.

To make things a little more complicated (or interesting), it is worth remembering that

some amino acids are specified by more than one codon (termed redundancy; there are 20

basic amino acids and 64 available codons). Quite often, where an amino acid has

associated codon redundancy, an organism will more frequently use one or more codons

over the others to specify that particular amino acid (codon bias).


The assignment of amino acids to codons, sometimes called the universal genetic code, has

been developed by comparison of many DNA and protein sequences. However, some

organisms may routinely assign ‘unexpected’ amino acids to codons (nonstandard genetic

code). Both codon bias and nonstandard genetic code can present problems when trying to get

expression of genes in foreign hosts. As a general rule, for a cloned gene, if two nonstandard

codons are contiguous then protein expression will be suppressed to very low levels.

phen

yl-

alan

ine

leuc

ine

STOP

tyrosin

e

STOP

serin

e

cysteine

tryptophan

proline

arginine

histidineleucine

glutamine

isol

euci

ne

met

hion

ine

valine

lysine

aspara

gine

serine

arginineth

reon

ine

alanine

aspartic

acid

glutamic

acid

glycine

C UAG

CU

A GCU

CU A GCU

AG

C

U

AG

CU

AG

CU

AG

CU

AG

C

U

AG

A G

CU

AG

CU

AG

C UAGC U

A

GC

UA

GC

UA

G

G U

A C

G UC

AG

UC

AGU

C

AG

UC

A

Figure 2.5 The genetic code. The code has been illustrated by various formats: in tables,charts or, as shown here, as a wheel. This chart shows codons in RNA format (i.e. as they wouldbe represented in mRNA). The nucleotides are represented in sequence by the three shadedconcentric rings. The first nucleotide in the triplet is represented by the centre four sectors, thesecond by the next 16 sectors in the middle shaded ring, and the final position is designated bythe outermost shaded ring. Some amino acids have unique codons (methionine AUG, trypto-phan UGG), while others, such as arginine and threonine, are encoded by multiple codons(‘degeneracy’). Most organisms do not apply the code randomly when translating nucleic acidsto proteins, but instead ‘prefer’ to use a subset of codons for some amino acids; this is called‘codon bias’ and can sometimes cause difficulties in producing active enzymes from clonedDNA. The chosen host cell may have a codon bias which is not reflected in the clonedsequence; this can slow down or ‘stall’ the translation process and yield truncated or misfoldedpolypeptides (inactive enzyme). When this happens, various remedies can be adopted, includ-ing changing the host or changing the sequence artificially to suit the host’s codon preference(codon matching). The code illustrated here is often called the ‘universal code’; however, thereare differences sometimes seen in various classes of organism, or by certain individual species,or even in different subcellular compartments of eukaryotic cells. It is recognized, for example,that human mitochondrial DNA uses a slightly different code from that found in humangenomic DNA. Bioinformatics programs usually allow the researcher to specify which recog-nized version of the code is to be used when analysing DNA or RNA sequences to determinethe likely associated protein sequence from coding regions


Some DNA is referred to as noncoding, as it does not appear to specify a polypeptide. Large

regions of eukaryotic genome comprise noncoding intergenic sequences and noncoding

sequences which are present within gene sequences (such as introns). The protein-coding

regions of the mouse and human genomes is about 3 % of the total genome. Prokaryotes

typically have more ‘compact’ genomes, where protein-coding regions account for about 90 %

of the genome. Other types of genetic information are specified by the noncoding regions –

these are often related to regulatory functions, such as determining when the protein will be

produced and how much protein will be produced in a given set of conditions. The term motif

is often used for nucleic acid sequence specific information (both coding and noncoding

regions). However, the function of much of the noncoding DNA in eukaryotes is not known,

sometimes leading to it being referred to rather disparagingly as junk DNA.

2.3.1.4 Molecular Cloning and rDNA

When genetic information (nucleic acid) is transferred between different cells, species or

genera it is often carried by a specialized DNA molecule called a vector. Viruses are

natural vectors, as are some kinds of small independently replicating circular extra-

chromosomal DNA molecules (plasmids). A few of the basic features of plasmids used

in molecular biology are reviewed in Figure 2.6.

Commercial vectors are often derived from viruses or plasmids and now usually contain

highly modified control systems for maintaining, amplifying or expressing gene sequences

(‘cloned DNA’) in foreign host cells. Some vectors (shuttle plasmids) have been developed

to be transferred between different kinds of host. This can be very useful when genetic

manipulation is easy in one host (such as the laboratory favourite E. coli), but the functional

protein (enzyme) can only be expressed in another type of system (such as in a yeast).

Usually, the objective is to obtain a host cell which maintains the stability of the foreign

gene. This is achieved either by using a vector which the host will continue to replicate

separately from the host’s genomic (chromosomal) DNA or integrating the foreign DNA

into the host’s genome at a target site by a process known as homologous recombination.

When a stable genetic change has occurred in a cell due to the uptake of nucleic acid, this is

often referred to as transformation. If a virus were involved in the process of transferring the

information, this is sometimes termed transfection. rDNA is any form of DNA which has been

produced by the combining of DNA sequences which would not be found together in nature.

The resulting hybrid or chimeric DNA molecule is often simply called a construct. Organisms

which contain rDNA molecules are referred to as genetically manipulated organisms. Usually,

microbes are used for the production of enzymes using rDNA technology, and these then are

sometimes called genetically modified microorganisms. Cloning simply refers to making

many copies of something; informally, it can refer to rDNA technology which transfers copies

of foreign genes to another DNA molecule and/or biological host cell (gene cloning). A clone

usually refers to a subset of viable cells derived from a selection procedure to identify those

individual cells which contain the desired rDNA construct.

2.3.1.5 Basic Gene Cloning from DNA Templates

Perhaps one of the simplest methods for obtaining an enzyme by rDNA technology is by

using the well-established techniques of basic gene cloning, sometimes called shotgun

cloning. The method is relatively simple but can be time consuming. Although many of the


PLASMID1234bp

Multiple cloning site

ORI

AntibioticR

Promoter

Figure 2.6 Basic features of bacterial plasmids used in rDNA technology. An origin ofreplication (ORI) allows the plasmid DNA to be replicated by the host cell to ensure theplasmid’s propagation and survival. Plasmids can vary in copy number (number of plasmidsper cell). The copy number can in turn affect the amount of product expressed by a gene dosageeffect. The copy number of a vector can vary, and often the presence of a very large clonedinsert can severely reduce the plasmid copy number. Selectable markers (such as drug-resistance genes) are usually incorporated so that transformants can be selected on solidmedia and plasmids retained in culture. Controlled expression (production of protein fromcloned gene) can be achieved using a vector promoter. The promoters are usually inducible, sothat expression of the foreign gene is tightly controlled. This is very important in the early stagesof a gene cloning experiment, as uncontrolled or excessive expression of a foreign gene productcan sometimes be toxic for the host cell, resulting in loss of the clone. In order to be expressedunder the control of the vector promoter, the cloned sequence has to be inserted in the correctorientation (as indicated by the arrowhead) usually with the start codon a short defined distanceaway from the promoter sequence. Multiple cloning sites allow accurate insertion of foreignDNA. The sites often contain several restriction sites (short target sequences recognized byrestriction enzymes). Directional cloning is possible by using pairs of restriction sites to insertthe foreign DNA in the required orientation (usually so it can be under the control of the vectorpromoter). Most plasmids can accept inserts in the range 1–10 kbp, although often clones withsmaller inserts prove to be more stable than those containing large inserts. Where largefragments of DNA are to be cloned, then alternative vectors could include cosmids, yeastartificial chromosomes and bacterial artificial chromosomes. Other features often found inplasmids (or other vectors) include sites which allow direct cloning of PCR products oradditional genes or sequences which can be fused to the foreign gene (to generate taggedproteins with additional amino acids, or even fuse the target gene with another enzyme). Someplasmids also have additional features, such as mechanisms for colorimetric detection ofclones bearing inserts. ‘Suicide vectors’, which cannot be maintained autonomously in thecell, are also used to transform bacteria, yeasts and filamentous fungi. Plasmid components(including inserted genes) can only survive in transformants if they have successfully integratedinto the host cell’s genome. ‘Shuttle vectors’ contain multiple origins of replication andselectable markers which allow them to be maintained in different hosts. This could be twobacterial species or bacteria plus yeast (or fungus). Copy number, selectable marker andpromoter type are all important features to consider when choosing a plasmid for productionof an enzyme


procedures have been superseded by other more modern techniques, it still serves to

illustrate the basic cloning methods and is also still worth considering as a starting point

in gene discovery. This is especially relevant where there is information on an enzyme’s

activity but little or no sequence information available for the gene of interest.

The DNA template in this case is total DNA from the original organism which produced

the enzyme (for simplicity, preferably this will be a bacterium) (Figure 2.7). Usually, the

bacterial cell walls are digested to release multiple copies of very high molecular weight

DNA (genomic DNA). This genomic DNA is then broken up by physical shear force

(passing through a syringe needle, for example), or digested by special enzymes which cut

DNA at specific sites (restriction enzymes). The fragments are annealed (ligated) to a

linearized vector DNA molecule using an enzyme from E. coli (DNA ligase). The ligation

process can be more efficient if restriction enzymes have been used to generate the fragments

and the cut vector, as this generates complementary short DNA single-strand ‘overhangs’ at

the cut ends. These complementary ‘sticky ends’ can then anneal, stabilizing the hybrid

construct prior to ligation. The population of hybrid molecules obtained is called a library.

When the generation of inserted DNA fragments has been by random cutting or shearing it is

sometimes called a shotgun library and the whole process is called shotgun cloning.

Genomic DNAPhysicalshear Enzymatic digest

Vector

Enzymaticdigest

Enzymaticligation

Recombinant DNAMolecule

Cell

BASIC GENE CLONINGShotgun cloning experiment

Figure 2.7 Basic gene cloning I: shotgun cloning. Fragments of genomic DNA generated byphysical shear or enzymatic digest (using restriction enzymes) are mixed with predigestedplasmid vector. The ends of the linearized digested vector can be modified to reduce religationof the vector without insert. The religated vector (containing insert(s)) forms a library of rDNAmolecules. Similar methods can be used to clone DNA fragments obtained by other methods –those recovered directly from the environment, sequences modified by mutagenesis, or cDNAgenerated from mRNA by reverse transcription


Genes from eukaryotes (yeast, fungi, plants and animals) and some Archaea pose some

special cloning problems, as previously noted. The structure of genes in these organisms is

fundamentally different from those in most bacteria. The coding sequences of eukaryote

genes (‘exons’) are frequently interrupted by noncoding regions (called ‘introns’). Instead

of using the technique of direct shotgun cloning of genomic DNA, the mature mRNA from

these eukaryotic cells, which contains no introns, is converted to equivalent DNA mole-

cules called complementary DNA (cDNA) in vitro using reverse transcriptase. Libraries of

these cDNAs are then used in cloning experiments. The method has the advantage that

only protein-coding regions are cloned (avoiding the cloning of introns and other noncod-

ing DNA). On the downside, there is a bias towards cloning abundant transcripts (genes

which are actively undergoing transcription when the mRNA was harvested). Poorly

expressed genes might be difficult to detect, as there may be few or no copies of the

particular mRNA present under some circumstances. Sometimes, a large proportion of

transcripts are truncated and the upstream regulatory regions of the gene are not recovered

in the cDNA cloning process.

The DNA or cDNA library is then introduced into a preparation of bacterial host cells.

Usually, the first host selected is a laboratory strain of E. coli which has been grown and

pretreated with inorganic salts to make uptake of DNA easier. The ability to take up foreign

DNA is called competence; cells which have been specially prepared for the purpose are

called competent cells. Other methods to transfer DNA into cells include electroporation

(application of an external electric field to permeabilize the cell wall), transfection (where

a recombinant bacterial virus is used to transfer the DNA to the target cell) or ballistic

methods (by using DNA-coated particle projectiles). The last method has been used to

introduce foreign DNA into plant cells and mammalian cells.

After a brief incubation of the competent cells in contact with the DNA library to allow

uptake of the DNA, the bacterial cells are spread on Petri dishes containing sterile agar

which usually also contains an antibiotic for selection. The objective of selection is to

permit only those cells which have taken up the vector to grow; these cells are also more

likely to contain ‘foreign’ cloned genes. A common method to select colonies of bacteria

(clones) which contain the plasmid is to include a drug (antibiotic)-resistance marker on

the plasmid vector and then add the antibiotic to the cell growth such that all cells without a

plasmid are killed (Figure 2.8). Variations on this technique can be used to distinguish

between clones with or without inserts in the vector, or to select for clones with DNA

inserts of a defined size range.

The colonies which grow on the agar are selected for individual culture (‘purifica-

tion’) and subjected to various tests to check whether the gene of interest is present. The

most direct form of test is for enzyme activity (functional test), but this could prove

tedious, as a successful shotgun cloning can produce hundreds or thousands of colonies,

each to be tested for the presence of functional enzyme. Where it is possible to include

an enzyme assay directly (perhaps a colorimetric or fluorescence-based assay) in agar

culture plate or multiwell format, the screening process can be significantly easier. These

kinds of assay can be based on actual substrates reacting and inducing a measurable

change (such as a pH change) or on related ‘artificial’ substrates which can emit a signal

colour change or fluorescence when acted on by the enzyme. Artificial colorimetric

substrates are attractive because a huge number of colonies can be screened relatively

quickly, but they have the serious drawback that you are not directly selecting for the


activity of choice; and, as ‘you get what you screen for’, you may not be identifying the

optimal clones at this stage.

However, if the gene has been cloned, but the required activity is not produced, then the

functional test will fail to pick up the target gene. In this case, if some gene sequence

information is available, then it may be possible to test for the presence of DNA with the

expected sequence by hybridization with radio-labelled ‘probe’ DNA or, more usually, by

PCR. This sequence-based screening test could pick up ‘positives’ which have been

missed in the initial screen because the gene has been successfully cloned but the enzyme

has not been produced in an active form (perhaps because expression has not occurred or

because E. coli is a poor host to support production of active enzyme), or where there is no

convenient function-based assay available.

Other methods include detection of production of mRNA or protein in response to the

presence of the substrate (substrate-induced gene-expression screening). The success of

this method is dependent on the retention of native upstream regulatory regions of genes in

the clones which can switch on production of mRNA or enzyme in response to the stimulus

of the presence of substrate. The method is used in screening libraries derived by ‘random’

Vector + insert

(Library of recombinantDNA molecules)

Competent host cells

+

Growth onselective medium

Bacterial colonies containing plasmids(library of transform)

Transformation

BASIC GENE CLONINGTransformation and selection

Figure 2.8 Basic gene cloning II: transformation and selection. The library of rDNA moleculesgenerated in vitro needs to be introduced into a suitable host by a process known as transfor-mation. The uptake of DNA by the cells is enhanced by chemical treatments or by usingelectrical pulses. The cells now house the library. In order to distinguish between cells whichdo or do not contain the plasmid, most plasmids carry a selectable marker – a gene whichmodifies a characteristic of the host cells. Commonly used plasmids in basic cloning experi-ments often use antibiotic resistance as the marker. By incorporating the antibiotic into solidmicrobial growth medium (agar plates), only transformed cells containing the plasmid cangrow into colonies. Some other useful features of plasmid biology ensure that each survivingcell in a single colony will have the same individual recombinant plasmid. Individual coloniesare further analysed by functional tests of enzyme activity and/or by molecular screening for therequired insert. The transformed cells are a convenient storage for the library and for anyindividual clone selected for further study. Cultures (individual or pooled) are often routinelystored at �70 �C for several years without appreciable plasmid loss


cutting/shearing of genomic DNA into relatively large fragments, or where the cloned

DNA fragments have been directly isolated from the environment.

Both types of test could prove laborious, as many thousands of clones might be

produced in a single experiment. However, there are several strategies available which

could make the process simple, from using an initial selective growth test (functional

test – determine conditions where survival of positive clone depends on production of

active enzyme) to using pooled sub-libraries to reduce the number of sequence-based

tests needed to identify a positive clone. Where a colorimetric or fluorescent-based assay

is available it is also possible to use high-throughput screen methods based on cell sorting

or automated colony picking. These facilities are expensive, but are available on a

commercial basis for clone selection and are important when any high-throughput

experiments are being considered.

Some of the materials and techniques used in molecular biology may attract royalties if

used for commercial purposes. Vectors, host strains and off-the-shelf DNA manipulation

methods are usually readily available for modest licence fees for research purposes, but

additional licences would need to be sought (and fees paid) if these systems were used in a

commercial process. Where commercial exploitation is planned, the researchers should be

prepared to switch to royalty-free genetic systems and avoid the use of costly and

potentially toxic materials, such as artificial inducers or substrates, as gene expression

regulators.

2.3.1.6 PCR

PCR (polymerase chain reaction) is a technique widely used in molecular biology. Its

versatility as a technique for the manipulation of DNA and RNA sequences can probably

not be exaggerated. In its simplest form, a chain reaction can be developed in which a DNA

sequence template is exponentially amplified. PCR derives its name from one of its key

components, a DNA polymerase used to amplify a target section of DNA by in vitro

enzymatic replication. dsDNA recovered from genomic DNA or from vectors (plasmids)

normally has two complementary strands that can be separated into single-stranded DNA

(ssDNA) by heating (‘melting’) beyond the melting temperature Tm of the double-stranded

form. When the DNA is cooled, eventually the complementary bases find each other and

the double-stranded form is recovered (annealing). In PCR, short double-stranded sections

of DNA are generated using synthetic oligomers (primers) complementary to the

sequences flanking the target DNA sequence. The primers are typically 15–30 bases

long and are designed to anneal to the ssDNA. The synthetic primers will also eventually

form the termini of the amplified DNA segment. The DNA polymerase recognizes the

partial dsDNA sequences formed on annealing of the primers to the single-stranded

template and the enzyme initiates DNA replication, forming double-stranded strands

using the exposed single strand as the template. As the PCR cycles progress, the short

sections of DNA generated in previous cycles of replication are themselves used as

templates in successive cycles. ‘Normal’ bacterial DNA polymerase would not be able

to retain activity over the multiple temperature shifts of the reaction cycles, so the

sustained reaction is made possible by the use of a thermostable DNA polymerase with

a temperature optimum of about 70 �C (such as Taq polymerase from the thermophilic

bacterium T. aquaticus). Excess primer and deoxyribonucleotide triphosphates (dNTPs)


are also required in the reaction mix. Repeated cycles of denaturation, annealing and

extension are required to generate a population of DNA fragments, all of which are copies

of the target sequence flanked by the primer sequences.

Early PCR experiments were performed by manual or robotic transfer of reaction vials

between water baths or heating blocks. These have been superseded by several generations

of programmable thermal cyclers. Successful PCR amplifies a single or a few copies of a

target sequence of DNA by many orders of magnitude. The process is described schema-

tically in Figure 2.9.

The relatively simple basic format of PCR can be extensively modified to perform a

wide array of genetic manipulations. The reaction products can be directly modified by

altering the primer design (to introduce mutations or add ‘adaptamers’ to the ends of

sequences, for example). Control of the reaction conditions and choice of polymerase

enzymes can also alter the fidelity of the replication. The method can also be adapted to

anneal sections of DNA which have overlapping sequences. The ability to amplify specific

sequences based on a relatively small amount of sequence information (to enable primer

design) means that DNA fragments can be easily amplified from many environments and

from samples which may only have few copies of the target sequences – from genomic

DNA, from purified plasmids, directly from cells and from environmental samples (e.g.

clinical specimens, water, soil). As only a few copies of the target sequence are required,

the method is very sensitive, making it useful in amplification of sequences from mixed

populations (as in metagenomics) or for molecular fingerprinting (speciation, genotyping

and forensic applications). Careful experimental design can also modify the sensitivity and

Tem

per

atu

re0

5010

0

Repeat n times

Denature DNA

Anneal Primers

Extend primers

.

Figure 2.9 Schematic representation of a typical simple PCR reaction. The starting reactionmix contains dsDNA template, a pair of short ssDNA oligonucleotide primers (complimentaryto ends of target DNA sequence), a pool of the four dNTPs and a heat-resistant DNA poly-merase, e.g. Taq polymerase. dsDNA containing the target sequence is denatured to the single-stranded form by heating to �90–100 �C. The reaction is cooled to a few degrees below thecalculated annealing temperatures of the synthetic primers to the target DNA sequence. At theextension temperature (often 72 �C), Taq polymerase initiates the extension of the partialdsDNA segment using the single strand as a template. Successive denaturation, primer anneal-ing and extensions produces copies of the target. The process is repeated n times (typically n¼20–30 for a simple amplification experiment). Amplification factor: 2n


specificity of PCR. The versatility of the basic PCR technique in manipulation and analysis

of DNA make it central to many molecular biology procedures.

Many innovations in PCR technique have increased it versatility in identifying and

recovering enzyme genes. For example, it is possible to devise PCR-based methods to

recover genes based on limited information relating to short internal gene DNA sequences,

or by using amino acid sequences from purified proteins to predict and detect the actual

DNA sequence (this technique is sometimes called ‘reverse genetics’). Bioinformatics

tools can be used to generate ‘consensus sequences’ for the active site of specific families

of enzymes and the target DNA can then be probed for the presence of these sequences.

One of the most problematic steps encountered when generating libraries of DNA

sequences is often the enzymatic ligation of the DNA into vectors. Ligation can be the

most inefficient step in the entire process, and the enzymes (ligase) and cofactors required

are also relatively expensive. As a result, some of the diversity and complexity of the

library can easily be lost at this stage. Short (usually two to four nucleotides) complemen-

tary single-strand overhangs are generated by restriction enzymes. When these are

annealed together the hybrid molecules require in vitro stabilization by DNA ligase

prior to transformation of the microbial host. Ligation-independent cloning methods

based on modified PCR techniques generate complementary long sticky ends of 12–15

nucleotides on both the plasmid and the insert. The annealed insert/vector molecules

generated are sufficiently stable to be transformed directly into E. coli, where the DNA

backbone can be efficiently repaired by ligases in the host cell.25

2.3.2 Mutagenesis

2.3.2.1 Directed Evolution of Enzymes

The huge diversity of enzymes observed in nature is attributed to evolutionary processes

where diverse populations of gene variants (with sequence variations generated by muta-

tion and recombination) are subjected to selection of the ‘fittest’ enzyme functions. Those

genes which confer advantageous traits to their hosts are more likely to be maintained and

disseminated in populations than those that do not.

This process can be mimicked experimentally to modify enzymes by generating or

enhancing enzyme gene diversity via mutation and recombination and then devising

specific selection methods to identify ‘improved’ versions of the enzyme, which are

then amplified for further analysis and manipulation. The diversification, selection and

amplification can be thought of as the basic processes of directed evolution. Directed

evolution provides a powerful tool for the development of biocatalysts with novel proper-

ties, without requiring knowledge of enzyme structures or catalytic mechanism.26 The

initial step involves the identification and isolation of a ‘wild-type’ naturally occurring

gene responsible for encoding the desired enzyme. This requires the cloning of the relevant

gene into an efficient expression system before this target gene is subjected to random (or

rational) mutagenesis using the methods such as those described in these sections.

Improved variants are identified through screening of the function of the expressed

enzymes (preferably by a carefully designed high-throughput screening method). The

inferior enzymes and their genes are discarded and the improved genes are used as parents

for the next round of evolution by repeating the diversification and selection process as


often as necessary. The basis steps of the process are summarized in Figure 2.10. Applying

different techniques has resulted in the development of enzymes with improved properties

and production of ‘new’ enzymes with diverse properties, such as improved enantioselec-

tivity, activity, thermostability, protein solubility and expression.

2.3.2.2 Error-prone PCR

One of the commonest methods to achieve gene diversity by random mutagenesis is error-

prone PCR (epPCR). This technique exploits the fact that the thermostable polymerase

used for PCR has relatively low replication fidelity. It lacks a 30 to 50 exonuclease proof-

reading activity mechanism to replace any accidental mismatch in the newly synthesized

DNA strand and has an error rate measured at about 1 in 10 000 nucleotides. To enhance

this mutagenic effect, protocols have been developed with the aim of deliberately increas-

ing the error rate of Taq polymerase, which can be varied by increasing the concentration

of MgCl2, by addition of MnCl2 or by using unbalanced dNTP concentrations in the

reaction mix to achieve higher rates of mutations. Point mutations are the most common

types of mutation in epPCR, but deletions and frameshift mutations are also possible,

although rarer. Other ways of increasing the mutation rate can include the use of natural or

proprietary polymerases with enhanced mutation frequency27 and by incorporating syn-

thetic mutagenic dNTPs, such as 8-oxo-dGTP, which are then eliminated in a subsequent

ParentalGenes

Randommutagenesis

Bacterialtransformation

Error prone pcrMutator strainsrecombination

Mutant library

Screening and selection forimproved enzyme function

1St Generation mutated gene

Repeat to generateImproved enzymevariants

Figure 2.10 Generation of improved enzymes by directed evolution. The process starts withone or more parental genes which are subjected to mutagenesis – either by generation ofrandom point mutations or small insertions/deletions, and/or by recombination of gene frag-ments to generate libraries of mutants. In the next stage this library is screened for the desiredenzyme function. Ideally, this is combined with a selection or discrimination process as part ofa high-throughput process. In the third stage, the selected mutants are isolated and amplified(by propagation of the expression clone or by PCR). These selected first-generation mutantsthen undergo successive rounds of directed evolution, each time selecting for the favourablefeatures in the expressed enzymes. There are many variations on the technique: sometimes,selected gene domains (cassettes) are targeted for mutagenesis; iterative processes could beused in the generation of the mutant library, and specific sequence information may be used inthe targeting of the mutagenesis or in the screening protocols


PCR reaction using natural dNTPs.28 However, there are some drawbacks to epPCR.

Generally, the technique produces libraries of DNA fragments which need to be ligated

into expression plasmids (this can be a limiting step), and some of the methods do not

produce random mutations.

2.3.2.3 Cassette Mutagenesis and Mutator Strains

Another approach to mutagenesis is to restrict the mutagenesis of the gene to defined

regions or ‘cassettes’. These may have been targeted as key domains by other bioinfor-

matics or mutagenesis studies. Typically, a target region is excised by cutting the DNA at

two constructed or naturally occurring restriction enzyme sites that flank this region and

the excised portion is replaced with oligonucleotide(s) containing the desired mutation.

The mutagenesis target could be as small as a single codon (thus, changing a single amino

acid), and the type and degree of mutagenesis can be varied depending on the technique

used (epPCR, synthetic oligonucleotides or use of in vivo mutagenesis techniques, such as

mutator strains). Complete saturation of one or more positions in a protein with all possible

amino acid substitutions can be achieved with this method.

The technique has the disadvantage of being relatively expensive if large quantities of

synthetic nucleotides are required; and, depending on the methodology employed, ligation

steps may be required to recover the reconstructed gene. It can also be used in directed-

evolution experiments to generate libraries of genes which contain both conserved and

highly mutated domains.

Other useful random mutagenesis methods are based on introducing the target DNA into

a host cell which has error-prone DNA replication processes. The most popular of these is

probably the mutator strain method. Commercial strains, such as the E. coli XL1-Red

(Stratagene, La Jolla, CA), lack three of the primary DNA repair pathways, MutS, MutD

and MutT, resulting in a random mutation rate �5000-fold higher than in wild type. The

protocol for using the mutator strain is composed of two steps: transformation of the

mutator strain and recovery of the mutant from the transformant. In some ways the

technique is much simpler than epPCR, and ligation steps are eliminated as the mutated

gene is recovered in a plasmid vector. However, the actual mutation frequency can be

fairly low under the standard conditions (0.5 mutations per kilobase), and extended

cultivation periods are often required to introduce multiple mutations.

2.3.2.4 DNA Shuffling

In studying the mechanisms of gene evolution it is important to recognize the importance

of recombination of blocks of sequence, rather than point mutagenesis alone, in generating

sequence (and function) diversity. The DNA shuffling approach involves mixing of a

family of homologous sequences obtained from nature (typically the same gene from

related species or related genes from a single species) or from libraries of artificially

mutated genes which creates a large diversity of novel structure–function proteins. The

method shares some of the features of natural genetic recombination, in that new genes are

generated by combining sections from the ‘parental’ genes. A basic feature of gene

shuffling is that genes or fragments of genes are cut into fragments and reassembled as

chimeric molecules. The library of chimeric sequences is inserted into expression plasmids

and screened for desirable traits. The development of DNA shuffling by Stemmer in 1994


overcame some of the various drawbacks of random mutagenesis alone, in that a much

greater diversity of useful mutants could be generated by making chimeric genes.29

A simplified representation of DNA shuffling is given in Figure 2.11.

2.3.2.5 Combinatorial Methods

There has been a rapid expansion in the past decade in methods for creating libraries using

directed evolution by gene mixing techniques. Many combinations of mutagenesis and

shuffling have been developed. More sophisticated approaches have also included appli-

cation of statistical analysis of protein sequence–activity relationships to identify bene-

ficial mutations in early round variants (including variants with reduced activity) and then

combining these mutations by the incorporation of synthetic oligonucleotides with DNA

Starting genes

Gene fragments

‘Shuffled’ genelibrary

Denature and anneal

Extend

DNase 1

Extend

Figure 2.11 Gene shuffling to generate chimeric gene libraries. A ‘classical’ DNA-shufflingstrategy begins by fragmenting a pool of double-stranded parent genes randomly using partialenzymatic digest with DNase I. A further refinement can be included by selection of smallfragments by size fractionation to maximize the probability of multiple recombination eventsoccurring. The fragments are recombined in vitro by allowing annealing of homologoussequences. The short sections of dsDNA formed then can act in a similar way to conventionalPCR primers, allowing the fragments to ‘cross-prime’ each other in a round of primer-independent PCR amplification. Successive rounds of product annealing and amplificationgenerate a library of ‘shuffled’ genes. In order to facilitate expression cloning of the gene library,the full-length, diversified products are usually then modified by an additional round of PCRamplification with terminal primers to allow insertion of the sequences into expression vectors


shuffling. A multitude of techniques have been described based on combinations of

procedures to generate mutations, recombine gene fragments and isolate improved

mutants. Techniques include: staggered extension protocol, (StEP), iterative truncation

for the creation of hybrid enzymes (ITCHY), degenerate oligonucleotide gene shuffling

(DOGS), sequence homology-independent protein recombination (SHIPREC),

random chimeragenesis on transient template (RACHITT), synthetic shuffling, sequence-

independent site-directed chimeragenesis (SISDC), combination libraries enhanced by

recombination in yeast (CLERY) and THIO-ITCHY to name a few. Several methods

have been patented and the processes commercialized. We recommend the interested

reader to consult more specialist literature to get a deeper understanding of these techni-

ques. See Ref. 30, for example, for further reading.

2.3.2.6 Emerging Methods in Directed Evolution: Neutral Drift and Indels

In directed evolution the target mutation rates are orders of magnitude higher than those of

nature (greater than one mutation per gene per generation from approximately 1 in every

10�6 in most natural organisms). Enzymes tolerate most single mutations with no loss of

function, but their stability and the ability to tolerate more mutations is often severely

compromised. Enzymes used as starting points for laboratory evolution were never

evolved to withstand high mutational loads. Using more-stable enzymes (perhaps those

from thermophiles) or laboratory-evolved stabilized enzymes is predicted to give better

quality libraries. Another way is the neutral drift technique, where a starting library is first

evolved with mutations selected to maintain the protein’s original function; this has been

shown to be a way of generating small and highly effective libraries for directed evolu-

tion.31 Neutral-drift library sequence analysis has suggested that these mutations act by

enriching ‘global suppressor’ mutations which increase the enzyme stability and suppress

the effect of a broad range of otherwise destabilizing mutations. These mutations often

involve sequence changes which result in drift to ‘back-to-consensus’/ ‘ancestral enzyme’

sequences.

All these techniques create genetic diversity by recombination and point mutations and

are well developed. However, insertions and deletions (indels) are also important types of

mutation which are probably underrepresented in many conventional mutagenesis strate-

gies. Methods for incorporation of indels in predefined positions in a combinatorial

manner have been developed.32 Although there are some published studies on their use

in the directed evolution of biocatalysts,33 the full potential of these newer methods of gene

mutation for enzyme improvement are yet to be demonstrated.

2.3.2.7 Rational Enzyme (Re)design

The application of random mutagenesis or recombinatorial DNA shuffling methods to

genes coupled with screening and selection has frequently been successfully applied to

generate mutated enzymes with improved characteristics. These methods often do not

require any specific knowledge of the enzymes’ tertiary structure. However, a completely

‘random’ mutagenesis approach can generate huge numbers of variants to be screened in

directed-evolution experiments. Other approaches use a rational or semi-rational techni-

que to ‘target’ mutations to subsections of sequence or individual codons which are

associated with critical amino acids, such as those in or near the enzyme active site, for


example. In several instances, critical amino acid changes which can alter enzyme

performance have been first identified by random mutagenesis strategies. Often

these amino acids are at locations which are distant from the active site and their

importance would not have been inferred by knowledge of the protein’s structure. The

role of these distal residues in enzyme activity could then be further investigated by

saturation mutagenesis at these sites.34 However, where changes in the enzyme’s substrate

specificity or kinetics are sought, investigators often target the active site residues. The

modification of these limited numbers of residues generates much smaller libraries of

mutants. Individual amino acid changes can be investigated, or, as has been recently

described, combinations of relevant residues can be simultaneously mutated.35

In a technique termed ‘CASTing’ (combinatorial active site saturation test), small

subsets of active site residues (typically three residues) are subjected to saturation muta-

genesis (where every possible protein amino acid is substituted for the native one).

Beneficial mutations are re-entered into successive rounds of iterative saturation mutagen-

esis based on the same principles, with selection for improved performance. By this means,

several small catalytically diverse libraries can readily be generated and subjected to

successive rounds of directed evolution. Recent advances in enzyme engineering have

used a combination of the more ‘random’ methods of directed evolution with elements of

rational enzyme modification to try to overcome the limitations of both directed evolution

(very large libraries, hard to screen) and rational design (based on knowledge of enzyme

tertiary structure, which is frequently limited).36 Semi-rational approaches that target

multiple, specific residues to mutate on the basis of prior structural or functional knowl-

edge have the potential to create ‘smart’ libraries that are more likely to yield positive

results. Emerging techniques combine bioinformatics to model protein sequence–function

relationships and provide a rational basis for identifying ‘beneficial diversity’ to be

investigated in further rounds of enzyme evolution.37

2.3.3 Gene Synthesis

The entire process of identifying a gene based on sequence information, using PCR or

shotgun cloning to recover the entire sequence from DNA to generate a construct suitable

for expression and troubleshooting expression, can be a time-consuming and expensive

process. The increased availability of commercial oligonucleotide synthesis and sequen-

cing facilities has accelerated the pace of molecular biology research significantly in the

past decade, and this has recently been supplemented by the availability of relatively low-

cost gene synthesis services. No intact template DNA is required in this case, but it is

necessary to have actual sequence information (perhaps from a search of existing data-

bases or as a result of sequencing experiments) or a novel ‘hypothetical’ sequence based on

bioinformatics.

2.3.4 Overcoming Problems of Codon Bias

A good example where access to synthetic gene technology superseded laborious conven-

tional genetic manipulation is demonstrated in the case of Candida rugosa lipase 1

(lip1).38 C. rugosa (also known as Candida cylindracea) is a yeast which produces a

mixture of lipases (known as isoforms). In order to get access to a single pure lipase (lip1)

in significant quantities it was necessary to isolate the gene for this enzyme, clone and


express it in a suitable host. Unfortunately, C. rugosa has unconventional codon usage and,

more unfortunately, this yeast frequently uses the codon CUG to represent serine,39 even in

the Ser209 at the catalytic site. In most organisms this codon represents leucine; so, when

the unmodified lip1 gene was cloned and transferred into S. cerevisiae, an inactive enzyme

was produced – presumably because 17 out of the total 47 serines (including that at the

active site) had been replaced by leucine residues. The options for changing the sequence

of the gene to introduce serine codons which would be correctly translated by S. cerevisiae

or most other conventional hosts were (i) the laborious site-directed mutagenesis of the

individual serine codons in the cloned gene until active enzyme could be produced or (ii)

the generation of a new completely codon-optimized sequence by splicing synthetic

oligomers together to reconstruct the 1.7 kbp gene. Option (ii) proved to be the more

successful, not least because the ‘synthetic gene’ approach allowed several other mod-

ifications to be included which assisted in the cloning and expression of the enzyme (codon

optimization to match the host preference, removal of unwanted restriction sites).

A decade on, option (ii) would prove even more rapid and less expensive, as commercial

gene synthesis has become widely available.

Provided with DNA or protein sequence, commercial gene synthesis companies can

rapidly generate the complete sequence as dsDNA. The sequence can be optimized to

overcome codon bias and can be supplied inserted in an appropriate vector for the selected

expression host – bacterial, fungal, insect or mammalian. Site-specific mutagenesis is

usually also an option during the gene synthesis (by incorporating a range of nucleotides at

any specific site or sites), so a ‘library’ of specific mutants can also be supplied at the

outset. At the time of writing, a ‘typical’ enzyme gene of 2–3 kbp could be produced in an

expression-ready vector in a matter of days or weeks for under US$5000.

2.4 Microbiology and Fermentation

2.4.1 E. coli as an Expression Host: the Pros and Cons

E. coli is often the first choice as host for many research purposes due to the large amount

of commercial vector, transformation and expression systems available and because gene

expression is relatively well understood in this organism. It is also easy to cultivate, with

visible colonies growing from single cells in less than 24 h. However, E. coli frequently

presents several problems when it comes to producing large quantities of functional

purified enzyme. Foreign proteins are usually not exported by E. coli, so in order to obtain

purified enzyme it may be necessary to lyse the cells by using detergents and/or physical

shock (sonication, shear stress). Inside the cell, overexpressed proteins may have accu-

mulated in insoluble aggregates (inclusion bodies). Often, the inclusion bodies contain

misfolded (inactive) protein.

Various strategies can be employed to overcome the limitations of E. coli as an

expression host. These include increasing the copy number of the vector (and, therefore,

increasing the number of copies of the cloned gene per cell) and using various induction

methods and cell culture strategies to increase expression of functional protein.

Occasionally, the expressed foreign protein is itself toxic for the host, so expression

must be very carefully controlled. Various strategies can be adopted to overcome these

2.4 Microbiology and Fermentation 111

problems, including reducing the rate of production of proteins (lowering temperature,

using stricter induction controls). Proprietary strains of E. coli have also been developed

which have been genetically modified to overcome codon bias, promote correct folding,

post-translational modification and promote secretion of active enzyme.40 Inactive

enzyme derived from purified inclusion bodies can sometimes be refolded in vitro to

produce active enzyme. It may also be important to deal with issues of codon bias or

nonstandard genetic code, if necessary by codon matching of the introduced sequence

(substituting mutated or synthesized gene sequences with the appropriate ‘host’ codons for

those found in the ‘native’ sequence).

In addition, there are problems sometimes encountered when trying to maintain plas-

mids in scaled-up or extended cultures. These can be due to the need to avoid the use of

allergenic antibiotics and/or inherent plasmid instability (often accompanied by multi-

merization of the plasmids). These issues can sometimes be overcome by use of specia-

lized hosts, by careful selection of culture conditions to optimize plasmid retention or by

adoption of alternative selection mechanisms; for example, those based on complementing

auxotrophy. Auxotrophy is the inability of an organism to synthesize a particular organic

compound (often an amino acid or vitamin) required for its growth. By replacing the

missing gene on the transforming expression plasmid, the autotrophic cell acquires the

ability to grow in the absence of the amino acid or vitamin.41

2.4.2 Alternative Host Expression Systems: Bacteria, Yeasts, Filamentous Fungi

Alternative hosts are sometimes chosen for their ability to produce secreted products

(E. coli cells tend to produce intracellular proteins), to improve the probability of ‘correct’

translation and to incorporate post-translational processing of polypeptides (such as

glycosylation, folding, cross-linking and export). However, initial genetic manipulations

are often carried out in E. coli and the relevant construct then transferred into the

‘production’ host. For certain applications, such as when the product may be used in

pharmaceutical, food, feed or personal care markets, it may be preferable to use hosts

which are not associated with pathogenicity or toxicity (‘generally recognized as safe’–

examples include S. cerevisiae and Bacillus subtilis). Where large quantities of recombi-

nant enzyme are required, successive rounds of cloning experiments may be carried out to

place the gene in the appropriate vector, design a controlled expression system and choose

a host cell which can be grown on the appropriate industrial scale.

In order to overcome some of the difficulties associated with heterologous gene

expression in E. coli, various alternative microbial hosts have been developed (e.g.

Bacillus, Pseudomonas (DowPharma PfenexTM Expression Technology)). Various yeasts

(S. cerevisiae, Hansenula polymophia, Pichia pastoris, Yarrowia lipotytica) are frequently

used as an expression system for the production of proteins. A number of properties that

make Pichia suited for this task include its high growth rates and an ability to grow on a

simple, inexpensive medium. Pichia can be readily grown in either shake flasks or

fermenters, which makes it suitable for both small- and large-scale production.

Filamentous fungi such as Aspergillus species are also excellent hosts for the production

and export of enzymes. Although their genetic manipulation is initially a bit more time

consuming compared with yeasts or bacteria, they readily form stable transformants which

can be grown at large scale for industrial enzyme production.42


2.5 Summary, Overview and Future

While reactions catalysed by ‘wild-type’ enzymes will remain a basic tool kit for bioca-

talysis, the increased accessibility of gene manipulation (expression optimization, func-

tional optimization by mutagenesis) will make lower cost ‘bespoke’ enzymes more readily

available. Not only will a wider range of enzymes be available in the future, but the lead

times in process development may also be reduced as the catalyst (the enzymes and/or the

cell) can be more rapidly modified to suit the desired process conditions. Advances in

bioinformatics and improvements in reaction modelling could mean that, from biocatalyst

screening (discovery) right through to enzyme optimization, process scale-up could be

contracted into a few weeks. In conjunction with improvements in enzyme function, there

have been significant recent advances in improving host cells. Progress has already been

made in the redesign of microbial metabolic networks to generate strains optimized for

production of small molecules.43 Other research is focusing on the radical rational ‘strip-

ping down’ of microbial genomes to generate ‘simplified’ cells with the minimum number

of characterized functional genes – usually with the aim of enhancing protein (enzyme)

production by removing bottlenecks. These ‘minimum genome factories’ have been

proposed for various platform biotechnology strains, including those of E .coli and B.

subtilis.44 Further advancements in systems biology (advanced studies of complex biolo-

gical systems) could in the future help to establish ‘synthetic biology’ programmes –

construction and design of artificial biological system parts or whole organisms. Delivery

of these technological advances will depend on the multidisciplinary efforts of scientists

and engineers, but could produce reliable and efficient commodity and biocatalyst

manufacturing platforms of the future.

References

1. Peterson, D.H., Murray, H.C., Eppstein, S.H., Reineke, L.M., Weintraub, A., Meister, P.D. andLeigh, H.M., Microbiological transformations of steroids. I. Introduction of oxygen at carbon-11of progesterone. J. Am. Chem. Soc., 1952, 74, 5933–5936.

2. Woodward, R.B., Sondheimer, F., Taub, D., Heusler, K. and McLamore, W.M., The totalsynthesis of steroids. J. Am. Chem. Soc., 1952, 74, 4223–4251.

3. Roberts, S.M. and Wiletts, A.J., Introduction to Biocatalysis Using Enzymes and Micro-organisms, 2nd edition. Cambridge University Press, 1995.

4. Savile, C.K, Magloire, V.P. and Kazlauskas, R.J., Subtilisin-catalyzed resolution of N-acylarylsulfinamides. J. Am. Chem. Soc., 2005, 127, 2104–2113. Dawson, M.J., Mahmoudian, M.and Wallis, C.J., Process for preparing enantiomerically enriched N-derivatised lactams. 1999,WO/1999/010519.

5. Schomburg I., Chang A., Hofmann O., Ebeling C., Ehrentreich F. and Schomburg, D.,BRENDA: a resource for enzyme data and metabolic information. Trends Biochem. Sci.,2002, 27, 54–56.

6. Ellis L.B.M., Roe, D. and Wackett, L.P., The University of Minnesota Biocatalysis/Biodegradation Database: the first decade. Nucleic Acids Res., 2006, 34, 517–521.

7. Altschul, S.F., Gish, W., Miller, W., Myers, E.W. and Lipman, D.J., Basic local alignmentsearch tool. J. Mol. Biol., 1990, 215, 403–410.

8. Sigrist C.J.A., Cerutti L., Hulo N., Gattiker A., Falquet L., Pagni M., Bairoch A. and Bucher, P.,PROSITE: a documented database using patterns and profiles as motif descriptors. BriefBioinform., 2002, 3, 265–274.

References 113

9. Handelsman, J., Rondon, M.R., Brady, S.F., Clardy, J. and Goodman, R.M., Molecular biolo-gical access to the chemistry of unknown soil microbes: a new frontier for natural products.Chem. Biol., 1998, 5, 245–249. Handelsman, J., Metagenomics: application of genomics touncultured microorganisms. Microbiol. Mol. Biol. Rev., 2004, 68, 669–685.

10. Ferrer M., Golyshina O., Beloqui A. and Golyshin P.N., Mining enzymes from extreme environ-ments. Curr. Opin. Microbiol., 2007, 10, 207–214; Liu J.R., Yu B., Lin S.H., Cheng K.J. and ChenY.C., Direct cloning of a xylanase gene from the mixed genomic DNA of rumen fungi and itsexpression in intestinal Lactobacillus reuteri. FEMS Microbiol. Lett., 2005, 251, 233–241.

11. Bull, A.T., Microbial Diversity and Bioprospecting. ASM Press, 2004.12. Yun, J. and Ryu, S., Screening for novel enzymes from metagenome and SIGEX, as a way to

improve it. Microb. Cell Fact., 2005, 4, 8.Uchiyama, T., Abe, T., Ikemura, T. and Watanabe, K.,Substrate-induced gene-expression screening of environmental metagenome libraries forisolation of catabolic genes. Nat. Biotechnol., 2005, 23, 88–93.

13. Ji, S.C., Dockyu, K., Jung-Hoon,Y., Oh, T.-K. and Choong-Hwan,L., Sequence-based screeningfor putative polyketide synthase gene-harboring clones from a soil metagenome library.J. Microbiol. Biotechnol., 2006, 16, 153–157; Ginolhac, A., Jarrin, C., Gillet, B., Robe, P.,Pujic, P., Tuphile, K., Bertrand, H., Vogel, T.M., Perriere, G., Simonet, P. and Nalin, R.,Phylogenetic analysis of polyketide synthase I domains from soil metagenomic libraries allowsselection of promising clones. Appl. Environ. Microbiol., 2004, 70, 5522–5527.

14. Gabor, E.M., de Vries, E.J. and Janssen, D.B., Construction, characterization, and use of small-insert gene banks of DNA isolated from soil and enrichment cultures for the recovery of novelamidases. Environmental Microbiol., 2004, 6, 948–958.

15. Littlechild, J.A., Guy, J., Connelly, S., Mallett, L., Waddell, S., Rye, C.A., Line, K. and Isupov,M., Natural methods of protein stabilization: thermostable biocatalysts. Biochem. Soc. Trans.,2007, 35, 1558–1563.

16. Dalmau, E., Montesinos, J.L., Lotti, M. and Casas, C., Effect of different carbon sources onlipase production by Candida rugosa. Enzyme Microb. Technol., 2000, 26, 657–663.

17. Gonzalez-Lopez,C., Szabo, R., Blanchin-Rolanda,S. and Gaillardina, C., Genetic control ofextracellular protease synthesis in the yeast Yarrowia lipolytica. Genetics, 2002, 160, 417–427.Voigt, B., ThiHoi, L., Jurgen,B., Albrecht, D., Ehrenreich, A., Veith, B., Evers, S., Maurer, K.H.,Hecker, M. and Schweder, T., The glucose and nitrogen starvation response of Bacilluslicheniformis. Proteomics, 2007, 7, 413–423.

18. Better, M.D., Improved methods and bacterial cells for expression of recombinant proteinproducts. PCT Int. Appl., 2001, WO 2001073082.

19. Chater, K.F., Bruton, C.J., O’Rouke, S.J. and Wietzorrek, A.W., Methods and materials relatingto gene expression. 2002, EP1244799 (A1).

20. Hermanna, M., Kietzmanna, M.U., Ivancica, M., Zenzmaiera, C., Luitenb, R.G.M., Skrancc, W.,Wubbolts, M., Winklere, M., Birner-Gruenbergerf, R., Pichlera, H. and Schwab, H., Alternativepig liver esterase (APLE) – cloning, identification and functional expression in Pichia pastorisof a versatile new biocatalyst. J. Biotechnol., 2008, 133, 301–310.

21. Osprian, I., Fechter, M.H. and Griengl, H., Biocatalytic hydrolysis of cyanohydrins: an efficientapproach to enantiopure �-hydroxy carboxylic acids. J. Mol. Catal. B: Enzymatic, 2003, 24–25,89–98.

22. Reisinger, C., Osprian, I., Glieder, A., Schoemaker, H.E, Griengl, H. and Schwab, H., Enzymatichydrolysis of cyanohydrins with recombinant nitrile hydratase and amidase from Rhodococcuserythropolis. Biotechnol. Lett., 2005, 26, 1675–1680.

23. Burton, S.G., Cowan, D.A. and Woodley, J.M., The search for the ideal biocatalyst. Nat.Biotechnol., 2002, 20, 37–45.

24. Allison, L.A. Fundamental Molecular Biology, Wiley-Blackwell, 2006. Kreuzer, H. and Massey,A., Molecular Biology and Biotechnology: A Guide for Students, 3rd edition. John Wiley &Sons, Ltd, 2008.

25. Aslanidis, C. and de Jong, P.J., Ligation-independent cloning of PCR products (LIC-PCR).Nucleic Acids Res., 1990, 18, 6069–6074.

26. Turner, N., Directed evolution of enzymes for applied biocatalysis Trends Biotechnol., 2003, 21,474–478; Johannes, T.W. and Zhao, H., Directed evolution of enzymes and biosyntheticpathways. Curr. Opin. Microbiol., 2006, 9, 261–267.


27. Examples of proprietary enzymes include Mutazyme II DNA Polymerase (www.stratagene.com). Other companies have developed ‘kits’ containing selections of reagentswhich can be combined to yield controlled error rates in PCR reactions.

28. Zaccolo, M., Williams, D.M., Brown D.M. and Gherardi, E., An approach to random mutagen-esis of DNA using mixtures of triphosphate derivatives of nucleoside analogues. J. Mol. Biol.,1996, 255, 589–603.

29. Stemmer, W.P., DNA shuffling by random fragmentation and reassembly: in vitro recombina-tion for molecular evolution. Proc. Natl. Acad. Sci. U. S. A., 1994, 91, 10747–10751.

30. Arnold, F.H. and Volkov, A.A., Directed evolution of biocatalysts. Curr. Opin. Chem. Biol.,1999, 3, 54–59. Thomas, J.M. and Raja, R., Designing catalysts for clean technology, greenchemistry, and sustainable development. Annu. Rev. Mater. Res., 2005, 35, 315–350. Kaur, J.and Sharma, R., Directed evolution: an approach to engineer enzymes. Crit. Rev. Biotechnol.,2006, 26, 165–199.

31. Gupta, R.D. and Tawfik, D.S., Directed enzyme evolution via small and effective neutral driftlibraries. Nat. Methods, 2008, 5, 939–942.

32. Fujii, R., Kitaoka, M. and Hayashi, K., RAISE: a simple and novel method of generating randominsertion and deletion mutations. Nucleic Acids Res., 2006, 34, 30.

33. Bershtein, S. and Tawfik, D.S., Advances in laboratory evolution of enzymes. Curr. Opin.Chem. Biol., 2008, 12, 151–158.

34. Parikh, M.R. and Matyoumara, I., Site-saturation mutagenesis is more efficient thanDNA shuffling for the directed evolution of �-fucosidase. J. Mol. Biol., 2005, 352,621–628.

35. Reetz, T., Wang, L.-W. and Bocola, M., Directed evolution of enantioselective enzymes:iterative cycles of CASTing for probing protein sequence space. Angew. Chem. Int. Ed., 2006,45, 1236–1241.

36. Chica1, R.A., Doucet, N. and Pelletier, J.N., Semi-rational approaches to engineering enzymeactivity: combining the benefits of directed evolution and rational design. Curr. Opin.Biotechnol., 2005, 16, 378–384.

37. Fox, R.J. and. Huisman, G.W., Enzyme optimization: moving from blind evolution to statisticalexploration of sequence–function space. Trends Biotechnol., 2008, 26, 132–138.

38. Brocca, S., Schmidt-Dannert, C., Lotti, M., Alberghina, L. and Schmid, R.D., Design, totalsynthesis, and functional overexpression of the Candida rugosa lip1 gene coding for a majorindustrial lipase. Protein Sci., 1998, 7, 1415–1422.

39. Kawaguchi, Y., Honda, H., Taniguchi-Morimura,J. and Iwasaki, S., The codon CUG is read asserine in an asporogenic yeast Candida cylindracea. Nature, 1989, 341, 164–166.

40. Choi, J.H. and Lee, S.Y., Secretory and extracellular production of recombinant proteins usingEscherichia coli. Appl. Microbiol. Biotechnol., 2004, 64, 625–635.

41. Vidal, L., Pinsach, J., Striedner, G., Caminal, G. and Ferrer, P. Development of an antibiotic-freeplasmid selection system based on glycine auxotrophy for recombinant protein overproductionin Escherichia coli. Biotechnol., 2008, 134, 127–36.

42. Punt, P., van Biezen, N., Conesa, A., Albers, A., Mangnus, J. and van den Hondel, C.,Filamentous fungi as cell factories for heterologous protein production. Trends Biotechnol.,2002, 20, 200–206.

43. Pharkya, P., Burgard, A.P. and Maranas, C.D., OptStrain: a computational framework forredesign of microbial production systems. Genome Res., 2004, 14, 2367–2376.

44. Ara, K., Ozaki, K., Nakamura, K., Yamane, K., Sekiguchi, J. and Ogasawara, N., Bacillusminimum genome factory: effective utilization of microbial genome information. Biotechnol.Appl. Biochem., 2007, 46, 169–178; Mizoguchi, H., Mori, H. and Fujio, T., Escherichia coliminimum genome factory. Biotechnol. Appl. Biochem., 2007, 46, 157–167; Mizoguchi, H.,Mori, H., Fujio, T., Posfai, G., Plunkett, G., Feher, T., Frisch, D., Keil, G.M., Umenhoffer, K.,Kolisnychenko, V., Stahl, B., Sharma, S.S., de Arruda, M., Burland, V., Harcum, S.W. andBlattner, F.R., Emergent properties of reduced-genome Escherichia coli. Science, 2006, 312,1044–1046.Morimoto, T., Kadoya, R., Endo, K., Tohata, M., Sawada, K., Liu, S., Ozawa, T.,Kodama, T., Kakeshita, H., Kageyama, Y., Manabe, K., Kanaya, S., Ara, K., Ozaki, K. andOgasawara, N., Enhanced recombinant protein productivity by genome reduction in Bacillussubtilis. DNA Res., 2008, 15, 73–81.

References 115

3

Kinetic Resolutions UsingBiotransformations

3.1 Stereo- and Enantio-selective Hydrolysis of rac-2-Octylsulfate UsingWhole Resting Cells of Pseudomonas spp.Petra Gadler and Kurt Faber

Sulfatases are a heterogenic group of hydrolytic enzymes which catalyse the cleavage of

the sulfate ester bond yielding the corresponding alcohol and hydrogen sulfate. In contrast

to the more commonly employed hydrolytic enzymes, such as proteases, esterases

and lipases, they show not only enantioselectivity – by preference of a given substrate

enantiomer over its mirror-image counterpart – but also stereoselectivity with respect to

the stereochemical course of their action. Depending on the enzyme, sulfate ester hydro-

lysis may proceed either through retention or inversion of configuration at the chiral

carbon centre (Scheme 3.1). Whereas breakage of the S�O bond leads to retention, C�O

bond cleavage results in inversion of configuration (Scheme 3.1).1 The rare feature of

double selectivities makes them top candidates for the deracemization6 of sec-alcohols via

enantio-convergent hydrolysis of their corresponding sulfate esters.2

Overall, retaining sec-alkylsulfatase activity has been detected in Planctomycetes spp.

(such as Rhodopirellula baltica DSM 10527;3 complementary inverting sulfatase activity

was found in Actinomycetes (e.g. Rhodococcus ruber DSM 445412,4), Archaea (e.g.

Sulfolobus spp.5) and pseudomonads.7

Among the latter group, Pseudomonas sp. DSM 6611 was identified as top candidate by

displaying excellent stereo- and enantio-selectivities for a range of sec-alkyl sulfate esters

by transforming the (R)-enantiomer of the rac-sulfate ester into the corresponding

(S)-alcohol (Scheme 3.2).7



3.1.1 Procedure 1: Biocatalyst Preparation

3.1.1.1 Materials and Equipment

• Pseudomonas spp. DSM 6611 and 6978 and Rhodococcus ruber DSM 44541 were

obtained from DSMZ (Deutsche Stammsammlung fur Mikroorganismen und

Zellkulturen, Braunschweig, Germany, www.dsmz.de)

• YPG medium comprising:

– yeast extract (10 g L�1)

– bacteriological peptone (10 g L�1)

– glucose (10 g L�1)

– MgSO4�2H2O (0.15 g L�1)

– NaCl (2 g L�1)

– K2PO4 (4.4 g L�1)

– Na2HPO4 (1.3 g L�1)

• phosphate buffer (50 mM, pH 7.5; 7.58 g L�1 Na2HPO4�2H2O and 1.01 g L�1

KH2PO4)

• agar plates

• freeze drier.

3.1.1.2 Procedure

1. Pseudomonas spp. DSM 6611 and 6978 and Rhodococcus ruber DSM 44541 were

cultivated in shaking flasks for 3 days at 30 �C with shaking at 120 rpm in YPG

medium containing 10 g yeast extract, 10 g bacteriological peptone, 10 g glucose,

R2R1

3–

R2R1R2R1

H O SO H OHHO H

SN at CarbonBreakage of

C-O bond

+ HSO4 + HSO4–

[OH–]

[OH–]

Inversion

SN at SulfurBreakage of S-O bond

Retention

–

Scheme 3.1 Enzymatic hydrolysis of alkylsulfate esters catalysed by alkylsulfatases proceed-ing through retention or inversion of configuration

(S)-2-octanol

OSO3– OH

OSO3–

Pseudomonas spp.

E >200Buffer pH 7.5

OSO3–

rac -2-octylsulfate (S)-2-octylsulfate

Scheme 3.2 Enantioselective microbial hydrolysis of rac-2-octyl sulfate using whole restingcells of Pseudomonas spp. through inversion of configuration

118 Kinetic Resolutions Using Biotransformations

0.15 g MgSO4�2H2O, 2 g NaCl, 4.4 g K2HPO4 and 1.3 g Na2HPO4 per litre.

Precultures were grown either on agar plates or in liquid medium for 2–3 days as

described above.

2. Cells were harvested by centrifugation for 20 min at 4 �C and 8000 rpm, washed

with 50 mM pH 7.5 phosphate buffer and lyophilized. Lyophilized cells were stored

at 4 �C.

3.1.2 Procedure 2: Microbial Hydrolysis of rac-2-Octylsulfate


• Lyophilized whole cells of Pseudomonas spp. DSM 6611, DSM 6978 or Rhodococcus

ruber DSM 44541 (50 mg)

• tris-HCl buffer (600 mL, pH 7.5, 100 mM)

• stock solution of substrate rac-2-octylsulfate4 (50 mg mL�1 in 100 mM tris-HCl buffer

pH 7.5)

• ethyl acetate (600 mL)

• stock solution of internal standard (10 mg mL�1 rac-2-dodecanol)

• Na2SO4 anhydrous.

• acetic anhydride (60 mL)

• 4-dimethylaminopyridine (DMAP) (cat.)

• Eppendorf vials (1.5 mL)

• thermoshaker

• gas chromatograph (GC) equipped with a flame ionization detector (FID).

3.1.2.2 Analytics

Progress of the reaction was monitored using a GC equipped with a FID on an

achiral CP 1301 capillary column (30 m � 0.25 mm � 0.25 mm film) and N2 as

carrier gas. Enantiomeric purity of 2-octanol was analysed after derivatization

with acetic anhydride (see below) using a CP-Chirasil Dex-CB column (25 m �0.32 mm � 0.25 mm film, column B) and H2 as carrier gas. Enantioselectivities

(expressed as the enantiomeric ratio E) were calculated from enantiomeric excess of

the product and conversion as previously reported.8 Retention times and methods are

listed in Table 3.1.

Table 3.1 GC methods and retention times of 2-octanol

Column Retention time (min)

rac-2-Octanol

(S)-2-Octanol

(R)-2-Octanol

rac-2-Dodecanol

(S)-2-Dodecanol

(R)-2-Dodecanol

CP1301a 4.4 — — 7.2 — —DEX-CBb — 10.8 12.8 — 18.6 18.9

a14.5 psi N2 100 �C/hold 3 min – 50 �C min�1 – 240 �C/hold 3 min.b14.5 psi H2 60 �C/hold 7 min, 4 �C min�1 – 80 �C, 10 �C min�1 – 160 �C, 10 �C min�1 – 170 �C/hold 5 min.

3.1 Hydrolysis of rac-2-Octylsulfate Using Pseudomonas 119

3.1.2.3 Procedure

1. Lyophilized whole cells (50 mg) of Pseudomonas spp. DSM 6611, DSM 6978 or

Rhodococcus ruber DSM 44541 were rehydrated in tris-HCl buffer (600 mL, pH 7.5,

100 mM) for 1 h at 30 �C with shaking at 120 rpm.

2. An aliquot (200 mL) from a substrate stock solution (50 mg mL�1) was added. The

mixture was incubated at 30 �C with shaking at 120 rpm for 24 h.

3. The samples were extracted with ethyl acetate (600 mL) and centrifuged at 13 000 rpm

for 2 min to separate the organic layer from the cell/buffer suspension. The organic

layer was dried over Na2SO4 and 100 mL of an internal standard (10 mg mL�1 rac-2-

dodecanol) was added.

4. Conversions were measured on an achiral GC column using calibration curves. For the

determination of the enantiomeric excess, the 2-octanol formed was derivatized into the

corresponding acetate ester using acetic anhydride (60 mL) and catalytic DMAP over-

night. The reaction was quenched with tap water (300 mL), centrifuged for 2 min at

13 000 rpm and the organic layer dried (Na2SO4) and analysed as described above.

Results are listed in Table 3.2.

Table 3.2 Conversion and enantioselectivities (E-values) for the microbial hydrolysisof rac-2-octylsulfate

Strain (whole cells) Time (h) Conversion (%) Ee (S)-2-octanol (%) E-value

Pseudomonas sp. DSM 6611 24 21 99 >200Pseudomonas sp. DSM 6978 24 7 99 >200Rhodococcus ruber DSM 44541 24 23 77 10

References

1. Gadler, P. and Faber, K., New enzymes for biotransformations: microbial alkyl sulfatasesdisplaying stereo- and enantioselectivity. Trends Biotechnol., 2007, 25, 83.

2. Wallner, S.R., Pogorevc, M., Trauthwein, H. and Faber, K., Biocatalytic enantioconvergentpreparation of sec-alcohols using sulfatases. Eng. Life Sci., 2004, 4, 512.

3. Wallner, S.R., Bauer, M., Wurdemann, C., Wecker, P., Gloeckner, F.O. and Faber, K., Highlyenantioselective sec-alkyl sulfatase activity of the marine planctomycete Rhodopirellula balticashows retention of configuration. Angew. Chem. Int. Ed., 2005, 44, 6381.

4. Pogorevc, M. and Faber, K., Enantioselective stereoinversion of sec-alkyl sulfates by an alkyl-sulfatase from Rhodococcus ruber DSM 44541. Tetrahedron Asymm., 2002, 13, 1435.

5. (a) Wallner, S.R., Nestl, B.M. and Faber, K., Highly enantioselective sec-alkyl sulfatase activityof Sulfolobus acidocaldarius DSM 639. Org. Lett., 2004, 6, 5009; (b) Wallner, S.R., Nestl, B.M.and Faber, K., Highly enantioselective stereo-inverting sec-alkylsulfatase activity of hyperther-mophilic Archaea. Org. Biomol. Chem., 2005, 3, 2652.

6. (a) Faber, K., Non-sequential processes for the transformation of a racemate into a singlestereoisomeric product: proposal for stereochemical classification. Chem. Eur. J., 2001, 7,5004; (b) Gadler, P., Glueck, S.M., Kroutil, W., Nestl, B.M., Larissegger-Schnell,B.,Ueberbacher, B.T., Wallner, S.R. and Faber, K., Biocatalytic approaches for the quantitativeproduction of single stereoisomers from racemates. Biochem. Soc. Trans., 2006, 34, 296.

7. Gadler, P. and Faber, K., Highly enantioselective biohydrolysis of sec-alkyl sulfate esters withinversion of configuration catalysed by Pseudomonas spp. Eur. J. Org. Chem., 2007, 5527.

8. Chen, C.-S., Fujimoto, Y., Girdaukas, G. and Sih, C.J., Quantitative analysis of biochemicalkinetic resolutions of enantiomers. J. Am. Chem. Soc., 1982, 104, 7294.


3.2 Protease-catalyzed Resolutions Using the 3-(3-Pyridine)propionylAnchor Group: p-ToluenesulfonamideChristopher K. Savile and Romas J. Kazlauskas

Proteases require water-soluble substrates and bind them in a shallow active site in an

extended conformation.1 The shallow active site allows proteases to accept sterically

hindered substrates2,3 and also polar substrates, since one substituent remains in water.4

To bind substrates, proteases contain a specificity pocket for the acyl group.5,6 For

example, subtilisins and chymotrypsin favour ester and amides of phenylalanine.5–7 The

3-(3-pyridine)propionyl group mimics phenylalanine and increases substrate binding

and solubility in water, thereby increasing the rates of protease-catalyzed reactions. In

addition, the 3-(3-pyridine)propionyl group eliminates chromatography, since mild acid

extraction separates the remaining starting material and product. To demonstrate the

synthetic usefulness of this strategy, we resolved multi-gram quantities of (R)- and

(S)-p-toluenesulfinamide with �-chymotrypsin.8

3.2.1 Procedure: Resolution of N-3-(3-Pyridine)propionyl-p-tolylsulfinamide

N

O

OH

N

O

O

O

N

H3C

S

O

N

NH

carbodiimide

H3C SO

NH2

NaH

20.2g

α-chymotrypsinH2O E=52

H3C S(R)

NH2

O

H3C

S

O

N

NH

O

+

(1) H Extraction(2) NH2NH2

H3C S(S) NH2

O

3.58 g 33 % yield98 % ee af terrecrystallization

3.81 g 35 % yield98 % ee af terrecrystallization

1

1a

(R)-1

(S)-1

O


• N-(3-Dimethylaminopropyl)-N0-ethylcarbodiimide hydrochloride (25.2 g, 132 mmol)

• triethylamine (37 mL)

• 3-(3-pyridine)propionic acid (40.0 g, 265 mmol)

• CH2Cl2 (3250 mL)

• p-toluenesulfinamide (15.5 g, 100 mmol)

3.2 Protease-catalyzed Resolutions Using p-Toluenesulfonamide 121

• sodium hydride (60 % dispersion in oil; 12.0 g, 300 mmol)

• tetrahydrofuran (THF, 750 mL)

• EtOAc (1550 mL)

• hexanes (150 mL)

• aqueous saturated NaHCO3 (1400 mL)

• MgSO4 anhydrous

• bovine �-chymotrypsin, Sigma (12 g)

• N,N-bis(2-hydroxyethyl)-2-aminoethanesulfonic acid (BES, 0.7 g)

• potassium chloride (23.4 g)

• dimethylformamide (350 mL)

• NaOH, 1.0 M (35 mL)

• aqueous saturated NaCl (750 mL)

• HCl, 0.10 M (200 mL)

• hydrazine hydrate (35 mL)

• HCl, 1.0 M (50 mL)

• three-neck round-bottom flask, 250 mL and 2 L with magnetic stirbar

• glass stopper (24/40)

• rubber septum (24/40)

• cannula (18 gauge)

• syringe and needle (18 gauge)

• ice bath

• magnetic stir plate

• graduated cylinder, 1 L

• separatory funnel, 500 mL and 2 L

• Erlenmeyer flask, 1 L and 2 L

• glass funnel

• filter paper

• round-bottom flask, 1 L and 2 L

• rotary evaporator

• beaker, 4 L with magnetic stir bar

• dropping funnel, 500 mL

• pH Stat (Radiometer Titralab TIM854 or equivalent).

3.2.1.2 Procedure

3-(3-Pyridine)propionic Acid Anhydride.

1. N-(3-Dimethylaminopropyl)-N0-ethylcarbodiimide hydrochloride (25.2 g, 132

mmol) was added to a solution of 3-(3-pyridine)propionic acid (40.0 g, 265

mmol) in CH2Cl2 (750 mL) and Et3N (37 mL) at 0 �C and stirred at room

temperature (RT).8,9

2. After 24 h, the reaction was washed with ice-cold saturated NaHCO3 (3 � 500 mL),

dried over MgSO4 and concentrated in vacuo to give a pale yellow oil (34.1 g, 91 %).1H NMR � 2.72 (t, J¼ 7.2, 2H, C(O)CH2), 3.03 (t, J¼ 7.5, 2H, CH2Ph), 7.24 (m, 1H,

pyridyl), 7.58 (m, 1H, pyridyl), 8.50 (m, 2H, pyridyl).


Racemic N-3-(3-Pyridine)propionyl-p-toluene sulfinamide 1a.

1. Sodium hydride (60 % dispersion in oil; 12.0 g, 300 mmol) was added portion-wise

over 15 min to a solution of p-toluenesulfinamide (15.5 g, 100 mmol) in THF (750 mL)

at 0 �C. The symmetrical anhydride of 3-(3-pyridine)propionic acid (32.1 g, 113 mmol)

was added drop-wise over 15 min at 0 �C and the reaction mixture was then stirred at

RT for 3 h.10

2. The reaction mixture was diluted with EtOAc (400 mL) and saturated NaHCO3 (400

mL) was added slowly. The layers were separated and the aqueous layer was extracted

with EtOAc (3 � 250 mL). The combined EtOAc layers were washed with saturated

NaHCO3 (500 mL) and dried over MgSO4. The aqueous layer was extracted with

CH2Cl2 (2 � 250 mL). The combined CH2Cl2 layers were washed with NaHCO3 (250

mL) and dried over MgSO4. The combined organic layers were concentrated in vacuo

to give a pale yellow solid. Trituration with hexane/ethyl acetate gave 1a as a white

powder (21.1 g, 73 %).

Mpt. 161–163 �C; 1H NMR � 2.39 (s, 3H, PhCH3), 2.72 (m, 2H, C(O)CH2), 3.01

(t, J ¼ 7.2, 2H, CH2Pyr), 4.78 (br s, 1 H, NH), 7.25–7.48 (m, 3H, phenyl or pyridyl),

7.47 (m, 2H, phenyl or pyridyl), 7.68 (m, 1H, phenyl or pyridyl), 8.25 (m, 2H, pyridyl);13C NMR (DMSO-d6) � 21.4 (PhCH3), 27.9 (CH2Pyr), 36.9 (C(O)CH2), 124.0, 125.4,

130.2, 136.4, 136.6, 140.9, 142.1, 147.9, 150.2 (phenyl or pyridyl), 173.8 (C¼O);

HRMS calc. for C15H17N2O2S [MþH]þ 289.1010. Found: 289.0989. The enantiomers

were separated using HPLC (Chiralcel OD-H column, 85:15 hexanes/EtOH, 0.75 mL

min�1, 254 nm; (R)-enantiomer tR ¼ 20.0 min; (S)-enantiomer, tR ¼ 22.5 min).

Resolution of N-3-(3-Pyridine)propionyl-p-tolylsulfinamide.

1. �-Chymotrypsin (12 g) was added to a solution of BES buffer (3.15 L, 1 mM, pH 7.2)

and 100 mM KCl and stirred for 15 min to ensure complete dissolution. Substrate 1a

(20.2 g, 70 mmol) was dissolved in dimethylformamide (350 mL) and added drop-wise

to the enzyme solution. The rate of hydrolysis was monitored by pH Stat, which

maintained the pH at 7.2 by automatic titration with 1 M NaOH.

(i) Subtilisin BPN0 or subtilisin E are better proteases for this resolution because they

are more enantioselective, but they require a fermentation to produce.11 As an

alternative, we chose a commercially available, but less enantioselective, protease –

�-chymotrypsin.

(ii) If a pH Stat is not available, increase the buffer concentration from 1 mM to 50 mM

and maintain the pH at 7.2 by manual addition of 1 M NaOH. Approximately 35 mL

will be required.

2. At �50 % conversion (4 days), the reaction was terminated by extraction of remaining

starting material and product with CH2Cl2 (3 � 500 mL). The combined organic layers

were washed with H2O (3 � 500 mL), saturated NaCl (1 � 500 mL), dried over

MgSO4 and concentrated in vacuo. The crude mixture was dissolved in EtOAc

(250 mL) and unreacted starting material was extracted with ice-cold 0.1 M HCl

(2 � 100 mL). The combined aqueous layers were then back-extracted with EtOAc

(50 mL). The combined EtOAc layers were washed with saturated NaHCO3 (100 mL),

saturated NaCl (100 mL), dried over MgSO4 and concentrated in vacuo to give (R)-1 as

a white solid (4.48 g, 41 % yield) with 87 % ee.

3.2 Protease-catalyzed Resolutions Using p-Toluenesulfonamide 123

3. The combined aqueous layers were neutralized with solid NaHCO3 and extracted with

CH2Cl2 (2 � 200 mL). The combined CH2Cl2 layers were washed with saturated

NaHCO3 (100 mL), saturated NaCl (100 mL) and dried over MgSO4. The solution was

concentrated in vacuo to give (S)-1a, which was subsequently treated with hydrazine

hydrate (35 mL).12 After stirring for 3 h, the reaction solution was diluted with CH2Cl2(100 mL) and washed with 1 M HCl (50 mL), saturated NaHCO3 (50 mL), saturated

NaCl (50 mL) and concentrated in vacuo to give (S)-1 (4.29 g, 40 % yield) with 92 % ee.

4. Recrystallization from hexanes/ethyl acetate gave (R)-1 (3.81 g, 35 % yield) with 98%

ee and (S)-1 (3.58 g, 33 % yield) with 98 % ee.

References

1. Tyndall, J.D.A., TessaNall, T. and Fairlie, D.P., Proteases universally recognize beta strands intheir active sites. Chem. Rev., 2005, 105, 973–1000.

2. (a) Muchmore, D.C., Enantiomeric enrichment of (R,S)-3-quinuclidinol. US Patent US5,215,918, 1993; (b) Savile, C.K., Magloire, V.P. and Kazlauskas, R.J., Subtilisin-catalyzedresolution of N-acyl arylsulfinamides. J. Am. Chem. Soc., 2005, 127, 2104–2113.

3. Mugford, P.F., Lait, S.M., Keay, B.A. and Kazlauskas, R.J., Enantiocomplementary enzymaticresolution of the chiral auxiliary cis,cis-6-(2,2-dimethylpropanamido)spiro-[4.4]nonan-1-ol andthe moleuclar basis for the high enantioselectivity of subtilisin Carlsberg. ChemBioChem, 2004,5, 980–987.

4. Savile, C.K. and Kazlauskas, R.J., How substrate solvation contributes to the enantioselectivityof subtilisin toward secondary alcohols. J. Am. Chem. Soc., 2005, 127, 12228–12229.

5. (a) Lin, Y.Y., Palmer, D.N. and Jones, J.B., The specificity of the nucleophilic site of �-chymo-trypsin and its potential for the resolution of alcohols. Enzyme-catalyzed hydrolyses of some(þ)-, (�)-, and (–)-2-butyl, -2-octyl, and -�-phenethyl esters. Can. J. Chem., 1974, 52, 469–476.

6. (a) Estell, D.A., Graycar, T.P., Miller, J.V., Powers, D.B., Burnier, J.P., Ng, P.G. and Wells,J.A., Probing steric and hydrophobic effects on enzyme–substrate interactions by proteinengineering. Science, 1986, 233, 659–663. (b). Wells, J.A., Powers, D.B., Bott, R.R., Graycar,T.P. and Estell, D.A., Proc. Natl. Acad. Sci. U. S. A., Designing substrate specificity by proteinengineering of electrostatic interactions. 1987, 84, 1219–1223.

7. Pohl, T. and Waldmann, H., Enhancement of the enantioselectivity of penicillin G acylase fromE. coli by ‘substrate tuning’. Tetrahedron Lett., 1995, 36, 2963–2966.

8. Savile, C. K., Kazlauskas, R.J., The 3-(3-pyridine)propionyl anchor group for protease-cata-lyzed resolutions: p-toluenesulfinamide and sterically hindered secondary alcohols. Adv. Synth.Catal., 2006, 348, 1183–1192.

9. Walker, F.A. and Benson, M., Entropy, enthalpy, and side arm porphyrins. 1. Thermodynamicsof axial ligand competition between 3-picoline and a series of 3-pyridyl ligands covalentlyattached to zinc tetraphenylporphyrin. J. Am. Chem. Soc., 1980, 102, 5530–5538.

10. Backes, B.J., Dragoli, D.R. and Ellman, J.A., Chiral N-acyl-tert-butanesulfinamides: the‘safety-catch’ principle applied to diastereoselective enolate alkylations. J. Org. Chem.,1999, 64, 5472–5478.

11. (a) Harwood, C.R. and Cutting, S.M., Molecular Biological Methods for Bacillus, John Wiley &Sons, Ltd, Chichester, 1990, pp. 33–35, 391–402; (b) Cho, S.-J., Oh, S.-H., Pridmore, R.D.,Juillerat, M.A. and Lee,C.[hyphen]H., Purification and characterization of proteases fromBacillus amyloliquefaciens isolated from traditional soybean fermentation starter. Agric. FoodChem., 2003, 51, 7664–7670.

12. Keith, D.D., Tortora, J.A. and Yang, R., Synthesis of L-2-amino-4-methoxy-trans-but-3-enoicacid. J. Org. Chem., 1978, 43, 3711–3713.


3.3 Desymmetrization of Prochiral Ketones Using EnzymesAndrew J. Carnell

Chiral enol esters are useful synthetic intermediates which can serve as chiral enolate

equivalents or undergo oxidative cleavage to produce reactive ester-aldehydes.1,2 Enolate

equivalents, such as silyl enol ethers and enol esters, can be made using chiral lithium

amide bases at low temperature, although poor selectivity can be a problem, particularly

with nonconformationally locked ketones.3 Lipase-catalysed resolution by enantioselec-

tive transesterification of racemic enol acetates derived from prochiral ketones can give

enol acetates in very high ee.4 Enol esters derived from 8-oxabicyclic ketones can be

resolved in excellent selectivity (E¼ 45–48) using silica-absorbed butanol-rinsed enzyme

preparation (BREP) Humicola sp. lipase.5 Similarly, enol esters derived from 4,4-disub-

stituted cyclohexanones can be resolved using Pseudomonas fluorescens lipase (PFL) in

tetrahydrofuran (THF).6,7 The prochiral ketone can be recycled, leading to a formal

desymmetrization of the ketone and good yield of the enantiomerically pure enol ester.

O

OAcR

RO

O R

RO

OAcR

RO

AcO R

R+

Humicola sp.lipase (BREP)

hexane/n-BuOH

+

1 R = Me, 2 R = Et3 R = CH2OMe

E = 45-48 (1R, 5S)

N

O

ArNR2

NK-2antagonists

O

Ar CN

OAc

Ar CN

OAc

Ar CN

OAc

Ar CN

+

recycle

+PFL, n-BuOH

THFE = 6.5–13

7 Ar = Ph8 Ar = 3,4-Cl2C6H39 Ar = 3,4-(MeO)2C6H3

steps

R1

4 R = Me, 5 R = Et6 R = CH2OMe

(S) (R)

10 Ar = Ph11 Ar = 3,4-Cl2C6H3 (82%, 99% e.e. after 3 recycles of ketone)12 Ar = 3,4-(MeO)2C6H3

(S)

91-99% e.e.

95–99% e.e.

3.3.1 Procedure 1: Humicola sp. Lipase BREP


• Humicola sp. lipase (Chirazyme L-8) (190 mg)

• silica gel (Merck Kieselgel 60 (230–400 mesh) (3.2 g)

• tris-HCl buffer (pH 7, 64 mL)

• dry n-butanol (120 mL)

• dry n-hexane (120 mL)

• one 250 mL conical flask

• orbital shaker.

3.3.1.2 Procedure

1. Humicola sp. lipase (190 mg) was dissolved in tris-HCl buffer (pH 7, 64 mL). To this

solution was added silica gel (3.2 g) and the mixture shaken at room temperature for 2 h.


2. The mixture was allowed to settle and the aqueous phase decanted ensuring that the

silica remained wet.

3. Dry n-butanol (40 mL) was added, swirled and decanted ensuring that the silica

remained wet. This was repeated twice (2 � 40 mL of n-butanol).

4. Dry hexane (40 mL) was added, swirled and decanted ensuring that the silica remained

wet. This was repeated twice (2 � 40 mL of hexane)

3.3.2 Procedure 2: Resolution of (–)-1,5-Dimethyl-3-acetyloxy-8-

oxabicyclo[3.2.1]oct-2,6-diene (1) using Humicola sp. Lipase BREP


• One quantity of Humicola sp. lipase BREP prepared as above

• (–)-1,5-dimethyl-3-acetyloxy-8-oxabicyclo[3.2.1]oct-2,6-diene 1 (760 mg, 3.92 mmol)

• dry n-butanol (0.72 mL, 7.84 mmol)

• dry n-hexane (40 mL)

• Celite (5 g)

• hexane for washing (30 mL)

• two 100 mL round-bottom flasks

• stirrer bar

• magnetic stirrer plate

• sintered vacuum filtration funnel


• equipment for column chromatography.

3.3.2.2 Procedure

1. Dry hexane (40 mL) was added to the Humicola sp. lipase BREP prepared above in

Procedure 1. This suspension was transferred to a 100 mL round-bottom flask contain-

ing the enol acetate 1 (760 mg, 3.92 mmol). Dry n-butanol (0.72 mL, 7.84 mmol) was

added and the mixture stirred at 25 �C for 1 h.

2. The mixture was filtered through a pad of Celite (5 g) using a vacuum sinter funnel and

washed through into a 100 mL flask with dry hexane (3 � 30 mL). The filtrate was

concentrated in vacuo to yield a crude mixture which was separated using flash column

chromatography on silica with 5 % ethyl acetate/40–60 % petroleum ether as eluent to

give the prochiral ketone 4 (290 mg, 49%) and enol acetate (�)-1 of >99.5 % ee (230

mg, 30 %) as a yellow oil [�]D �53.5 (c ¼ 8, CHCl3); (high-resolution mass spectro-

metry (HRMS): Found [M þ H]þ 195.10237. C11H15O3 requires 195.10212); �max

(neat)/cm�1 1759 (CO); �H (300 MHz; C6D6; Me4Si) 1.29 (3 H, s, CH3), 1.33 (3 H, s,

CH3), 1.66 (3 H, s, COCH3), 1.83 (1 H, dd, J 17 and 1.5, CH2endo), 2.45 (1 H, dd, J 17

and 1.8, CH2exo), 5.43 (1 H, d, J 5.7, CH¼), 5.81 (1 H, d overlapping, J 1.8 and 1.5,

CH¼CO) and 5.99 (1 H, d, J 5.4, CH¼); �C (75.5 MHz; C6D6; Me4Si) 20.8 (CH3), 22.0

(CH3), 24.4 (CH3), 36.8 (CH2), 82.1 (C), 83.3 (C), 121.9 (CH), 131.6 (CH), 141.4 (CH),

147.7 (C) and 168.6 (C); m/z (CI) 212 (100 %, [M þ NH4]þ), 195 (37, [M þ H]þ).

Chiral analysis was done using a Chiralpak AD column eluting with 5–10 % isopropa-

nol/n-hexane (1 mL min�1) and was recorded at 220 nm.


3.3.3 Procedure 3: Resolution of (–)-4-Cyano-4-(30,40-dichlorophenyl)cyclohex-1-

enyl Acetate (8)


• (–)-4-Cyano-4-(30,40-dichlorophenyl)cyclohex-1-enyl acetate (8) (10 g, 32.4 mmol)

• PFL (Amano AK) (8 g)

• n-butanol (5.21 mL, 64.7 mmol)

• THF (125 mL)

• THF for washing (100 mL)

• 250 mL round-bottom flask

• stirrer bar


• sintered vacuum filtration funnel



3.3.3.2 Procedure

1. The (–)-enol acetate 8 (1 g, 32.4 mmol), PFL (8 g) and n-BuOH (5.21 mL, 64.7 mmol)

were stirred in THF (125 mL) at room temperature for 9.5 h.

2. The solution was filtered through a glass sinter funnel under vacuum, the residual enzyme

washed with THF (100 mL) and the solvent removed by evaporation under reduced

pressure. The crude residue was purified by flash chromatography on silica using diethyl

ether/petroleum ether (1:2) as eluent to give the ketone (7 g) and the (S)-enol acetate 8

(2.8 g, 28 %, >99 % ee) as a white solid, m.p. 148–150 �C, [�]D ¼ þ11.5 (c ¼ 1.74 in

CHCl3). Chiral HPLC (Chiralpak AD) indicated 100 % ee for the enol acetate. RT (R)-8

15.5 min. (S)-8 20.9 min, eluent 100% EtOH, flow rate 0.5 mL min�1, l 220 nm. (Found:

C, 57.81; H, 4.18; N, 4.48. C15H13Cl2NO2 requires C, 58.08; H, 4.22; N, 4.52 %) (HRMS:

found Mþ 309.0322. C15H1335Cl2NO2 requires 309.0323); �max(neat) 2233 (CN), 1755

(CO); �H (300 MHz, CDCl3) 2.23 (3 H, s, AcCH3), 2.23–2.67 (6 H, m, 2 � H-3, 2 � H-5

and 2 � H-6), 5.46 (1H, s, H-2) 7.22–7.56 (3 H, Ar); �C (75 MHz, CDCl3) 20.9 (AcCH3),

24.6, 32.7, 35.4 (C-3, C-5, and C-6), 40.0 (C-4), 110.5 (C-2), 121.5 (CN), 125.3, 128.1

and 131.1 (C-20, C-50, C-60), 132.8, 133.5 and 139.6 (C-10, C-30 and C-40), 148.1 (C-1),

169.3 (C¼O); m/z (EI) 309 (3 %, Mþ), 267 (11), 70 (69), 43 (100).

3.3.4 Conclusion

This methodology was shown to work well for the desymmetrization of related ketones.

For example, oxabicyclic enol acetates 2 and 3 with other substituents (Et and CH2OMe) at

the bridge position were transformed more slowly but with similarly high enantioselec-

tivity (Table 3.3).3 For the cyclohexanone series, other aryl groups are tolerated at the

4-position as in substrates 7 and 9, although with PFL the cyano group is required for good

enantioselectivity (Table 3.4).1

An attractive feature of this type of resolution is that the prochiral ketone can be

recycled. The homochiral (S)-enol ester 8 was obtained in 82 % yield by recycling the

ketone without prior separation from the enantioenriched enol ester. For a cyclic enzyme


resolution of this type it can be derived that the maximum theoretical enantiomeric excess

for 100 % yield is eemax¼ (E� 1)/(Eþ 1).6 Thus, for an enzyme resolution with an E value

of 13, the eemax¼ 85.7 %. A higher ee in a stepwise process is possible only if the yield is

compromised. This ee corresponds to a conversion of 56 % and is the optimum point at

which to stop each kinetic resolution. Of course, with a mixture of ketone and enantioen-

riched ester for the start of each biotransformation after the first, the conversion required to

get to the 56 % (or 85.7 % ee) point is less. In the final biotransformation the conversion is

allowed to go beyond this point and the yield is compromised in order to get homochiral

ester. The enol ester 8 was subsequently used in a short and efficient four-step synthesis of

a nonpeptidic neurokinin NK-2 antagonist developed by Pfizer for the treatment of

neuroinflammatory conditions.1,2

Table 3.3 Reactions carried out in hexane using Humicola sp. lipase as described inProcedure 2

Substrate Reaction time (h) Conversion (%) ee of enol acetate (%) E

1 1 67 >99 452 19 51 91 473 48 53 96 48

Table 3.4 Reactions carried out in THF using freeze-dried Amano AK PFL as described inProcedure 3

Substrate Conversion to ketone (%) ee of enol acetate (%) E

7 68 >99 (S) 138 70 >99 (S) 119 71 95 7.4

References

1. Allan, G., Carnell, A.J, Escudero Hernandez, M.L. and Pettman, A., Chemoenzymatic synthesisof a tachykinin NK-2 antagonist. Tetrahedron, 2001, 57, 8193.

2. Carnell, A.J., Escudero Hernandez, M.L., Pettman, A. and Bickley, J., Chemoenzymatic synthesisof a non-peptide tachykinin NK-2 antagonist. Tetrahedron Lett., 2000, 41, 6929.

3. Allan, G., Carnell, A.J., Escudero Hernandez, M.L. and Pettman, A., Desymmetrisation of 4,4-disubstituted cyclohexanones by enzyme-catalysed resolution of their enol acetates. J. Chem. Soc.Perkin Trans. 1, 2000, 3382.

4. Carnell, A., Desymmetrisation of prochiral ketones using lipases. J. Mol. Catal. B Enzymatic,2002, 19–20, 83.

5. Carnell, A.J., Swain, S.A. and Bickley, J.F., Chiral enol acetates derived from prochiral oxabi-cyclic ketones using enzymes. Tetrahedron Lett., 1999, 40, 8633.

6. Carnell, A.J., Barkley, J. and Singh, A., Desymmetrisation of prochiral ketones by catalyticenantioselective hydrolysis of their enol esters using enzymes. Tetrahedron Lett., 1997, 38, 7781.

7. Allan, G., Carnell, A.J. and Kroutil, W., One-pot deracemisation of an enol acetate derived from aprochiral cyclohexanone. Tetrahedron Lett., 2001, 42, 5959.


3.4 Enzymatic Resolution of 1-Methyl-tetrahydroisoquinolineusing Candida rugosa LipaseGary Breen

Although secondary amines are common building blocks in the pharmaceutical industry,

there are few examples of the resolution of secondary amines in the literature. Preparation

of substituted phenyl allylcarbonates allowed the resolution of 1-methyl-tetrahydroisoqui-

noline (1-MTQ) to proceed with excellent enantioselectivity and recovery (Figure 3.1).

3.4.1 Procedure 1: Preparation of 3-Methoxyphenyl Allylcarbonate1

O O

O

MeO


• 3-Methoxyphenol (6.21 g)

• tetra-n-butylammonium chloride hydrate (100 mg)

• dichloromethane (40 mL)

• allyl chloroformate (6 mL)

• 4 M sodium hydroxide solution (30 mL)

• anhydrous magnesium sulfate

• N2 gas

• one 100 mL three-necked flask with a magnetic stirrer

• one magnetic stirring hotplate

• ice

• one 100 mL separating funnel

• filter paper


• Kugelrohr distillation equipment.

3.4.1.2 Procedure

1. 3-Methoxyphenol (6.21 g) and tetra-n-butylammonium chloride hydrate (100 mg) were

dissolved in dichloromethane (40 mL) in a 100 mL three-necked flask.

2. Sodium hydroxide solution (4 M, 20 mL) was added and the mixture cooled to 0–5 �C in

an ice bath with magnetic stirring under nitrogen.

3. Allyl chloroformate (6 mL) was added slowly, keeping the temperature between 0 and 5 �C.

4. After stirring for a further 1 h, the two layers were separated in a 100 mL separating

funnel and the dichloromethane layer was washed with 10 mL 4 M sodium hydroxide

solution. The organic layer was then dried with anhydrous magnesium sulfate and

concentrated using a rotary evaporator. The crude product (10.0 g) was purified using

vacuum distillation on a Kugelrohr apparatus (�3 mbar, 100 �C, cooling the recipient

flask with ice). (Yield 9.2 g, 88 %.)1H NMR (400 MHz, CDCl3) � 3.80 (3H, s), 4.75 (2H, d, J 8.3 Hz), 5.31–5.44 (2H, m),

5.95–6.05 (1H, m), 6.74 (1H, t, J 4.8 Hz), 6.77–6.81 (2H, m), 7.28 (1H, d, J 8.8 Hz).

3.4 Enzymatic Resolution of 1-MTQ using C rugosa Lipase 129

3.4.2 Procedure 2: Synthesis of (S)-1-Methyltetrahydroisoquinoline

NH

(S)-1-MTQ


• Racemic 1-methyltetrahydroisoquinoline (5 g)

• toluene (water saturated, 70 mL)2

• 3-methoxyphenyl allylcarbonate (4.65 g)

• ChiroCLEC-CR (100 mg)3

• saturated sodium chloride solution (50 mL)

• 2 M hydrochloric acid solution (50 mL)

• 10 M sodium hydroxide solution

• tert-butylmethylether (TBME, 100 mL)


• two 100 mL round-bottomed flasks

• two magnetic stirring hotplates

• one Buchner flask, 100 mL

• one Buchner funnel

• one 100 mL separating funnel

• rotary evaporator.

3.4.2.2 Procedure

1. Racemic 1-methyltetrahydroisoquinoline (5 g) and 3-methoxyphenyl allylcarbonate

(4.65 g) were stirred at 30 �C in a round-bottomed flask connected in a closed

system to another round-bottomed flask containing saturated sodium chloride

solution at 50 �C.4

2. ChiroCLEC-CR was added to the reaction flask and the reaction monitored for com-

pletion by high-performance liquid chromatography (HPLC).

3. After 8 h the enzyme was filtered off in a Buchner funnel and washed with toluene

(10 mL). The combined organic layers were washed with 2 M hydrochloric acid solution

(2 � 25 mL) in a 100 mL separating funnel. The combined acid layers were then

washed with toluene (10 mL) and the pH then adjusted to 12 with 10 M sodium

NH NH N

O

OO O

O

R +

1-MTQ (S)-1-MTQ

Candida rugosa lipase

Figure 3.1 Enzymatic resolution of 1-MTQ


hydroxide solution. The oil which formed was extracted with TBME (2 � 50 mL). The

organic portion was dried over anhydrous magnesium sulfate and concentrated using a

rotary evaporator. The product, (S)-1-MTQ was obtained as an oil with no further

purification. (Yield 2.3 g, 46 %, 99.6 % ee.)1H NMR (400 MHz, CDCl3) � 1.46 (3H, d, J 6.8 Hz), 1.90 (1H, br s), 2.73 (1H, dt, J

16.3, 4.8 Hz, 2.87 (1H, m), 3.02 (1H, m), 3.26 (1H, dt, J 12.8, 5.0 Hz), 4.10 (1H, q, J

6.8 Hz), 7.10 (4H, m).

HPLC analysis: Chiralcel OD column, 3 % hexane in methanol eluent, 1.5 mL min�1,

UV at 220 nm. Typical retention times: (S)-1-MTQ, 8.6 min; (R)-1-MTQ, 10.4 min.

3.4.3 Conclusion

This is a simple procedure for the enzymatic resolution of a secondary amine. The

acylating agent can be modified by altering the substitution on the phenol ring. This tuning

of the reactivity and selectivity should allow other amines to be resolved using a similar

approach.

References and Notes

1. This acylating agent has also been used in the resolution of indolines; see

Gotor-Fernandez, V., Rebolledo, F. and Gotor, V., Chemoenzymatic preparation of

optically active secondary amines: a new efficient route to enantiomerically pure

indolines. Tetrahedron Lett., 2006, 17, 2558.

2. This is prepared by stirring toluene with excess water and separating the two layers. The

saturated toluene layer contains 0.05 % w/w water. Water is important to maintain the

conformational integrity of the enzyme.

3. This enzyme is no longer commercially available, but other C. rugosa lipases were

also found to be active under these conditions, including Biocalysts L034P

(LipomodTM 34P).

4. The saturated salt solution maintains a constant water level in the toluene solution and

leads to faster reaction times.

Section 3.4 reprinted from Breen, G. F. Enzymatic resolution of a secondary amine using

novel acylating reagents. Tetrahedron Asymmetry 2004, 15(9), 1427–1430, with permis-

sion from Elsevier.

3.4 Enzymatic Resolution of 1-MTQ using C rugosa Lipase 131

4

Dynamic Kinetic Resolutionfor the Synthesis of Esters,

Amides and Acids Using Lipases

4.1 Dynamic Kinetic Resolution of 1-Phenylethanol by ImmobilizedLipase Coupled with In Situ Racemization over Zeolite BetaKam Loon Fow, Yongzhong Zhu, Gaik Khuan Chuah and Stephan Jaenicke

The one-pot dynamic kinetic resolution (DKR) of (–)-1-phenylethanol lipase esterifi-

cation in the presence of zeolite beta followed by saponification leads to (R)-1

phenylethanol in �70 % isolated yield at a multi-gram scale. The DKR consists of

two parallel reactions: kinetic resolution by transesterification with an immobilized

biocatalyst (lipase B from Candida antarctica) and in situ racemization over a

zeolite beta (Si/Al ¼ 150).1 With vinyl octanoate as the acyl donor, the desired

ester of (R)-1-phenylethanol was obtained with a yield of 80 % and an ee of 98 %.

The chiral secondary alcohol can be regenerated from the ester without loss of optical

purity. The advantages of this method are that it uses a single liquid phase and

both catalysts are solids which can be easily removed by filtration. This makes the

method suitable for scale-up. The examples given here describe the multi-gram

synthesis of (R)-1-phenylethyl octanoate and the hydrolysis of the ester to obtain

pure (R)-1-phenylethanol.



4.1.1 Procedure 1: Synthesis of (R)-1-Phenylethyl Octanoate

OH

C7H15

O

O

O C7H15

O

O

H

Catalyst:Novozym 435Zeolite Beta

Toluene, 60 °C80% yield98% ee

+ +


• (–)-1-Phenylethanol (1.22 g, 10 mmol)

• vinyl octanoate (2.04 g, 12 mmol)

• H-zeolite beta with Si/Al ¼ 150 (250 mg)

• immobilized lipase B from C. antarctica (E.C.3.1.1.3); tradename: Novozym 435 (150 mg)

• toluene (5 mL)

• hexane (20 mL)

• 0.1 M NaOH solution (60 mL)

• Na2SO4, anhydrous (�3 g)

• two-necked round-bottom flask, 25 mL capacity

• magnetic stirring bar

• hotplate with magnetic stirrer

• oil bath

• analytical balance

• separation funnel, 50 mL


Vinyl octanoate was obtained from TCI (Tokyo, Japan). All other chemicals with the

exception of the zeolite beta are available from Sigma Aldrich. The synthesis of a

particularly active modification of low-alumina zeolite beta has been described by us.2

Commercial material, available as samples from, for example, Zeolyst or Sudchemie can

be used, but because of excessive acidity may result in up to 15 % of styrene formation.

4.1.1.2 Procedure

1. (–)-1-Phenylethanol (1.22 g, 10 mmol), vinyl octanoate (2.04 g, 12 mmol) and toluene

(5 mL) were added into a 50 mL round-bottom flask and heated to 60 �C in an oil bath.

Zeolite beta with Si/Al ¼ 150 (250 mg) and the immobilized lipase Novozym 435

(150 mg) were added to the reaction mixture. The mixture was stirred for 6 h at 60 �C.

2. The mixture was left to cool to room temperature and the solid catalysts were removed

by filtration. Toluene and the by-product, acetaldehyde, were removed under reduced

pressure by using a rotary evaporator. The residue contains the desired product together

with unreacted vinyl octanoate and traces of octanoic acid, which are formed by

hydrolysis of the vinyl octanoate. The product can be purified by redissolving the

residue in hexane (20 mL) and washing it with 0.1 M NaOH (20 mL � 3). The organic

layer was dried with anhydrous Na2SO4 and the hexane was removed by a rotary

evaporator.

134 DKR for the Synthesis of Esters, Amides and Acids Using Lipases

3. The product is pure (R)-1-phenylethyl octanoate (2.01 g, �80 % yield).1H NMR (300 MHz, CDCl3) � 7.42–7.40 (m, 5H), 5.98 (q, J ¼ 6.6, 1H), 2.39

(t, J ¼ 7.1, 2H), 1.60 (d, J ¼ 6.8, 3H), 1.36-1.35 (m, 10H), 0.96 (t, J ¼ 6.6, 3H).13C NMR (75 MHz, CDCl3) � 173.53, 142.45, 129.00 (2 C), 128.31, 126.60 (2 C),

72.53, 35.15, 32.23, 29.61, 29.48, 25.55, 23.15, 22.80, 14.61.

The purity and ee were determined by gas chromatography (GC) with a chiral

column (Supelco Beta DEX 120 (90 �C isotherm)): major enantiomer Rt ¼ 119 min,

minor enantiomer Rt ¼ 120 min.

4.1.2 Procedure 2: Hydrolysis of (R)-1-Phenylethyl Octanoate

O C7H15

OOH

+ C7H15

O

Na+O–

reflux overnight

NaOH 1M


• (R)-1-Phenylethyl octanoate (�2 g, 8 mmol)

• hexane (90 mL)

• 1 M NaOH solution (30 mL)

• Na2SO4, anhydrous (�3 g)

• two-necked round-bottom flask, 25 mL capacity

• magnetic stirring bar


• oil bath

• analytical balance



4.1.2.3 Procedure

1. (R)-1-Phenylethyl octanoate (�2 g) was heated at reflux with 1 M NaOH (30 mL)

overnight.

2. The mixture was cooled to room temperature and extracted with hexane (30 mL � 3).

The combined organic layers were dried with anhydrous Na2SO4 and the hexane was

removed by rotary evaporation.

3. The product obtained is the pure (R)-1-phenylethanol (0.70 g, �70 % yield).1H NMR (500 MHz, CDCl3) � 7.34�7.30 (m, 5H), 4.82 (q, J¼ 6.5, 1H), 2.34 (s, 1H),

1.45 (d, J ¼ 6.5, 3H).13C NMR (125 MHz, CDCl3) � 145.78, 128.37 (2 C), 127.32, 125.32 (2 C), 70.21,

25.03.

The purity and ee were determined by GC with a chiral column (Supelco Beta DEX

325 (90 �C for 2 min, then 10 �C min�1 to 180 �C)): major enantiomer Rt ¼ 9.11 min,

minor enantiomer Rt ¼ 9.02 min.

4.1 DKR of 1-Phenylethanol by Immobilized Lipase Coupled 135

4.1.3 Conclusion

This DKR method can be applied to a variety of secondary alcohols. The size of the acyl

donor does have a major impact on the ee of the product, as shown in Table 4.1.

Table 4.1 Effect of the size of acyl donors on the ee of the product a

Entry

1

2

3

4

Acyl donor

CH3

O

OIsopropenyl acetate

CH3

O

OVinyl acetate

C3H7

O

OVinyl butanoate

C7H15

O

OVinyl octanoate

Conversion (%)

97

97

98

98

ee (%)

68

65

92

98

aReaction conditions (±)-1-Phenylethanol (0.122 g, 1 mmol), acyl donors (1.5 mmol),zeolite beta with Si/Al = 150 (50mg), Novozym 435 (30 mg) and toluene (5 mL) at 60 °C.

References

1. Zhu, Y-.Z., Fow, K.L., Chuah, G.K. and Jaenicke, S., Dynamic kinetic resolution of secondaryalcohols combining enzyme-catalyzed transesterification and zeolite-catalyzed racemisation.Chem. Eur. J. 2007, 13, 541.

2. Jaenicke, S., Chuah, G.K. and Fow, K.L., Dynamic kinetic resolution combining enzyme andzeolite catalysis. Stud. Surf. Sci. Catal. 2007, 172, 313.


4.2 Synthesis of the (R)-Butyrate Esters of Secondary Alcohols by DynamicKinetic Resolution Employing a Bis(tetrafluorosuccinato)-bridgedRu(II) ComplexS.F.G.M. van Nispen, J. van Buijtenen, J.A.J.M. Vekemans, J. Meuldijk

and L.A. Hulshof

Dynamic kinetic resolution (DKR) of secondary alcohols employing Novozym 435 and a

ruthenium complex as catalysts is a powerful method for the preparation of enantiomeri-

cally pure (R)-esters (Scheme 4.1).1 In this application of tandem catalysis, in situ

racemization of the slow-reacting enantiomer of the alcohol enables complete conversion

of the racemate into the desired enantiomer of the ester. The (R)-alcohol can be obtained by

subsequent hydrolysis of the ester. Various ruthenium catalysts were successfully

employed for the DKR of a broad range of secondary alcohols. Recently, we reported

the DKR of a series of alcohols employing bis(tetrafluorosuccinato)-bridged Ru(II) com-

plex (1) for the racemization (Figure 4.1, Table 4.2).2 Isopropyl butyrate was used as the

acyl donor and performing the reaction at reduced pressure (200 mbar) afforded excellent

yields of the (R)-butyrate esters, as was previously reported by Verzijl et al.3 employing a

different ruthenium catalyst.

R R'

OH

R R'

OCOR''

R R'

OH

R R'

OCOR''

lipase, acyl donor

fast

slowlipase, acyl donor

Ru-catalysedracemization

Scheme 4.1 DKR of secondary alcohols

P P= rac-BINAP

1

Ru

O

P

O

O

OC

P

O

Ru

P

CO

P

OO

O

O

FF

F

F

F

FFF

HH

HH

O

O

Figure 4.1 Bis(tetrafluorosuccinato)-bridged Ru(II) complex 1

4.2 Synthesis of (R)-Butyrate Esters of Secondary Alcohols by DKR 137

4.2.1 Materials and Equipment

General DKR procedure:

O HO

Ph

C3H7

3

O H

Ph

2

1 (0.1 mol%)Novozym 435,

isopropyl butyrate (2 eq.),K2CO3, toluene, 70 °C

+

O

Table 4.2 Dynamic kinetic resolution of various racemic alcohols. Reprintedfrom reference 2 with permission from Elsevier

Alcohol

OH

2a

OH

2b

OH

F3C

2c

OH

O

2d

OOH

2e

Time(h)

1024

23

30

31

23

7

Product

O C3H7

O

3a

O C3H

O

3b

O

F3C

C3H7

O

3c

O

O

C3H7

O

3d

OO

O

C3H7

3e

Yielda,b

(%)

95 >99 (87)

86

96 (79)f

98 (63)f

98

Eea

(%)

>99 >99

87

>99

98

79

aDetermined by chiral gas chromatography. bIsolated yield in parentheses. cThe ketone corresponding to the substrate (1.5 mmol) was added to the reaction mixture together with the substrate itself, isopropylbutyrate and toluene.

dNovozym 435: 0.05 g; Ru-catalyst: 0.4 mol%. eNovozym 435: 0.4 g. f Product not separated from ketone; calculated yield.

1

2d

3c,e

4c

5c

Entry


• Novozym 435 (0.1 g)

• complex 1 (0.018 g, 0.0093 mmol)

• K2CO3 (0.5 g)

• substrate (9 mmol)

• isopropyl butyrate (18 mmol)

• toluene (9 mL)

• Schlenk tube

• vacuum oven

• P2O5

• oil bath


• filter paper


• Kugelrohr apparatus.

4.2.2 Procedure

1. Novozym 435 (0.10 g), complex 1 (0.018 g, 0.0093 mmol) and K2CO3 (0.5 g, 3.8 mmol)

were dried overnight in a Schlenk tube under vacuum at 50 �C in the presence of P2O5.

2. The substrate (9 mmol), isopropyl butyrate (18 mmol) and toluene (9 mL) were then

added and the Schlenk tube was inserted in an oil bath at 73 �C, which indicated the

started of the reaction. The reaction mixture was stirred at 70 �C for 23 h at a pressure of

200 mbar. Small aliquots of reaction mixture were taken for gas chromatography

analysis. For preparative purposes, the reaction mixture was concentrated, filtered,

washed with toluene and concentrated in vacuo to yield the crude product.

3. Further purification of the crude product by distillation in a Kugelrohr apparatus

provided the (R)-esters.

(R)-1-Phenylethyl butyrate (3a). Yield: 87 %. 1H NMR (300 MHz, CDCl3) � (ppm)

0.95 (t, 3H, CH3), 1.55 (d, 3H, CH3), 1.63 (sextet, 2H, CH2), 2.32 (t, 2H, CH2), 5.92

(q, 1H, CH), 7.32 (5H, Ar-H). (R)-1-phenylethyl butyrate: 8.9 min. ½��25D ¼ þ91:3�

(c ¼ 0.98, CHCl3).

(R)-�-Methyl-4-(trifluoromethyl)benzyl butyrate (3c). 1H NMR (300 MHz, CDCl3)

� (ppm) 0.95 (t, 3H, CH3), 1.55 (d, 3H, CH3), 1.69 (sextet, 2H, CH2), 2.35 (t, 2H, CH2),

5.94 (q, 1H, CH), 7.48 (d, 2H, Ar-H-2,6), 7.61 (d, 2H, Ar-H-3,5).

(R)-�-Methyl-4-methoxybenzyl butyrate (3d). 1H NMR (300 MHz, CDCl3) � (ppm)

0.95 (t, 3H, CH3), 1.52 (d, 3H, CH3), 1.69 (sextet, 2H, CH2), 2.31 (t, 2H, CH2), 3.80

(s, 3H, CH3), 5.88 (q, 1H, CH), 6.89 (d, 2H, Ar-H-3,5), 7.32 (d, 2H, Ar-H-2,6).

References

1. Other groups’ works: (a) Larsson, A.L.E., Persson, B.A. and Backvall, J-E., Enzymatic resolutionof alcohols coupled with ruthenium-catalyzed racemization of the substrate alcohol. Angew.Chem. Int. Ed. Engl., 1997, 36, 1211. (b) Persson, B.A., Larsson, A.L.E., Le Ray, M. andBackvall, J.-E., Ruthenium- and enzyme-catalyzed dynamic kinetic resolution of secondaryalcohols. J. Am. Chem. Soc., 1999, 121, 1645. (c) Choi, J.H., Kim, Y.H., Nam, S.H.,Shin, S.T., Kim, M.-J. and Park, J., Aminocyclopentadienyl ruthenium chloride: catalytic

4.2 Synthesis of (R)-Butyrate Esters of Secondary Alcohols by DKR 139

racemization and dynamic kinetic resolution of alcohols at ambient temperature. Angew. Chem.Int. Ed., 2002, 41, 2373. (d) Choi, J.H., Choi, Y.K., Kim, Y.H., Park, E.S., Kim, E.J., Kim, M.-J.and Park, J., Aminocyclopentadienyl ruthenium complexes as racemization catalysts for dynamickinetic resolution of secondary alcohols at ambient temperature. J. Org. Chem., 2004, 69, 1972.(e) Martın-Matute, B., Edin, M., Bogar, K. and Backvall, J.-E., Highly compatible metal andenzyme catalysts for efficient dynamic kinetic resolution of alcohols at ambient temperature.Angew. Chem. Int. Ed., 2004, 43, 6535. (f) Martın-Matute, B., Edin, M., Bogar, K., Kaynak, F.B.and Backvall, J-E., Combined ruthenium(II) and lipase catalysis for efficient dynamic kineticresolution of secondary alcohols. Insight into the racemization mechanism. J. Am. Chem. Soc.,2005, 127, 8817. (g) Kim, N., Ko, S.B., Kwon, M.S., Kim, M.J. and Park, J., Air-stableracemization catalyst for dynamic kinetic resolution of secondary alcohols at room temperature.Org. Lett., 2005, 7, 4523.

2. Van Nispen, S.F.G.M., van Buijtenen, J., Vekemans, J.A.J.M., Meuldijk, J. and Hulshof, L.A.,Efficient dynamic kinetic resolution of secondary alcohols with a novel tetrafluorosuccinatoruthenium complex. Tetrahedron: Asymm., 2006, 17, 2299.

3. Verzijl, G.K.M., de Vries, J.G. and Broxterman, Q.B., Removal of the acyl donor residue allowsthe use of simple alkyl esters as acyl donors for the dynamic kinetic resolution of secondaryalcohols. Tetrahedron: Asymm., 2005, 16, 1603.


4.3 Dynamic Kinetic Resolution of 6,7-Dimethoxy-1-methyl-1,2,3,4-tetrahydroisoquinolineMichael Page, John Blacker and Matthew Stirling

Dynamic kinetic resolution is a technique that combines a racemization with a simulta-

neous resolution to overcome the inherent 50 % yield limit of kinetic resolution allowing a

theoretical 100 % yield. Recently, a novel chemoenzymatic system has been developed for

the dynamic kinetic resolution of 6,7-dimethoxy-1-methyl-1,2,3,4-tetrahydroisoquino-

line,1 building on kinetic resolution methodology developed by Breen.2 The corresponding

(R)-carbamate was isolated in high yield and enantiomeric excess (Figure 4.2).

4.3.1 Procedure 1: Synthesis of the Amine Racemization Catalyst

Pentamethylcyclopentadienyliridium(III) Iodide Dimer

IrI

IIr

I

I


• Pentamethylcyclopentadienyliridium(III) chloride dimer (4.57 g)

• sodium iodide (8.55 g)

• argon cylinder

• acetone (<100 ppm water) (525 mL)


• distilled water (750 mL)

• methanol

NH

MeO

MeO

O

O

OMeO

N

MeO

MeO

O

O

82% yield,96% ee

0.2 mol% [IrCp*I2]2,50 %w/w Candida rugosa,

Toluene, 40°C, 23 hrs+

Figure 4.2 Dynamic kinetic resolution of 6,7-dimethoxy-1-methyl-1,2,3,4-tetrahydroisoquinoline

4.3 DKR of 6,7-Dimethoxy-1-methyl-1,2,3,4-tetrahydroisoquinoline 141

• chloroform

• anhydrous sodium sulfate

• three-neck 1.0 L round-bottom flask

• condenser

• stirrer hotplate

• oil bath


• Buchi flask, 1.0 L

• glass funnel

• filter paper

• measuring cylinder, 500 mL

• 1.0 L separating funnel

• flow meter 0 –100 mL min�1.

4.3.1.2 Procedure

1. Pentamethylcyclopentadienyliridium(III) chloride dimer (4.57 g, 5.75 mmol) and

sodium iodide (8.55 g, 57.30 mmol) were added to a single-neck 1000 mL round-

bottom flask. A water condenser was fitted to the flask, the remaining necks were

stoppered and argon was sparged through the vessel at 500 mL min�1 for 30 min.

2. The purge of argon was then reduced to 20 mL min�1 and acetone (525 mL) was added,

the reaction flask was then placed in an oil bath at 60 �C and stirred using a magnetic

stirrer resulting in a dark orange solution containing some insoluble iridium dimer. The

reaction was heated at reflux under argon for 3 h before being cooled to room

temperature.

3. The reaction was concentrated to dryness under vacuum to yield a brown–red solid

that was dissolved in dichloromethane (500 mL) and washed with ultrapure water

(250 mL � 3) and the organic layer dried using sodium sulfate, filtered and concen-

trated to dryness under vacuum to yield a brown solid.

4. The solid was recrystallized from chloroform–methanol to yield brown needle-like

crystals. The filtrates were concentrated to dryness and the resulting residue was

recrystallized from chloroform–methanol. This was repeated a third time and the

three crops of catalyst combined to yield 5.102 g (78.2 % isolated yield assuming

98 % pure). The crystals were analysed by carbon and proton NMR and elemental

analysis.1H NMR (300 MHz, CDCl3) � 1.83 (s, Cp*—CH3).

C NMR (300 MHz, CDCl3) � 11.13 (Cp*—CH3), 89.3 (Cp*).

Elemental analysis. Calc.: C ¼ 20.7 %, H ¼ 2.6 %. Found: C ¼ 20.6 %, H ¼ 2.5 %.

4.3.2 Procedure 2: Synthesis of 3-Methoxyphenylpropyl Carbonate

O

O

OMeO



• 3-Methoxyphenol (6.47 g)

• tetrabutylammonium iodide (131.9 mg)


• 4.0 M sodium hydroxide (20 mL)

• propyl chloroformate (7.06 g)

• ethyl acetate

• hexane

• three-neck 100 mL round-bottom flask

• thermometer

• magnetic stirrer

• ice–salt bath

• filter paper

• filter funnel

• 100 mL separating funnel


• 500 mL Buchi flask

• flash chromatography column

• measuring cylinder, 100 mL.

4.3.2.2 Procedure

1. To a three-neck 100 mL round-bottom flask was added 3-methoxyphenol (6.47 g, 52.09

mmol), tetrabutylammonium iodide (131.9 mg, 0.36 mmol) and dichloromethane

(40 mL), resulting in a red–orange solution.

2. A thermometer was attached and magnetic agitation started. Sodium hydroxide solu-

tion (20 mL of 4.0 M) was added, which caused the reaction to turn a brown colour. The

reaction flask was then placed in an ice–salt bath and cooled to 0�5 �C.

3. Propyl chloroformate (7.06 g, 57.64 mmol) was added over an hour using a syringe

pump; the reaction solution was kept between 0 and 5 �C during the addition. A yellow

precipitate formed during the addition.

4. The reaction was stirred for a further hour at 0�5 �C and then allowed to separate. The

aqueous layer was a dark brown colour and the organic layer was yellow. The organic

layer was washed with sodium hydroxide solution (10 mL of 2.0 M) then dried using

magnesium sulfate and filtered. The filtrates were concentrated to dryness under

vacuum to leave a yellow oil.

5. The crude product was purified using flash column chromatography with a 70:30

hexane/ethyl acetate mobile phase, producing, after vacuum distillation, a yellow oil

(10.17 g ¼ 93 % isolated yield). The oil was analysed by gas chromatography–mass

spectrometry (GCMS) and NMR.

GCMS

Column: Varian VF-1MS (10 m � 150 mm � 0.12 mm); oven temperature: 35 �C for 1

min then ramp at 50 �C min�1 to 300 �C and hold for 2 min; inlet pressure: 23 psi. Rt

3-methoxyphenylpropyl carbonate: 3.6 min.


NMR (300 MHz, CDCl3) 3-Methoxyphenylpropyl Carbonate

CB

CGCF

CE

CD

CC OCH

OCI

CJO

HB

HE

HD

HC

O

CK

HF HF

HG

HH

HH

CA

HA

HA

HA

HG

HH

Proton ID Peak multiplicity Chemical shift/ppm No. of protons Coupling constant/Hz

A Singlet 3.8 3 N.A.B, C, E Multiplet 6.73 �6.80 3 N.A.D Multiplet 7.24 �7.27 1 N.A.F Triplet 4.21 2 6G Multiplet 1.77 2 N.A.H Triplet 1.01 3 6

4.3.3 Procedure 3: Dynamic Kinetic Resolution of 6,7-Dimethoxy-1-methyl-

1,2,3,4-tetrahydroisoquinoline

NH

MeO

MeO


• Toluene (30 mL)


• 6,7-dimethoxy-1-methyl-1,2,3,4-tetrahydroisoquinoline (3.093 g)

• Candida rugosa lipase (2.1 g, 1410 units/mg)

• pentamethylcyclopentadienyliridium (III) iodide (34.3 mg)

• saturated brine (40 mL)

• 3-methoxyphenylpropyl carbonate (4.657 g)

• Celite


Carbon IDa A B C D E F G H I J K

Chemicalshift/ppm 55.8 160.9 107.5 152.5 113.7 130.2 112.2 154.0 70.8 22.4 10.6

a The assignments of carbon atoms C, E, F and G were arbitrary.


• 1.0 M hydrochloric acid (100 mL)

• 1.0 M sodium hydroxide (100 mL)


• hexane

• ethyl acetate

• conical flask, 250 mL

• two stirrer hotplates

• two-neck 50 mL round-bottom flasks

• two oil baths

• one-neck 50 mL round-bottom flask



• Buchner flask, 250 mL

• Buchner funnel

• filter paper


• filter funnel

• Buchi flask, 250 mL

• flash chromatography column

• separating funnel, 250 mL.

4.3.3.2 Procedure

1. Prior to the reaction a stock solution of toluene saturated with water was prepared by

vigorously stirring a toluene–water mixture for 1 h and then allowing the two layers to

separate; the toluene layer was used for the reaction solvent.

2. To a two-neck 50 mL round-bottom flask was added 6,7-dimethoxy-1-methyl-1,2,3,4-

tetrahydroisoquinoline (3.093 g, 14.92 mmol), C. rugosa lipase (1.5 g, 1410 units/mg),

pentamethylcyclopentadienyliridium(III) iodide (34.3 mg, 0.030 mmol) and toluene satu-

rated with water (25 mL), resulting in an orange solution containing some insoluble amine.

3. The reaction vessel was placed in an oil bath at 40 �C and connected to a second flask

containing saturated brine at 50 �C. 3-Methoxyphenylpropyl carbonate (4.657 g, 22.15

mmol) was added and washed in using toluene saturated with water (5 mL), resulting in

an orange–brown solution containing brown insoluble material (enzyme).

4. The reaction flask was sealed so that the system was closed and then stirred overnight.

After 24 h, an additional aliquot of 3-methoxyphenylpropyl carbonate (1.55 g, 7.37

mmol) and C. rugosa lipase (600 mg of 1410 units/mg) were added.

5. Aftera total reaction time of48h the reactionsolution was filtered throughCelite to remove

any enzyme. The filtrates were diluted with dichloromethane (100 mL) and washed

with 1 M aqueous hydrochloric acid (2� 50 mL) and 1 M aqueous sodium hydroxide

solution (2� 50 mL). The organic layer was dried using sodium sulfate, filtered and

concentrated to dryness under vacuum to yield a brown oil which was purified through a

silica column using a hexane–ethyl acetate gradient elution system. The fractions contain-

ing only the product were combined and concentrated to dryness under vacuum to yield a

yellow oil (3.53 g, 81.6 %).

The product was analysed by gas chromatography (GC), dissolved in hexane/propan-

2-ol (70/30) and analysed by chiral high-performance liquid chromatography (HPLC)


and dissolved in CDCl3 for NMR analysis (H1, C13 and heteronuclear multiple quantum

coherence).

GC (for Conversion)

Column: HP5 (25.0 m � 320 mm � 0.52 mm); oven temperature: 150 �C for 25 mins then

ramp at 20 �C.min-1 to 300 �C and hold for 10 min; inlet pressure: 12.0 psi; Rt 6,7-

dimethoxy-1-methyl-1,2,3,4-tetrahydroisoquinoline: 22.2 min; Rt 6,7-dimethoxy-1-

methyl-1,2,3,4-tetrahydroisoquinoline: 20.5 min; Rt 3-methoxyphenol: 2.5 min; Rt 3-meth-

oxyphenylpropyl carbonate: 8.9 min; Rt propyl carbamate product: 31.8 min.

GC (for Enantiomeric Excess of Starting Material)

Samples were derivatized with trifluoroacetic anhydride prior to injection.

Column: CP-Chirasil-Dex-CB (25 m � 250 mm � 0.25 mm); oven temperature: 165 �C

for 45 min then ramp at 20 �C min�1 and hold for 25 min; inlet pressure: 10.0 psi. Rt

(R)-6,7-dimethoxy-1-methyl-1,2,3,4-tetrahydroisoquinoline: 40.9 mins; Rt (S)-6,7-

dimethoxy-1-methyl-1,2,3,4-tetrahydroisoquinoline: 41.6 min.

HPLC (for Enantiomeric Excess of Carbamate Product)

Column: Chiralpak AD (25 cm � 0.46 cm); mobile phase: isocratic 70 % hexane–30 %

propan-2-ol; flow rate: 1.0 mL min�1; detector wavelength: 220 nm. Rt (R)-carbamate: 6.3

min; Rt (S)-carbamate: 7.1 min.

NMR: Carbamate Product (300 MHz, CDCl3 Solvent, Reference SiMe4)

CC

CH

CG

CF

CE

CD

CK

N

CJ

CI

CL

HE HEHF

HF

HC

HD

O

OCA

HA

HA

HA

CB

HB

HB

HB

CM

O

CN

CO

CP

O HI HI

HJ HJ

HK

HK

HK

HH HHHH

HG

Proton ID Peak multiplicity Chemical shift/ppm No. of protons Coupling constant/Hz

A and B Two singlets 3.85, 3.86 2 � 3 N.A.C and D Singlet 6.59 2 N.A.E Triplet 2.62 1 3E Triplet 2.67 1 3F Broad multiplet 3.23 1 N.A.F Broad multiplet 4.23 1 N.A.G MissingH Doublet 1.45 3 7I Multiplet 4.10 2 N.A.J Multiplet 1.69 2 N.A.K Triplet 0.97 3 7


4.3.4 Conclusion

The procedure constitutes the first known example of a chemoenzymatic dynamic kinetic

resolution of a secondary amine. The operational simplicity of the procedure is exempli-

fied by the mild conditions, air-stable reagents and low catalyst loading.

Section 4.3 reprinted from Stirling, M., Blacker, J. and Page, M. I. Chemoenzymatic

dynamic kinetic resolution of secondary amines. Tetrahedron Letters 2007, 48, 1247–

1250, with permission from Elsevier.

Carbon ID A/B A/B C þ H D/G E/F E/F D/G H I Ja

Chemical shift/ppm 56.3 56.4 148.0 110.1 126.5 126.1 111.8 153.9 28.9 37.6

Carbon ID Ja K La La M N O PChemical shift/ppm 38.0 50.4 37.6 38.0 155.8 67.3 22.8 10.9

a Two peaks due to amide resonance.

References

1. Stirling, M., Blacker J. and Page M.I., Chemoenzymatic dynamic kinetic resolution of secondaryamines. Tetrahedron Lett., 2007, 48, 1247.

2. Breen, G., Enzymatic resolution of a secondary amine using novel acylating reagents.Tetrahedron: Asymm., 2004, 15, 1427.


4.4 Dynamic Kinetic Resolution of Primary Amines with a RecyclablePalladium Nanocatalyst (Pd/AlO(OH)) for RacemizationSoo-Byung Ko, Mahn-Joo Kim and Jaiwook Park

The complete transformation of a racemic mixture into a single enantiomer is one of the

challenging goals in asymmetric synthesis. We have developed metal–enzyme combina-

tions for the dynamic kinetic resolution (DKR) of racemic primary amines.1 This proce-

dure employs a heterogeneous palladium catalyst, Pd/AlO(OH),2 as the racemization

catalyst, Candida antarctica lipase B immobilized on acrylic resin (CAL-B) as the

resolution catalyst and ethyl acetate or methoxymethylacetate as the acyl donor.

Benzylic and aliphatic primary amines and one amino acid amide have been efficiently

resolved with good yields (85�99 %) and high optical purities (97�99 %). The racemiza-

tion catalyst was recyclable and could be reused for the DKR without activity loss at least

10 times.

R1 R2

NH2

R1 R2

NH2

R1 R2

NHCOCH2RPd/AlO(OH)CAL-B,RCH2COOR' (3.0 equiv)

(R)-enantiomer

R = H, OMe

4.4.1 Procedure 1: Synthesis of Pd/AlO(OH)

Pd(PPh3)4

+HO(CH2CH2O)3CH2CH2OH

Pd/AlO(OH)

i. (sec-BuO)3Al/BuOH, 120 °C, 10 hii. H2O, 120 °C, 0.5 h

iii. Filtrationiv. Wash with acetonev. Dry


• Pd(PPh3)4 (260 mg)

• tetra(ethylene glycol) (418 mg)

• aluminium sec-butoxide (9.50 g)

• 1-butanol (3 mL)

• water (2 mL)

• 5 mL syringe



• glass filter, (pore size: 20�30 mm)

• acetone (5 mL).

4.4.1.2 Procedure

1. Pd(PPh3)4 (260 mg, 0.225 mmol), tetra(ethylene glycol) (418 mg, 2.20 mmol), (sec-

BuO)3Al (9.50 g, 38.5 mmol) and 1-butanol (3 mL, 32.7 mmol) were added to a 50 mL

round-bottom flask equipped with condenser. The mixture was stirred for 10 h at 120 �C

to give a black suspension.


2. Water (2 mL) was added and the resulting mixture was stirred at 120 �C for 30 min.

3. The resulting black solid was filtered, washed with acetone and dried at room tempera-

ture in air to give Pd/AlO(OH) as a dark olive-green powder (2.75 g; 0.86 wt% of Pd).

4.4.2 Procedure 2: Dynamic Kinetic Resolution of 1-Phenylethylamine

NH2 NHO

Novozym 435, Pd/AlO(OH),

EtOAc, 4A molecular sieves,toluene, 70 oC, 3 d

97 % yield, 98 % ee


• Dried and degassed toluene (8 mL)

• 1-phenylethylamine (97 mg, 0.80 mmol)

• Pd/AlO(OH) (99 mg)

• Novozym 435 (96 mg)


• molecular sieves 4 A (560 mg)

• 50 mL round bottom flask

• glass filter (pore size: 20�30 mm)

• a balloon filled with argon gas

• silica gel (Merck: silica gel 60 A 200�400 mesh), 10 g


• high performance liquid chromatograph (HPLC) equipped with a chiral column ((R,R)

Whelk-O1, Merck).

4.4.2.2 Procedure

1. A suspension containing 1-phenylethylamine (97 mg, 0.80 mmol), Pd/AlO(OH) (99

mg), Novozym 435 (96 mg), ethyl acetate (234 mL) and 4 A molecular sieve (560 mg) in

dry and degassed toluene (8 mL) was stirred at 70 �C for 3 days under argon atmosphere.

2. The mixture was cooled to room temperature and filtered through a glass sinter. The

filtrate was concentrated and then purified by column chromatography on silica gel (1/1

n-hexane/ethyl acetate) to give N-((R)-1-phenylethyl)acetamide (119 mg, 0.73 mmol,

92 %, 98 % ee).1H NMR (300 MHz, CDCl3) � 7.34–7.26 (m, 5H), 5.85 (s, 1H), 5.15–5.09 (m, 1H),

1.98 (s, 3H), 1.49 (d, J¼ 6.88 Hz, 3H) ppm; 13C NMR (75 MHz, CDCl3) � 169.5, 143.7,

129.0, 127.7, 126.6, 49.2, 23.8, 22.1 ppm.

HPLC condition: (R,R)-Whelk-O1, 80/20 n-hexane/2-propanol, flow rate 1.0 mL

min�1, UV 217 nm. (S)-form: 8.19 min; (R)-form: 14.36 min.

M.p. 97–100 �C.

½��25D ¼ þ141 (c ¼ 1.0, CHCl3).

4.4 DKR of Primary Amines with a Recyclable Palladium Nanocatalyst 149

4.4.3 Procedure 3: Dynamic Kinetic Resolution of 1-Methyl-3-phenylpropylamine

NHONH2

H2 (1 atm), Novozym 435,

Pd/AlO(OH), EtOAc, toluene, 100 °C, 4 h

95 % yield, 98 % ee


• Dried and degassed toluene (3 mL)

• 1-methyl-3-phenylpropylamine (90 mg, 0.60 mmol)

• Pd/AlO(OH) (891 mg)

• Novozym 435 (72 mg)



• glass filter (pore size:20�30 mm)

• a balloon filled with H2 gas

• silica gel (Merck: silicagel 60 A 200�400 mesh), 10 g


• HPLC equipped with a chiral column ((R,R) Whelk-O1, Merck).

4.4.3.2 Procedure

1. A suspension containing 1-methyl-3-phenylpropylamine (90 mg, 0.60 mmol), Pd/

AlO(OH) (891 mg), Novozym 435 (72 mg) and ethyl acetate (176 mL) in dry and

degassed toluene (8 mL) was stirred at 100 �C for 4 h under H2 (1 atm).

2. The mixture was cooled to room temperature and filtered through a glass sinter. The

filtrate was concentrated and then purified by column chromatography on silica gel (1/1

n-hexane/ethyl acetate) to give N-((R)-4-phenylbutan-2-yl)acetamide (109 mg, 0.57

mmol, 95 %, 98 % ee).1H NMR (300 MHz, CDCl3) � 7.30–7.15 (m, 5H), 5.47 (br s, 1H), 4.10–4.00 (m, 1H),

2.67–2.62 (m, 2H), 1.93 (s, 3H), 1.79–1.71 (m, 2H), 1.71 (d, J ¼ 6.60 Hz, 3H) ppm; 13C

NMR (75 MHz, CDCl3) � 169.5, 142.0, 128.6, 128.5, 126.1, 45.4, 38.8, 32.7, 23.7, 21.2 ppm.

HPLC condition: (R,R)-Whelk-O1, 85/15 n-hexane/2-propanol, flow rate 0.5 mL

min�1, UV 217 nm. (S)-form ¼ 23.99 min, (R)-form ¼ 27.07 min.

M.p. 71–72 �C.

½��25D ¼ þ41:3 (c ¼ 1.0, CH2Cl2).

4.4.4 Conclusion

A heterogeneous and recyclable palladium catalyst, Pd/AlO(OH), is excellent for the

racemization of primary amines. We have demonstrated successful DKR of various primary

amines by combining the palladium catalyst and a lipase to produce the corresponding

(R)-acetamides in high yields and in high optical purities. Tables 4.3 and 4.4 show the

results of the DKR of benzylic and aliphatic primary amines.


Table 4.3 DKR of various benzylic amines

Entry

1

2

3

4

5

6

7

8

3

Product

NHCOMe

NHCOMe

NHCOMe

OMe

NHCOMe

CF

NHCOMe

MeCOHN

NHCOMe

O

NHCOMe

Yield (%)

92

91

93

95

90

88

86

94

Ee (%)

R) 98 (

R) 97 (

R) 98 (

R) 98 (

R) 98 (

R) 99 (

R) 97 (

98 (R)

4.4 DKR of Primary Amines with a Recyclable Palladium Nanocatalyst 151

References

1. Kim, M.-J., Kim,W.-H., Han, K., Choi, Y.K. and Park, J., Dynamic kinetic resolution of primaryamines with a recyclable Pd nanocatalyst for racemization. Org. Lett., 2007, 9, 1157–1159.

2. (a) Kim, N., Kwon, M.S., Park, C.M. and Park, J., One-pot synthesis of recyclable palladiumcatalysts for hydrogenations and carbon–carbon coupling reactions. Tetrahedron Lett., 2004, 45,7057–7059. (b) Kwon, M.S., Kim, N., Park, C.M., Lee, J.S., Kang, K.Y. and Park, J., Palladiumnanoparticles entrapped in aluminum hydroxide: dual catalyst for alkene hydrogenation andaerobic alcohol oxidation. Org. Lett., 2005, 7, 1077–1079. (c) Kwon, M.S., Kim, N., Seo, S.H.,Park, I.S., Cheedrala, R.K. and Park, J., Recyclable palladium catalyst for highly selective �alkylation of ketones with alcohols. Angew. Chem. Int. Ed., 2005, 44, 6913–6915.

Table 4.4 DKR of various aliphatic amines

Entry

1

2

3

4

Product

NHCOMe

NHCOMe

NHCOMe

CONH2

NHCOMe

Yield (%)

96

93

92

96

Ee (%)

98

99

98

98 (R)

(R)

(R)

(R)


4.5 Dynamic Kinetic Resolution of Amines Involving Biocatalysisand In Situ Free-radical-mediated RacemizationStephane Gastaldi, Gerard Gil and Michele P. Bertrand

Dynamic kinetic resolution enables the limit of 50 % theoretical yield of kinetic resolution

to be overcome. The application of lipase-catalyzed enzymatic resolution with in situ thiyl

radical-mediated racemization enables the dynamic kinetic resolution of non-benzylic

amines to be obtained. This protocol leads to (R)-amides with high enantioselectivities.

It can be applied either to the conversion of racemic mixtures or to the inversion of

(S)-enantiomers.

4.5.1 Procedure 1: Synthesis of (R)-N-(Octan-2-yl)dodecanamide from Racemic

2-Aminooctane

Novozym 435C11H23CO2Et

Heptane, AIBN, 80°CEt2NOC SH

74% yield, ee>99%

C11H23HN

O

(R)

NH2


• Novozym 435 (CAL-B, gift from Novo Nordisk, Denmark) (1.5 g)

• ethyl laurate (2.64 g, 11.6 mmol)

• 2,20-azobis(2-methylpropionitrile) (AIBN) (430 mg, 2.7 mmol)

• heptane (77 mL)

• N,N-diethyl-2-mercapto-propionamide1 (1.49 g, 9.2 mmol)

• 2-aminooctane (1 g, 7.7 mmol)

• thin-layer chromatography (TLC) plates (silica gel 60 F254, Macherey-Nagel)

• one-necked reaction flask equipped with a magnetic stirrer bar, 250 mL


• cooling equipment.

4.5.1.2 Procedure

1. A solution of 2-aminooctane (1 g, 7.7 mmol), ethyl laurate (2.64 g, 11.6 mmol), N,N-

diethyl-2-mercapto-propionamide (1.49 g, 9.2 mmol) and Novozym 435 (1.5 g) in

heptane (77 mL) was heated for 8 h at 80 �C in the presence of AIBN (the overall

quantity of 30 mol% of AIBN (430 mg, 2.7 mmol) was divided into four equal portions

that were added successively every 2 h).

2. After filtration of the enzyme, the solution was cooled in a freezer to precipitate the

lauramide. The cold suspension was filtered to give pure (R)-N-(octan-2-yl)dodecana-

mide. Three precipitations were necessary to obtain the pure product (1.775 g, 5.7

mmol, enantiomeric excess (ee) > 99 %, 74 % yield).

(R)-N-(Octan-2-yl)dodecanamide. 1H NMR (CDCl3, 300 MHz): � 5.24 (br d, J¼ 6.6,

NH), 3.98 (sept, J ¼ 6.6, 1H), 2.08�2.18 (m, 2H), 1.54�1.72 (m, 2H), 1.18�1.37

(m, 26H), 1.11 (d, J ¼ 6.6, 3H), 0.88 (t, J ¼ 6.8, 6H).

4.5 DKR of Amines by Biocatalysis and In Situ Free-radical-mediated Racemization 153

13C NMR (CDCl3, 75 MHz): � 14.8 (CH3), 14.9 (CH3), 21.8 (CH3), 23.4 (CH2), 26.7

(CH2), 26.8 (CH2), 30.0 (CH2), 30.1 (2 � CH2), 30.2 (2 � CH2), 30.3 (CH2), 30.4

(2 � CH2), 32.5 (CH2), 32.6 (CH2), 37.7 (CH2), 37.8 (CH2), 45.8 (CH), 173.2 (C¼O).

½��25D ¼ þ1:5 (c ¼ 1, CHCl3).

HRMS (TOF MS ESþ) MHþ Calc. [M þ 1] for C20H41NO: 312.3266; found:

312.3256.

The ee was determined by high-performance liquid chromatography with a

Chiralpak AS, hexane/isopropanol (97/3), 1 mL min�1. Detectors: UV (220 nm) and

circular dichroism (220 nm). Rt (S) form: 9.43; Rt (R) form: 10.71.

4.5.2 Procedure 2: Synthesis of (R)-N-(Octan-2-yl)dodecanamide

from (S)-2-aminooctane

62% yield, ee>99%

HN

Novozym 435C11H23CO2Et

Heptane, AIBN, 80°C

NH2

O

C11H23

Et2NOC SH(R)(S)


• Novozym 435 (CAL-B, gift from Novo Nordisk, Denmark) (134 mg)

• ethyl laurate (228 mg, 1.00 mmol)

• AIBN (40 mg, 0.24 mmol)

• heptane (6.7 mL)

• N,N-diethyl-2-mercapto-propionamide1 (129 mg, 0.80 mmol)

• (S)-2-aminooctane (86 mg, 0.67 mmol)

• silica gel (MN Kieselgel 60, 63–200 mm, Macherey-Nagel)

• TLC plates (silica gel 60 F254, Macherey-Nagel)



• cooling equipment



4.5.2.2 Procedure

1. A solution of commercially available (S)-2-amino-octane (86 mg, 0.67 mmol), lauric

acid ethyl ester (228 mg, 1.00 mmol), N,N-diethyl-2-mercapto-propionamide (129 mg,

0.80 mmol) and Novozym 435 (134 mg) in heptane (6.7 mL) was heated for 8 h at 80 �C

in the presence of AIBN (the overall quantity of 30 mol% of AIBN (40 mg, 0.24 mmol)

was divided into four equal portions that were added successively at 2 h intervals).

2. After filtration of the enzyme and concentration, the crude product was purified by flash

chromatography on silica gel (0/100 to 15/85 EtOAc/pentane) to give the correspond-

ing lauramide (130 mg, ee > 99 %, 62 % yield).


Et2NOC SH

NH2

Me R

NHCOR1

Me R

CAL-B, R1CO2R2

1(R)-2

8h, 80 °C

AIBN, heptane,

4.5.3 Conclusion

This dynamic kinetic resolution (DKR) process,1 associating lipase-catalyzed enzymatic

resolution with in situ thiyl radical-mediated racemization,2 applies only to aliphatic

amines. The racemization process involves the reversible generation an �-aminoalkyl

radical. In the case of benzylic radicals, oxidative degradation competes with hydrogen

atom transfer from the thiol. The efficiency of the methodology relies on the very high

selectivity of the enzymatic kinetic resolution at 80 �C involving the thermostable lipase

Novozym 435. The selectivity is intimately connected to the number of carbon atoms in the

acyl donor (Table 4.5, entry 1 versus entries 2 and 3).3 Most DKR processes, associating

enzymatic resolution to metal-transition catalyzed racemization apply mainly to benzylic

amines.4

Table 4.5 Other examples of amine DKR a

Amine R R1CO2R2 Product Yield (%) Ee (%)

Isolated NMR

1 rac-1a Ph(CH2)2 CH3CO2Etb 2a 31 95 742 rac-1a Ph(CH2)2 C11H23CO2Et 2a 70 71 993 rac-1a Ph(CH2)2 C11H23CO2H 2a 71 69 >994 (S)-1a Ph(CH2)2 C11H23CO2Et 2a 58 nd 995 rac-1c t-BuOCOCH2 C11H23CO2Et 2c 47 54 926 rac-1d Me2C¼CH(CH2)2 C11H23CO2Et 2d 68 70 947 rac-1e Et C11H23CO2H 2e 57 nd 86

aConditions: amine (1 mmol, 0.063 M), N,N-diethyl-2-mercapto-propionamide (1.2 equiv), acyl donor (1.5 equiv),Novozym 435 (200 mg).bEtOAc/heptane (2.6:8 by vol.).

References

1. Gastaldi, S., Escoubet, S., Vanthuyne, N., Gil, G. and Bertrand, M.P., Dynamic kinetic resolutionof amines involving biocatalysis and in situ free radical mediated racemization. Org. Lett., 2007,9, 837–839.

2. (a) Escoubet, S., Gastaldi, S., Vanthuyne, N., Gil, G., Siri, D. and Bertrand, M.P., Thiyl radicalmediated racemization of nonactivated aliphatic amines. J. Org. Chem., 2006, 71, 7288–7292.(b)Escoubet, S., Gastaldi, S., Vanthuyne, N., Gil, G., Siri, D. and Bertrand, M., Thiyl radicalmediated racemization of benzylic amines. Eur. J. Org. Chem., 2006, 3242–3250.

3. Nechab, M., Azzi, N., Vanthuyne, N., Bertrand, M., Gastaldi, S. and Gil, G., Highly selectiveenzymatic kinetic resolution of primary amines at 80�C: a comparative study of carboxylic acidsand their ethyl esters as acyl donors. J. Org. Chem., 2007, 72, 6918–6923.

4.5 DKR of Amines by Biocatalysis and In Situ Free-radical-mediated Racemization 155

4. (a) Paetzold, J. and Backvall, J.E., Chemoenzymatic dynamic kinetic resolution of primaryamines. J. Am. Chem. Soc., 2005, 127, 17620–17621. (b) Reetz, M.T. and Schimossek, K.,Lipase-catalyzed dynamic kinetic resolution of chiral amines: use of palladium as the racemiza-tion catalyst. Chimia, 1996, 50, 668–669. (c) Parvulescu, A., DeVos, D. and Jacobs, P., Efficientdynamic kinetic resolution of secondary amines with Pd on alkaline earth salts and a lipase.Chem. Commun. 2005, 5307–5309. (d) Roengpithya, C., Patterson, D.A., Livingston, A.G.,Taylor, P.C. Irwin, J.L. and Parrett, M.R., Towards a continuous dynamic kinetic resolution of1-phenylethylamine using a membrane assisted, two vessel process. Chem. Commun., 2007,3462–3463. (e) Kim,M.-J., Kim,W.-H., Han, K., Choi, Y.K. and Park, J., Dynamic kineticresolution of primary amines with a recyclable Pd nanocatalyst for racemization. Org. Lett.,2007, 9, 1157–1159. (f) Stirling, M., Blacker, J. and Page, M.I., Chemoenzymatic dynamickinetic resolution of secondary amines. Tetrahedron. Lett., 2007, 48, 1247–1250.


4.6 Chemoenzymatic Dynamic Kinetic Resolution of (S)-IbuprofenA.H. Kamaruddin and F. Hamzah

Dynamic kinetic resolution (DKR) is a process in which the resolution process is

coupled with in situ racemization of unreacted substrate. This has been shown to be a

potential and feasible method to produce 100 % theoretical yield. We have developed a

chemo-enzymatic DKR to obtain higher desired yield for (S)-ibuprofen.1 The combined

base catalyst with lipase has resulted in high conversion and excellent ee of the product.

4.6.1 Procedure 1: Chemical Synthesis of (R,S)-2-Ibuprofen Ethoxyethyl Ester

O

HO

(R,S)-ibuprofen acid

HOO

2-ethoxyethanol

p-toluenesulfonic acid

isooctane

O

OO

+

(R,S)-2-ethoxyethyl-2-(4-isobutyl-phenyl)propionate ester


• (R,-S)Ibuprofen (20 g, 96.9 mmol)

• p-toluenesulfonic acid (0.44 g, 2.27 mmol)

• 2-ethoxyethanol (21.9 mL, 224 mmol)

• isooctane (180 mL)

• 5 % sodium hydroxide solution (25 mL, 3�)

• water (25 mL, 3�)

• sodium sulfate anhydrous (10 g)

• boiling chips as anti-bumping agent

• Dean–Stark apparatus

• separator funnel

• Erlenmeyer flask


• micro-distillation unit.

4.6.1.2 Procedure

1. (R,S)-Ibuprofen acid (20 g, 96.9 mmol), p-toluenesulfonic acid (0.44 g, 2.27 mmol) and

2-ethoxyethanol (21.9 mL, 224 mmol) were dissolved in the organic solvent, isooctane

(180 mL). The boiling chips were added into the reaction flask as anti-bumping agent.

Then the reaction mixtures were refluxed for 8�9 h using a Dean–Stark apparatus.

2. The reaction mixtures were cooled to room temperature and the contents of the flask

were poured into a separating funnel for extraction purposes. The mixture was washed

twice with 5 % sodium hydroxide solution (25 mL) then water (25 mL) and the organic

layer was transferred into an Erlenmeyer flask and dried using anhydrous sodium

sulfate (10 g).

3. The drying step was repeated several times in order to achieve a translucent organic

layer. The organic solvent (isooctane) was removed from the (R,S)-ibuprofen ester


mixture using a rotary evaporator with a bath temperature of 100 �C. In order to obtain

the pure (R,S)-ibuprofen ester, the residue was purified further using a micro-

distillation unit to give the pure (R,S)-2-ethoxyethyl ibuprofen ester.2 The ibuprofen

ester which was synthesized chemically can be used in the kinetic resolution or the

DKR without being purified further by the micro-distillation unit.

4. The racemic (R,S)-ibuprofen ester obtained from chemical synthesis was characterized

by FTIR and 1H NMR.

The FTIR analysis showed that the ester functional group appeared at 1736 cm�1.1H NMR (400 MHz, CDCl3) �¼ 0.91 (9H, t), 1.13 (3H, t), 1.35 (3H, t), 1.85 (1H, m),

2.45 (2H, d), 3.41 (2H, q), 3.95 (2H, t), 3.73 (1H, q), 4.10 (2H, t), 7.15(4H, q).

4.6.2 Procedure 2: Immobilized Lipase Preparation


• Lipase from Candida rugosa, EC 3.1.1.3 (type IV), 724 U/mg solids (2.2 g)

• phosphate buffer solution (50 mL, pH 7.0)

• Amberlite XAD7 (2 g)

• ethanol (for washing)

• Erlenmeyer flask

• incubator shaker (200 rpm)

• filter paper of 0.45 mm pore size

• oven (60 �C).

4.6.2.2 Procedure

1. The immobilized lipase was prepared by adsorption of the lipase onto Amberlite

XAD7. The lipase solution (50 mL) was prepared by dissolving 2.2 g of crude lipase

in 50 mL of phosphate buffer solution, pH 7. The lipase solution was gently stirred for a

few minutes until dissolved.

2. 2 g of support (Amberlite XAD7) prewashed with ethanol and phosphate buffer

solution was prepared. The washed Amberlite was dried in the oven for 6 h at 60 �C.

Then the dry Amberlite XAD7 was added into the lipase solution in an Erlenmeyer

flask and the mixture was gently shaken for 24 h at room temperature. After 24 h

agitation, the immobilized lipase was filtered using filter paper of 0.45 mm pore size.

The residual immobilized lipase preparation was subsequently dried in a vacuum

oven until a constant weight was achieved and stored at 4 �C for further use.

3. High-performance liquid chromatography (HPLC) was used to determine the product and

remaining substrate concentration. HPLC analysis was carried out using the chiral column

from Regis, (R,R)-Whelk-01. The column is capable of separating R and S acid derivatives

and also R and S ester derivatives. The performance of the HPLC analysis was enhanced

using this new combination in mobile phases which are: n-hexane: 2-propanol: ethanol +

ammonium acetate in the ratio 47:47:6 (v/v/v). This combination gives better separation

of S-ibuprofen and appeared at earlier retention time compared to the previous mobile

phases mentioned in the manuscript BBE 28:227–233.The expected retention times were

4.12 min and 4.66 min for (S)-ibuprofen acid and (R)-ibuprofen acid respectively.


4.6.3 Procedure 3: Enzymatic Kinetic Resolution of (R,S)-2-Ethoxyethyl Ibuprofen

Ester with Immobilized Lipase

where KS�KR.


• (R,S)-2-Ethoxyethyl ibuprofen ester (10 mL)

• isooctane (250 mL) ester stock solution

• lipase immobilized on Amberlite XAD7 (0.1 g)

• phosphate buffer at pH 7.0 (25 mL of 50 mm phosphate buffer at pH 7.0)

• Erlenmeyer flasks, 250 mL

• incubator shaker, 200 rpm at 40 �C.

4.6.3.2 Procedure

1. (R,S)-2-Ethoxyethyl ibuprofen ester (10 mL) was dissolved in isooctane (250 mL) to

prepare a substrate stock solution of 20 mM.

2. The immobilized lipase (0.1 g) in pH 7 phosphate buffer (25 mL) was added to 25 mL

(20 mM) of ester stock solution in a 250 mL Erlenmeyer flask (reaction flask). The

reaction flask was incubated in an incubator shaker at 40 �C with the agitation speed set

to 200 rpm. Samples from the organic phase and aqueous phase were withdrawn at 24 h

intervals over a 5-day reaction period. The samples collected were filtered using 0.45

mm nylon filter and injected into the HPLC system to determine the rate of resolution by

monitoring both substrate ((R,S)-2-ethoxyethyl ibuprofen ester) and product

(S-ibuprofen acid concentration).

4.6.3.3 Analytical Results

The results obtained from the studies are shown in Table 4.6. The process parameters for

the kinetic resolution process were optimized using the one factor at a time procedure. The

highest enantiomeric excesses achieved for eep and ees were 96.3 % and 84.9 %

respectively.

KS

COOH

HCH3

(S)-Ibuprofen acid

Lipase, H2O

O

OO

(S)-Ibuprofen ester

+HO

O

2-ethoxyethanol

KR

COOH

HCH3

(R)-Ibuprofen acid

Lipase, H2O

O

OO

(R)-Ibuprofen ester

+ HOO

2-ethoxyethanol


4.6.4 Procedure 4: DKR of (R,S)-2-Ethoxyethyl Ibuprofen Ester


• 0.5 M sodium hydroxide solution (25 mL)

• (R,S)-ibuprofen ester (0.27 g, 40 mmol)

• immobilized lipase (0.2 g)

• isooctane (25 mL)

• 5 M HCl

• 100 mL Erlenmeyer flask

• orbital shaker.

4.6.4.2 Procedure

1. To isooctane (25 mL) was added (R,S)-ibuprofen ester (0.27 g, 40 mM), 0.5 M sodium

hydroxide (25 mL) and immobilized lipase (0.2 g). The reaction medium consisted of

two layers of solution, namely ibuprofen ester in isooctane and aqueous NaOH, where

the reaction only occurred at the interface between these two layers. The mixture was

agitated in an orbital shaker at constant temperature of 45 �C and at 200 rpm.

Table 4.6 The optimum conditions for kinetic resolution

Parameter Optimum conditions/highestprocess performance

Enantiomericexcess (%)

Eep Ees

Enzyme loading 0.2 g 93.5 68.5Temperature 45 �C 94.5 73.0Agitation speed 200 rpm 92.9 73.0Substrate concentration 40 mmol L�1 96.3 84.9

COOH

HCH3

COOH

HCH3

(R)-Ibuprofen acid

(S)-Ibuprofen acid

Lipase, H2O

Lipase, H2O

O

OO

(S)-Ibuprofen ester

O

OO

(R)-Ibuprofen ester

+

+

HOO

2-ethoxyethanol

HOO

2-ethoxyethanol

NaoH

Fast

Slow

Rapidly equilibrating(racemizing substrate)


2. Samples of the aqueous phase were withdrawn at different time intervals for analysis

purposes. The sample was extracted with isooctane and the pH was adjusted to pH 1–2

using 5 M HCl before HPLC analysis was carried out.

4.6.4.3 Analytical Results

The overall results for DKR process are shown in Table 4.7. Various process parameters

were optimized. The final quantity of the S-enantiomer increased in the DKR process from

34.85 mmol L�1 to 68.63 mmol L�1.

4.6.5 Conclusion

The procedure shows that it is feasible to combine racemization with the kinetic resolu-

tion process (hence the DKR) of (R,S)- ethoxyethyl ibuprofen ester. The chemical

synthesis of the ester can be applied to any esters, as it is a common procedure. The

immobilized lipase preparation procedure can also be used with any enzymes or support

of choice. However, the enzyme loading will need to be optimized first. The procedures

for the enzymatic kinetic resolution and DKR will need to be adjusted accordingly with

different esters. Through this method, the enantiopurity of (S)-ibuprofen was found to be

99.4 % and the conversion was 85 %. It was demonstrated through our work1 that the

synthesis of (S)-ibuprofen via DKR is highly dependent on the suitability of the reaction

medium between enzymatic kinetic resolution and the racemization process. This is

because the compatibility between both processes is crucial for the success of the DKR.

The choice of base catalyst will vary from one reaction to another, but the basic

procedures used in this work can be applied. DKRs of other profens have been reported

by Lin and Tsai3 and Chen et al.4

Table 4.7 Results for DKR process

Parameter Optimum conditions [(S)-Ibuprofen acid] (mmol L�1) Eep (%)

NaOH 0.5 M 27.70 98.40Temperature 45 �C 31.64 98.58Dimethylsulfoxide 20 % 35.07 98.71Substrate concentration 40 mmol L�1 68.63 99.36


1. Fazlena, H., Kamaruddin, A.H. and Zulkali, M.M.D., Dynamic kinetic resolution: alternativeapproach in optimizing S-ibuprofen production. Bioprocess Biosyst. Eng., 2006, 28, 227–233.

2. Long, W.S. Kamaruddin, A.H. and Bhatia, S., Chiral Resolution of Racemic Ibuprofen Ester in anEnzymatic Membrane Reactor. Journal of Membrane Science., 2005, 247, 185–200.

3. Lin, C.N. and Tsai, S.W., Dynamic kinetic resolution of suprofen thioester via coupled triocty-lamine and lipase catalysis. Biotechnol. Bioeng., 2000, 69, 31–38.

4. Chen, C.Y, Cheng, Y.C. and Tsai, S.W., Lipase-catalyzed dynamic kinetic resolution of (R,S)-fenoprofen thioester in isooctane. J. Chem. Technol. Biotechnol., 2002, 77, 699–705.


4.7 Dynamic Kinetic Resolution Synthesis of a Fluorinated Amino AcidEster Amide by a Continuous Process Lipase-mediated Ethanolysis ofan AzalactoneMatthew Truppo, David Pollard, Jeffrey Moore and Paul Devine

Chiral g-fluoroleucine-�-amino acid pharmaceutical intermediates may be synthesized

enzymatically in a dynamic kinetic resolution through an azalactone ring-opening coupled

with spontaneous racemization of the azalactone through enol tautomerization.1 For this

reaction system to maintain good selectivity and productivity, competing side reactions,

including background hydrolysis and nucleophilic alcohol addition, must be minimized.

Kinetic modeling of these reactions was recently used to guide process optimization and

dramatically reduced the amount of enzyme required to catalyze the resolution.2 The

reaction has been demonstrated in both fed-batch and continuous operations, and has

been shown to be readily scaled for large productions (Figure 4.3).

4.7.1 Procedure 1: Fed-batch Operation


• Azalactone (320 g)

• ethanol (137.6 g)

• triethylamine (7.6 g)

• methyl tert-butyl ether (MTBE) (1 L)

• immobilized Candida antarctica lipase B Novozym 435 (Amano, 80 g).

ON

O

F NH

CO2Et

O

F

NH

CO2Et

O

F

NH

CO2H

O

F

12

3

4

NH

CO2H

O

F

5

EDC

backgroundethanolysis

+EtOH

immob CalB, EtOH

backgroundhydrolysis

+H2O

Figure 4.3 Enzymatic chiral synthesis of fluoroleucine derivative (EDC ¼ 1-(3-dimethylami-nopropyl)-3-ethylcarbodiimide hydrochloride)


4.7.1.2 Procedure

1. Azalactone substrate (80 g, 0.37 mol), ethanol (86 g, 1.87 mol, 5 equiv) and triethyla-

mine (7.6 g, 75 mmol, 0.2 equiv) were added to 1 L MTBE that was stirring at 50 �C.

The reaction was started with the addition of immobilized lipase Novozym 435 (80 g).

2. At time points 0.5, 1.5 and 3.0 h after the initiation of the reaction, more azalactone

(80 g, 0.37 mol) and ethanol (17.2 g, 0.37 mol) were added to the mixture. Conversion

was complete after 6 h total reaction time.

3. The mixture was filtered and the enzyme washed with MTBE until the filtrate turned

colorless. The filtrate was then successively washed with 1 M aqueous HCl, saturated

NaHCO3 and brine and then concentrated and deprotected directly without further

purification (typical yield: 80 %).

1H NMR (400 MHz; CDCl3) � 6.16 (1H, br), 5.82 (1H, ddt, J¼ 17.1, 10.3, 6.1 Hz), 5.07

(1H, dq, J¼ 17.1, 1.5 Hz), 5.00 (1H, dq, J¼ 10.3, 1.5 Hz), 4.19 (2H, mq, J¼ 7.2 Hz), 2.39

(2H, m), 2.32 (2H, m), 2.12 (1H, ddd, J¼ 25.2, 15.2, 5.2 Hz), 2.05 (1H, ddd, J¼ 19.2, 15.2,

8.4 Hz), 1.42 (3H, d, J ¼ 21.6 Hz), 1.40 (3H, d, J ¼ 21.5 Hz), 1.28 (3H, t, J ¼ 7.0 Hz).

Product ester ee was determined by isocratic normal-phase high-performance liquid

chromatography using a Chiralcel OD-H (250 mm � 4.6 mm) column and a 98 %

hexanes/2 % isopropanol mobile phase at 1.75 mL min�1 and 25 �C. The undesired (R)-

ester and desired (S)-ester were quantified using their characteristic retention times of 10.3

min and 21 min respectively during elution.

4.7.2 Procedure 2: Continuous Operation


• Azalactone (1 kg)

• ethanol (1.075 kg)

• triethylamine (95 g)

• MTBE (13 L)

• immobilized C. antarctica lipase B Novozym 435 (Amano, 50 g)

• jacketed column

• sulfuric acid (0.5 M).

4.7.2.2 Procedure

1. A slurry of immobilized lipase Novozym 435 in MTBE was prepared and packed into a

jacketed column at atmospheric pressure (Figure 4.4).

2. A substrate solution containing azalactone (1 kg, 4.67 mol) in MTBE (6.25 L) and an

alcohol solution containing ethanol (1.075 kg, 23.37 mol) and triethylamine (95 g,

0.94 mol) in MTBE (6.25 L) were prepared.

3. The substrate and alcohol solutions were then passed through the column, which was

maintained at 50 �C, at rates of 312 mL h�1 each. Mixing of the two solutions occurred

immediately prior to the column to minimize nonenzymatic background reactions.

Flow exiting the column was passed through a back-pressure regulator set at 20 psi to

prevent any boiling from taking place. The exiting mixture was then fed into a quench

vessel containing 0.5 M sulfuric acid.

4.7 DKR Synthesis of a Fluorinated Amino Acid Ester Amide 163

4.7.3 Conclusion

Previously described batch processes for synthesizing fluoroleucine ethyl esters were

impractical because of their high biocatalyst loading levels required to minimize the

background reactions. By utilizing enol tautomerization to racemize the azalactone in

the kinetic resolution, increasing the temperature to 50 �C and implementing a substrate

feeding strategy, the biocatalyst requirements are reduced and the economics of the

process are made much more attractive. By further improving the process to incorporate

a continuous plug flow rather than fed-batch reactor, enzyme deactivation through shear is

rendered insignificant, and the enzyme-to-substrate ratio and associated cost drop still

more. The data presented in Table 4.8 clearly show the benefit of the plug flow reactor

relative to alternate reactor configurations. This continuous process is highly scalable and

has been demonstrated in operations generating multiple kilograms f product.

Table 4.8 Reactor configuration data

Enzyme:substrate Undesired acid (wt%) Ester Reactor volume (l)

Yield (wt%) Ee (%)

Batch 1:1 17 79 78 1250Fed batch 1:4 6 84 78 383Column 1:>20 2 90 86 24

Figure 4.4 Continuous operation setup

References

1. Turner, N.J., Winterman, J.R., McCague, R., Parratt, J.S. and Taylor, S.J.C., Synthesis ofhomochiral l-(S)-tert-leucine via a lipase catalysed dynamic resolution process. TetrahedronLett., 1995, 36, 1113.

2. Limanto, J., Shafiee, A., Devine, P.N., Upadhyay, V., Desmond, R.A., Foster, B.R., Gauthier,D.R., Reamer, R.A. and Volante, R.P., An efficient chemoenzymatic approach to(S)-g-fluoroleucine ethyl ester. J. Org. Chem. 2005, 70, 2372. Truppo, M.D. and Moore, J.C.,Process for making fluoroleucine ethyl esters. US Pat. Appl., 2007, US 2007/0059812 A1.


5

Enzymatic Selectivity in SyntheticMethods

5.1 Alcalase-catalysed Syntheses of Hydrophilic Di- and Tri-peptides inOrganic SolventsXue-Zhong Zhang, Rui-Zhen Hou, Li Xu and Yi-Bing Huang

A practical enzymatic procedure using alcalase as biocatalyst has been developed

for the synthesis of hydrophilic peptides.1–3 Alcalase is an industrial alkaline pro-

tease from Bacillus licheniformis produced by Novozymes that has been used as a

detergent and for silk degumming.4 The major enzyme component of alcalase is the

serine protease subtilisin Carlsberg, which is one of the fully characterized bacterial

proteases. Alcalase has better stability and activity in polar organic solvents, such as

alcohols, acetonitrile, dimethylformamide, etc., than other proteases.5 In addition,

alcalase has wide specificity and both L- and D-amino acids that are accepted as

nucleophiles at the p-10 subsite.6 Therefore, alcalase is a suitable biocatalyst to

catalyse peptide bond formation in organic solvents under kinetic control without

any racemization of the amino acids (Scheme 5.1).

R OR'

O

R O

O

Enz

H2Ohydrolysis

R''NH2aminolysis

R OH

O

R NHR''

O

Scheme 5.1 The principle of protease-catalysed kinetically controlled peptide synthesis.



5.1.1 Procedure 1: Synthesis of Bz-Arg-Gly-NH2

NH

OPh

O

O

EtCONH2

HN

NH

Ph

O

O

CONH2H2NEt3N

+AlcalaseNH

NH

H2N

HCl

NH

NH

H2N


• Alcalase (0.3 mL) (Novo Industrial, Denmark; 2.5 AU mL�1)

• anhydrous ethanol (2 mL)

• Bz-Arg-OEt�HCl (34.3 mg, 0.1 mmol)

• Gly-NH2 (51.8 mg, 0.7 mmol)

• triethylamine (28 mL, 0.2 mmol)

• acetonitrile (1.8 mL)

• pH 10.0, 0.1 M Na2CO3–NaHCO3 buffer (0.2 mL)

• Sephadex G-10 column (16� 1000 mm) (Pharmacia)

• high-performance liquid chromatography (HPLC; Agilent 1100N-1946C), C18 col-

umn (Zorbax Extend C18, 150 mm� 3 mm). Mobile phase A: 0.1 % TFA; mobile

phase B: acetonitrile. Flow rate: 0.5 mL min�1. Gradient: start with 8 % B, at 10 min

48 % B, post time 3 min. The elution was monitored at 220 nm. Oven temperature

was 25 �C

• HPLC–mass spectrometry (MS). MS conditions: ionization mode, atmospheric pressure

ionization–electrospray (API–ES); polarity, positive; Vcap, 4000 V; nebulizer pressure,

35 psig; drying gas, 10 L min�1; gas temperature, 350 �C; fragmentor, 70 V; scan range,

120–600 atm.

5.1.1.2 Procedure

1. For pretreatment of the enzyme, alcalase (0.3 mL) and anhydrous ethanol (2 mL)

were added to a centrifuge tube and the mixture was agitated for 5 min. The resulting

mixture was centrifuged at 3000 rpm for 15 min to separate the enzyme from the

solvent and the ethanol was removed by decantation. This procedure was repeated

three times. 0.2 mL of a 0.1 M Na2CO3–NaHCO3 buffer solution (pH 10.0) was

added to the pretreated enzyme obtained from 0.3 mL of untreated alcalase and

incubated for 10 min at 45 �C.

2. Bz-Arg-OEt�HCl (34.3 mg, 0.1 mmol), Gly-NH2 (51.8 mg, 0.7 mmol) and triethyla-

mine (28 mL, 0.2 mmol) were dissolved in acetonitrile (1.8 mL). The above enzyme

solution was added to start the enzymatic reaction. Aliquots (0.5 mL) were periodically

taken from the reaction mixture, quenched by adding 10 % trichloroacetic acid (TCA,

0.25 mL) and after centrifugation (8000 rpm for 10 min), the supernatant was analysed

by HPLC.

166 Enzymatic Selectivity in Synthetic Methods

3. The target dipeptide product was purified on a Sephadex G-10 column (16 mm � 1000

mm) equilibrated and eluted with water at the elution rate of 1.0 mL min�1. The elution

process was monitored at 220 nm. The fractions collected were lyophilized to afford the

desired product (29.6 mg). The HPLC purity and the yield of the product were 93.5 %

and 82.9 % respectively.

4. The reaction products were identified by HPLC–MS. MS conditions: ionization mode,

API–ES; polarity, positive; Vcap, 4000 V; nebulizer pressure, 35 psig; drying gas, 10 L

min�1; gas temperature, 350 �C; fragmentor, 70 V; scan range, 120–600 atm.

5.1.2 Procedure 2: Synthesis of Z-Asp-Ser-NH2

HNO

HO2C

O

Me + CONH2HNAlcalase

Et3N

OHO OPh

HN

HO2C

OO OPh

CONH2H2N

OH



• Z-Asp-OMe (28.1 mg, 0.1 mmol)

• Ser-NH2 (72.8 mg, 0.7 mmol)


• acetonitrile (1.7 mL)

• pH 10.0, 0.1 M Na2CO3–NaHCO3 buffer (0.3 mL)

• Sephadex G-10 column (16 mm � 1000 mm) (Pharmacia)

• HPLC (see Procedure 1, Section 5.1.1)

• HPLC–MS (see Procedure 1, Section 5.1.1).

5.1.2.2 Procedure

5. 0.3 mL of 0.1 M Na2CO3–NaHCO3 buffer (pH 10.0) was added to the pretreated enzyme

obtained from 0.3 mL of untreated alcalase (see Procedure 1, step 1, Section 5.1.1) and

the mixture incubated for 10 min at 35 �C.

6. Z-Asp-OMe (28.1 mg, 0.1 mmol), Ser-NH2 (72.8 mg, 0.7 mmol) and triethylamine

(56 mL, 0.4 mmol) were dissolved in acetonitrile (1.7 mL). The above enzyme

solution was added to start the enzymatic reaction. Aliquots (0.5 mL) were taken

from the reaction mixture periodically, quenched by adding 10 % TCA (0.25 mL)

and, after centrifugation (8000 rpm for 10 min), the supernatant was analysed by

HPLC.

7. The target dipeptide product was purified on a Sephadex G-10 column (16 mm � 1000

mm) equilibrated and eluted with water at an elution rate of 1.0 mL min�1. The elution




8. The reaction product Z-Asp-Ser-NH2 was identified by HPLC–MS.

5.1 Alcalase-catalysed Syntheses of Hydrophilic Di- and Tri-peptides 167

5.1.3 Procedure 3: Synthesis of Bz-Arg-Gly-Asp(-NH2)-OH

NH

OPh

O

O

Et

HN

NH

Ph

O

O OHN

CONH2

Et3N

NH2

O

HN CONH2

CONH2

COOH

+ Alcalase

NH

NH

H2N

HCl NH

NH

H2N

Note. NMR analyses revealed that the major component of the tripeptide product synthe-

sized from Bz-Arg-OEt and Gly-Asp-(NH2)2 was Bz-RGD(-NH2)-OH rather than Bz-

RGD-(NH2)2. The primary amide group adjacent to the secondary carbon centre is

hydrolysed during the reaction, whereas the primary amide adjacent to the primary carbon

centre remains intact, as determined by heteronuclear multiple bond correlation.3



• Bz-Arg-OEt�HCl (34.3 mg, 0.1 mmol)

• Gly-Asp-(NH2)2 (94 mg, 0.5 mmol)


• absolute ethanol (1.7 mL)

• pH 8.0, 0.1 M tris-HCI buffer (0.3 mL)

• Sephadex G-10 (Pharmacia) column (16 mm� 1000 mm)

• HPLC (see Procedure 1, Section 5.1.1)

• HPLC–MS (see Procedure 1, Section 5.1.1)

• NMR (Bruker av 600 spectrometer).

5.1.3.2 Procedure

9. 0.3 mL of 0.1 M tris-HCI buffer (pH 8.0) was added to the pretreated enzyme obtained

from 0.3 mL untreated alcalase (see Procedure 1, step 1, Section 5.1.1) and the mixture

incubated for 10 min at 35 �C.

10. Bz-Arg-OEt�HCl (34.3 mg, 0.1 mmol), Gly-Asp-(NH2)2 (94 mg, 0.5 mmol) and

triethylamine (70 mL, 0.5 mmol) were dissolved in absolute ethanol (1.7 mL) and

incubated for 10 min at 35 �C. The above enzyme solution was added to start the

enzymatic reaction. Aliquots (0.5 mL) were taken from the reaction mixture periodi-

cally, quenched by adding 10 % TCA (0.25 mL) and (after centrifugation at 8000 rpm

for 10 min) the supernatant was analysed by HPLC.

11. The target tripeptide product was purified on a Sephadex G-10 column (16 mm� 1000

mm) equilibrated and eluted with water at an elution rate of 1.0 mL min�1. The elution





12. The target tripeptide product was identified by HPLC–MS.

13. NMR data of Bz-Arg-Gly-Asp(-NH2)-OH1H NMR (dimethylsulfoxide (DMSO); 600.13 MHz) � 12.65 (br, 1H,�COOH),

8.55 (d, 1H,�NH�), 8.26 (t, 1H,�NH�), 8.11 (d, 1H,�NH�), 7.91 (d, 2H, ArH), 7.55

(t, 1H, ArH), 7.47 (t, 2H, ArH), 7.44�6.53 (s, br, 6H,�NH�,�NH2), 4.52 (m, 1H,�CH<), 4.45 (m, 1H,�CH<), 3.77 (dd, 1H,�CH2�), 3.71 (dd, 1H,�CH2�), 3.11

(m, 2H,�CH2�), 2.55 (dd, 1H,�CH2�), 2.46 (dd, 1H,�CH2�), 1.84 (m, 1H,�CH2�),

1.71 (m, 1H,�CH2�), 1.55 (m, 2H,�CH2�).13C NMR (DMSO; 150.92 MHz) � 172.68 (s, 1C,�CO�), 171.78 (s, 1C,�CO� ),

171.12 (s, 1C, �CO�), 168.44 (s, 1C, �CO�), 166.58 (s, 1C, �CO�), 156.67

(s, 1C, >C¼), 133.92 (s, 1C, ArC), 131.40 (d, 1C, ArCH), 128.19 (d, 2C, ArCH),

127.57 (d, 2C, ArCH), 53.06 (d, 1C,�CH<), 48.64 (d, 1C,�CH<), 41.69 (t, 1C,�CH2�), 40.43 (t, 1C,�CH2�), 36.74 (t, 1C,�CH2�), 28.60 (t, 1C,�CH2�), 25.28

(t, 1C,�CH2�).

5.1.4 Conclusion

The industrial alkaline protease alcalase has been used to synthesize hydrophilic peptides

in organic solvents under kinetic control. For the synthesis of Bz-Arg-Gly-NH2, the

optimum conditions were pH 10.0, 45 �C, in an acetonitrile/0.1 M Na2CO3–NaHCO3

buffer system (90:10, v/v), 1 h with a dipeptide yield of 82.9 %. For the synthesis of Z-

Asp-Ser-NH2, the optimum conditions are pH 10.0, 35 �C, in an acetonitrile/Na2CO3–

NaHCO3 buffer system (85:15, v/v), 6 h with a dipeptide yield of 75.5 %. For the synthesis

of Bz-Arg-Gly-Asp(-NH2)-OH from Bz-Arg-OEt�HCl and Gly-Asp-(NH2)2, the optimum

conditions are pH 8.0, 35 �C, in an ethanol/tris-HCl buffer system (85:15, v/v), 8 h with a

tripeptide yield of 73.6 %. The structure of the tripeptide was confirmed to be Bz-Arg-Gly-

Asp(-NH2)-OH by NMR.

References

1. Hou, R.-Z., Yang, Y., Li, G., Huang, Y.-B., Wang, H., Zhang, N., Liu, Y.-J., Li, X. and Zhang, X.-Z., Synthesis of a precursor dipeptide of RGDS (Arg-Gly-Asp-Ser) catalysed by the industrialprotease alcalase. Biotechnol. Appl. Biochem., 2006, 44, 73.

2. Hou, R.-Z., Yang, Y., Li, G., Huang, Y.-B., Wang, H., Zhang, N., Liu, Y.-J. and Zhang, X.-Z. ,Alcalase[hyphen]catalyzed, kinetically controlled synthesis of a precursor dipeptide of RGDS inorganic solvents. Prep. Biochem. Biotechnol., 2006, 36, 93.

3. Hou, R.-Z., Yang, Y., Li, G., Huang, Y.-B., Wang, H., Zhang, N., Liu, Y.-J., Li, X. and Zhang, X.-Z., Synthesis of tripeptide RGD amide by a combination of chemical and enzymatic methods. J.Mol. Catal. B: Enzym., 2005, 37, 9.

4. (a) Gupta, R., Beg, Q.K. and Lorenz, P., Bacterial alkaline proteases: molecular approaches andindustrial applications., Appl. Microbiol. Biotechnol., 2002, 59, 15; (b) Maurer, K.-H., Detergentproteases. Curr. Opin. Biotechnol., 2004, 15, 330.

5. Zaks, A. and Klibanov, A.M., Enzymatic catalysis in nonaqueous solvents. J. Biol. Chem., 1988,263, 3194.

6. Chen, S.-T., Chen, S.-Y. and Wang, K.-T., Kinetically controlled peptide bond formation inanhydrous alcohol catalyzed by the industrial protease alcalase. J. Org. Chem., 1992, 57, 6960.

5.1 Alcalase-catalysed Syntheses of Hydrophilic Di- and Tri-peptides 169

5.2 Selective Alkoxycarbonylation of 1a,25-Dihydroxyvitamin D3 DiolPrecursor with Candida antarctica Lipase BMiguel Ferrero, Susana Fernandez and Vicente Gotor

Selective protection/deprotection of compounds containing multiple hydroxyl groups is a

challenging problem in organic synthesis. For the manipulation of protecting groups,

enzymatic esterification reactions have been commonly used, whereas the alkoxycarbo-

nylation reaction has scarcely been investigated. Previously, we reported regioselective

enzymatic alkoxycarbonylation of 1�,25-dihydroxyvitamin D3 A-ring precursor using

lipase-mediated reaction in organic solvent.1 Candida antarctica lipase B was found to

be the best catalyst in toluene, affording transformation exclusively at the C-5-(R)-hydro-

xyl group. Excellent yield of the carbonate derivative was achieved.

5.2.1 Procedure 1: Synthesis of Acetone O-[(Vinyloxy)carbonyl]oxime

O ON

O


• Vinyl chloroformate (50 mmol)

• acetone oxime (35 mmol)

• anhydrous pyridine (10 mL)

• dichloromethane (3 � 20 mL)

• brine


• N2 gas

• filter paper

• one 25 mL Schlenk

• one addition funnel

• one separatory funnel


• vacuum pump

• Kugelrohr distillation equipment.

OHHO OHOO

O

35

C. antarctica lipase B(Novozym 435)

O ON

O

+Toluene, 30 °C, 4 h

98% yield

Scheme 5.2


5.2.1.2 Procedure

1. In a 25 mL Schlenk with magnetic stirrer and addition funnel, vinyl chloroformate (4.25 mL,

50 mmol) was added dropwise to a solution of acetone oxime (2.56 g, 35 mmol) in

anhydrous pyridine (10 mL) at 0 �C. The reaction was stirred overnight at room temperature.

2. Pyridine was evaporated under vacuum (10�2–10�5 mmHg) and the residue was

extracted with dichloromethane (3 � 20 mL). The combined organic fractions were

washed with brine (15 mL), dried over Na2SO4, filtered and concentrated using a rotary

evaporator. The crude residue was purified by vacuum distillation using Kugelrohr

apparatus. 4.5 g (90 % yield).1H NMR (CDCl3, 300 MHz): 2.36 (s, 3H, Me), 2.37 (s, 3H, Me), 5.03 (dd, 1H, CH2-

cis, 3JHH 6.1, 2JHH 1.6 Hz), 5.31 (dd, 1H, CH2-trans, 3JHH 13.8, 2JHH 1.6 Hz), and 7.50

(dd, 1H, CH, 3JHH 13.9, 3JHH 6.2 Hz).

5.2.2 Procedure 2: Synthesis of (3S,5R)-1-Ethynyl-3-hydroxy-2-methyl-5-

[(vinyloxy)carbonyl]-1-cyclohexene

OHOO

O


• C. antarctica lipase B (Novozym 435, 7300 U/g) (570 mg)

• acetone O-[(vinyloxy)carbonyl]oxime (8.2 mmol)

• (3S,5R)-1-ethynyl-3,5-dihydroxy-2-methyl-1-cyclohexene2 (0.8215 mmol)

• anhydrous toluene (30 mL)

• dichloromethane

• ethyl acetate

• hexanes

• thin-layer chromatography (TLC) plates (silica gel 60 F254)

• silica gel 60 (230-400 mesh)

• N2 gas

• one 100 mL Erlenmeyer

• one Buchner funnel with joint

• one flash chromatography column

• orbital shaker


• vacuum pump.

5.2.2.2 Procedure

1. A solution of (3S,5R)-1-ethynyl-3,5-dihydroxy-2-methyl-1-cyclohexene (125 mg, 0.82

mmol) and acetone O-[(vinyloxy)carbonyl]oxime (1.17 g, 8.2 mmol) in toluene

5.2 Selective Alkoxycarbonylation of 1a,25-Dihydroxyvitamin D3 Diol Precursor 171

(30 mL) was added to a 100 mL Erlenmeyer containing C. antarctica lipase B (570 mg)

under N2 gas. The suspension was shaken in an orbital shaker at 30 �C for 4 h. The

progress of the reaction was followed by TLC (50 % EtOAc/hexane).

2. The mixture was filtered, the enzyme washed with CH2Cl2 and the filtrate concentrated.

3. The crude residue was purified by flash chromatography (gradient elution with 10–30 %

EtOAc/hexanes). 179 mg (98 % yield).1H NMR (CDCl3; 300 MHz) � 2.01 (s, 3H, H9), 2.05 (m, 2H, H4), 2.16 (br s, 1H, OH),

2.31 (dd, 1H, H6, 2JHH 17.2, 3JHH 7.5 Hz), 2.67 (dd, 1H, H6, 2JHH 17.2, 3JHH 3.7 Hz),

3.10 (s, 1H, H8), 4.29 (br s, 1H, H3), 4.58 (dd, 1H, H12-cis, 3JHH 6.4, 2JHH 2.1 Hz), 4.91

(dd, 1H, H12-trans, 3JHH 13.8, 2JHH 2.1 Hz), 5.07 (m, 1H, H5) and 7.06 (dd, 1H, H11,3JHH 13.8, 3JHH 6.4 Hz).

13C NMR (CDCl3, 75.5 MHz) � 18.32 (C9), 34.66 (C4), 35.99 (C6), 68.03 (C3), 71.29

(C5), 80.93 (C8), 82.37 (C7), 97.88 (C12), 113.56 (C1), 142.37 (C11), 142.81 (C2), and

152.00 (C10).

High-resolution mass spectrometry (m/z). Calc. for C12H14O4: 222.0892. Found:

222.0895

5.2.3 Conclusion

An efficient and high-yielding enzymatic protocol for regioselective alkoxycarbonylation

of the diol precursor of 1�,25-dihydroxyvitamin D3 has been accomplished. The proce-

dure provided a convenient synthesis of the A-ring vinyl carbonate derivative, which is a

useful synthon of vitamin D3 analogues for pharmaceutical research.3

References

1. Ferrero, M., Fernandez, S. and Gotor, V., Selective alkoxycarbonylation of A-ring precursors ofvitamin D using enzymes in organic solvents. Chemoenzymatic synthesis of 1�,25-dihydroxyvi-tamin D3 C-5 A-ring carbamate derivatives. J. Org. Chem., 1997, 62, 4358.

2. Previously synthesized in: Okamura, W.H., Aurrecoechea, J.M., Gibbs, R.A. and Norman, A.W.,Synthesis and biological activity of 9,11-dehydrovitamin D3 analogues: stereoselective prepara-tion of 6�-vitamin D vinylallenes and a concise enynol synthesis for preparing the A-ring. J. Org.Chem., 1989, 54, 4072.

3. (a) Gotor-Fernandez, V., Fernandez, S., Ferrero, M., Gotor, V., Bouillon, R. and Verstuyf, A.,Chemoenzymatic synthesis and biological evaluation of C-3 carbamate analogues of 1�,25-dihydroxyvitamin D3. Bioorg. Med. Chem., 2004, 12, 5443. (b) Oves, D.; Fernandez, S.;Verlinden, L.; Bouillon, R.; Verstuyf, A.; Ferrero, M.; Gotor, V., Novel A-ring homodimericC-3-carbamate analogues of 1�,25-dihydroxyvitamin D3: synthesis and preliminary biologicalevaluation. Bioorg. Med. Chem., 2006, 14, 7512.


5.3 The Use of Lipase Enzymes for the Synthesis of Polymers and PolymerIntermediatesAlan Taylor

The control of the synthesis of polymers is crucial to obtain the final bulk properties of the

polymers needed for the end application. The use of enzymes in polymer synthesis has

been demonstrated to allow control of polymer properties such as average molecular

weight and dispersity, avoid the use of toxic intermediates, enable the selective reaction

of functional groups and allow the use of unstable intermediates.

This section describes the synthesis of oxazolidine esters used as polymer hardeners that

cannot be synthesized using chemical catalysis, the synthesis of polyurethane polymers

with methods that avoid the use of isocyanates and the enzymatic synthesis of polyesters

with low molecular weight dispersity.

5.3.1 Synthesis of Oxazolidine Esters

One of the best examples of the utility of enzymatic synthesis in catalyzing reactions that

cannot be accomplished by any other route is the synthesis of substituted oxazolidine

diesters. The oxazolidine ring is extremely water sensitive, the oxazolidine rapidly revert-

ing back to the alkanolamine and aldehyde in the presence of water. Bis-oxazolidines have

been used as hardeners for polymer coatings but the diester based on the hydroxyethyl

oxazolidine and adipic acid cannot be synthesized directly with chemical catalysis because

of the rapid rate of reaction of the oxazolidine ring with either the water from the

esterification or the alcohol from transesterification.1

The advent of the low temperature, enzymatic esterification process offered the oppor-

tunity to manipulate the various reaction rates so that the ester might be formed keeping the

oxazolidine ring intact (Figure 5.1).

The dimethyl ester of adipic acid, rather than adipic acid, was used as a transesterifica-

tion substrate. Reaction rate studies had shown that the transesterification would be much

faster than the esterification reaction. It was considered that the rate of attack on the

oxazolidine ring by methanol would be slower than the rate of attack by water and that the

ring opening would not be catalysed by the enzyme, whereas the rate of the transester-

ification would be increased significantly, particularly at the low temperature of the

enzymatic esterification.

O N

R OH

CO2R1R1O2C

+

cat Novozym 43560 °C

O

O

O

O

N

N O

O

R

R

+2R1OH

Figure 5.1 Synthesis of di-[2-(2-isopropyl-1,3-oxazolidin-3-yl)ethyl] hexane-1,6-dioate

5.3 Use of Lipase Enzymes for the Synthesis of Polymers 173

5.3.2 Procedure 1: Synthesis of Di-[2-(2-isopropyl-1,3-oxazolidin-3-yl)ethyl]

Hexane-1,6-dioate

O

O

O

O

N

N O

O


• 2-(2-iso-Propyl-1,3-oxazolidin-3-yl)ethanol (60 g, 0.38 mol)

• Dimethyl adipate (33.03 g, 0.19 mol)

• Novozym 435 (2.02 g)

• four-necked flange flask with stirrer bar

• oil bath


• water-cooled condenser

• vacuum pump with pressure control system.

5.3.2.2 Procedure

1. Novozym 435 (2.02 g) was added to the 2-(2-iso-propyl-1,3-oxazolidin-3-yl)ethanol

(60 g, 0.38 mol) and dimethyl adipate (33.03 g, 0.19 mol).

2. The reaction was maintained at a temperature of 60 �C and a pressure of 400 mmHg with

stirring for 8 h. The pressure was then reduced to 185 mmHg for 16 h then further reduced

to 100 mmHg for 24 h and finally reduced to 10 mmHg for 24 h. Evolved methanol (11 g)

was collected in a liquid-nitrogen trap. Analysis by gas chromatography–mass spectrometry

showed less than 0.1 % of unreacted dimethyl adipate remained. Gel permeation chroma-

tography (GPC) showed a single peak for the white crystalline product (86.2 g, 99 %)

3. Elemental analysis. C22H40N2O6 requires: C 61.66 %, H 9.41 %, N 6.54 %; found:

C 60.83 %, H 9.71 %, N 6.54 % for the crude product without recrystallization.1H NMR (CDCl3; 250 MHz) � 0.93 (12H, bm, (�CH(CH3)2)2), 1.66 (4H, bm,

�CO�CH2�CH2�CH2�CH2�CO�), 2.65 (4H, bm,�CO�CH2�CH2�CH2�CH2�CO�),

3.20 (2H, bm, (�C�CH(CH3)2)2), 3.81 (4H, bm, (�CH2�CH2�O)2), 3.83 (4H, bm,

(�N�CH2�CH2)2), 3.91 (4H, bm, (�N�CH2�CH2�O)2), 4.15 (4H, bm,

(�CH2�CH2�O�CO)2), 4.99 (2H, s, (�O�CH�N)2.

5.3.3 Enzymatic Synthesis of Novel Urethane Polyesters

There have been many attempts to develop non-phosgenation routes to synthesize diiso-

cyanates and for the synthesis of polyurethanes without the use of isocyanates. None of

these has been applied commercially. In the conventional process, the addition of the


isocyanate must occur after esterification because the carbamate group begins decompos-

ing at 160–180 �C, well below the esterification temperature used in polymer production

(typically 220 �C). However, the use of enzymatic methods allows us to reverse the

conventional process by creating the urethane first and then using a low-temperature

enzymatic polyester synthesis to build the polymer. Thus, we were able to synthesize a

novel series of bis-carbamate esters and polyesters.2 It was known from the work of Delaby

et al.3,4 in the 1950s that the carbamate group could be synthesized by the ring-opening

addition of a cyclic carbonate, such as ethylene carbonate with a primary diamine, the

product of this reaction being the bis-(hydroxyethyl) carbamate.

5.3.4 Procedure 2: Synthesis of a Polyester Containing Di(hydroxyethyl)hexamethylene

Bis-carbamate under Solvent-free Conditions

HN

HNO O

O O

O6

O

O

O

4O

O

O

O

4O

Diacid LinkerLinker Diol

n

The dihydroxyethyl hexamethylene bis-carbamate was synthesized by the published

method.5

HN

HNO O

O O

HO OH6

This product was recrystallized from ethanol and dried to give the bis-carbamate as

white crystals (m.p. 94 �C).1H NMR (CDCl3, 250 MHz), � 1.19 (4H, bm,�(NH�CH2�CH2�CH2)2), 1.60 (4H, bm,

�(NH�CH2�CH2�CH2)2), 3.26 (4H, bm, �(NH�CH2�CH2�CH2)2), 3.74 (4H, bm,

�O�CH2�CH2�OH), 4.18 (4H, bm, �O�CH2�CH2�OH), 5.24 (2H, bm,

�(NH�CH2�CH2�CH2)2). 13C (CDCl3, 63 MHz), � 26.10 (t, NH�CH2� CH2�CH2)2),

29.12 (t, �(NH�CH2�CH2�CH2)2), 40.70 (t, �(NH�CH2� CH2�CH2)2), 61.76

(t,�O�CH2�CH2�OH), 66.64 (t,�O�CH2�CH2�OH), 157.30 (s,�O�CO�N).


• Dihydroxyethyl hexamethylene bis-carbamate (7.25 g, 0.0248 mol)

• 1,6-butanediol (22.72 g, 0.252 mol)

• adipic acid (40.17 g, 0.274 mol)

• Novozym 435 (1.2 g)

• nitrogen

• four-necked flange flask with Heidolf mechanical stirrer

• oil bath

• hotplate




5.3.4.2 Procedure

1. Dihydroxyethyl hexamethylene bis-carbamate (7.25 g, 0.0248 mol) and 1,4-butanediol

(22.72 g, 0.252 mol) were placed in a flask and heated to 90 �C under an atmosphere of

nitrogen. Adipic acid (8 g, 0.055 mol) was added and stirred until dissolved.

2. The reactants were cooled to 60 �C and Novozym 435 (0.7 g) was added. The pressure

was reduced to 400 mmHg after 2 h, then further adipic acid (25 g, 0.17 mol) was added

and left for 16 h. The remaining adipic acid (7.17 g, 0.049 mol) was added and the

pressure was reduced to 100 mmHg and left for 24 h. A further amount of Novozym 435

(0.5 g) was added.

3. The reaction temperature was raised to 70 �C and the pressure reduced to 50 mmHg

for a further 24 h. The reaction was stopped and the polyester product sampled.

The molecular weight was determined by GPC: Mw was 9350, Mn 5345 and the

dispersity 1.75.

5.3.5 Enzyme-catalysed Polyurethane Solution Polymerization

using Diphenyl Ether

Polyesters of molecular weight up to Mw 30 000 have been synthesized by Mahapatro

et al.,6 although only in quantities of less than 10 g. Attempts to increase the

molecular weight and batch size showed the importance of the absence of water in

the reactants. It was found necessary to dry the diphenyl ether over 3 A molecular

sieves and then distil under reduced pressure before using in the polymerization

experiments.

5.3.6 Procedure 3: Synthesis of a Polyester Containing

Di(hydroxyethyl)hexamethylene Bis-carbamate in Diphenyl Ether



• 1,6-hexanediol (50.37 g, 0.43 mol)


• Novozym 435 (1.3 g)

• diphenyl ether (136 mL)

• nitrogen


• oil bath

• hotplate

• water-cooled condenser.

5.3.6.2 Procedure (R. Van Calck and A. Taylor, Unpublished Results)

1. A flange flask was loaded with 1,6-hexanediol (50.37 g 0.43 mol), dihydroxyethyl

hexamethylene bis-carbamate (15.23 g, 0.05 mol) and diphenyl ether (50 % equal

weight) and was heated to 80 �C until a solution was obtained.

2. The mixture was stirred with a mechanical anchor stirrer at 50 rpm at 60 �C until

completely dissolved. A stoichiometric amount of adipic acid (69.92 g, 0.48 mol) was


added in three steps together with Novozym 435 (1.0 g) over 12 h, the pressure being

kept at 600 mmHg in this period.

3. After the third addition of adipic acid, an extra aliquot of Novozym 435 (0.3 g)

was added and the pressure was gradually reduced to 10 mbar over a period of

48 h. The reaction was allowed to stand at this pressure for a further 48 h and then

cooled.

4. The polymer solution was then washed with diethyl ether, the polymer precipitated and

the diphenyl ether removed in the diethyl ether.

5. The molecular weight was determined by GPC analysis: Mn 20 600, Mw 35 200 with a

dispersity of 1.7

5.3.7 Use of Heptane Azeotrope in a Dean–Stark Apparatus (Z. Liu and A. Taylor,

Unpublished Results)

Although similar to the solution polymerization in diphenyl ether, the use of Dean–

Stark distillation has advantages over both bulk and pure solution polymerization. The

use of heptane, although not a true solvent, reduces the viscosity of the reaction mixture

and, at the same time, provides an excellent means of removing the water formed during

the reaction. The azeotropic behaviour of the water–heptane system allows for easy

removal of water and a reaction temperature comparable to the high-vacuum reaction

described earlier. The boiling point of the (minimum) azeotrope of water and heptane is

83 �C, whereas the normal boiling point of heptane is 99 �C, and this temperature

compared very well with the reaction conditions of the high-vacuum work described

earlier. The temperature was constant during the reaction, but increased towards the end

of the reaction, which was due to the removal of the water. This eventually led to the

complete disappearance of the azeotrope and the subsequent increase of reaction

temperature to the normal boiling point of heptane. This increase of the temperature

brings the reaction in the optimum temperature range for Candida antarctica lipase B.

5.3.8 Procedure 4: Synthesis of a Polyester Containing

Di(hydroxyethyl)hexamethylene Bis-carbamate in n-Heptane



• 1,6-hexanediol (29.39 g, 0.249 mol)


• Novozym 435 (1.2 g)

• n-heptane (40 mL)

• nitrogen


• oil bath


• water-cooled condenser with Dean–Stark trap



5.3.8.2 Procedure

1. The dihydroxyethyl hexamethylene bis-carbamate (7.31 g, 0.025 mol), adipic acid

(40.17 g, 0.274 mol) and 1,6-hexanediol (29.39 g, 0.249 mol) were added to the

n-heptane (40 mL), then Novozym 435 (1.2 g) was added and heated to 85 �C with

stirring for 3 h; the temperature was then raised until reflux occurred and the heptane

returned to the flask via the Dean–Stark trap.

2. After 1 day, the molecular weight of the polyester was Mw 10 000, Mn 2340 and the

dispersity 4.3 as determined by GPC.

3. After 2 days, the Mw had increased to 25 000 and the Mn to 9000. The reaction was

continued for 5 days, when the Mw was 36 000, the Mn 14 800 with a dispersity of 2.45.

4. The pressure was decreased to 10 mbar and left for 8 h at 85 �C to ensure the removal of

all remaining solvent; however, there was no further increase in molecular weight.

5. End group analysis: acid number 3 mg KOH/g; hydroxyl number 13 mg KOH/g.

5.3.9 Combined One-pot Process Using Chemical and Enzymatic Processes

In this procedure, the diol to be used in the synthesis of the polyester is used as the inert

diluent in the synthesis step of the bis-carbamate. This avoids the requirement to use a

solvent that then has to be removed prior to the polyester synthesis. The 1,4-butanediol is

then incorporated into the polyester in the second stage of the reaction.

HN

HNO O

O O

O6

O

O

O

4O

O

O

O

4O

Diacid LinkerLinker Diol

n

O

OO H2N

NH2+

60 °C

NH

HN O

O

O

O

HOOH

AdipicAcidNovozym 435 HO

OH

5.3.10 Procedure 5: Two-stage, One-pot Synthesis of a Polyester Containing

Di(hydroxyethyl)hexamethylene Bis-carbamate


• Ethylene carbonate (44.32 g, 0.50 mol)

• 1,4-butanediol (40 g, 0.44 mol)

• 1,6-hexamethylenediamine (29 g, 0.25 mol)

• adipic acid (25.29 g , 0.173 mol)


• Novozym 435 (1.04 g)

• four-necked flange flask with stirrer bar

• oil bath




5.3.10.2 Procedure

1. Ethylene carbonate (44.32 g, 0.50 mol) and 1,4-butanediol (40 g, 0.44 mol) were added

to a reactor and heated to 60 �C. 1,6-Hexamethylenediamine (29 g, 0.25 mol) was added

over 1 h making sure the exotherm did not exceed 88 �C. The reaction was maintained

at 60 �C for 16 h. The product is a clear liquid at 60 �C, but it crystallizes rapidly on

cooling to a white waxy solid. GPC showed that the reaction had gone to completion,

with only the peaks of the diol and the bis-carbamate remaining; the composition was

64.7 % bis-carbamate and 35.3 % 1,4-butanediol.

2. A portion of this mixture (25 g) was heated at 100 �C with adipic acid (10.42 g,

0.071 mol) until the acid had dissolved. The reactants were cooled to 60 �C and

Novozym 435 (1.04 g) added and the pressure maintained at 200 mmHg.

3. The remaining adipic acid (14.86 g, 0.102 mol) was added in three equal amounts over

5 h, the temperature being maintained at 60 �C and the pressure 200 mmHg for a further

11 h. The pressure was then reduced to 80 mmHg for 8 h and finally to 2 mmHg for 24 h.

4. The resulting polyester (43 g, 97.5 %) had a molecular weight Mw of 9350 Da and a

dispersity of 1.75. End-group analysis: acid number 3 mg KOH/g; hydroxyl number of

29 mg KOH/g.

5.3.11 Enzymatic Synthesis of Aliphatic Polyesters7,8

The conventional synthesis of aliphatic polyesters based on adipic acid and a range of

diols, such as 1,4-butanediol or 1,6-hexanediol, involves a high-temperature esterification

reaction typically at 240–260 �C and an organometallic catalyst such as stannous octano-

ate. The use of enzyme catalysis results in a much lower reaction temperature, but also the

possibility of removing the esterification catalyst, giving the polyester significantly

improved hydrolysis resistance.

5.3.12 Procedure 6: Synthesis of a Poly(hexanediol-adipate) Polyester

O

O

O

O

On


• 1,6-Hexanediol (42.84 g, 0.363 mol)


• toluene (20 g)


• Novozym 435 (1.44 g)

• nitrogen


• oil bath

• hotplate



5.3.12.2 Procedure

1. 1,6-Hexanediol (42.84 g, 0.363 mol) and toluene (20 g) were charged into a flask and

allowed to mix at 60 �C.

2. A stoichiometric amount of adipic acid (52.97 g, 0.363 mol) was added in three steps.

The first aliquot (16 g) was added to the mechanically stirred mixture and the tempera-

ture was increased to 80 �C until the adipic acid was completely dissolved. The mixture

was allowed to cool to 70 �C and 1.5 % Novozym 435 (1.44 g) was added.

3. The pressure was reduced to 600 mmHg and after 2 h the second portion of adipic acid

(20.0 g) was added. The reaction was continued overnight at 400 mmHg before the third

aliquot (16.97 g) was added. The volume of water in the distillate was measured as an

indication of the extent of the reaction. The pressure was reduced to 10 mmHg over a

period of 24 h and the reaction was left for another 48 h at this pressure to remove all

traces of water and toluene and give the product polymer as a viscous oil.

4. The molecular weight was determined by GPC as Mw 37 000 with a dispersity of 1.9.

End-group analysis: acid number 2 mg KOH/g; hydroxyl number 13 mg KOH/g.1H NMR (400 MHz; CDCl3) � 1.61–1.72n (m CH2CH2COOH, CH2CH2OCOR,

CH2CH2COOR, HOCH2CH2), 2.24–2.40 (m CH2COOH, CH2COOR), 3.68 (t J¼ 6

Hz, CH2OH) 4.06–4.13 (m CH2OCOR).13C NMR (100 MHz; CDCl3) � 24.12, 24.30, 24.35, 24.39, 25.09, 25.31, 29.06, 29.68

(CH2CH2COOH, CH2CH2OCOR, CH2CH2COOR, HOCH2CH2), 33.41 (CH2COOH),

33.87–33.92 (CH2COOR), 62.31 (HOCH2), 63.89–64.21 (CH2OCOR), 173.32,

173.45, 173.49 (COOR).

References

1. Blum, H., Pedain, J. and Hentschel, K.H., Bis-oxazolidines, oxazolidine mixtures consistingessentially thereof and their use as hardeners for plastics precursors containing isocyanate groups.US Patent Appl., 1993, US 5,189,176.

2. McCabe R.W. and Taylor, A., Synthesis of novel polyurethane polyesters using the enzymeCandida antarctica lipase B. Green Chem., 2004, 6, 151.

3. Delaby, R., Chabrier, P. and Najer, H., Synthese de quelque diiodures de bis-(carbamoylcholine)douse d’activite curarisante. Mem. Present. Soc. Chim., 1956, 1616.

4. Delaby, R. Sekera, A., Chabrier, P. and Pignaniol, P., Synthese de biscarbamat des amines. Bull.Soc. Chem., 1953, 20, 278.

5. Gross, R., Kumar, A. and Kalra, B., Polymer synthesis by in vitro enzyme catalysis Chem. Rev.,2001, 101, 2097.


6. Mahapatro, A., Kalra, B,, Kumar, A. and Gross, R.A., Lipase-catalyzed polycondensations: effectof substrates and solvent on chain formation, dispersity, and end-group structure.Biomacromolecules, 2003, 4, 544.

7. Binns, F., Harffey, P., Roberts, S.M. and Taylor, A., Studies of lipase-catalyzed polyesterificationof an unactivated diacid/diol system. J. Polym. Sci. A: Polym. Chem., 1998, 36, 2069.

8. Binns, F., Harffey, P., Roberts, S.M. and Taylor, A., Studies leading to the large scale synthesis ofpolyesters using enzymes. J. Chem. Soc. Perkin Trans. 1999, 2671.


5.4 Bioconversion of 3-Cyanopyridine into Nicotinic Acid withGordona terrae NDB1165Tek Chand Bhalla

Nitrilase-mediated conversion of 3-cyanopyridine into nicotinic acid is an attractive

alternative to chemical methods of nicotinic acid synthesis.1,2 It has been synthesized

using whole cells (containing nitrilase) of some microorganisms3 and involves the follow-

ing reaction:

N

N

N

OH

O

NH3+Gordona terrae nitrilase

OH2

Based on this reaction, a method for the production of nicotinic acid using whole cells of

Gordona terrae NDB 1165 (previously Rhodococcus sp. NDB 1165) is described.4

5.4.1 Procedure 1: Cultivation of G. terrae NDB 1165


• Yeast extract (6.5 g)

• Bacto-peptone (6.5 g)

• dipotassium hydrogen phosphate (K2HPO4) (6.5 g)

• disodium hydrogen phosphate (Na2HPO4) (2.5 g)

• potassium dihydrogen phosphate (KH2PO4) (2.6 g)

• magnesium sulfate (MgSO4�7H2O) (0.26 g)

• ferrous sulfate (FeSO4�7H2O) (0.039 g)

• sodium chloride (NaCl) (1.3 g)

• calcium chloride (CaCl2�2H2O) (0.078 g)

• glucose (13 g)


• propionitrile (5 mL)

• 3-cyanopyridine (100 mL 5 M solution in distilled water)

• nicotinic acid (100 mL 0.4 M solution in distilled water)

• 0.1 M potassium phosphate buffer pH 8.0 (100 mL)

• stored culture of G. terrae NDB 1165

• distilled water 1500 mL

• test tubes (15 mL � 25)

• Erlenmeyer flasks (250 mL � 25)

• Eppendorf tubes (1.5 mL x � 3)

• cotton plugs for culture tubes and flasks

• weighing balance

• pH meter

• electronic balance

• autoclave


• laminar flow cabinet

• bacteriological loop

• incubator shaker

• refrigerated centrifuge

• high-performance liquid chromatography (HPLC) system equipped with C-18 reverse-

phase column and UV–vis detector

• HPLC elution buffer (containing 25 % v/v acetonitrile in water plus 0.1 % v/v H3PO4)

(500 mL).

5.4.1.2 Procedure

Preparation of Preculture of G. terrae NDB 1165

1. Prepare 50 mL of preculture medium (in distilled water) containing yeast extract (0.25 g),

Bacto-peptone (0.25 g), K2HPO4 (0.25 g), KH2PO4 (0.1 g), MgSO4�7H2O (0.01 g),

FeSO4�7H2O (0.0015 g), NaCl (0.05 g), CaCl2�2H2O (0.003 g), glucose (0.5 g), adjust

to pH 7.5 and dispense 2 mL into each of 25� 15 mL test tubes.

2. Plug these tubes with cotton plugs and autoclave at 15 psi pressure for 20 min. Transfer

a bacteriological loop full of cells from the slants of G. terrae NDB 1165 culture to each

of the sterile preculture medium tubes and incubate at 30 �C in a gyratory shaker

(180 rpm) for 24 h.

Induction of Nitrilase Enzyme in G. terrae NDB 1165 Culture

1. Prepare 1.25 L nitrilase production medium (in distilled water) containing yeast extract

(6.25 g), Bacto-peptone (6.25 g), K2HPO4 (6.25g), KH2PO4 (2.5 g), MgSO4�7H2O

(0.25 g), FeSO4�7H2O (0.0375 g), NaCl (1.25 g), CaCl2�2H2O (0.075 g), glucose (12.5

g), adjust to pH 7.5 and dispense 50 mL into each of 25� 250 mL Erlenmeyer flasks.

2. Plug these flasks with cotton plugs and autoclave at 15 psi pressure for 20 min. Transfer

2 mL of preculture as prepared above into each of the flasks. Add 0.2 mL of filter sterile

propionitrile (filter 5 mL of propionitrile using 0.22 mm filter into a sterile tube) into

each flask as inducer for the induction of nitrilase in G. terrae NDB 1165 cells. Incubate

these at 30 �C in a gyratory shaker (180 rpm) for 24 h.

Preparation of Resting Cells of G. terrae NDB 116

1. Centrifuge the culture at 5000g for 10 min in a refrigerated centrifuge at 4 �C and

discard the supernatant and suspend the cell pellet in 100 mL of 0.1 M potassium

phosphate buffer (pH 8).

2. Centrifuge the cell suspension at 5000g for 10 min at 4 �C and discard the supernatant

and suspend the cell pellet in 40 mL of 0.1 M potassium phosphate buffer (pH 8). The

cell suspension is now resting cells. The cells yield will be 2 g dry cell weight (DCW).

In order the measure the dry weight of cells, take a preweighed Eppendorf tube (W1),

add 0.5 mL of resting cell suspension. Centrifuge it at 5000g for 10 min, discard

the supernatant and leave the Eppendorf tube lid open in an oven at 80 �C for 24 h

and record the weight (W2). The dry cell weight in 1 mL of the resting cell suspension

will be

ðW2 �W1Þ � 2

5.4 Bioconversion of 3-Cyanopyridine into Nicotinic Acid 183

Assay of Nitrilase Activity of Resting Cells

1. Add 0.1 M potassium phosphate buffer pH 8.0 (985 mL), resting cell suspension (5 mL)

and substrate (5 M 3-cyanopyridine 10 mL, 50 mmol of 3-cyanopyridine) into an

Eppendorf tube. Incubate at 40 �C for 20 min and stop the reaction by adding 100 mL

of 1 M HCl. Centrifuge at 10 000g for 10 min at 0–4 �C.

2. Determine the concentration of nicotinic acid formed during the reaction using HPLC

equipped with a C-18 reverse-phase column (250 mm � 4.6 mm) at ambient tempera-

ture 25 �C and 210 nm with a flow rate of 1 mL min�1 of elution buffer (containing 25 %

v/v acetonitrile in water plus 0.1 % v/v H3PO4). Prepare a standard curve using different

concentrations of nicotinic acid (0.04–0.4 mM) and calculate the nicotinic acid formed in

the reaction mixture from the graph. Dilute the samples for analysis of nicotinic acid to

keep in the range of 0.04–0.4 mM. One unit of nitrilase activity is defined as the amount

of enzyme that catalyses the conversion of 1 mmol of 3-cyanopyridine to nicotinic acid

per minute under these assay conditions. Normally, the specific activity of the cells

(U mg�1 DCW) may vary from 2.25 to 2.50 U mg�1 DCW.

5.4.2 Procedure 2: Conversion of 3-Cyanopyridine to Nicotinic Acid


• 3-Cyanopyridine (166.4 g)

• 0.1 M potassium phosphate buffer pH 8.0 (1000 mL)

• hydrochloric acid (conc.) (100 mL)


• fermentor/ bioreactor (1.5 L capacity)

• rotary vacuum evaporator/oven

• ice

• refrigerator.

5.4.2.2 Procedure

1. Prepare a solution of 5 M 3-cyanopyridine (320 mL) to be used as substrate for nicotinic

acid preparation in 0.1 M potassium phosphate buffer pH 8.0.

2. Take 0.1 M potassium phosphate buffer (640 mL) and add resting cell suspension

(40 mL, containing 2 g dry cell weight) into a 1.5 L vessel or a fermentor (reactor

with temperature control, impeller and addition port to add substrate periodically).

3. Set the temperature to 40 �C and add 10 mL of 5 M 3-cyanopyridine solution to start the

bioconversion reaction in fed-batch mode with mixing of the contents at impeller speed

of 180 rpm.

4. Add substrate 10 mL (5 M 3-cyanopyrindine solution) to the reaction every 20 min and

add 32 feeds over a period of 10 h and 20 min.

5. Allow the reaction to proceed for another 40 min, centrifuge the reaction mixture at

5000g for 10 min at 4 �C and collect the supernatant.

6. While stirring slowly, add concentrated HCl to the supernatant and bring down the pH

to 4–5 and crystals of nicotinic acid start appearing.


7. Keep the solution overnight at 0–4 �C and allow the nicotinic acid crystals to settle

down to the bottom of the container. Separate the crystals and dissolve in distilled

water, adjust the pH to 4–5 and cool to 0–4 �C. Decant the liquid and dry the crystals in

vacuum evaporator or at 50 �C in an oven (90% yield).

5.4.3 Conclusion

A simple method for the conversion of 1.6 M 3-cyanopyridine to nicotinic acid in fed-batch

mode of addition of substrate into the reaction catalysed by resting cells of G. terrae NDB

1165 has been described. A procedure for cultivation of G. terrae NDB 1165 and recovery

of nicotinic acid from the reaction mixture is described. Using these procedures, 196 g of

nicotinic acid (90 % recovery) can be practically produced in a reaction volume of 1 L.

References

1. Offermanns, H., Kleeman, A., Tanner, H., Beschke, H. and Friedrich, H., Vitamins. In Kirk–Othmer Encyclopaedia of Chemical Technology, Vol. 24, Mark, H.F., Othmer, D.F., Overberger,C.G., Seaborg, G.T. (eds). John Wiley & Sons, Inc.: New York, 1984, pp. 1–226.

2. Finar, I.L. Alkaloids. In Organic Chemistry, vol. 2: Stereochemistry and the Chemistry of NaturalProducts. Addison Wesley Longman Limited, UK. 1997, pp. 702–768.

3. Mathew, C.D., Nagasawa, T., Kobayashi, M. and Yamada, H., Nitrilase-catalyzed production ofnicotinic acid from 3-cyanopyridine in Rhodococcus rhodochrous J1. Appl. Environ. Microbiol.1988, 54, 1030.

4. Prasad, S., Misra, A., Jangir, V.P., Awasthi, A., Raj, J. and Bhalla, T.C., A propionitrile-inducednitrilase of Rhodococcus sp. NDB 1165 and its application in nicotinic acid synthesis. World J.Microbiol. Biotechnol. 2007, 23, 345.

5.4 Bioconversion of 3-Cyanopyridine into Nicotinic Acid 185

5.5 Enzyme-promoted Desymmetrization of Prochiral DinitrilesMarloes A. Wijdeven, Piotr Kiełbasinski and Floris P.J.T. Rutjes

Nitriles constitute a synthetically versatile class of compounds due to the ease of cyanide

introduction and subsequent conversion into various other functional groups, such as

amines, amides and acids.1 The hydrolysis, however, usually requires rather harsh condi-

tions, involving either concentrated acid or base, heavy metals or elevated temperatures.

Furthermore, where multiple cyanide groups are present no chemoselectivity is observed.

A viable alternative to chemical hydrolysis is found in the use of nitrile-hydrolyzing

enzymes. First of all, enzymatic nitrile hydrolysis proceeds under intrinsically mild

conditions (neutral pH, 25–37 �C, aqueous buffer). More importantly, however, as an

additional benefit, one may profit from the fact that cyanide-hydrolyzing enzymes often

display high levels of chemo- and enantio-selectivity. This is exemplified by whole cells

from the bacterium strain Rhodococcus erythropolis NCIMB 11540, which have been

successfully used for the enantioselective monohydrolysis of prochiral dinitriles such as

2-heptyl-2-methylmalononitrile2 and 3-benzyloxypentanedinitrile.3 An additional exam-

ple that we wish to highlight is the successful desymmetrization of 2,20-sulfinyldiacetoni-

trile.4 The latter is the first example of an enzymatic desymmetrization where the chirality

resides in a heteroatom.

5.5.1 Procedure 1: Synthesis of 2-Cyano-2-methylnonamide2

C C

C7H15MeC

C7H15Me

O

H2NRhodococcus erythropolis

NCIMB 11540

pH 7, 28 °C

98% ee85% yield

NNN


• 2-Heptyl-2-methylmalononitrile (25 mg)

• freeze-dried whole cells of R. erythropolis NCIMB 11540 (50 mg)

• phosphate buffer solution of pH 7 (50 mL, 100 mM)

• phosphoric acid (85%)


• heptane (200 mL)

• MgSO4 (500 mg)

• incubator

• flask (100 mL)

• glass fritt

• separation funnel


• equipment for column chromatography

• thin-layer chromatography (TLC) plates (silica-gel-coated glass plates Merck 60 F254)


• silica gel (Acros Organics silica gel, 0.035–0.070 mm)

• Celite.

5.5.1.2 Procedure

1. Freeze-dried whole cells of R. erythropolis NCIMB 11540 (50 mg) were added to a

mixture of 2-heptyl-2-methylmalononitrile (25 mg) in a phosphate buffer solution of

pH 7 (50 mL, 100 mM). The reaction was incubated at 28 �C and 125 rpm, and

monitored by TLC.

2. After complete conversion of the substrate, the reaction was acidified to pH 2 with

phosphoric acid (85 %). The reaction was filtered over Celite and the filtrate was

extracted with EtOAc (3� 15 mL). The organic layers were combined, dried over

MgSO4, filtrated and concentrated in vacuo.

3. Purification by flash column chromatography (heptane/EtOAc 4:1) provided 2-cyano-

2-methylnonamide (24 mg, 85 %, 98 % ee) as a white solid.

2-Cyano-2-methylnonamide: ee 98 %, determined by high-performance liquid chroma-

tography, Kromasil-5CHI-TBB column, eluent: n-hexane/isopropanol 95:5; flow: 1 mL

min�1; detection: UV 210 nm. Rt¼ 6.8 min (major ent) and 8.6 min (minor ent). Melting

point 78.1–78.5 �C. IR (neat): 3403, 3178, 2975, 2923, 2854, 2232, 1687, 1631 cm�1. 1H

NMR (300 MHz; CDCl3): � 6.28 (br s, 1H), 5.87 (br s, 1H), 1.99–1.89 (m, 1H), 1.72–1.64

(m, 1H), 1.57 (s, 3H), 1.53–1.51 (m, 10H), 0.88 (t, J¼ 6.6 Hz, 3H); 13C NMR (75 MHz;

CDCl3): � 107.2, 121.7, 44.0, 38.2, 31.8, 29.3, 29.1, 25.7, 24.0, 22.7, 14.2. High-resolution

mass spectrometry (MS) (electron impact) calculated for C11H21N2O 197.1654, found

197.1652.

5.5.2 Procedure 2: Synthesis of (S)-Methyl 3-Benzyloxy-4-cyanobutanoate3

NC

Rhodococcus erythropolisNCIMB 11540

pH 7, 30 °C

OBnCO2H NC

OBnCO2MeNC

OBnCN

CH2N2, EtOH

Et2O

96% ee70% yield


• 3-Benzyloxypentanedinitrile (1.0 g, 5.0 mmol)

• lyophilized whole cells of R. erythropolis NCIMB 11540 (2.0 g)

• phosphate buffer of pH 7 (400 mL, 100 mM)

• 2 M HCl


• MgSO4

• diazomethane solution in diethyl ether

• acetic acid

• ethanol

• petroleum ether


• incubator

• flask (500 mL)

• glass filter



• TLC plates (silica gel-coated glass plates Merck 60 F254)

• silica gel (Acros Organics silica gel, 0.035–0.070 mm).

5.5.2.2 Procedure

1. Lyophilized whole cells of R. erythropolis NCIMB 11540 (2 g) were added to a mixture

of 3-benzyloxypentanedinitrile (1.0 g, 5.0 mmol) in a phosphate buffer solution of pH

7 (400 mL, 100 mM). The reaction was incubated at 30 �C and 125 rpm, and monitored

by TLC.

2. After complete conversion of the substrate, the reaction was adjusted to pH 1 by the

addition of aqueous 2 M HCl. The aqueous layer was extracted with EtOAc (3 � 100

mL) and the combined organic layers were dried over MgSO4, filtrated and concen-

trated in vacuo.

3. The crude product was immediately esterified to the corresponding methyl ester by

treatment with an excess of a diazomethane solution in diethyl ether, excess diazo-

methane was reacted with a few drops of acetic acid and the resulting mixture was

concentrated in vacuo.

4. Purification with flash column chromatography (petroleum ether/EtOAc 2:1) gave (S)-

methyl 3-benzyloxy-4-cyanobutanoate (766 mg, 70% (hydrolysis and esterification),

96 % ee).

(S)-3-Benzyloxy-4-cyanobutyric acid methyl ester: [�]D¼þ12.1 (CHCl3, c¼ 1.0).

ee¼ 96 % (heptane/isopropanol 9:1, flow: 1 mL min�1, Rt¼ 17.2 min (R-ent), 23.4 min

(S-ent). IR (neat): 3546, 3024, 3012, 2952, 2254, 1735 cm�1. 1H NMR (200 MHz; CDCl3):

� 7.37–7.28 (m, 5H), 4.63 (d, J¼ 11.5 Hz, 1H), 4.59 (d, J¼ 11.5 Hz, 1H), 4.11 (quintet,

J¼ 5.9 Hz, 1H), 3.68 (s, 3H), 2.77–2.59 (m, 4H); 13C NMR (50 MHz; CDCl3): � 170.3,

137.0, 128.3, 127.8, 127.6, 116.8, 72.0, 71.0, 51.7, 38.5, 22.8.

5.5.3 Procedure 3: Desymmetrization of 2,20-Sulfinyldiacetonitrile4

Nitrilase 104

pH 7.2, 30 °CSNC CNO

SNCO

SNCO

NH2

O O

OH+

99% ee57% yield

77% ee20% yield


• 2,20-Sulfinyldiacetonitrile (0.10 g, 0.78 mmol)

• Nitrilase 104 (purchased from Biocatalytics, Pasadena, USA, 10 mg)

• phosphate buffer of pH 7.2

• incubator


• flask (100 mL)



• TLC plates (silica-gel-coated glass plates Merck 60 F254)

• silica gel (Merck 60 silica gel)

• strongly acidic ion exchange resin (Dowex�

50W).

5.5.3.2 Procedure

1. 2,20-Sulfinyldiacetonitrile (0.10 g, 0.78 mmol) was dissolved in a phosphate buffer

of pH 7.2, nitrilase 104 (10 mg) was added and the reaction was incubated at 30 �C

for 48 h.

2. The reaction mixture was concentrated in vacuo and the residue was purified using

column chromatography.

3. The resulting product was dissolved in water and this solution was eluted through a

strongly acidic ion-exchange resin (Dowex�

50W). Lyophilization of the fractions

provided the products cyanomethylsulfinylacetamide (70 mg, 57%) and cyanomethyl-

sulfinylacetic acid (25 mg, 20%).

Cyanomethylsulfinylacetamide: white crystals (MeCN), m.p. 130–133 �C. 1H NMR

(200 MHz; D2O): � 4.05 (br. s, 4H); 13C NMR (50 MHz; D2O): � 167.8, 112.4, 56.2,

42.7. MS (chemical ionization (CI)): m/z¼ 147 (MþH); Anal. calc. for C4H4N2O2S: C

32.88, H 4.11, N 19.18, S 21.92 %; found: C 32.17; H 4.11; N 19.61; S 21.62 %.

Cyanomethylsulfinylacetic acid: colorless oil. 1H NMR (200 MHz; CD3COCD3):

� 4.25 (AB, 2H), 4.07 (AB, 2H); 13C NMR (50 MHz; CD3COCD3): � 165.7, 112.4, 56.0,

39.07; MS (CI): m/z¼ 148 (M þ H); Anal. calc. for C4H5NO3S: C 32.65, H 3.40, N

9.25, S 21.77 %; found: C 32.82; H 3.60; N 9.60; S 21.34 %.

5.5.4 Conclusions

The enzymatic hydrolysis of nitriles provides a viable alternative for the generally harsh

chemical conditions that are most often used. As a result of the ability of many nitrile-

hydrolyzing enzymes to give selective monohydrolysis, in the case of dinitriles, additional

opportunities such as desymmetrization can be explored. With the previous examples, we

have shown that, for several substrate classes, enzymatic desymmetrization of dinitriles is

indeed a synthetically viable option.

References

1. For an excellent overview, see for example: Drauz, K. and Waldmann, H. (eds.), EnzymeCatalysis in Organic Synthesis, Vol. 2, Wiley–VCH Verlag, Weinheim, 2002, pp. 699–716.

2. Vink, M.K.S., Wijtmans, R., Reisinger, C., vanden Berg, R.J.F., Schortinghuis, C.A., Schwab, H.,Schoemaker, H.E. and Rutjes, F.P.J.T., Nitrile hydrolysis activity of Rhodococcus erythropolisNCIMB 11540 whole cells. Biotechnol. J., 2006, 1, 569.

3. Vink, M.K.S., Schortinghuis, C.A., Luten, J., Maarseveen, J.H., Schoemaker, H.E., Hiemstra, H.and Rutjes, F.P.J.T., A stereodivergent approach to substituted 4-hydroxypiperidines. J. Org.Chem., 2002, 67, 7869.

4. Kiełbasinski, P., Rachwalski, M., Mikołajczyk, M., Szyrej, M., Wieczorek, M.W., Wijtmans, R.and Rutjes, F.P.J.T., Enzyme-promoted desymmetrization of prochiral bis-(cyanomethyl) sulf-oxide. Adv. Synth. Catal., 2007, 349, 1387.


5.6 Epoxide Hydrolase-catalyzed Synthesis of (R)-3-Benzyloxy-2-methylpropane-1,2-diolTakeshi Sugai, Aya Fujino, Hitomi Yamaguchi and Masaya Ikunaka

Bacillus subtilis, engineered to overproduce epoxide hydrolase, was used as a whole-cell

biocatalyst to resolve racemic 1-benzyloxymethyl-1-methyloxirane with high

(S)-selectivity.1 The remaining (R)-epoxide was subsequently ring opened in situ, with

inversion of stereochemistry, to obtain highly enantiomerically enriched (R)-3-benzyloxy-

2-methylpropane-1,2-diol in greater than 50 % theoretical yield (Figure 5.2).

5.6.1 Procedure 1: Synthesis of 3-Benzyloxy-2-methylpropene

BnOCl

benzylalcohol

NaH, DMF0 °C to r.t.


• 55 % Sodium hydride (4.49 g, 0.103 mmol)

• benzyl alcohol (10.1 g, 93.5 mmol)

• methallyl chloride (10.2 g, 0.112 mol)

BnOO

(±)–

OBnO

OBnHOHO

BnO OHHO

(R)–

recrystal-lization

(mp 30-31 °C, 100% ee)

+

(mother liquor,68.1% ee; recyclable

in 5.6.4)

(82.3% ee)

OH2

BnO OHHO

(R)–

BnO OHHO

(R)–

BnO OHHO

OBnO

BnOO

B. subtilisepoxidehydrolase

(R)–

(S)– (fast)

(slow)

OH2

dil.H2SO4

Figure 5.2 Asymmetric synthesis of (R)-3-benzyloxy-2-methylpropane-1,2-diol.


• dimethyl formamide (DMF; anhydrous, 65 mL)

• NH4Cl solution (saturated aqueous, 35 mL)


• hexane (1000 mL)

• brine (35 mL)

• Na2SO4 (anhydrous, 10 g)

• silica gel 60 (spherical; 100–210 mm, 37558-79, KANTO Chemical Inc., 450 g)

• three-necked reaction flask (300 mL) equipped with a magnetic stirrer, an argon balloon,

a dropping funnel (50 mL) and a thermometer (�100 �C to þ50 �C)


• one 300 mL separatory funnel




5.6.1.2 Procedure

1. To a mixture of 55 % sodium hydride (4.49 g, 0.103 mmol) in anhydrous DMF (18 mL)

was added dropwise a solution of benzyl alcohol (10.1 g, 93.5 mmol) in anhydrous

DMF (30 mL) under argon. The mixture was stirred at room temperature for 90 min.

2. A solution of methallyl chloride (10.2 g, 0.112 mol) in anhydrous DMF (17 mL) was

added dropwise at 0 �C. The mixture was stirred at room temperature for 24 h.

3. The reaction was quenched by the addition of saturated aqueous NH4Cl solution (35

mL) and the mixture was extracted with hexane (200 mL). The organic layer was

washed with brine (35 mL), dried over Na2SO4 (10 g) and concentrated in vacuo.

4. The crude residue (15.2 g) was purified by silica gel column chromatography. Elution

with hexane/ethyl acetate (10:1, 880 mL) gave 3-benzyloxy-2-methylpropene (13.9 g,

85.7 mmol, 92 %) as a colorless oil.

1H NMR (270 MHz, CDCl3): � 7.26–7.15 (5H, m), 4.90 (1H, s), 4.83 (1H, s), 4.40 (2H,

s), 3.84 (2H, s), 1.67 (3H, s). NMR spectral data were identical with that reported

previously.2

5.6.2 Procedure 2: Synthesis of Racemic 1-Benzyloxymethyl-1-methyloxirane from

3-Benzyloxy-2-methylpropene

BnOCH3CNr.t.

BnOO

(±)-

H2O2


• 3-Benzyloxy-2-methylpropene (5.00 g, 30.8 mmol)

• MeCN (7.5 mL)

• EtOH (12.5 mL)

5.6 Catalyzed Synthesis of (R)-3-Benzyloxy-2-methylpropane-1,2-diol 191

• KHCO3 (0.925 g, 9.24 mmol)

• H2O2 (30 % in H2O, 5.98 mL, 77 mmol)

• Na2S2O3 solution (12.5 g in H2O, 30 mL)

• brine (30 mL)


• hexane (1100 mL)



• two-necked reaction flask (100 mL) equipped with a magnetic stirrer bar





5.6.2.2 Procedure

1. To a solution of 3-benzyloxy-2-methylpropene (5.00 g, 30.8 mmol) in CH3CN (2.5 mL)

and ethanol (12.5 mL) was added an aqueous solution of H2O2 (30 % in H2O, 4.78 mL,

61.6 mmol) containing KHCO3 (0.925 g, 9.24 mmol). CH3CN (5 mL) was added and

the mixture was stirred at room temperature for 24 h. An aqueous solution of H2O2 (30

% in H2O, 1.2 mL, 15.4 mmol) was further added and the stirring was continued at room

temperature for 2 days.

2. The reaction was quenched by the addition of an aqueous solution of Na2S2O3 (12.5

g in H2O, 30 mL) and the mixture was extracted with cold hexane. The organic layer

was washed with brine (30 mL), dried over Na2SO4 (10 g) and concentrated in

vacuo.


with hexane/ethyl acetate (10:1, 990 mL) gave 1-benzyloxymethyl-1-methyloxirane

(5.05 g, 28.4 mmol, 92 %) as a colorless oil.

1H NMR (270 MHz, CDCl3): � 7.54–7.16 (5H, m), 4.50 (1H, d, J¼ 12.0 Hz), 4.44 (1H,

d, J¼ 12.0 Hz), 3.49 (1H, d, J¼ 11.1 Hz), 3.35 (1H, d, J¼ 11.1 Hz), 2.66 (1H, d, J¼ 4.8

Hz), 2.54 (1H, d, J¼ 4.8 Hz), 1.31 (3H, s). These NMR spectral data were identical with

those reported previously.3

5.6.3 Procedure 3: Synthesis of (R)-3-Benzyloxy-2-methylpropane-1,2-diol from

Racemic 1-Benzyloxymethyl-1-methyloxirane

OBnO

OHBnO OH

i. B. subtilis epoxide hydrolase

ii. H2SO4

iii. Crystallize


• (–)-1-Benzyloxymethyl-1-methyloxirane (4.5 g, 25.2 mmol)

• B. subtilis Tamy-2 strain cell suspension (19.5 mL)


• glycerol (6.0 mL)

• NaCl (3 g)

• Celite (20 g)

• acetone (80 mL)

• H2O (86.4 mL)

• concentrated H2SO4 (6.7 mL)

• Na2CO3 solution (saturated aqueous, 50 mL)

• brine (100 mL)

• Et2O (100 mL)


• hexane (600 mL)



• round-bottomed reaction flask (50 mL) equipped with a magnetic stirrer bar

• round-bottomed reaction flask (50 mL) equipped with a magnetic stirrer bar and a

thermometer (�100 �C to þ50 �C)







• cooling equipment

• motor-driven centrifuge.

5.6.3.2 Procedure

1. Pre-incubation of the engineered B. subtilis Tamy-2 strain overexpressing epoxide

hydrolase was conducted according to the procedures developed at the Research &

Development Centre, Nagase & Co., Ltd. To request freeze-dried cell bodies of the B.

subtilis strain engineered to overproduce epoxide hydrolase, please contact H.Y.

([email protected]).

2. A mixture of the cell suspension (19.5 mL), glycerol (6.0 mL), (–)-1-benzyloxymethyl-

1-methyloxirane (4.5 g, 25.2 mmol) was stirred at room temperature for 7 days.

3. The broth was centrifuged (3000 rpm) and the separated supernatant was saturated with

NaCl (3 g). After ethyl acetate (50 mL) was added, the mixture was stirred for 1 h and

filtered through a pad of Celite (10 g). The organic layer of the filtrate was separated

and the aqueous layer was further extracted with ethyl acetate (200 mL). The cell debris

precipitated by centrifugation was mixed with acetone (80 mL). The mixture was

stirred for 1 h and filtered through a pad of Celite (10 g). The combined organic extracts

were washed with brine (50 mL), dried over Na2SO4 (10 g) and concentrated in vacuo.

4. The workup provided a crude mixture: 4.74 g, (R)-3-benzyloxy-2-methylpropane-1,2-diol

(88.3 % ee) and (R)-1-benzyloxymethyl-1-methyloxirane (97.9 % ee), 47.4:52.6. The

conversion was estimated by reverse-phase high-performance liquid chromatography

(HPLC) analysis (Senshu Pack PEGASIL ODS, 0.46 cm � 15 cm; MeOH/H2O (3:2),

1.0 mL min�1), TR (min)¼ 3.8 for 3-benzyloxy-2-methylpropane-1,2-diol, 6.8 for


benzyloxymethyl-1-methyloxirane. HPLC analysis of 3-benzyloxy-2-methylpropane-

1,2-diol: 88.3 % ee (Chiralcel OD-H, 0.46 cm � 25 cm; hexane/i-PrOH (15:1), 0.5 mL

min�1), TR (min)¼ 29.1 ((S)-, 5.85 %), 31.1 ((R)-, 94.15 %). HPLC analysis of benzy-

loxymethyl-1-methyloxirane: 97.9 % ee (ChiralPak AS-H, 0.46 cm � 25 cm; hexane/i-

PrOH (90:1), 0.5 mL min�1), TR (min)¼ 19.3 ((R)-, 98.95 %), 20.1 ((S)-, 1.05 %).

5. The mixture obtained above was diluted with H2O (86.4 mL) and ice-cooled.

Concentrated H2SO4 (6.7 mL) was added dropwise and the mixture was stirred at

0 �C for 10 min and then at room temperature for 30 min.3,4

6. The reaction was quenched by neutralization with saturated aqueous Na2CO3 solution

(50 mL) and the mixture was extracted with ethyl acetate (250 mL). The organic layer

was washed with brine (50 mL), dried over Na2SO4 (10 g) and concentrated in vacuo.

7. The crude residue was purified by silica gel column chromatography (160 g). Elution

with hexane/ethyl acetate (4:1, 750 mL) gave (R)-3-benzyloxy-2-methylpropane-

1,2-diol (4.10 g, 83 %, 82.3 % ee) as a colorless solid.

8. This was dissolved with Et2O (82 mL), cooled slowly to �30 �C and kept at that

temperature for 6 h. Mother liquor was decanted off under suction and the crystals

collected were rinsed twice with cold Et2O. The crystals were directly dried in vacuo

without any washing to afford (R)-3-benzyloxy-2-methylpropane-1,2-diol (2.1 g, 52 %

recovery) as colorless fine needles; m.p. 30–31 ��C.

Attention. When the mixture is kept at a temperature lower than �30 �C for a

prolonged period, crystals of the (R)-isomer would suffer from contamination with

the (S)-isomer. As the crystal has a low melting point and is highly soluble in hexane at

ambient temperature, one should avoid not only filtration, but also reslurrying.

1H NMR (CDCl3): � 7.31–7.19 (5H, m), 4.49 (2H, s), 3.58 (1H, dd, J¼ 4.6, 11.0 Hz),

3.45 (1H, d, J¼ 9.1 Hz), 3.40 (1H, dd, J¼ 7.8, 11.0 Hz), 3.36 (1H, d, J¼ 9.1 Hz), 2.71 (1H,

s), 2.26 (1H, dd, J¼ 4.6, 7.8 Hz), 1.08 (3H, s). These NMR spectral data were identical

with those reported previously;3 ½��27D ¼ �7:03 (c¼ 0.965, CH2Cl2) for the recrystallized

material (lit.5 [�]D¼�6.30 (c¼ 0.87, CH2Cl2)); 100 % ee for the recrystallized material

and 68.1 % ee for the material recovered from the mother liquor were confirmed by HPLC

under the above-mentioned conditions.

5.6.4 Procedure 4: Synthesis of (S)-1-Benzyloxymethyl-1-methyloxirane (68.1 %

ee) from (R)-3-Benzyloxy-2-methylpropane-1,2-diol (68.1 % ee)

BnO OHHO

(R )–(68.1% ee)

BnOO

(S)–(68.1% ee)

1) TsCl pyridine

2) K2CO3 MeOH


• (R)-3-Benzyloxy-2-methylpropane-1,2-diol (1.51 g, 7.69 mmol)

• pyridine (anhydrous, 31 mL)

• molecular sieves 4 A (2.55 g)


• p-toluenesulfonyl chloride (6.57 g, 34.4 mmol)

• H2O (5 mL)

• hydrochloric acid (1 M, 60 mL)

• NaHCO3 solution (saturated aqueous, 60 mL)

• brine (60 mL)


• hexane (500 mL)



• two-necked reaction flask (100 mL) equipped with a magnetic stirrer bar, an argon

balloon and a thermometer (�100 �C to þ50 �C)






• (S)-3-Benzyloxy-2-hydroxy-2-methylpropyl p-toluenesulfonate (1.51 g, 4.31 mmol)

• K2CO3 (5.01 g, 36.2 mmol)

• MeOH (87 mL)

• sodium phosphate buffer (0.2 M, pH 7.5, 10 mL)

• brine (30 mL)

• NaHCO3 solution (saturated aqueous, 60 mL)


• hexane (600 mL)



• round-bottomed reaction flask (200 mL) equipped with a magnetic stirrer bar and a

thermometer (�100 �C to þ50 �C)






5.6.4.2 Procedure

1. (R)-3-Benzyloxy-2-methylpropane-1,2-diol (1.51 g, 7.69 mmol, 68.1 % ee) was dis-

solved in dry pyridine (31 mL). 4 A molecular sieves (2.55 g) were added to the solution

under argon and the mixture stirred for 2 h and then cooled to 0 �C. Then p-

toluenesulfonyl chloride (6.57 g, 34.4 mmol) was added to the solution at 0 �C.

2. After stirring for 24 h at room temperature, the reaction was quenched by the addition of

water (5 mL) and the mixture was extracted with ethyl acetate (350 mL). The organic

layer was washed with 1 M hydrochloric acid (60 mL), saturated aqueous NaHCO3

solution (60 mL) and brine (60 mL). The solution was dried over Na2SO4 (10 g) and

concentrated in vacuo.



with hexane/ethyl acetate (2:1, 750 mL) yielded (S)-3-benzyloxy-2-hydroxy-2-methyl-

propyl p-toluenesulfonate (2.67 g, 7.61 mmol, 99 %) as a colorless oil.1H NMR (270 MHz, CDCl3): � 7.71 (2H, d, J¼ 8.1 Hz), 7.26–6.96 (7H, m), 4.40

(2H, s), 3.92 (1H, d, J¼ 9.4 Hz), 3.79 (1H, d, J¼ 9.4 Hz), 3.34 (1H, d, J¼ 9.1 Hz), 3.23

(1H, d, J¼ 9.1 Hz), 2.35 (3H, s), 1.09 (3H, s).

4. A suspension of (S)-3-benzyloxy-2-hydroxy-2-methylpropyl p-toluenesulfonate (1.51

g, 4.31 mmol) and K2CO3 (5.01 g, 36.2 mmol) in MeOH (87 mL) was stirred at room

temperature for 2 h.

5. The reaction was quenched by the addition of sodium phosphate buffer (0.2 M, pH 7.5,

10 mL) and the mixture was extracted with ethyl acetate (300 mL). The organic layer

was washed with brine (30 mL), dried over Na2SO4 (15 g) and concentrated in vacuo.


with hexane/ethyl acetate (4:1, 750 mL) yielded (S)-benzyloxymethyl-1-methyloxirane

(0.73 g, 4.09 mmol, 95 %) as a colorless oil.

Its 1H NMR spectrum was identical with that recorded in Section 5.6.2.2 for the

racemate; ½��18D ¼ þ7:74 (c¼ 0.96, MeOH); 68.1 % ee was confirmed by chiral HPLC

under the conditions specified in Section 5.6.3.2, step 4.

5.6.5 Procedure 5: Synthesis of (R)-3-Benzyloxy-2-methylpropane-1,2-diol from

(S)-1-Benzyloxymethyl-1-methyloxirane of 68.1 % ee


• (S)-1-Benzyloxymethyl-1-methyloxirane (0.50 g, 2.81 mmol, 68.1 % ee)

• B. subtilis Tamy-2 strain cell suspension (2.2 mL)

• glycerol (0.67 mL)

• NaCl (0.5 g)

• Celite (5 g)

• acetone (20 mL)


• hexane (500 mL)








5.6.5.2 Procedure

1. A mixture of the cell suspension (2.2 mL), glycerol (0.67 mL) and (S)-1-benzylox-

ymethyl-1-methyloxirane (0.50 g, 2.81 mmol) was stirred at room temperature. The

progress of the actual enzymatic reaction was monitored occasionally by HPLC as

described in Section 5.6.3.2, step 4.


2. After 2 days, the reaction was stopped at 82 % conversion. The extractive workup was

conducted in the same manner as described in Section 5.6.3.2, step 3.

3. Chromatographic purification was conducted in the same manner as described in

Section 5.6.3.2, step 7, to provide (R)-3-benzyloxy-2-methylpropane-1,2-diol (0.45 g,

2.30 mmol, 82 %, 100 % ee) and (R)-1-benzyloxymethyl-1-methyloxirane (90.1 mg,

0.506 mmol, 18 %, 68.1 % ee).

Attention. In pursuing high ee of the digested products (more reactive enantio-

mers) under kinetically resolving conditions, termination of the reaction at the

proper conversion is very important. When the relationship between conversion

and ees of the digested product and of the unaffected substrate was calculated

using the mathematical model of Chen et al.,6 it was predicted that �80 % conver-

sion should be the critical point, as depicted in Figure 5.3, which corroborated the

empirical results mentioned in steps 2 and 3.

5.6.6 Conclusion

(R)-3-Benzyloxy-2-methylpropane-1,2-diol, a desymmetrized form of 2-methylpropane-

1,2,3-triol with its terminal hydroxy being protected as a benzyl ether, was prepared using

the B. subtilis epoxide hydrolase-catalyzed enantioselective hydrolysis of the racemic

benzyloxymethyl-1-methyloxirane readily available from methallyl chloride and benzyl

alcohol. The preparation of the racemic epoxide, a key intermediate, was described in

Procedures 1 and 2 (Sections 5.6.1 and 5.6.2), its overall yield being 78 %. The combined

yield of enantiomerically pure (R)-3-benzyloxy-2-methylpropane-1,2-diol was 74 % from

(–)-benzyloxymethyl-1-methyloxirane, as described in Procedures 3–5 (Sections 5.6.3 and

5.6.5), with the overall procedures leading to ‘the biocatalytic dihydroxylation of benzyl

methallyl ether’.

100

%ee

60

60

product

80 100

unaffected recovery

conversion (%)

0

–60

Figure 5.3 Relationship between conversion and product ee.


References

1. Fujino, A., Asano, M., Yamaguchi, H., Shirasaka, N., Sakoda, A., Ikunaka, M., Obata, R.,Nishiyama, S. and Sugai, T., Bacillus subtilis epoxide hydrolase-catalyzed preparation of enan-tiopure 2-methylpropane-1,2,3-triol monobenzyl ether and its application to expeditious synthesisof (R)-bicalutamide. Tetrahedron Lett., 2007, 48, 979.

2. Tietze, L.F. and Gorlitzer, J., Preparation of chiral building blocks for a highly convergent vitaminE synthesis. Systematic investigations on the enantioselectivity of the Sharpless bishydroxilation.Synthesis, 1998, 873.

3. Avenoza, A., Cativiela, C., Peregrina, J.M., Sucunza, D. and Zurbano, M.M., An alternativeapproach to (S)- and (R)-2-methylglycidol O-benzyl ether derivatives. Tetrahedron Asymm.,2001, 12, 1383.

4. Orru, R.V.A., Mayer, S.F., Kroutil, W. and Faber, K., Tetrahedron, Chemoenzymatic deracemi-sation of (–)-2,2-disubstituted oxiranes. 1998, 54, 859. Steinreiber, A., Hellstrom, H., Mayer, S.F.,Orru, R.V.A., Faber, K., Chemo-enzymatic enantio-convergent synthesis of C4-building blockscontaining a fully substituted chiral carbon center using bacterial epoxide hydrolases. Synlett,2001, 111.

5. Tanner, D. and Somfai, P., Asymmetric synthesis of (R)-(þ)- and (S)-(�)-2,2,4-trimethyl-4-(hydroxymethyl)-1,3-dioxolane of high enantiomeric purity. Tetrahedron, 1986, 42, 5985.

6. Chen, C.-S., Fujimoto, Y., Girdaukas, G. and Sih, C.J., Quantitative analyses of biochemicalkinetic resolutions of enantiomers. J. Am. Chem. Soc., 1982, 104, 7294.


5.7 One-pot Biocatalytic Synthesis of Methyl (S)-4-Chloro-3-hydroxybutanoate and Methyl (S)-4-Cyano-3-hydroxybutanoateMaja Majeric Elenkov, Lixia Tang, Bernhard Hauer and Dick B. Janssen

The kinetic resolution of methyl 4-chloro-3-hydroxybutanoate by enantioselective epox-

ide ring opening with cyanide and a mutant halohydrin dehalogenase1 afforded two

versatile building blocks in highly enantioenriched form (>95 % ee) and in high yield

(81 % total yield).2 The transformation of halohydrin to cyanohydrin proceeds via an

epoxide intermediate that is transiently present in low amount during the reaction. This is

an example of a sequential kinetic resolution in which two steps are catalysed by a single

halohydrin dehalogenase.

5.7.1 Procedure 1: Expression and Purification of a Mutated Halohydrin

Dehalogenase HheC-W249F

Cl CO2CH3

OH

CO2CH3O

NC CO2CH3

OH

buffer

Cl CO2CH3

OH

+


NaCN, buffer


40% yield, 96.8% ee

41% yield, 95.2% ee


• 1000� stock solution of isopropyl �-D-1-thiogalactopyranoside (IPTG) (0.4 M): 1 g

IPTG dissolved in 10 ml bidest water; sterilize by filtering through a 0.22 mm filter

• 1000� stock solution of ampicillin (50 mg ml-1): 1 g ampicillin dissolved in 20 ml

sterilized bidest water

• Luria–Bertani (LB) liquid medium and LB agar plates containing ampicillin

(50 mg ml�1)

• tris-SO4 buffer (50 mM, pH 8.0) and tris-SO4 buffer (10 mM, pH 7.5)

• buffer A: tris-SO4 buffer (10 mM, pH 7.5) containing ethylenediaminetetraacetic acid

(1 mM), �-mercaptoethanol (1 mM) and 10 % of glycerol (v/v)

• buffer B: buffer A containing 0.45 M (NH4)2SO4 (pH 7.5)

• halide reagent I: NH4Fe(SO4)2 (0.25 M) in HNO3 (9 M)

• halide reagent II: a saturated solution of Hg(SCN)2 in absolute ethanol

• plasmid pGEFHheC-W249F

• calcium-competent cells of Escherichia coli strain BL21(DE3)

• one 100 ml and one 2.5 l Erlenmeyer flask

• five 1 l Erlenmeyer flasks

• thermostatted shaking incubator

• UV–vis spectrophotometer

• sonicator

5.7 Synthesis of Methyl (S)-4-Chloro-3- and Methyl (S)-4-Cyano-3-hydroxybutanoate 199

• Q Sepharose column (40 ml, Pharmacia Biotech) and gradient liquid chromatography

system

• equipment and materials for sodium dodecyl sulfate (SDS)–polyacrylamide gel

electrophoresis

• ultracentrifuge with rotor and tubes.

5.7.1.2 Procedure

1. To transform the E. coli cells, a 0.1 ml aliquot of competent cells was mixed with 1 ml

plasmid DNA (1–10 ng). The mixture was left on ice for 10–30 min and then heated at

42 �C for 90 s. After adding 0.4 ml of LB, the cell suspension was incubated at 37 �C

and then dilutions were plated on LB agar plates containing 50 mg ml�1 of ampicillin.

Incubation was then overnight at 30 �C. For more details, consult the Stratagene manual

(http://www.stratagene.com/manuals/200133.pdf).

2. A preculture was prepared by inoculating 30 ml of LB sterile medium containing

ampicillin with several colonies of transformed cells. The preculture was incubated

under shaking (130 rpm) at 37 �C for 2–3 h. The preculture was then diluted in 1 l of

sterile LB containing ampicillin. This cell suspension was distributed to five 1000 ml

Erlenmeyer flasks (200 ml each) without baffles and the cultures were incubated at

20 �C overnight with rotational shaking at 130 rpm. Expression was induced by adding

1 ml IPTG stock solution when the optical density at 600 nm (OD600) reached around

1–1.2. Incubation was continued for another 2.5 h.

3. The cells were harvested by centrifugation (10 min, 6500 g), washed with tris-SO4 (10 mM,

pH 7.5), and resuspended in 10 ml of buffer A. All further steps were carried out at 4 �C to

protect the enzyme against inactivation. A cell extract was prepared by ultrasonic treatment

(8–10 times for 10 s) of the cell suspension, followed by centrifugation (40 000g for 1 h).

4. The supernatant was collected and applied to a Q Sepharose column. The column was

washed with buffer A and bound proteins were eluted with a gradient of buffer A to

buffer B. Activity (measured as described in step 5) eluted at 100–150 mM (NH4)2SO4

and active fractions were pooled, yielding a pure protein as judged by SDS–polyacry-

lamide gel electrophoresis.

5. Halohydrin dehalogenase activity was determined by monitoring halide liberation3 at

30 �C in tris-SO4 buffer (50 mM, pH 8.0) containing 5 mM 1,3-dichloropropanol or 1,3-

dibromopropanol as the substrate. All buffers used for activity assay were prepared with

bidest water. From the incubation mixture, 0.5 ml samples were taken and mixed with

1.6 ml of H2O, 0.2 ml or halide reagent I and 0.2 ml of halide reagent II. Absorbances

were read at 460 nm. A calibration curve of 0–1 mM of chloride or bromide was used to

calculate the concentration of halide. The extinctions at 460 nm should be below 0.4

(for chloride) or 0.8 (for bromide).

5.7.2 Procedure 2: Synthesis of Methyl (S)-4-chloro-3-hydroxybutanoate and

Methyl (S)-4-cyano-3-hydroxybutanoate


• Methyl 4-chloro-3-hydroxybutanoate (0.50 g, 3.3 mmol)

• tris-SO4 buffer (62 ml, 0.5 M, pH 7.5)


• NaCN (322 mg, 6.6 mmol)

• HheC-W249F (15 mg purified mutant halohydrin dehalogenase in 3 ml buffer)

• NaCl

• ethyl acetate


• thin-layer chromatography (TLC) mixture: phosphomolybdic acid (25 g):cerium(II)

sulfate (7.5 g):sulfuric acid (25 ml):water (495 ml)

• hexane, ethyl acetate

• one round-bottom flask equipped with a magnetic stirrer bar and a stopper, 100 ml


• a 250 ml separatory funnel

• a conical funnel

• filter paper

• TLC plates (silica gel 60 F254, Merck).

• silica gel (Merck type 9385, 230–400 mesh), 10 g


• silica chromatography column.

5.7.2.2 Procedure

1. To a solution of methyl 4-chloro-3-hydroxybutanoate (0.50 g, 3.3 mmol) in 62 ml tris-

SO4 buffer (0.5 M, pH 7.5) NaCN was added (322 mg, 6.6 mmol), followed by addition

of purified HheC-W249F halohydrin dehalogenase (15 mg in 3 ml buffer). The result-

ing mixture was stirred at ambient temperature (22 �C) for 5 h.

2. The reaction mixture was then saturated with NaCl and extracted with ethyl acetate

(4 � 70 ml). The combined organic extracts were dried with Na2SO4 and concentrated

using a rotary evaporator.

3. The mixture of products was separated by column chromatography on a silica gel

column with hexane/ethyl acetate (7:3) as the eluent. (S)-4-Chloro-3-hydroxybutanoate

eluted first from the column followed by (S)-4-cyano-3-hydroxybutanoate.

4. Methyl (S)-4-cyano-3-hydroxybutanoate was obtained in 40 % yield (190 mg, 96.8 % ee).

½��24D ¼ þ26:3� (c¼ 1.1, CHCl3).

1H NMR (CDCl3) � 2.61–2.64 (4H, m), 3.72 (3H, s), 4.30–4.39 (1H, m).13C NMR (CDCl3) � 25.1, 39.9, 52.1, 64.0, 117.1, 171.8

Retention times: Rt¼ 14.7 min (S-ent), 15.2 min (R-ent).

5. The remaining methyl (S)-4-chloro-3-hydroxybutanoate was obtained in 41 % yield

(208 mg, 95.2 % ee).

½��24D ¼�18.9� (c¼ 1.27, CHCl3).

1H NMR (CDCl3) � 2.60 (1H, d, J¼ 7.5 Hz), 2.62 (1H, d, J¼ 4.5 Hz), 3.56 (1H, d,

J¼ 1 Hz), 3.58 (1H, d, J¼ 1 Hz), 3.69 (3H, s), 4.19–4.27 (1H, m).13C NMR (CDCl3) � 38.3, 48.1, 51.9, 67.8, 172.1.

Retention times: Rt¼ 10.5 min (R-ent), 10.6 min (S-ent).

6. The ee was determined by chiral gas chromatography analysis on a Chiraldex GT-A

column (30 m � 0.25 mm � 0.25 mm, Astec). The temperature was isothermal at

100 �C for 6 min, then increased at 10 �C min�1 to 170 �C, and finally kept for 3 min at

170 �C.

5.7 Synthesis of Methyl (S)-4-Chloro-3- and Methyl (S)-4-Cyano-3-hydroxybutanoate 201

5.7.3 Conclusion

The mutated halohydrin dehalogenase was used for catalysing two consecutive reactions:

ring closure of halohydrin to epoxide and epoxide ring opening by cyanide ion. Both

reactions occur in an enantioselective manner, resulting in two highly enantiomerically

enriched products that can easily be isolated from the reaction mixture and separated by

column chromatography.

References

1. Tang, L., Torres Pazmino, D.E., Fraaije, M.W., de Jong, R.M., Dijkstra, B.W. and Janssen, D.B.,Improved catalytic properties of halohydrin dehalogenase by modification of the halide-bindingsite. Biochemistry, 2005, 44, 6609.

2. Majeric Elenkov, M., Tang, L., Hauer, B. and Janssen, D.B., Sequential kinetic resolutioncatalyzed by halohydrin dehalogenase. Org. Lett., 2006, 8, 4227.

3. Bergmann, J.G. and Sanik, J., Determination of trace amounts of chlorine in naptha. Anal.Chem., 1957, 29, 241.


6

Aldolase Enzymes for ComplexSynthesis

6.1 One-step Synthesis of L-Fructose Using Rhamnulose-1-phosphateAldolase in Borate BufferWilliam A. Greenberg and Chi-Huey Wong

Dihydroxyacetone phosphate (DHAP)-dependent aldolases are powerful catalysts for

synthesis of carbohydrates and related compounds, such as iminocyclitols. Their

utility is limited by the strict requirement for DHAP, an expensive, unstable

compound, as donor substrate. The DHAP-dependent rhamnulose-1-phosphate aldo-

lase (RhaD) was able to catalyze aldol reactions with readily available dihydrox-

yacetone (DHA) as the donor when borate buffer was used, presumably by

reversible in situ formation of borate esters that mimicked DHAP.1 This effect was

used in a facile, inexpensive one-step synthesis of L-fructose, a valuable chiral synthon

and promising noncalorific sweetener (Figure 6.1). In addition to using inexpensive DHA

as the donor, this procedure used racemic glyceraldehyde as the acceptor.

L-Glyceraldehyde was preferentially accepted by the enzyme. If D-glyceraldehyde had

competed as a substrate, then D-sorbose would have been the expected aldol product; it

was not observed by high-performance liquid chromatography or NMR analysis of the

crude reaction mixture.



6.1.1 Procedure 1: Expression of RhaD


• Seed culture of Escherichia coli BL21 (DE3) harboring pETRhaD

• Luria–Bertani (LB) media (500 mL)

• shake flask

• shaker

• centrifuge

• phosphate-buffered saline solution (200 mL).

6.1.1.2 Procedure

1. A seed culture of E. coli BL21 (DE3) cells harboring the plasmid pETRhaD1 was added

to 500 mL LB media containing 50 mg mL�1 carbenicillin. After 3 h growth at 37 �C,

isopropyl �-D-1-thiogalactopyranoside was added to a final concentration of 10 mm.

After 16 h at 37 �C, sodium dodecyl sulfate–polyacrylamide gel electrophoresis analy-

sis showed RhaD to be the major band in the soluble fraction. Cells were harvested by

centrifugation, washed once with 200 mL phosphate-buffered saline and centrifuged

again. Whole cells were used for subsequent biocatalytic reactions, without purification

of the enzyme.

6.1.2 Procedure 2: Synthesis of L-Fructose

OHO

OHOH

OH

OH


• Crude RhaD preparation (2.4 g wet weight)


• dihydroxyacetone, (3.60 g, 40 mmol, sold as solid dimer which spontaneously dispro-

portionates in water)

• DL-glyceraldehyde (1.80 g, 20 mmol)

• 1 M sodium borate aqueous buffer, pH 7.6 (40 mL)

• toluene (0.4 mL)

• shake flask (500 mL)

OOHHO

+O

H ROH

RhaD

sodium boratepH 7.6

OHO

OHOH

OH

OH

L-fructose, 92%

Figure 6.1 Synthesis of L-fructose (R ¼ OH).

204 Aldolase Enzymes for Complex Synthesis

• Amberlite IR-120 (Hþ form) resin (50 mL)

• Amberlite IRA-743 resin (120 mL)

• silica gel

• ethyl acetate

• methanol

• silica gel thi-layer chromatography plates


6.1.2.2 Procedure

1. Dihydroxyacetone (3.60 g, 40 mmol) and DL-glyceraldehyde (1.80 g, 20 mmol) were

dissolved in 160 mL water. Sodium borate buffer (1 M, pH 7.6, 40 mL) was added to reach

a final borate concentration of 200 mM. Toluene was added (0.4 mL), and RhaD-

containing E. coli cell pellet (2.4 g, wet weight after centrifugation) was suspended in

the solution. The mixture was stirred at 37 �C for 16 h and then centrifuged to remove the

cells. The supernatant was acidified by passing through a column (50 mL) of Amberlite

IR-120 (Hþ form) resin. The column was washed with 50 mL additional water.

2. The solution from step 1 was passed through a column (120 mL) of Amberlite IRA-743

resin to remove borate from the mixture. The resulting mixture was concentrated on a

rotary evaporator and the product was purified by silica gel chromatography, using ethyl

acetate/methanol/water (40/10/7) as the mobile phase. Unreacted D-glyceraldehyde and

excess DHA eluted before L-fructose. Rf DHA, glyceraldehyde:~0.7, 0.6; Rf fructose:

�0.35. Concentration on a rotary evaporator yielded 1.66 g L-fructose (9.2 mmol, 92 %

based on L-glyceraldehyde. Before NMR analysis, the sample was allowed to equili-

brate for 1 h in D2O in order to reach equilibrium between furanose and pyranose forms.1H and 13C spectra matched those of authentic samples of D-fructose and L-fructose.

½��23D ¼þ93:1� (c ¼ 3, H2O).

6.1.3 Conclusion

A practical, inexpensive one-step procedure was developed for the RhaD-catalyzed gram-

scale synthesis of L-fructose. The requirement for DHAP as the donor substrate was

circumvented by use of borate buffer, presumably by in situ formation of borate esters as

a phosphate ester mimic. Racemic glyceraldehyde was also used, as the enzyme preferen-

tially accepted the L-enantiomer as a substrate. The method can also be applied to other

products, including L-rhamnulose, and towards a two-step synthesis of L-iminocyclitols.1

Reference

1. Sugiyama, M., Hong, Z., Whalen L.J., Greenberg, W.A. and Wong,C.-H., Borate as a phosphateester mimic in aldolase-catalyzed reactions: practical synthesis of L-fructose and L-iminocycli-tols. Adv. Synth. Catal., 2006, 348, 2555.

6.1 Synthesis of L-Fructose Using Rhamnulose-1-phosphate Aldolase 205

6.2 Straightforward Fructose-1,6-bisphosphate Aldolase mediatedSynthesis of AminocyclitolsMarielle Lemaire and Lahssen El Blidi

Fructose-1,6-bisphosphate aldolase (RAMA) is a useful enzyme for catalysis of carbon–

carbon bond formation with 3S,4R stereoselectivity.1 Recently, we have developed a highly

stereoselective methodology for nitro and aminocyclitols preparation using RAMA which

catalysed the condensation of dihydroxyacetone phosphate (DHAP) on nitrobutyraldehydes

(Figure 6.2).2 The key step is based on a one pot–two enzymes process where three reactions

take place: the aldolisation catalysed by RAMA, the phosphate hydrolysis catalysed by a

phosphatase and an intramolecular Henry reaction (nitroaldolization). Two families of

nitrocyclitols were obtained, depending on the carbon configuration in � position to the

nitro group. When an optically pure compound was used, only one isomer was isolated.

6.2.1 Procedure 1: Synthesis of (1S,2S,3R,5S,6R)-1-Hydroxymethyl-

6-nitrocyclohexane-1,2,3,5-tetraol and (1R,2S,3R,5R,6S)-

1-Hydroxymethyl-6-nitrocyclohexane-1,2,3,5-tetraol

HO

HO

OH

OHNO2 HO

HO

OH

OHNO2

OH

+

OH

1S, 2S, 3R, 3S, 6R 1R, 2S, 3R, 3R, 6S

12

A B


• Distilled water

• racemic 4,4-diethoxy-1-nitrobutan-2-ol2b (400 mg, 1.93 mmol)

• Dowex� 50� 8, Hþ form (1.5 g)

OEt

DHAP

O

OH OPO3

O2NOEt

R

R =

HO

HO

OH

OHNO2

1) Dowex 50x8, H+, 45°C2) RAMA pH 7.5, 25°C3) Phytase pH 3.9, 25°C HO

HO

OH

OHNO2

OH

+

R = , 35 % yield 29 % yield

OH

R = OH

R = OH 50 % yield

orOH rac

(R) OH

(R )

1S, 2S, 3R, 5S, 6R 1R, 2S, 3R, 5R, 6S

12

Figure 6.2 Enzymatic synthesis of nitrocyclitols.


• NaOH (1 M, for pH adjustment)

• HCl (1 M, for pH adjustment)

• DHAP3 (1.61 mmol)

• RAMA from rabbit muscle, 60 U (suspension in ammonium sulfate from

Sigma)

• phytase from Aspergillus ficuum, 108 U (crude from Sigma)

• argon



• methanol (80 mL)

• silica gel 60 (40–63 mm, Merck) (25 g)

• thin-layer chromatography (TLC) plates (silica gel 60 F254, Merck)




• sintered glass funnel

• filtered flask

• water bath and thermostat

• microtube Eppendorf, 1.5 mL

• microcentrifuge

• pH meter


• equipment for flash column chromatography.

6.2.1.2 Procedure

1. To a solution of racemic 4,4-diethoxy-1-nitrobutan-2-ol2b (400 mg, 1.93 mmol) in 5 mL

water was added cation exchange resin (Dowex 50x8, Hþ form, 1.5 g). The suspension

was stirred at 45 �C for 2.5 h (quantitative by TLC).

2. Resin was filtered off and pH was adjusted to 7.5 with 1 M NaOH.

3. To this solution was added DHAP3 (1.61 mmol) followed by 30 mL water, and the pH

was adjusted to 7.5 with 1 M NaOH. The mixture was bubbled with Ar and previously

centrifuged aldolase4 (60 U) was added.

4. The volume of DHAP solution depends on the concentration, which is generally around

400 mM.

5. After stirring 24 h at 25 �C, the mixture was washed with 3� 20 mL EtOAc. The

water phase pH was adjusted to 3.9 with 1 M HCl and phytase (108 U) was added.

The resulting solution was stirred at 25 �C for 24 h and then concentrated under

vacuum.

6. Purification by silica gel flash column chromatography (eluent: CH2Cl2/MeOH 9/1 and

then 8/2) gave the nitrocyclitols in 64 % yield (B: 104 mg, 29 %) and (A: 126 mg, 35 %)

as brown solids.

Data for A. Rf ¼ 0.18 (CH2Cl2/MeOH: 85/15). M.p. 136 �C. ½��23D ¼þ27 (c ¼

2.67, CH3OH). 1H NMR (400 MHz, CD3OD): � 4.65 (d, 1H, J ¼ 10.2 Hz); 4.5

(ddd, 1H, J ¼ 10.2, 5.1, 12 Hz); 3.8 (ddd, 1H, J ¼ 9.3, 12, 4.7 Hz); 3.78 (d, 1H, J

¼ 11 Hz); 3.4 (d, 1H, J ¼ 9.3 Hz); 3.25 (d, 1H, J ¼ 11 Hz); 2.27 (ddd, 1H, J ¼

6.2 Straightforward RAMA-mediated Synthesis of Aminocyclitols 207

4.7, 5.1, 12 Hz); 1.43 (ddd, 1H, J ¼ 12, 12.2, 12 Hz). 13C NMR (100 MHz,

CD3OD): � 93.3; 76.6; 75; 69; 66.7; 62; 39.1. IR (KBr) � (cm�1) 3480; 1545; 1380;

1063.

Data for B. Rf ¼ 0.36 (CH2Cl2/MeOH: 85/15). M.p. 153 �C. [a]23D¼þ 33.3 (c ¼

4.25, CH3OH). 1H NMR (400 MHz, CD3OD): � 4.67 (ddd, 1H, J ¼ 11, 4, 11 Hz); 4.63

(d, 1H, J ¼ 11 Hz); 4.03 (ddd, 1H, J ¼ 3, 3, 3 Hz); 3.96 (d, 1H, J ¼ 3 Hz); 3.77 (d, 1H,

J¼ 11.5 Hz); 3.42 (d, 1H, J¼ 11.5 Hz); 2.15 (ddd, 1H, J¼ 3, 3, 13.5 Hz); 1.98 (ddd, 1H,

J ¼ 3, 11, 13.5 Hz). 13C NMR (100 MHz, CD3OD): � 93.5; 78.4; 72; 70.3; 65.4; 64.9;

36.4. IR (KBr) � (cm�1) 3480; 1544; 1378; 1063.

6.2.2 Procedure 2: Lipase Kinetic Resolution of 4,4-Diethoxy-1-nitrobutan-2-ol

O2NOEt

OH OEt

CAL-B Novozym 435

DIPE, rt

O2N O2NOEt

OH OEt

OEt

OEt

+

PrOCO

vinylbutyrate

(R)49 % yield92 % ee non isolated: decomposes into alkene


• 4,4-Diethoxy-1-nitrobutan-2-ol2b (0.5 g, 2.41 mmol)

• Candida antarctica lipase B (CAL-B), Novozyme 435 (from Sigma) (2 g)

• vinylbutyrate (630 ml, 4.96 mmol)

• diisopropylether (DIPE) (30 mL)

• 1,3,5-trimethoxybenzene (50 mg, 10 % of the alcohol mass)


• cyclohexane (400 mL)

• hexane (490 mL)

• isopropanol (10 mL)

• silica gel 60 (40–63 mm, Merck) (25 g)

• TLC plates (silica gel 60 F254, Merck)



• sintered glass funnel

• filtered flask

• syringe (10 ml)

• Daicel Chiracel OD column (25 cm � 4.6 mm ID)


• equipment for flash column chromatography

• equipment for high-performance liquid chromatography (HPLC) with UV detector.

6.2.2.2 Procedure

1. Racemic alcohol2b (0.5 g, 2.41 mmol), CAL-B (2 g) in DIPE (20 mL) and vinylbutyrate

(630 ml, 4.96 mmol) were shaken at room temperature following the progress of the

reaction by chiral HPLC (Chiracel OD) and 1,3,5-trimethoxybenzene as internal standard.

2. After 48 h,5 the reaction was quenched by filtration over a sintered glass funnel and the

solid was rinsed with DIPE (2 � 5 mL). The filtrate was concentrated in vacuo.


3. Purification by silica gel flash column chromatography (eluent: ethyl acetate/cyclohex-

ane, 2:8) gave (R)-4,4-diethoxy-1-nitrobutan-2-ol (245 mg, 1.18 mmol) in 49 % yield,

and 92 % ee.

[a]23D¼�10.4 (c¼ 1.22, CHCl3). 1H NMR (400 MHz, CDCl3): � 4.74 (t, 1H, J ¼ 5

Hz); 4.57 (m, 1H); 4.45 (dd, 2H, J¼ 2.3, 7 Hz); 3.72 (m, 2H); 3.58 (m, 1H); 3.55 (m, 2H);

1.88 (m, 2H); 1.22 (m, 6H). 13C NMR (100 MHz, CDCl3): � 101.3; 80.4; 65.8; 62.8; 62.5;

37.2; 15.2. IR (thin film) � (cm�1) 3435; 1550; 1376; 1125. Ee was determined by HPLC

with a Chiracel OD column (hexane/isopropanol 98:2), 0.7 mL min�1; major enantiomer

(R) Rt ¼28 min; minor enantiomer (S) Rt ¼32 min.

6.2.3 Procedure 3: Synthesis of (1S,2S,3R,5S,6R)-6-Amino-

1-hydroxymethylcyclohexane-1,2,3,5-tetraol

HO

HO

OH

OHNO2

OH

12

H2/PtO2,MeOH/AcOH: 95/5 HO

HO

OH

OHNH2

OH

1S, 2S, 3R, 5S, 6R

12

78 %


• (1S,2S,3R,5S,6R)-1-Hydroxymethyl-6-nitrocyclohexane-1,2,3,5-tetraol (80 mg, 0.32 mmol)

• PtO2 (20 mg)

• MeOH/AcOH, 95:5 (40 mL)

• hydrogen

• Dowex� 50WX8, 200–400 mesh, Hþ form (10 g)

• NH4OH (1 M, 60 mL)



• Parr apparatus

• ultrafiltration membrane

• glass funnel filtration system with sintered glass membrane support

• filtered flask



6.2.3.2 Procedure

1. To a solution of nitrocyclitol (80 mg, 0.32 mmol) in MeOH/AcOH (95:5, 40 mL) was

added PtO2 (20 mg). The mixture was submitted to 50 psi of H2 in a Parr apparatus.

2. After stirring for 48 h at room temperature, the catalyst was removed by ultrafiltration

and washed with MeOH (20 mL). The filtrate was concentrated under vacuum.

3. Purification by cation exchange chromatography (Dowex� 50WX8, 200–400 mesh,

Hþ form) eluted with 1 M NH4OH afforded the amine as a white solid in 78 % yield

(48 mg, 0.25 mmol).


Rf¼ 0.32 (CH2Cl2/MeOH/NH4OH, 8/1/1). M.p. 101 �C. ½��23D ¼�7:9 (c¼ 1.1, H2O).

1H NMR (400 MHz, CD3OD): � 4.65 (d, 1H, J¼ 10.2 Hz); 4.5 (ddd, 1H, J¼ 10.2, 5.1,

11.7, 5.1 Hz); 3.8 (ddd, 1H, J¼ 9.3, 12, 4.7 Hz); 3.78 (d, 1H, J¼ 11 Hz); 3.4 (d, 1H, J¼9.4 Hz); 3.25 (d, 1H, J¼ 11 Hz); 2.27 (ddd, 1H, J¼ 4.7, 5.1, 12 Hz); 1.43 (ddd, 1H, J¼12, 12.2, 12 Hz). 13C NMR (100 MHz, CD3OD): � 74.4; 70.1; 69.9; 65.7; 63.6; 57.7;

33.9. IR (KBr) � (cm�1) 3414; 1110.

6.2.4 Conclusion

This highly stereoselective procedure where four stereocenters are fixed in one pot can

be applied to other nitroaldehydes. Thus, the nitrocyclitols are easily isolated in good

yields. The reduction of the nitro group provides the corresponding amines. Tables 6.1

and 6.2 give details of some other examples, in particular the condensation of (3R)-

3-hydroxy-4-nitrobutyraldehyde.

Table 6.1 Synthesis of nitrocyclitols using RAMA

Entry

1

2

Substrate

OEt

O2NOEt

OH

(R)

O2NO

Product

HO

HO

OH

OHNO2

OH

1R, 2S, 3R, 5R, 6S

HO

HO

OH

OHNO2

Yield(%)

50

69

Reference

2b

2a

Table 6.2 Table 6.2 Reduction of nitrocyclitols using Procedure 3

Entry

1

2

Nitrocyclitol

HO

HO

OH

OHNO2

OH

1R, 2S, 3R, 5R, 6S

HO

HO

OH

OHNO2

Product yield (%)

80

80

Reference

2b

2a



1. Wong,C.-H., Formation of C–C bonds. In Enzyme Catalysis in Organic Synthesis, vol. II, Drauz,K. and Waldmann, H. (eds), Wiley–VCH, Weinheim, 2002, pp. 931–943.

2. (a) El Blidi, L., Crestia, D., Gallienne, E., Demuynck, C., Bolte, J. and Lemaire, M.,A straightforward synthesis of an aminocyclitol based on an enzymatic aldol reaction and ahighly stereoselective intramolecular Henry reaction. Tetrahedron Asymm., 2004, 15, 2951. (b) ElBlidi, L., Ahbala, M., Bolte, J. and Lemaire, M., Straightforward chemo-enzymatic synthesis ofnew aminocyclitols, analogues of valiolamine and their evaluation as glycosidase inhibitors.Tetrahedron Asymm., 2006, 17, 2684.

3. Charmantray, F., El Blidi, L., Gefflaut, T., Hecquet, L., Bolte, J. and Lemaire, M., Improvedstraightforward chemical synthesis of dihydroxyacetone phosphate through enzymatic desym-metrization of 2,2-dimethoxypropane-1,3-diol. J. Org. Chem., 2004, 69, 9310.

4. Ammonium salts must be removed to prevent side reactions, in particular the formation ofiminium on aldehydes.

5. Reaction time depends on HPLC results; the reaction was quenched when 51 % conversion wasreached.


6.3 Synthesis of D-Fagomine by Aldol Addition of Dihydroxyacetone toN-Cbz-3-Aminopropanal Catalysed by D-Fructose-6-phosphate AldolaseJose A. Castillo, Teodor Parella, Tomoyuki Inoue, Georg A. Sprenger, Jesus Joglar

and Pere Clapes

D-Fructose-6-phosphate aldolase (FSA) mediated a chemo-enzymatic synthesis of

D-fagomine 3 in 51 % isolated yield and 99 % diastereomeric excess (de).1 The

key step was the FSA-catalysed aldol addition of simple dihydroxyacetone (DHA) to

N-Cbz-3-aminopropanal. FSA is a novel class I aldolase from Escherichia coli

related to a novel group of bacterial transaldolases, which catalyses the aldol

addition of DHA to glyceraldehyde-3-phosphate.2 The cloning and overexpression

in E. coli DH5a of the gene encoding FSA and the biochemical characterization was

carried out for the first time by Schurmann and Sprenger.2 The use of FSA provides

a greatly simplified alternative to the chemo-enzymatic procedures that use DHA

phospate.3

6.3.1 Procedure 1: Production of FSA


• Yeast extract (4 g)

• tryptone (8 g)

• glyclyglycine (Gly-Gly) Guffer

• 1,4-dithio-D,L-threithol (DTT, 7.7 mg)

• isopropyl-�-D-thiogalacto-pyranoside (IPTG, 0.19 g)

• ampicillin (0.081 mg)

• sodium chloride (8 g)

• stored culture of E. coli DH5a pJF119fsa

• distilled water 800 mL

• shot cell disrupter system (1.37 kbar)

• UV–vis spectrophotometer (l ¼ 340 and 595 nm)

• centrifuge (20 000 rpm)

• autoclave

• shaker

• phosphoglucose isomerase from baker’s yeast (Saccharomyces cerevisiae) Type III,

ammonium sulfate suspension, 500–800 units mg�1

• glucose-6-phosphate dehydrogenase from Leuconostoc mesenteroides, Type XXIII,

ammonium sulfate suspension, 550–1100 units/mg protein (biuret)

• DL-Glyceraldehyde 3-phosphate solution 45–55 mg�1 in H2O.

6.3.1.2 Procedure

1. Recombinant E. coli cells were cultivated at 37 �C in Luria–Bertani (LB) medium

(1 % w/v tryptone, 0.5 % w/v yeast extract, 1 % w/v NaCl, pH 7.2) containing

ampicillin (100 mg L�1). Culture medium (LB) was prepared by dissolving yeast

extract (4 g), tryptone (8 g) and NaCl (8 g) in distilled water (800 mL) and

sterilized by autoclaving at 125 �C for 30 min. Preinoculum cultures of E. coli


DH5a (pJF119fsa) were grown from a frozen stock4 (1 mL) overnight in LB (45

mL) with ampicillin (100 mg L�1) at 37 �C on an orbital shaker (250 rpm). An

aliquot of preinoculum (15 mL) was transferred to a 2000 mL Erlenmeyer flask

containing LB (800 mL) with ampicillin (0.1 mg L�1) and the mixture was

incubated at 37 �C with shaking at 250 rpm. When the absorbance of the medium

was 0.5 at 600 nm (i.e. �2 h), IPTG (0.19 g, 1 mM final concentration) was added

and the incubation left to proceed for 16 h at 37 �C with shaking at 250 rpm.

2. Cells from 3� 800 mL induced culture broths were withdrawn and centrifuged at

12 000g for 10 min at 4 �C. The pellet was resuspended with glycylglycine (Gly-Gly)

buffer (100 mL, 50 mM, 1 mM DTT, pH 8.0) and the cells disrupted with a shot cell

disrupter system (1.37 kbar). Cellular debris was removed by centrifugation at 12 000g

for 30 min at 4 �C.

3. The clear supernatant thus obtained (150 mL) was incubated at 75 �C for 40 min. After

that a precipitate appeared which was separated by centrifugation (12 000g for 30 min

at 4 �C). The clear supernatant was dialysed against glycylglycine (Gly-Gly) buffer (10

L, 10 mM, pH 8.0) and lyophilized (2.8 g).

4. Protein content: Bradford’s test was performed to determine the protein content (mg g�1)

of the lyophilisates. A sample of the lyophilisate (2.5 mg) was dissolved in glycylglycine

(Gly-Gly) buffer (1.5 mL). An aliquot of this solution (20 mL) was diluted with glycyl-

glycine (Gly-Gly) buffer (30 mL) and Bradford’s solution (950 mL) was added. After 5

min, the absorbance was measured at 595 nm. The concentration of FSA is calculated

from the interpolation of the calibration curve of bovine serum albumin (BSA): 437 mg

total protein/gram of lyophilisate.

5. The activity of the lyophilisate was measured by a multienzymatic test based on

the formation of D-fructose-6-phosphate (Figure 6.3) Phosphoglucose isomerase

(6 mL, commercial solution), glucose-6-phosphate dehydrogenase (0.4 mL, commercial

solution), NADP (10 mL of a 50 mM solution), FSA (5 mL of an FSA solution 1.67 mg

mL�1), 50 mM Gly-Gly buffer pH 8, DTT, 1 mM (965 mL) and DHA (10 mL of a 2.5 M

solution) were mixed and placed at 30 �C for 5 min. Finally, DL-glyceraldehyde-3-

phosphate (10 mL, commercial solution) was added and the reaction was followed

spectrophotometrically. The increment of absorbance at 340 nm due to the NADPH

production is proportional to D-fructose-6-phosphate formation (30 �C, pH 8.5

FSA+

DHA G3P

OHO

OHOH

GPD

NADPHNADP

OH

2–O3PO

O

OH

OH O

OH

OPO32–

OHHO

OHOH

O

2–O3PO

OHHO

OH

2–O3PO

PGI

F6P G6P Gluconolactone-6-P

O

O

Figure 6.3 Multienzymatic activity test for FSA. G3P: D-glyceraldehyde-3-phosphate; F6P:fructose-6-phosphate; PGI: phosphoglucose isomerase; G6P: glucose-6-phosphate; GPD: glu-cose-6-phosphate dehydrogenase.

6.3 Catalysed Synthesis of D-Fagomine 213

(glycylglycine (Gly-Gly) 50 mM buffer) from D-glyceraldehyde-3-phosphate and DHA.

Unit definition: one unit (U) will synthesize 1 mmol of D-fructose-6-phosphate per

minute at pH 8.5 (glycylglycine 50 mM buffer) and 30 �C. The enzymatic activity of the

lyophilized powder was 1.7 U mg�1.

6.3.2 Procedure 2: Synthesis of N-Cbz-3-aminopropanal

NH

OCbz

1

NH

CbzOHH2N OH

HCbz-OSu

Dioxane:H2O4:1

IBX

EtOAc, reflux


• 3-Aminopropanol (4.2 g)

• 2-iodoxybenzoic acid (IBX, 9.4 g)

• N-(benzyloxycarbonyloxy)succinimide (CBz-OSu, 12.3 g)

• sodium hydrogen carbonate

• distilled water

• ethyl acetate


• filter paper

• one 500 mL round-bottom flask

• one Buchner funnel, diameter 10 cm


• one 500 mL separation funnel


6.3.2.2 Procedure

1. To a solution of 3-amino-1-propanol (4.2 g, 55.5 mmol) in dioxane/H2O 4:1 (150 mL),

CBz-OSu (12.3 g, 49.4 mmol) dissolved in dioxane/H2O 4:1 (150 mL) was added

dropwise. The reaction mixture was stirred overnight at room temperature and then

evaporated to dryness under reduced pressure. The residue was dissolved in EtOAc and

washed with an aqueous solution of NaHCO3 (5 % w/v, 2 � 50 mL), an aqueous

solution of citric acid (5 % w/v, 2 � 50 mL) and brine (2 � 50 mL). The organic layer

was dried with anhydrous Na2SO4, filtered and evaporated under vacuum to dryness

furnishing N-benzyloxycarbonyl-3-amino-1-propanol (9.28 g, isolated yield 90 %) as a

white solid. 1H NMR (500 MHz; CDCl3) � (ppm) ¼ 7.35 (m, Ar, 5H), 5.10 (s, PhCH2,

2H), 5.07 (s, NH, 1H), 3.67 (t, J¼ 5.5,�CH2OH, 2H), 3.36 (dd, J¼ 12.3, 6.1, NHCH2,

2H), 1.70 (td, J ¼ 11.9, 5.8 and 5.8, CH2CH2CH2, 2H).

2. IBX (9.38 g, 34.49 mmol) was added to a solution of N-benzyloxycarbonyl-3-amino-

1-propanol (4.67 g, 22.3 mmol) in EtOAc (150 mL) and the mixture was heated at

reflux for 6 h and then cooled to room temperature and filtered. The filtrate was washed

with a 5 % w/v aqueous solution of NaHCO3 (2 � 90 mL) and brine (2 � 90 mL) and

the organic phase was dried with anhydrous sodium sulfate, filtered and the solvent

evaporated under vacuum to dryness affording N-benzyloxycarbonyl-3-amino-1-propanal


(1) (3.77 g, isolated yield 82 %) as a white solid. 1H NMR (500 MHz; CDCl3) � (ppm)¼9.80 (s, CHO, 1H), 7.34 (m, Ar, 5H,), 5.17 (s, PhCH2, 2H), 5.08 (s, NH, 1H), 3.49 (dd,

J ¼ 11.9, 6.0, NHCH2, 2H), 2.75 (t, J ¼ 5.6,�CH2CHO, 2H).

6.3.3 Procedure 3: Synthesis of D-Fagomine

OH

NH

Cbz OHNH

OCbz

H

OH

O

1 2

HN

OH

OHOH

3

HO OHO

FSA DMF:Buffer 1:4

H2, Pd/C

H2O:EtOH 9:1

Cbz : O

O


• N-Benzyloxycarbonyl-3-amino-1-propanal 1 (4.7 g)

• DHA (2.1 g)

• dimethylformamide (DMF, 40 mL)


• lyophilisate with FSA activity (2.1 g, 3445U)

• 10% palladium over charcoal (100 mg)

• ethanol

• distilled water

• acetonitrile (MeCN)

• trifluoroacetic acid (TFA)

• Celite� filter aid

• neutral aluminium oxide

• borate buffer pH 7

• one Buchner funnel 150 mL with 60 mm diameter coarse porosity (4) fritted disc



• analytical and preparative high-performance liquid chromatography (HPLC) system

6.3.3.2 Procedure

1. (3S,4R)-6-[(Benzyloxycarbonyl)amino]-5,6-dideoxyhex-2-ulose (2): DHA (2.1 g, 22.9

mmol) and FSA lyophilisate powder (2.1 g, 3445 U) were dissolved in boric/borate

buffer 50 mM pH 7.0 (155 mL) and cooled to 4 �C. N-Cbz-aminoaldehyde 1 (4.7 g, 22.9

mmol) dissolved in DMF (40 mL) at 4 �C was added to this mixture. The reaction

mixture was then placed in a reciprocal shaker (120 rpm) at 4 �C. After 24 h, MeOH (200

mL) was added to the mixture to stop the reaction and centrifuged (1000 rpm) at 10 �C

for 40 min.

HPLC reaction monitoring: HPLC analyses were performed on an RP-HPLC

cartridge, 250 mm � 4 mm filled with Lichrosphere� 100, RP-18, 5 mm from Merck

(Darmstadt, Germany). Samples (50 mL) were withdrawn from the reaction medium,

dissolved in MeOH to stop the reaction and analysed by HPLC. The solvent system


used was solvent (A) (H2O 0.1 % (v/v) TFA) and solvent (B) (MeCN/H2O 4/1 0.095 %

(v/v) TFA), gradient elution from 10 % to 70 % B in 30 min, flow rate 1 mL min�1,

detection 215 nm. Retention time for 2 was 3.7 min.

2. The supernatant was filtered on a 0.45 mm filter and purified by preparative HPLC as

follows. The crude 2 was loaded onto a preparative column (47 mm i.d. � 300 mm)

filled with Bondapack C18 (Waters), 300 A, 15–20 mm stationary phase. Products were

eluted using a gradient from 0 % to 28 % MeCN in 35 min. The flow rate was 100 mL

min�1 and the products were detected at 215 nm. Analysis of the fractions was

accomplished under isocratic conditions (33 % of solvent B) on the analytical HPLC.

Pure fractions were pooled and lyophilized to give 2 (4.7 g, 69 %) as a white solid.

½��22D ¼ þ9:0 (c¼ 1.0 in MeOH); 1H NMR (500 MHz; D2O): � (ppm)¼ 7.47–7.35 (5H,

Ar), 5.09 (s, 2H, 8-H), 4.54 (d, J ¼ 19.4 Hz, 1H, 1-H), 4.43 (d, J ¼ 19.4 Hz, 1H, 1-H),

4.28 (s, 1H, 3-H), 4.02 (dt, J ¼ 6.8, 6.8, 2.2 Hz, 1H, 4-H), 3.2 (m, 2H, 6-H) and 1.75

(q, J ¼ 6.8 Hz, 2H, 5-H); 13C NMR (100 MHz; D2O,): � (ppm) ¼ 212.9 (C-2), 158.6

(C-7), 136.7 (Ar), 128.9 (Ar), 128.5 (Ar), 127.8 (Ar), 77.7 (C-3), 69.5 (C-1), 67.0 (C-8),

66.1(C-4), 37.2 (C-6), 32.5 (C-5).

3. D-Fagomine [(2R,3R,4R)-2-hydroxymethylpiperidine-3,4-diol] (3): Pd/C (100 mg)

was added to a solution of the aldol adduct (373 mg, 1.26 mmol) in H2O/EtOH 9:1

(50 mL). The reaction mixture was stirred under hydrogen gas (50 psi) overnight at

room temperature. After removal of the catalyst by filtration through neutralized and

deactivated aluminium oxide, the solvent was evaporated under reduced pressure.

D-Fagomine (164 mg, yield 89 %) was afforded as a brown solid ½��20D ¼þ20:4 (c¼ 1.0

in H2O) (lit.5 ½��20D ¼þ19:5 (c ¼ 1.0, H2O)), 93 % pure by NMR.

4. A fraction of the crude D-fagomine (24 mg) was purified by weak cation-exchange

chromatography on a fast protein liquid chromatography system. Bulk stationary-

phase CM-Sepharose CL-6B (Amersham Pharmacia) in NH4þ form was packed into a

glass column (120 mm � 10 mm) to a final bed volume of 8 mL. The flow rate was

0.7 mL min�1. The CM-Sepharose was equilibrated initially with H2O. Then, an

aqueous solution of the crude material (24 mg) at pH 7 was loaded onto the column.

Minor coloured impurities were washed away with H2O (two or three column

volumes) and then D-fagomine was eluted with 0.01 M NH4OH and lyophilized to

afford a pale brown solid (20 mg, 83 %). 1H NMR (500 MHz; D2O), � (ppm) ¼ 3.86

(dd, J ¼ 11.8, 3.0 Hz, 1H, 7-H), 3.66 (dd, J ¼ 11.8, 6.5 Hz, 1H, 7-H), 3.56 (ddd,

J ¼ 11.5, 9.0, 5.0, 1H, 4-H), 3.21 (t, J ¼ 9.5 Hz, 1H 3-H), 3.06 (ddd, J ¼ 12.9, 4.4, 2.3

Hz, 1H, 6-H), 2.68 (dt, J ¼ 12.9, 12.9, 2.7 Hz, 1H, 6-H), 2.61 (ddd, J ¼ 9.7, 6.4, 3.0,

1H, 2-H), 2.01 (tdd, J ¼ 13.0, 4.9, 2.5,2.5, 1H, 5-H), 1.53-1.43 (m, 1H, 5-H);13C NMR (101 MHz; D2O), � (ppm) ¼ 72.9 (C-4), 72.7 (C-3), 61.1 (C-2), 60.9

(C-7), 42.6 (C-6), 32.1 (C-5). D-2,4-Di-epi-fagomine was eluted in later fractions,

affording a brown solid (<1 mg) after lyophilization.

6.3.4 Conclusion

In summary, D-fagomine can be obtained stereoselectively in two chemo-enzymatic

asymmetric steps using inexpensive achiral DHA and N-Cbz-3-aminopropanal as the

starting materials and FSA from E. coli as biocatalyst in 51 % isolated yield and �99 %

de. The strategy offers a fresh and novel approach to tackle the problem of the previous


enzymatic approaches that used DHA phosphate,3,6 a reactant prepared chemically in

several steps7 or generated in situ by a multienzymatic procedure,8 which hampered the

general use of these enzymes. FSA is a robust biocatalyst offering a simple and highly

stereoselective aldol reaction starting from achiral reactants.


1. Castillo, J.A., Calveras, J., Casas, J., Mitjans, M., Vinardell, M. P., Parella, T., Inoue, T., Sprenger,G.A., Joglar, J. and Clapes, P., Fructose-6-phosphate aldolase in organic synthesis: preparation ofD-fagomine, N-alkylated derivatives, and preliminary biological assays. Org. Lett., 2006, 8, 6067.

2. Schurmann,M. and Sprenger, G.A., Fructose-6-phosphate aldolase is a novel class I aldolase fromEscherichia coli and is related to a novel group of bacterial transaldolases. J. Biol. Chem., 2001,276, 11055.

3. Von derOsten, C.H., Sinskey, A.J., Barbas III, C.F., Pederson, R.L., Wang, Y.F. and Wong, C.H.,Use of a recombinant bacterial fructose-1,6-diphosphate aldolase in aldol reactions: preparativesyntheses of 1-deoxynojirimycin, 1-deoxymannojirimycin, 1,4-dideoxy-1,4-imino-D-arabinitol,and fagomine. J. Am. Chem. Soc., 1989, 111, 3924.

4. E. coli DH5a (pJF119fsa) strain was conserved at �80 �C in glycerol stocks prepared fromexponential-phase cultures.

5. Kato, A., Asano, N., Kizu, H., Matsui, K., Watson, A.A. and Nash, R.J., Fagomine isomers andglycosides from Xanthocercis zambesiaca. J. Nat. Prod., 1997, 60, 312.

6. Espelt, L., Parella, T., Bujons, J., Solans, C., Joglar, J., Delgado, A. and, Clapes, P.,Stereoselective aldol additions catalyzed by dihydroxyacetone phosphate-dependent aldolasesin emulsion systems: preparation and structural characterization of linear and cyclic iminopolyolsfrom aminoaldehydes. Chem. Eur. J., 2003, 9, 4887.

7. Jung,S.-H., Jeong, J.-H., Miller, P. and Wong,C.-H., An efficient multigram-scale preparation ofdihydroxyacetone phosphate. J. Org. Chem., 1994, 59, 7182.

8. Fessner, W.D., and Sinerius, G., Synthesis of dihydroxyacetone phosphate (and isosteric analo-gues) by enzymatic oxidation; sugars from glycerol. Angew. Chem. Int. Ed. Engl, 1994, 33, 209.


6.4 Chemoenzymatic Synthesis of 5-Thio-D-xylopyranoseFranck Charmantray, Philippe Dellis, Virgil Helaine, Soth Samreth and Laurence

Hecquet

Recently, we investigated a two-step chemo-enzymatic procedure to synthesize 5-thio-

D-xylopyranose,1 both an inhibitor of �-D-xylosidase and as a useful chiral building block

for the synthesis of D-xylopyranosides, which display antithrombotic activity. We performed

the synthesis of 5-thio-D-xylulofuranose through both the stereospecific C�C bond forma-

tion catalysed by transketolase (TK, EC 2.2.1.1) from Saccharomyces cerevisiae from the

recombinant yeast strain H402 x pTKL1, route A; and by commercially available fructose-

1,6-bisphosphate aldolase from rabbit muscle (FruA, EC 4.1.2.13), route B (Scheme 6.1).

The second step consisted of an enzymatic isomerization of 5-thio-D-xylulofuranose

(ketose), using glucose isomerase (GlcI, EC 5.3.1.5), into 5-thio-D-xylopyranose (aldose).

5-Thio-D-xylopyranose was isolated in good yield and high reproducibility.

6.4.1 Procedure 1: Cultivation of Saccharomyces cerevisiae Recombinant Strain

H402 x pTKL1 for TK Expression


Ingredients for Culture Medium

• Galactose (100 g)

• yeast nitrogen base without amino acids and ammonium sulfate (33.5 g)

• ammonium sulfate (25.2 g)

• adenine sulfate (0.10 g)

• uracil (0.10 g)

O

OH

SH

O–, Li +

O

HO

HPA

route A

CO2

O

X

route B

OP

O

HO

DHAP

O

HO

OH

OH

SH

OHO

OH

OH

SH

Glucoseisomerase

5-D-thioxylulofuranose 5-D-thioxylopyranose

X: Cl,Br

HPA: hydroxypyruvate, lithiumsaltDHAP: dihydroxyacetone phosphate

O

then NaSH

Scheme 6.2


• L-tryptophan (0.10 g)

• L-histidine (0.10 g)

• L-arginine (0.10 g)

• L-methionone (0.10 g)

• L-tyrosine (0.15 g)

• L-isoleucine (0.15 g)

• L-lysine (0.15 g)

• L-phenylalanine (0.25 g)

• L-glutamic acid (0.50 g)

• L-valine (0.75 g)

• L-serine (2.00 g)

• L-threonine (1.00 g)

• distilled water (5 L).

Other

• Autoclave

• rotary shaker

• one shot cell disrupter

• centrifuge

• syringe filter 0.22 mm

• tubes for centrifugation, 500 mL and 50 mL

• spectrophotometer

• Q Sepharose Fast Flow resin (Pharmacia Biotech)

• tris buffer

• L-erythrulose (120 mg)

• D-ribose 5-phosphate (25 mg)

• thiamine pyrophosphate (ThDP) (10 mg)

• alcohol dehydrogenase from S. cerevisiae (E.C. 1.1.1.1) (25 units)

• MgCl2 (10 mg)

• reduced nicotinamide adenine dinucleotide (NADH, 10 mg)

• Coomassie blue (100 mg)

• ethanol (50 mL)

• phosphoric acid 85% (100 mL)

• serum albumin (1 mg).

6.4.1.2 Procedure

The culture of H402 x pTKL1 yeast (given by Professor Gunter Schneider) was

prepared according to the procedure described in the literature with some

modifications.2

1. The culture medium was prepared by dissolving all the compounds for the culture

medium except L-threonine in an Erlenmeyer flask containing 5 L of water. The pH was

adjusted to 5.5 with NaOH and then sterilized for 20 min at 120 �C. L-Threonine was

filtered (syringe sterile filter: Sartorius Minisart� 0.2 mm) and was added to the

sterilized culture medium.


2. Preculture (100 mL) was prepared in the liquid medium containing an inoculum of

H402 x pTKL1 yeast,2a that was stored at�80 �C, and was grown for 48 h at 30 �C in an

Erlenmeyer flask (500 mL) under rotatory shaking (200 rpm).

3. The culture of cells was conducted in a 500 mL Erlenmeyer flask containing 100 mL

of liquid medium inoculated with 1.5 mL of preculture. The cultures were incubated

at 30 �C for 40 h and then harvested. The cells were collected by centrifugation at

2500g for 4 min. Approximately 10 g of packed cells were obtained from 1 L of

culture.

4. The yeast TK extraction was conducted from the packed cells. 10 g of cells were

resuspended in 100 mL of tris buffer (0.1 M) at pH 7.4 and placed in a one shot cell

disrupter at 2 kbar. The solution was centrifuged at 15000 rpm for 20 min. The pH of the

supernatant was adjusted to 7.4. 10 mL of Q Sepharose resin per 50 mg of proteins was

added and stirred for 2 h.

(Protein concentration was determined using the Bradford assay at 595 nm. 100 mL of

the sample were introduced into a cuvette containing 5 mL of Bradford solution (100

mg of Coomassie blue, 50 mL of ethanol and 100 mL of 85 % phosphoric acid dissolved

in 850 mL of H2O). The solutions were incubated for 5 min at room temperature. The

absorbance was measured at 595 nm. The protein concentration in the sample was

determined using a calibration curve plotted with serum albumin (1 mg mL�1) as a

standard.)

The resin was removed by centrifugation at 5000 rpm at 0 �C for 5 min. TK specific

activity was 2.6 units/mg. The filtrate was used for enzymatic synthesis.

5. TK activity was determined by a spectrophotometric assay at 340 nm. In 0.5 mL of

tris buffer (50 mM) pH 7.6, were added 50 mL of L-erythrulose from a 1 M stock

solution in water (0.05 mmol), 25 mL of D-ribose-5-phosphate from a 160 mM stock

solution in water (4.0 mmol), 5 mL of ThDP from a 21 mM stock solution in water

(0.1 mmol), 10 mL of MgCl2 from a 50 mM stock solution in water (0.5 mmol),

10 mL of NADH from a 14 mM solution in water (0.14 mmol), alcohol dehydro-

genase (12 units).

6.4.2 Procedure 2: Chemo-enzymatic Synthesis of 5-Thio-D-xylulofuranose. Route

A: TK-catalysed C�C Bond Formation

O

HO

OH

OH

SH


• Round-bottom flask (100 mL)

• tris buffer pH 7.5 (200 mM)

• lithium hydroxypyruvate (154 mg, 1.4 mmol)

• 3-thioglyceraldehyde (297 mg, 2.8 mmol)

• TK crude extract, yeast strain H4021 (200 units)

• distilled water


• MgCl2 (14 mg, 0.15 mmol)

• ThDP (47 mg, 0.10 mmol)

• NADH (10mg)

• triethanolamine buffer

• L-lactate dehydrogenase from rabbit muscle (E.C. 1.1.1.27) (25 units)

• methanol (analytical grade)

• methylene chloride (analytical grade)

• Dowex 50WX8 Hþ (10 mL)

• Dowex 1X8 HCO3� (10 mL)

• Merck 60 F254 silica gel thin-layer chromatography (TLC) plates

• Merck 60/230–400 and 60/40–63 mesh silica


• centrifuge

• heating block

• shaker


6.4.2.2 Procedure

1. In a 100 mL flask, tris buffer (1.36g, 10 mmol), lithium hydroxypyruvate (154 mg,

purity 60 %, 0.84 mmol), 3-thioglyceraldehyde (650 mg, 6.1 mmol), MgCl2 (21 mg,

0.105 mmol,) and 32 mg of ThDP (0.07 mmol) were added to 35 mL of distilled water.

The pH was adjusted to 7.5 and the volume adjusted to 50 mL with distilled water.

Then, 200 units of yeast TK were added. The reaction was stirred at 30 �C until

complete disappearance of a-hydroxypyruvate.3

2. The concentration of hydroxypyruvate was determined by spectrophotometry at 340 nm.

A 20 mL aliquot from the reaction mixture was introduced into 1 mL of triethanolamine

buffer (0.1 M) at pH 7.6 containing 20 mL of NADH solution in water (14 mm, 0.28 mmol)

and 2 units of L-lactate dehydrogenase. The absorbance due to NADH is proportional to

the concentration of hydroxypyruvate.

3. Proteins were precipitated with 170 mL of methanol and removed by centrifugation at

8000 rpm for 15 min. Then, 10 mL of Dowex 50WX8 Hþ form resin was added, stirred

for 30 min and removed by filtration. 10 mL of Dowex 1X8 HCO3� form resin was

added to the filtrate and stirred for 30 min. After filtration, the solution was concen-

trated and crude 5-thio-D-xylulofuranose was purified by flash column chromatography

using 8/2 CH2Cl2/MeOH.

5-Thio-D-xylulofuranose. � isomer: 1H NMR (CD3OD) � 2.57 (dd, J¼ 9, 10 Hz, 1H, 5-H),

3.03 (dd, J¼ 8, 10 Hz, 1H, 50-H), 3.59 (d, J¼ 13 Hz, 1H, 1-H), 3.62 (d, J¼ 13 Hz, 1H, 10-H),

3.73 (d, J ¼ 9 Hz, 1H, 3-H), 4.23–4.32 (m, 1H, 4-H) ppm; 13C NMR (100 MHz, CD3OD)

� 31.2 (C-5), 67.5 (C-1), 77.0 (C-4), 79.8 (C-3), 89.7 (C-2) ppm. a isomer: 1H NMR (CD3OD)

� 2.94 (dd, J¼ 5, 11 Hz, 1H, 5-H), 3.03 (dd, J¼ 5, 11 Hz, 1H, 50-H), 3.73 (d, J¼ 11 Hz, 1H,

1-H), 3.81 (d, J¼ 11 Hz, 1H, 10-H), 4.03 (d, J¼ 5 Hz, 1H, 3-H), 4.23–4.32 (m, 1H, 4-H) ppm;13C NMR (CD3OD) � 35.9 (C-5), 66.8 (C-1), 79.0 (C-4), 83.9 (C-3), 96.0 (C-2) ppm.

High-resolution electrospray ionization mass spectrometry (HR-ESI-MS) calculated for

C5H10O4NaS [MþNa]þ: 189.0198 ; found: 189.0192.

½��20D ¼�114� (0.3, H2O); (lit.4: �116� (0.3, H2O)).


6.4.3 Procedure 2: Chemo-enzymatic Synthesis of 5-Thio-D-xylulofuranose. Route

B: FruA-catalysed C–C Bond Making

Route B is shown in Scheme 6.2.

6.4.3.1 FruA-catalysed One-pot Synthesis of 5-Chloro-D-xylulose from rac-Glycidol

Materials and Equipment


• two-neck round-bottom flask (50 mL)

• Na2HPO4 (3.71 g, 25 mmol)

• rac-2,3-epoxypropanol (1.9 g, 25 mmol)

• L-glycerol-3-phosphate oxidase, L-GPO (EC 1.1.3.21, from Thermophilus bacillus)

(45 units)

• catalase (EC 1.11.1.6, from bovine liver) (1800 units)

• O2 cylinder

• fructose-1,6-bisphosphate aldolase, FruA (EC 4.1.2.13, from rabbit muscle) (20 units)

• acid phosphatase type XA, Pase (EC 3.1.3.2, from sweet potato) (50 units)

• 2-chloroacetaldehyde (0.13 mL, 2 mmol) 1 m HCl (5 mL)

• 1 M NaOH (5 mL)

• distilled water

• MeOH (analytical grade)


• Merck 60 F254 silica gel TLC plates


• heating block

• shaker



Procedure

1. Na2HPO4 (3.71 g, 25 mmol) was added to a 50 mL solution of rac-2,3-epoxypropanol

(1.9 g, 25 mmol) in distilled water. The mixture was heated at 100 �C for 3 h and assayed

for l-glycerol-3-phosphate content.5 The yield was 61 % from (S)-2,3-epoxypropanol.

2. To a 10 mL solution of 155 mm l-glycerol-3-phosphate at pH 6.8, 0.1 mL GPO–catalase

mixture (45 units–1800 units), 0.07 mL of FruA (20 units) and aldehyde

OHONa2HPO4

OHOPHO

OHO OP

O

HX

DHAP

OHOPHO

D

H2O2

H2O

OOHX

OH

OH

NaSHO

OHHSOH

OHL

L-GPO, pH 6.8

1/2 O2

1. FruA, pH 6.82. Pase, pH 4.7

X = Cl, Br

5-halo-D-xylulose 5-thio-D-xylulofuranoseglycidol

Scheme 6.2


(2-chloroacetaldehyde, 0.13 mL, 2 mmol) was added and then a stream of O2 (50 mL

min�1) was bubbled through the solution overnight. The pH was then adjusted to 4.7

with 1 m HCl and 50 units of acid phosphatase (Pase) was added. The reaction mixture

was shaken for another 24 h at room temperature.

3. The pH was raised to 7.0 with 1 m NaOH and methanol (30 mL) was poured into the

solution. The resulting precipitate was removed by filtration through Celite. The filtrate

was concentrated under vacuum and the brown residue purified by silica gel chromato-

graphy in 9/1 CH2Cl2/MeOH. 5-Chloro-d-xylulose (160 mg, 0.95 mmol) was recovered

as a light yellow oil with 47 % yield.

5-Chloro-D-xylulose. 1H NMR (400 MHz, CD3OD): � 3.54 (dd, J¼ 8, 10 Hz, 1 H, 5-H),

3.70 (dd, J ¼ 8, 10Hz, 1 H, 50-H), 4.10 (td, J ¼ 2, 8 Hz, 1 H, 4-H), 4.39 (d, J ¼ 2 Hz, 1 H,

3-H), 4.47 (d, J ¼ 19 Hz, 1 H, 1-H), 4.54 (d, J ¼ 19Hz, 1 H, 10-H) ppm. 13C NMR (100

MHz, CD3OD): � 45.1 (C-5), 67.9 (C-1), 73.6 (C-4), 76.7 (C-3), 212.5 (C-2) ppm. HR-ESI-

MS calculated for C5H9ClNaO4 [M þ Na]þ: 191.0087; found: 191.0097.

6.4.3.2 Halogen Displacement

Materials and Equipment


• sodium hydrosulfide hydrate (NaSH�xH2O) from Aldrich (0.532 g)


• methanol (analytical grade)




• centrifuge

• heating block

• shaker



Procedure

NaSH�xH2O (0.532 g, 9.5 mmol) was added to a solution of 5-chloro-d-xylulose (0.8 g,

4.75 mmol) in distilled water (70 mL). The solution was shaken at room temperature for 3

h and kept at 4 �C overnight. The solvents were removed under reduced pressure and the

crude product was purified by flash column chromatography on silica gel. 5-Thio-D-

xylulofuranose was eluted with 8/2 CH2Cl2/MeOH in 61 % yield (0.48 g, 2.9 mmol)

starting from 5-chloro-D-xylulose.

6.4.4 Procedure 3: Glucose-isomerase-catalysed Isomerization of 5-Thio-d-

xylulofuranose to 5-Thio-d-xylopyranose

OH

O

OH

OH

SH



• 5 mL capped vials


• 5-thio-d-xylulofuranose (0.4 or 0.8 mmol)

• 85 % phosphoric acid (5 ml)

• immobilized GlcI (EC 5.3.1.5, from Streptomyces murinus), Sweetzyme� (45 unit

aliquots)

• 2 m NaOH

• 4 mL of CoCl2 (100 mm), 4 mL of MgCl2 (100 mm), 2 mL of MnCl2 (50 mm)



• Millipore� ultrafiltration system

• heating block

• shaker


• Milli-Q water

• acetonitrile, high-performance liquid chromatography (HPLC) grade

• mBondapak NH2 column (4.0 mm � 150 mm).

6.4.4.2 Procedure

1. Into a 5 mL vial, was introduced a 100 mM or 200 mM aqueous solution of 5-thio-

D-xylulofuranose (4 mL), followed by addition of 5 mL of 85 % phosphoric acid.

The pH of the solution was adjusted to 7.5 with 90 mL of 2 M NaOH. Stock

solutions of Co2þ (100 mM, 38 mL), Mg2þ (100 mM, 38 mL) and Mn2þ (50 mM,

19 mL) were then added.

2. Process A: 45 units of GlcI were added in one portion and the reaction mixture kept at

48 �C for 8 days under stirring (200 rpm).

Process B: as process A, but 45 units of GlcI were added every 24 h.

3. After removal of the protein by ultrafiltration, the reaction mixture was subjected to

semi-preparative HPLC. Purification was performed using a mBondapak NH2 column

(4.0 mm � 150 mm) with the following conditions: CH3CN/H2O, 5 mL min�1, 25 �C;

injection volume 10 mL, 4 mg.

4. 5-Thio-D-xylopyranose was recovered with 60 % yield following the best experimental

conditions (Table 6.3). The structure was confirmed by 1H and 13C NMR.6

6.4.5 Conclusion

We performed a new chemoenzymatic synthesis of 5-thio-D-xylopyranose on a preparative

scale. Our approach was based on GlcI-catalysed enzymatic isomerization of 5-thio-D-

xylulofuranose, itself obtained through enzymatically controlled C�C bond formation

catalysed by TK or FruA enzymes. This chemoenzymatic strategy offers an attractive

alternative to conventional chemical methods because of its stereochemical control, mild

conditions and no requirement for protecting groups. To improve the yield of isomeriza-

tion of thioketose into thioaldose, recycling of unreacted starting material could be

considered, as in industrial production of d-fructose from d-glucose with GlcI.


References

1. Charmantray, F., Dellis, P., Helaine,V., Samreth, S., and Hecquet, L., Chemoenzymatic synthesisof 5-thio-d-xylopyranose. Eur. J. Org. Chem, 2006, 24, 5526.

2. (a) Nehlin, J.O., Carlberg, M. and Ronne, H., Yeast galactose permease is related to yeast andmammalian glucose transporters. Gene (Amst.), 1989, 85, 313. (b) Sundstrom, M., Lindquist, Y.,Schneider, G., Hellman, U., and Ronne, H., Yeast TKL1 gene encodes a transketolase that isrequired for efficient glycolysis and biosynthesis of aromatic amino acids. J. Biol. Chem., 1993,268, 24346. (c) Wikner, L., Meshalkina, U., Nikkola, M., Lindqvist, Y. and Schneider, G.,Analysis of an invariant cofactor–protein interaction in thiamin diphosphate-dependent enzymesby site-directed mutagenesis. Glutamic acid 418 in transketolase is essential for catalysis. J. Biol.Chem., 1994, 269, 32144.

3. (a) Hecquet, L., Lemaire, M., Bolte, J. and Demuynck, C., Chemo-enzymatic synthesis ofprecursors of fagomine and 1,4-dideoxy-1,4-imino-d-arabinitol. Tetrahedron Lett., 1994, 35,8791. (b) Hecquet, L., Bolte, J. and Demuynck, C., Enzymatic synthesis of [lsquo]natural-labeled[rsquo] 6-deoxy-l-sorbose precursor of an important food flavor. Tetrahedron, 1996, 52,8223. (c) Crestia, D., Guerard,C., Veschambre, H., Hecquet, L., Demuynck, C. and Bolte, J.,Chemoenzymatic synthesis of chiral substituted acrylate and acrylonitrile precursors for thesynthesis of 3-deoxy-2-ulosonic acids and [alpha]-methylene-[gamma]-lactones. TetrahedronAsymm., 2001, 12, 869.

4. Effenberger, F., Null, V. and Ziegler, T., Preparation of optically pure l-2-hydroxyaldehydes withyeast transketolase. Tetrahedron Lett., 1992, 33, 5157.

5. Bergmeyer, H.U., Methods in Enzymatic Analysis, vol. IV, Verlag Chemie, 1984, p. 342.6. Lalot, J., Stasik, I., Demailly, G. and Beaup[egrave]reD., Efficient synthesis of 5-thio-D-arabi-

nopyranose and 5-thio-d-xylopyranose from the corresponding D-pentono-1,4-lactones.Carbohydr. Res. 2003, 338, 2241.

Table 6.3 Glci-catalysed Isomerisation of 5-Thio-D-xylulofuranose into 5-Thio-D-xylopyranose.

Experiment Substrate thioketose(mm)

GlcI (units)a Processb Reactiontime (days)

Ketose/aldose

1 100 45 A 4 77/222 100 315 B 4 60/403 100 45 A 8 65/354 100 315 B 8 40/60c

5 200 45 A 4 84/166 200 315 B 4 71/287 200 45 A 8 77/238 200 315 B 8 57/43

aunit is defined as the amount of enzyme that converts d-glucose to d-fructose at an initial rate of 1 mmol min�1 in standardanalytical conditions.bProcess A: addition of 45 units GlcI at once. Process B: portion-wise addition of 45 units GlcI every 24 h.cThe highest ratio in favour of aldose (the desired product).


7

Enzymatic Synthesis of Glycosidesand Glucuronides

7.1 Glycosynthase-assisted Oligosaccharide SynthesisAdrian Scaffidi and Robert V. Stick

Glycosidases are enzymes that effectively catalyse the cleavage of the glycosidic linkage

of some of the most structurally diverse substrates in nature. Removal of the catalytic

nucleophile of retaining glycosidases, through site-directed mutagenesis, results in an

enzyme unable to cleave glycosidic linkages but with the ability to form them.1 There

have been numerous glycosynthases produced over the years and they have become

versatile tools in preparing complex oligosaccharides that could otherwise not be easily

achieved using traditional methods.2 Recently, preparation of the biologically active

glycosylated derivatives of 2-deoxy-2-fluoro-�-laminaribiosyl fluoride, along with the

use of inositols as substrates, has been reported.3,4

7.1.1 Procedure 1: Glycosynthase-assisted Synthesis of Glycosylated Derivatives of

2-Deoxy-2-fluoro-b-laminaribiosyl Fluoride

O

OAcOO

OAcAcO

AcOF

F

OAc OAcO

AcOAcO

AcO

OAc

F

+

O

OAcOO

OAcAcOO

FF

OAc OAc

O

AcOAcO

O

OAcAc

n

n = 1,2,3

i. Na, MeOHii. Abg E358G NH4HCO3

iii. Ac2O, DMAPpyridine




• Tetra-O-acetyl-�-D-glucopyranosyl fluoride1 (130 mg, 0.37 mmol)

• 4,6-di-O-acetyl-2-deoxy-2-fluoro-3-O-(tetra-O-acetyl-�-D-glucopyranosyl)-�-D-

glucosyl fluoride3 (44 mg, 0.07 mmol)

• sodium

• Amberlite resin IR-120 (Hþ)

• 150 mM NH4HCO3 solution (25 mL)

• Agrobacterium sp. E358G1 (0.2 mg)

• acetic anhydride (2.0 mL, 21 mmol)

• pyridine (5 mL, 62 mmol)

• 4-N,N-dimethylaminopyridine (DMAP, 5 mg, 0.04 mmol)



• saturated NaHCO3 solution (25 mL)

• anhydrous MgSO4 (3 g)



• silica gel 60 (particle size 40–63 mm, Merck)


• 50 mL reaction flask

• 50 mL separatory funnel



7.1.1.2 Procedure A

1. A small piece of Na was added to MeOH (5 mL) and the resulting solution was added to

tetra-O-acetyl-�-D-glucopyranosyl fluoride (130 mg, 0.37 mmol) and 4,6-di-O-

acetyl-2-deoxy-2-fluoro-3-O-(tetra-O-acetyl-�-D-glucopyranosyl)-�-D-glucosyl fluor-

ide (44 mg, 0.07 mmol) in MeOH (5 mL) and the solution stirred (1 h) before being

neutralized with Amberlite resin IR-120 (Hþ), filtered and concentrated to afford a

colourless residue.

2. This residue was then dissolved in NH4HCO3 solution (150 mM, 25 mL), followed by

the addition of Abg E358G (0.2 mg), and the solution kept at 25 �C for 2 days. The

solvent was removed under reduced pressure and the residue treated with Ac2O (2 mL,

21 mmol) and pyridine (5 mL) containing DMAP (5 mg) and stirred at 50 �C for 3 h.

The reaction was quenched with MeOH (10 mL) and concentrated. The residue was

dissolved in saturated NaHCO3 solution (25 mL) and dichloromethane (25 mL). The

dichloromethane layer was separated, dried over anhydrous MgSO4, filtered and con-

centrated using a rotary evaporator.

3. Purification by flash chromatography (EtOAc/petrol, 2:3–3:2) furnished the trisacchar-

ide as a powder (35 mg, 54 %), m.p. 93–95 �C, [�]D¼�4.5�. 1H NMR (500 MHz) �1.99–2.14 (27H, 8s, CH3), 3.62 (ddd, J40,50 9.9, J50,60 4.7, 1.9, H50), 3.65 (dd, J400,500 9.9,

J500,600 4.4, 2.2, H500), 3.79 (dd, J30,40 � J40,50, H40), 3.81 (dd, J4,5 9.5, J5,6 4.9, 3.1, H5),

3.97 (ddd, J3,F2 15.3, J2,3� J3,4 9.0, H3), 4.05 (dd, J600,600 12.5, H600), 4.12 (dd, J60,60 12.1,

H60), 4.14–4.16, 4.18–4.21 (2H, 2m, H6), 4.30–4.45 (m, H2), 4.35 (dd, H600), 4.47

228 Enzymatic Synthesis of Glycosides and Glucuronides

(dd, H60), 4.52 (d, J100,200 7.9, H100), 4.65 (d, J10,20 7.9, H10), 4.87 (dd, J20,30 9.6, H20), 4.91

(d, J200,300 9.3, H200), 5.02 (dd, H4), 5.05 (dd, J300,400 9.8, H400), 5.13 (dd, H300), 5.17 (dd,

H30), 5.35 (ddd, J1,F1 51.8, J1,2 6.2, J1,F2 4.2, H1). 13C NMR (125.8 MHz) � 20.40–20.74

(CH3), 61.49, 61.79, 61.88 (C6, C60, C600), 66.91 (d, J4,F2 7.5, C4), 67.72–75.98

(C20–C50, C200–C500), 72.20 (dd, J5,F1 3.5, C5), 78.92 (dd, J3,F2 18.9, J3,F1 8.9, C3),

91.09 (dd, J2,F2 187.9, J2,F1 26.8, C2), 100.59, 101.23 (C10, C100), 105.90 (dd, J1,F1

219.3, J1,F2 27.3, C1), 168.96–170.52 (9C, C¼O). m/z (high resolution mass spectro-

metry fast atom bombardment (HR-MS FAB)) 887.2632; [MþH]þ requires 887.2633.

4. Next to elute was the tetrasaccharide as a powder (25 mg, 29 %), m.p. 108–110 �C,

[�]D ¼ �6.8�. Partial 1H NMR (500 MHz)[a] � 1.97–2.15 (36H, 11s, CH3), 3.57 (ddd,

J5,6 5.1, 1.9, H5), 3.61 (ddd, J5,6 5.6, 1.9, H5), 3.65 (ddd, J4,5 9.9, J5,6 5.1, 1.9, H5), 3.82

(ddd, J4,5 9.45, J5,6 3.1, 4.8, H5), 3.96 (ddd, J3,F2 15.2, J2,3� J3,4 8.4, H3), 4.04 (dd, J6,6

12.4, J5,6 2.2, H6), 4.14 (dd, J3,4� J4,5 7.2, H4), 4.36 (dd, J6,6 12.6, J5,6 4.5, H6), 4.64

(d, J1,2 8.0, H1), 4.84 (dd, J2,3 9.4, J1,2 7.9, H2), 4.86 (dd, J2,3 9.5, J1,2 8.0, H2), 4.90 (dd,

J2,3 9.2, J1,2 7.9, H2), 5.34 (ddd, J1,F1 51.9, J1,2 6.3, J1,F2 4.2, H1). 13C NMR (125.8

MHz) � 20.40–20.78 (CH3), 61.49, 61.80, 61.85, 62.05 (C6, C60, C600, C6000), 66.92

(d, J4,F2 7.7, C4), 67.72–76.07 (C20–C50, C200–C500, C2000–C5000), 72.19 (d, J5,F1 4.3, C5),

78.94 (dd, J3,F2 19.0, J3,F1 9.5, C3), 90.96 (dd, J2,F2 188.1, J2,F1 26.7, C2), 100.31,

100.79, 101.21 (C10, C100, C1000), 105.89 (dd, J1,F1 218.9, J1,F2 27.2, C1), 169.00–170.53

(C¼O). m/z (HR-MS FAB) 1175.3488; [MþH]þ requires 1175.3478.

[a] Some of the ring protons could not be unambiguously assigned.

7.1.1.3 Procedure B

5. Tetra-O-acetyl-�-D-glucopyranosyl fluoride (150 mg, 0.41 mmol) and 4,6-di-O-

acetyl-2-deoxy-2-fluoro-3-O-(tetra-O-acetyl-�-D-glucopyranosyl)-�-D-glucosyl fluor-

ide (50 mg, 0.08 mmol) were treated first with Na in MeOH and then with Abg E358G

as for Procedure A for 4 days to furnish the tetrasaccharide as a powder (50 mg, 70 %).

6. Next to be obtained was the pentasaccharide as a powder (20 mg, 17 %), m.p.

126–128 �C, [�]D¼�8.2�. Partial 1H NMR (600 MHz) � 3.95 (ddd, J3,F2 15.1,

J2,3� J3,4 8.6, H3), 5.32 (ddd, J1,F1 51.8, J1,2 6.2, J1,F2 4.2, H1). Partial 13C NMR

(150.9 MHz) � 61.47, 6185, 6198, 62.04 (C6–C60000), 66.91 (d, J4,F2 7.6, C4),

67.71–76.09 (C20–C50, C200–C500, C2000–C5000), 72.19 (J5,F1 4.2, C5), 78.92

(dd, J3,F2 19.0, J3,F1 9.2, C3), 91.04 (dd, J2,F2 187.9, J2,F1 26.9, C2), 100.22,

100.52, 100.77, 101.19 (C10–C10000), 105.81 (dd, J1,F1 218.9, J1,F2 27.2, C1). m/z

(HR-MS FAB) 1463.4319; [MþH]þ requires 1463.4323.

7.1.2 Procedure 2: Glycosynthase-assisted Synthesis of Glycosylated

scyllo-Inositols

O

HOHO

OH

F

OH

HOOH

HOOBn

OH+

OAc

AcOOAc

OBnOAc

O

AcOOAc

OAc

OO

HO

Ac

n

n = 1,2

i. Abg E358G NH4HCO3

ii. Ac2O, DMAPpyridine



• �-D-Glucopyranosyl fluoride1 (40 mg, 0.22 mmol)

• 1-O-Benzyl-scyllo-inositol4 (30 mg, 0.11 mmol)

• 150 mM NH4HCO3 solution (5 mL)

• Millipore water (15 mL)

• Agrobacterium sp. E358G1 (0.2 mg)

• acetic anhydride (2.0 mL, 21 mmol)

• pyridine (5 mL, 62 mmol)

• DMAP (2 mg, 0.02 mmol)



• saturated NaHCO3 solution (25 mL)




• silica gel 60 (particle size 40–63 mm, Merck)


• 50 mL reaction flask




7.1.2.2 Procedure

1. �-D-Glucopyranosyl fluoride (40 mg, 0.22 mmol) in aqueous NH4HCO3 solution

(150 mM, 5 mL) was added to 1-O-benzyl-scyllo-inositol (30 mg, 0.11 mmol) in

H2O (15 mL). The glycosynthase, Abg E358G (0.2 mg) was then added and the

solution kept at 25 �C for 7 days. The solvent was removed under reduced

pressure and the resulting residue was dissolved in Ac2O (2 mL) and pyridine

(5 mL) containing DMAP (2 mg) and the solution stirred at 50 �C for 3 h. The

reaction was quenched with MeOH (10 mL) and concentrated. The residue was

dissolved in saturated NaHCO3 solution (25 mL) and dichloromethane (25 mL).

The dichloromethane layer was separated, dried over anhydrous MgSO4, filtered

and concentrated using a rotary evaporator.

2. Purification by flash chromatography yielded the pseudo-disaccharide as a glass

(32 mg, 38 %), [�]D¼�17.5�. 1H NMR (600 MHz) � 1.93–2.09 (24H, 7s, CH3),

3.64–3.67 (m, H50), 3.68 (dd, J3,4� J4,5 9.5, H4), 3.88 (dd, J1,2� J1,6 9.5, H1), 4.03

(dd, J60,60 12.4, J50,60 2.1, H60), 4.36 (dd, J50,60 4.7, H60), 4.58 (d, J10,20 8.1, H10), 4.60

(s, CH2Ph), 4.86 (dd, J20,30 9.5, H20), 4.98–5.17 (m, H2, H3, H30, H40, H5, H6),

7.17–7.34 (Ph). 13C NMR (150.9 MHz) � 20.23–20.77 (8C, CH3), 61.63 (C60),67.96–77.71 (C1–C6, C20–C50), 75.21 (CH2Ph), 100.65 (C10), 127.66–137.41 (Ph),

169.15–170.45 (8C, C¼O). m/z (HR-MS FAB) 769.2551; [MþH]þ requires

769.2555.

3. Next to elute was the pseudo-trisaccharide as a glass (47 mg, 40 %), [�]D¼�13.8�.

Partial 1H NMR (600 MHz) � 1.90–2.11 (33H, 11s, CH3), 3.56 (ddd, J400,500 9.8, J500,600

5.5, 1.8, H500), 3.60 (ddd, J40,50 9.9, J50,60 4.2, 2.2, H50), 4.00 (dd, J60,60 12.4, H60), 4.16


(dd, J600,600 12.1, H600), 4.34 (dd, H60), 4.36 (dd, H600), 4.43 (d, J100,200 7.9, H100), 4.54 (dd,

J10,20 8.1, H10), 4.57 (s, CH2Ph), 4.77 (dd, J30,40 8.3, H40), 4.87 (dd, J300,400 8.1, H400),7.15–7.31 (Ph). 13C NMR (150.9 MHz) � 20.19–20.74 (CH3), 61.45, 62.06 (C60, C600),67.67–77.62 (C1–C6, C20–C50, C200–C500), 75.14 (CH2Ph), 100.56, 100.75 (C10, C100),127.61–137.37 (Ph), 169.22–170.40 (C¼O). m/z (HR-MS FAB) 1057.3419; [MþH]þ

requires 1057.3400.

7.1.3 Conclusion

The use of glycosynthases as biocatalysts for the formation of glycosidic linkages has

proven to be an effective and versatile tool for the preparation of complex oligosacchar-

ides. The procedure is simple and reproducible and can be applied to a number of

substrates involving carbohydrate and noncarbohydrate moieties.

References

1. Mackenzie, L.F., Wang, Q., Warren, R.A.J. and Withers, S.G., Glycosynthases: mutant glycosi-dases for oligosaccharide synthesis. J. Am. Chem. Soc., 1998, 120, 5583.

2. Perugino, G., Trincone, A., Rossi, M. and Moracci, M., Oligosaccharide synthesis byglycosynthases. Trends Biotechnol., 2004, 22, 31.

3. Scaffidi, A., Stubbs, K.A. and Stick, R.V., Synthesis of some glycosylated derivatives of2-deoxy-2-fluoro-�-laminaribiosyl fluoride: another success for glycosynthases. Aust. J.Chem., 2007, 60, 83.

4. Scaffidi, A. and Stick, R.V., Glycosynthase-assisted synthesis of some glycosylated scyllo-inositols. Aust. J. Chem., 2006, 59, 894.


7.2 Glycosyl Azides: Novel Substrates for Enzymatic TransglycosylationsVladimır Kren and Pavla Bojarova

Enzymatic transglycosylation catalysed by glycosidases is a respected method in

carbohydrate synthesis. The spectrum of acceptors is practically infinite, contrary to

glycosyl donors, usually nitrophenyl glycosides. We have developed glycosyl azides1,2

as novel, efficient and easily prepared glycosyl donors for glycosidases. Their high

water solubility facilitates the synthesis of disaccharides by transglycosylations with

comparable or better yields than using traditional O-glycosides. Using the azide moiety,

such products can simply be conjugated to complex structures after reduction of the

azide to an amine.

7.2.1 Procedure 1: Synthesis of 1-azido-disaccharides

HO

AcHNO

N3

NHAcHO

O OH

OHOH

OOH

N3NHAc

HOHO

OOH

N3

NHAcHO

OOH

NHAcHO

HOO

pH 5.0, 35 °C

β-N-Acetylhexosaminidase(T.flavus CCF 2686)

+

1 2 3yield 16%yield 32%

O

7.2.1.1 Cultivation Procedure for �-N-Acetylhexosaminidase from Talaromyces flavus

CCF 2686

The fungal strain producing �-N-acetylhexosaminidase (EC 3.2.1.52) originated from

the Culture Collection of Fungi (CCF), Department of Botany, Charles University in

Prague. The strains were cultivated as described previously.3 Flasks (500 mL) with

peptone medium (100 mL) were inoculated with the suspension of spores in 0.1 %

Tween 80 and cultivated on a rotary shaker at 200 rpm and 28 �C. Peptone medium

contained yeast extract (0.5 g L�1), mycological peptone (5 g L�1), KH2PO4 (3 g

L�1), NH4H2PO4 (5 g L�1) and crude chitin hydrolysate (2 g L�1), pH 6.0. After

sterilization, each flask was supplemented with MgSO4�7H2O to the final concentra-

tion of 0.5 g L�1. The cultivation time was 10–12 days to reach the maximum �-N-

acetylhexosaminidase activity in the medium. Enzymes were isolated from the

cultivation medium by fractional precipitation by (NH4)2SO4 (80 % sat.). The

precipitate was stable for several years at 4 �C. One unit of enzyme activity is the

amount of enzyme that releases 1 mmol of p-nitrophenol per minute in a standard

assay1 with p-nitrophenyl 2-acetamido-2-deoxy-�-D-glucopyranoside (2 mM) in

McIlvaine buffer pH 5.0 at 37 �C.


• Substrate 1 (180 mg, 731 mmol)4

• �-N-acetylhexosaminidase from Talaromyces flavus CCF 2686 (12 U)

• McIlvaine buffer pH 5.0 (1218 mL) prepared by mixing 0.1 M citric acid (24.3 mL)

and 0.2 M Na2HPO4 (25.7 mL), by diluting with water to 100 mL and adjusting to

pH 5.0


• propan-1-ol:H2O:NH3 aq., 7:2:1

• thin-layer chromatography (TLC) silica gel plates (Merck F254, DE)

• thermomixer for Eppendorf tubes

• centrifuge for Eppendorf tubes

• rotary vacuum evaporator

• column chromatography equipment, column 3 cm � 100 cm

• Bio Gel P2 (BioRad, USA)

• high-performance liquid chromatography (HPLC) instrument with a UV–vis detector

• Lichrospher 100-5 NH2 column 250 mm � 8 mm (Watrex, CZ).

7.2.1.3 Procedure

1. Substrate 1 (180 mg, 731 mmol)4 was dissolved in McIlvaine buffer pH 5.0 (1.2 mL).

The �-N-acetylhexosaminidase from T. flavus CCF 2686 (12 U) was added and the

reaction mixture was shaken at 35 �C and its course was followed by TLC (propan-1-

ol:H2O:NH3 aq., 7:2:1).

2. After 7.5 h, the reaction was stopped by heating (100 �C, 2 min). After centri-

fugation (13 500 rpm, 10 min) and concentration in vacuo, the reaction mixture

was purified on a Bio Gel P2 (BioRad, USA) column (water, flow rate

15.4 mL h�1) to afford recovered starting material 1 (71 mg) and a mixture of

products 2 and 3.

3. The product mixture was further purified by preparative HPLC (Lichrospher

100-5 NH2 column 250 mm � 8 mm (Watrex, CZ), mobile phase 75:25

MeCN:H2O, flow rate 2.5 mL min�1, ambient temperature) to afford compound

2 (11 mg, 24.5 mmol; 32 % based on consumed donor); ½��20D ¼ �54:4 (c¼ 0.23 in

water); mass spectrometry (MS) (matrix-assisted laser desorption/ionization–time

of flight (MALDI-TOF)): m/z (%): 472.0 (92) [MþþNa] 488.0 (31) [MþþK];

and compound 3 (5.6 mg, 12.5 mmol; 16 % based on consumed donor); ½��20D ¼

�131:2 (c¼ 0.17 in water); MS (MALDI-TOF): m/z (%): 472.0 (69)

[MþþNa] 488.0 (22) [MþþK]) as white solids.

4. NMR data are shown in Table 7.1. Both �-D-GlcpNAc units (1H NMR, correlation

spectroscopy) were distinguished using chemical shifts of C-1. The heteronuclear

Table 7.1 1H and 13C NMR (399.87 and 100.55 MHz, D2O, 30 �C) data for compounds2 and 3.a

Compound 2b Compound 3b

�-D-GlcpNAc �-D-GlcpNAcN3 �-D-GlcpNAc �-D-GlcpNAcN3

Proton �H (ppm)

1 4.37 4.53 4.34 4.522 3.52 3.52c 3.51 3.463 3.35 3.52c 3.34 3.474 3.24 3.52c 3.24c 3.215 3.28 3.37 3.24c 3.42

(continued overleaf )

7.2 Glycosyl Azides: Novel Substrates for Enzymatic Transglycosylations 233

coupling of H-10 to C-4 determined the �(1!4) linkage in 2, those of H-10 to C-6 and of

H-6 to C-10 together with downfield resonating C-6 (68.9 ppm) led to the �(1!6)

structure of 3.

7.2.2 Conclusion

Glycosyl azide 1 proved to be an easily prepared and efficient glycosyl donor in �-N-

acetylhexosaminidase-catalysed transglycosylation, as demonstrated by the preparation of

two novel disaccharides 2 and 3. The present examples are autocondensation products;

however, azide glycosyl donors are applicable in the glycosylation of many other accep-

tors carrying suitable hydroxyl groups, analogously to nitrophenyl glycosides. This con-

cept of glycosyl azide donors can be used with other glycosidases, such as

�-galactosidases, �-glucosidases and �-mannosidases, as shown by Bojarova et al.2

Glycosyl azides represent highly prospective glycosyl donors that are a valuable alter-

native to traditional nitrophenyl glycosides.

Table 7.1 (continued )

Compound 2b Compound 3b

�-D-GlcpNAc �-D-GlcpNAcN3 �-D-GlcpNAc �-D-GlcpNAcN3

6d 3.70 3.65 3.71 3.986u 3.52 3.45 3.54 3.57

Coupling J(i,j) (Hz)

1,2 8.5 9.2 8.5 9.12,3 10.3 n.d. 10.3 10.33,4 8.6 n.d. 8.8 8.94,5 n.d. 9.6 n.d. 9.95,6d 2.0 2.0 1.7 2.05,6u 5.2 5.0 5.6 5.76d,6u 12.4 12.2 12.3 11.7

Carbon �C (ppm)d

1 101.8 88.8 102.0 89.02 55.8 54.7 55.8 55.43 73.8 72.7 74.0 74.04 70.0 79.3 70.2 69.95 76.3 76.8 76.2 77.06 60.8 60.3 61.0 68.9

an.d.: not determined; u: upfield; d: downfield.bAdditional signals. 2: �¼1.8 s, 1.9 s (2 � Ac), 22.3, 22.4 (2 � Ac), 174.9, 175.0 ppm (2 � C¼O); 3: �¼1.8 s, 1.8 s(2 � Ac), 22.4 (2 � Ac), 174.9, 175.1 ppm (2 � C¼O); 7: �¼1.8 s. 1.8 s (2 � Ac), 22.2 ppm (2 � Ac).cA strongly coupled spin system.d13C NMR data are heteronuclear multiple quantum coherence and heteronuclear multiple bond coherence readouts; thecarbon atoms at sites of glycosylation are given in bold.


References

1. Fialova, P., Carmona, A.T., Robina, I., Ettrich, R., Sedmera, P., Prikrylova, V., Husakova, L. andKren, V., Glycosyl azide – a novel substrate for enzymatic transglycosylations. TetrahedronLett., 2005, 46, 8715.

2. Bojarova, P., Petraskova, L., Ferrandi, E., Monti, D., Pelantova, H., Kuzma, M., Simerska, P. andKren, V., Glycosyl azides – an alternative way to disaccharides. Adv. Synth. Catal., 2007, 349,1514.

3. (a) Hunkova, Z., Kren, V., Scigelova, M., Weignerova, L., Scheel, O. and Thiem, J., Induction of�-N-acetylhexosaminidase in Aspergillus oryzae. Biotechnol. Lett., 1996, 18, 725;(b) Hunkova, Z., Kubatova, A., Weignerova, L. and Kren, V., Induction of extracellular glyco-sidases in filamentous fungi and their potential use in chemotaxonomy. Czech Mycol. 1999,51, 71; (c) Weignerova, L., Vavruskova, P., Pisvejcova, A. Thiem, J. and Kren, V., Fungal �-N-acetylhexosaminidases with high �-N-acetylgalactosaminidase activity and their use for synth-esis of �-GalNAc-containing oligosaccharides. Carbohydr. Res., 2003, 338, 1003.

4. Shulman, M.L., Lakhtina, O.E. and Khorlin, A.Y., Specific irreversible inhibition of human andboar N-acetyl-�-D-hexosaminidase by 2-acetamido-2-deoxy-�-D-glucopyranosyl isothiocyanate.Biochem. Biophys. Res. Commun., 1977, 74, 455.

7.2 Glycosyl Azides: Novel Substrates for Enzymatic Transglycosylations 235

7.3 Facile Synthesis of Alkyl b-D-Glucopyranosides from D-Glucose and theCorresponding Alcohols Using Fruit Seed MealsWen-Ya Lu, Guo-Qiang Lin, Hui-Lei Yu, Ai-Ming Tong and Jian-He Xu

The suitability of �-glucosidase for synthesis of alkyl �-glucosides from glucose and the

corresponding alcohols in one step has made this enzyme attractive for this synthetic

application. Various alkyl �-D-glucopyranosides were synthesized via a very simple

procedure, by using Prunus dulcis (almond) kernel meal as an inexpensive biocatalyst

(Table 7.2). P. dulcis (almond) meal is a robust and recyclable catalyst (Table 7.3). Some

other popular fruits seed, including Prunus persica (peach), Prunus armeniaca (apricot),

Malus pumila (apple) and Eriobotrya japonica (loquat), were tested as potential sources of

the glucosidase in the form of meal (Tables 7.4 and 7.5). It was found that P. persica kernel

meal and M. pumila seed meal not only had a higher activity, but also showed some

complementary substrate specificities to that of almond �-glucosidase.1

Table 7.2 Synthesis of alkyl �-glucosides using P. dulcis kernel meal

Entry Substrate [glu]0 (M) Time (h) Yield 1a (%) Yield 2b (%)

1 1a 0.50 72 45 422 1b 0.50 72 52 433 1c 0.28 72 39 384c 1d 0.30 72 13 125c 1e 0.28 60 12 116c 1f 0.30 72 0 07 1g 0.22 48 38 398 1h 0.30 72 10 119 1i 0.30 48 49 55

10 1j 0.30 72 45 4711d 1k 0.30 72 25 2712 1l 0.22 72 51 4013e 1m 0.30 48 12 014e 1n 0.30 48 13 1415f 1o 0.30 48 14 19

aIsolation yield, using 60 mg (1.9 U)2 of home-made acetone powder of P. dulcis kernel meal per millilitreof reaction mixture.bIsolation yield, using 5 mg (29.5 U) of commercially supplied glucosidase preparation from almond(Sigma, G-0395) per millilitre of reaction mixture.c50 % (v/v) CH3CN was added as cosolvent.dThe product was slightly in favour of the (R)-isomer (R/S ¼3/2, according to 1H NMR spectrum).eDioxane was added as cosolvent.ftert-Butyl alcohol was added as cosolvent.

OHOH

HO1n 1o

OH OH

OH OH OH OHOH OH

OH OH

( )3 ( )5

( )2 ClOH

OH( )9

1a 1b 1c 1d 1e 1f 1g 1h

1i 1j 1k 1l 1m

OH


Table 7.3 Repeated use of P. dulcis kernel meal in the enzymatic synthesis ofalkyl �-D-glucoside

Run Substrate [glucose]0 (M) Yield (%)a

1 1g 0.22 372 1g 0.22 343 1g 0.22 304 1g 0.22 285 1g 0.22 236 1g 0.22 211 1i 0.30 492 1i 0.30 363 1i 0.30 324 1i 0.30 28

aIsolation yield. The reaction was conducted under 50 �C for 48 h. After the reaction mixture was filtered,the retrieved P. dulcis kernel meal was washed with corresponding alcohol and reused immediately.

Table 7.4 The �-glucosidase activity and protein content of several fruit seed meals

Enzyme source (100 mg) Total activity (U) Total solubleprotein (mg)a

Specific activity(U mg�1)

P. dulcis kernel 3.13 21.54 0.15M. pumila seed 3.36 3.36 1.00P. persica kernel 2.15 11.06 0.19P. armeniaca kernel 1.72 3.63 0.47E. japonica kernel 0.10 0.30 0.33

a Protein was determined according to Bradford (1976)3 using ovoalbumin as a standard.

Table 7.5 Synthesis of alkyl �-glucosides using fruit seed meal

Entry Substrate [glu]0 (M) Time (h) Yielda (%)

M. pumila P. persica P. armeniaca E. japonica

1 1a 0.50 72 72 64 36 352 1b 0.50 72 60 52 52 323 1c 0.28 72 44 46 39 224b 1d 0.30 72 15 12 13 65b 1e 0.30 72 12 12 12 06b 1f 0.30 72 6 3 0 07 1g 0.22 48 29 29 29 218 1h 0.30 72 7 10 10 n.d.9 1i 0.30 48 33 67 45 2210 1l 0.22 72 25 47 47 2311c 1o 0.30 48 29 18 12 /

a Isolation yield. Conditions were not fully optimized. All glucosides obtained are in � form and are anomerically pure. Thereactions were conducted at 50 �C, each with 60 mg fruit seed meal per millilitre of reaction mixture.b 50 % (v/v) CH3CN was added as cosolvent.c tert-Butyl alcohol was added as cosolvent.

7.3 Facile Synthesis of Alkyl �-D-Glucopyranosides 237

7.3.1 Procedure 1: The Preparation of Fruit Seed Meal


• Almond (200 g)

• apricot kernel (200 g)

• peach kernel (200 g)

• apple seed (100 g)



• acetone (500 mL)

• homogenizer

• one 100 mL Buchi funnel

• filter paper.

7.3.1.2 Procedure

1. The P. dulcis (almond) kernels were soaked in distilled water for 2 h, peeled, air-dried

and then powdered in cold (0 �C) ethyl acetate with a homogenizer. The powder was

defatted by a further three washes with ethyl acetate and two washes with acetone and

then stored at 4 �C.

2. The fresh fruit kernels or seeds were peeled and treated in the same manner.

7.3.2 Procedure 2: General Procedure for Fruit Seed Meal-catalyzed

Glucosylation

R OHOHO

HOOH

OH+

1

OHO

HOOH

OR

2

reverse hydrolysis+H2O

OH OHfruit seed meal


• Glucose

• almond meal (200 g) or apricot kernel (200 g) or peach kernel (200 g) or

• apple seed (100 g)


• dioxane

• tert-butyl alcohol

• CH3CN



• CH3OH (100mL)

• thin-layer chromatography plates (silica gel 60 F254)

• silica gel (300-400 mesh)

• acetone (500mL)



• one 50 mL Buchi funnel


7.3.2.2 Procedure

1. Glucose was dissolved in the corresponding alcohols (see Tables 7.2–7.5) containing

10 % v/v of water and the fruit seed meal was then added. The mixture was stirred for

48–72 h at 50 �C and then filtered and concentrated under vacuum.

2. The resultant syrup was applied to flash column chromatography (eluent 15–10/1

EtOAc/MeOH or 5/1 CH2Cl2/MeOH). The corresponding �-D-glucopyranosides were

collected as white solids or clear syrups.

7.3.3 Conclusion

We have developed some novel, cheap, and green biocatalysts for the facile synthesis of

various alkyl �-D-glucopyranosides. Those easily available biocatalysts enable large-scale

preparation of physiologically active �-D-glucopyranosides (H.L. Yu, J.H. Xu, W.Y. Lu

and G.Q. Lin, unpublished results).


1. Lu, W.Y., Lin, G.Q., Yu, H.L., Tong, A.M. and Xu, J.H., Facile synthesis of alkyl �-D-glucopyr-anosides from D-glucose and the corresponding alcohols using fruit seed Meals. J Mol Catal B:Enzym., 2007, 44, 72–77.

2. The �-glucosidase activity was determined by measuring the release of p-nitrophenol fromp-nitrophenyl-�-D-glucopyranoside; one unit of �-glucosidase activity (U) is defined as theamount of enzyme that releases 1 [mu]mol p-nitrophenol per minute. All samples were assayedin potassium phosphate buffer (50 mM, pH 7.0) at 50 �C under conditions that activity wasproportional to enzyme concentration.

3. Bradford, M.M., A rapid and sensitive method for the quantitation of microgram quantities ofprotein utilizing the principle of protein–dye binding. Anal. Biochem., 1976, 72, 248–254.

7.3 Facile Synthesis of Alkyl �-D-Glucopyranosides 239

7.4 Laccase-mediated Oxidation of Natural GlycosidesCosimo Chirivı, Francesca Sagui and Sergio Riva

Laccases are oxidoreductases that oxidize a wide range of organic compounds using

molecular oxygen as the oxidant.1 Typical substrates of these enzymes are amines and

phenols, which undergo chemical coupling to give dimeric and oligomeric derivatives.

Laccase oxidation of primary alcohols is also possible, but it is necessary to make use of

ancillary chemical ‘mediators’, i.e. 2,2,6,6-tetramethyl-1-piperidinyloxy (TEMPO). These

compounds (used in catalytic amounts) are initially oxidized by the laccase and then the

mediator-catalyzed oxidation of primary alcohols takes place. With sugar substrates the

intermediate aldehydes are not stable under these reaction conditions and undergo sub-

sequent further oxidation to give the corresponding glucuronides (Figure 7.1).

Recently, the regioselective oxidations of the primary OH groups of natural glycosides

have been performed on a preparative scale by exploiting this methodology.2 Here, we

report on the experimental protocols related to the laccase-catalyzed oxidation of thiocol-

chicoside (1) and amygdalin (2), using either the native or the immobilized enzyme.

7.4.1 Procedure 1: Laccase-mediated Oxidation of Thiocolchicoside (1)

OOHO

OHOH

OH

NHAc

O

SMe

MeO

OMe

OOHO

OHOH NHAc

O

SMe

MeO

OMe

OHC

OOHO

OHOH NHAc

O

SMe

MeO

OMe

HOOC

TEMPO O2

Laccase

1

1a

1b

OROHO

OHOH

OH

OROH O

OHOH

OHC

OROH O

OHOH

HOOCTEMPO TEMPOox

Laccase

O2H2O

Figure 7.1 Laccase-mediated oxidation of glycosides.



• Thiocolchicoside (1, 200 mg, 0.354 mmol)

• TEMPO (11 mg, 0.07 mmol)

• 50 mM acetate buffer, pH 4.5 (13 mL)

• Laccase from Trametes versicolor purchased from Fluka (30 mg, 440 U)

• 2,20-azino-bis-(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS, 5.8 mg, 0.01 mmol)


• MeOH (300 mL)

• trifluoroacetic acid (TFA, 250 mL)

• acetonitrile, high-performance liquid chromatography (HPLC) grade (300 mL)

• deionized water (1 L)

• (NH4)6Mo7O24�4H2O (42 g)

• Ce(SO4)2 (2 g)

• H2SO4 concentrated (62 mL)

• 50 mL vial

• shaker


• Merck 60 F254 thin-layer chromatography (TLC) plates


• silica gel (Merck 60, 230–400 mesh)

• HPLC instrument equipped with a UV–vis detector

• column chromatography equipment.

7.4.1.2 Procedure

1. Thiocolchicoside (1 200 mg, 0.354 mmol) was dissolved in a 50 mM, pH 4.5 acetate

buffer solution (13 mL). Then TEMPO (11 mg, 0.07 mmol) and laccase (30 mg, 440 U)

were added.

2. The reaction was gently shaken at 30 �C for 36 h, monitoring the conversion by TLC

then HPLC upon completion.

3. After 36 h (85 % conversion according to HPLC analysis) the solvent was evaporated in

vacuo and the residue purified by flash chromatography (eluent AcOEt/MeOH/H2O,

8:3:0.5 and then 7:3:1) to give 143 mg (0.247 mmol, 70 % yield) of the glucuronide 1b

as a yellow solid (Rf¼ 0.23).

1H NMR (D2O) � 7.37 (2H, s, H-11 and H-12); 7.20 (1H, s, H-8); 6.97 (1H, s,

H-4); 5.03 (1H, d, J¼ 7.24 Hz, H-1); 4.49 (1H, dd, J1¼ 6.25 Hz, J2¼ 11.9 Hz, H-7);

3.97 (3H, s, CH3O in C-2); 3.80 (1H, d, J¼ 9.06 Hz, H-5); 3.63 (3H, s, CH3O in

C-1); 3.50–3.70 (3H, m, H-2, H-3 and H-4); 2.67 (1H, dd, J1¼ 5.65 Hz, J2¼ 12.8

Hz, H-5a); 2.49 (3H, s, CH3S); 2.25 (2H, m, H-6a and H-5b); 2.01 (3H, s, CH3CO);

1.80 (1H, m, H-6b).

Eluent for TLC was 8:4:1 AcOEt/MeOH/H2O. Substrates and products were visualized

at 254 nm and with the molybdic reagent ((NH4)6Mo7O24�4H2O, 42 g; Ce(SO4)2, 2 g;

H2SO4 concentrated, 62 mL; made up to 1 L with deionized water).

HPLC analyses were performed using a Jasco 880-PU pump equipped with a Jasco 870-

UV detector and a LiChrospher 100 RP-18 (5 mm) in LiChroCART 125-4 column along


with a Alusper 100-RP select B (5 mm) guard column. Substrates and products were

analyzed at 254 nm using as eluent H2O (0.05 % v/v TFA)–acetonitrile, 87:13.

Residual enzymatic activity was monitored during reaction progression using the

following process. A sample (50 mL) of the reaction solution was used to evaluate the

initial enzymatic activity; the enzymatic solution (10 mL) was added to a 1 mL cuvette

containing pH 4.5 acetate buffer solution (890 mL) and ABTS (100 mL of a 10 mM solution

in H2O). Laccase activity was evaluated by monitoring the oxidation of ABTS at 436 nm

(EABTS¼ 29.3 mM�1 cm�1). It was found that 440 enzymatic units were initially present, a

unit being the amount of laccase that oxidizes 1 mmol min�1 of ABTS under these

conditions. There were 83 residual units present after 21 h.

7.4.2 Procedure 2: Laccase-mediated Oxidation of Amygdalin (2)

OOH O

OHOH

O

CN

OH O

OHOH

OH

OOH O

OHOH

O

CN

OH O

OHOH

HOOC

2

2b

TEMPO, O2laccase


• Amygdalin (200 mg, 0.437 mmol)

• TEMPO (13.6 mg, 0.087 mmol)


• Laccase from T. versicolor purchased from Fluka (30 mg, 440 U)

• ABTS (5.8 mg, 0.01 mmol)


• MeOH (400 mL)

• (NH4)6Mo7O24 x 4H2O (42 g)

• Ce(SO4)2 (2 g)


• vial (50 mL)

• shaker


• Merck 60 F254 TLC plates


• silica gel (Merck 60, 230–400 mesh)



7.4.2.2 Procedure

1. Amygdalin (2, 200 mg, 0.437 mmol) was dissolved in a 50 mM, pH 4.5 acetate buffer

(13 mL). Then TEMPO (13.6 mg, 0.087 mmol) and laccase (30 mg, 440 U) were added.

2. The reaction was gently shaken at 30 �C for 36 h, monitoring the conversion by TLC

(AcOEt/MeOH/H2O, 8:4:1, molybdic reagent) and checking the residual enzymatic

activity in time using the method shown in Procedure 1 (Section 7.4.1.2) (22 U after 26 h).

3. After 36 h the solvent was evaporated in vacuo and the residue purified by flash

chromatography (eluent AcOEt/MeOH/H2O, 8:4:1 and then 7:3:1) to give 93 mg

(0.247 mmol, 45 % yield) of the glucuronide 2b as a pale yellow solid (Rf¼ 0.10).

1H NMR (CD3OD) � 7.64 (2H, m, ArH); 7.45 (3H, ArH); 5.92 (1H, s, CHCN); 4.59 (1H,

d, J¼ 7.8 Hz, H-1); 4.37 (1H, d, J¼ 7.5 Hz, H-1); 4.26 (1H, dd, J1¼ 12.0 Hz, J2¼ 2.0 Hz,

H-6b); 3.87 (1H, dd, J1¼ 12.0 Hz, J2¼ 6.8 Hz, H-6a); 3.67 (1H, d, J¼ 9.3 Hz, H-5); 3.3

(3H, m); 3.5 (4H, m).

7.4.3 Procedure 3: Immobilization of the Laccase from T. versicolor


• Laccase from T. versicolor purchased from Fluka (70 mg)

• 0.1 M phosphate buffer, pH 7.0 (0.6 mL)

• 1.17 M phosphate buffer, pH 7.0 (3.1 mL)

• Eupergit C250L (750 mg)

• ethanolamine (72 ml)

• orthophosphoric acid (200 ml)


• ABTS (5.8 mg, 0.01 mmol)

• doubly distilled water

• centrifuge flasks

• centrifuge

• UV–vis spectrophotometer.

7.4.3.2 Procedure

1. The laccase (70 mg) was dissolved in 0.1 M, pH 7.0 phosphate buffer solution (0.6 mL).

A 1.17 M, pH 7.0 phosphate buffer solution (3.1 mL) was then added to give a final 1.0 M

phosphate buffer solution.

2. 50 mL of the enzymatic solution were used to evaluate the initial total enzymatic

activity (5775 U). The remaining enzymatic solution was added dropwise to C250L

Eupergit (750 mg) in a centrifuge flask and the resulting slurry was stored at 4 �C for

16 h, being mixed at regular intervals.

3. The slurry was centrifuged (3000 rpm, 5 min) and washed three times with 0.1 M, pH 7.0

phosphate buffer solution (5 mL). The residual enzymatic activity was measured in

order to evaluate the amount of unbound laccase (725 U resulted unbound; therefore

5050 U were presumably immobilized).

4. The immobilized enzyme was resuspended in 0.3 M, pH 7.0 solution of ethanolamine in

1.2 M phosphate buffer (5 mL), and stored at 4 �C for 5 h, being mixed at regular intervals.


5. Finally, the suspension was centrifuged (3000 rpm, 5 min), washed three times with

0.1 M, pH 7.0 phosphate buffer solution (5 mL), and stored at 4 �C suspended in 50 mM,

pH 4.0 acetate buffer solution (5.8 mL), (863 U mL�1).

7.4.4 Procedure 4: Oxidation of Thiocolchicoside with the Immobilized Laccase


• Thiocolchicoside (200 mg, 0.354 mmol)

• TEMPO (11 mg, 0.07 mmol)


• immobilized laccase (500 mL of the stored solution)

• TFA (250 mL)

• acetonitrile (300 mL)

• doubly distilled water (1 L)

• (NH4)6Mo7O24�4H2O (42 g)

• Ce(SO4)2 (2 g)


• 50 mL vial

• shaker

• Merck 60 F254 TLC plates


• HPLC instrument equipped with an UV–vis detector.

7.4.4.2 Procedure

1. Thiocolchicoside (1, 200 mg, 0.354 mmol) was dissolved in 50 mM, pH 4.5 acetate

buffer (13 mL). Then TEMPO (11 mg, 0.07 mmol) and 500 mL of the immobilized

laccase stock suspension (435 U) were added.

2. The reaction was shaken at 30 �C for 36 h, monitoring the conversion by TLC (AcOEt/

MeOH/H2O, 8:4:1, molybdic reagent) then HPLC upon completion.

3. After 48 h, the conversion, evaluated by HPLC, was 85 %.

4. The enzyme was filtered, the solvent evaporated in vacuo and the residue purified by

flash chromatography (eluent AcOEt/MeOH/H2O, 8:3:0.5 and then 7:3:1) to give

135 mg (0.232 mmol, 66 % yield) of the glucuronide 1b as a yellow solid.

7.4.5 Conclusion

The reported procedure for the selective oxidation of natural glycosides is mild, conve-

nient and easily reproducible. The biotransformations are performed in mildly acidic water

solutions; therefore, this method is complementary to other chemical approaches for the in

situ regeneration of the oxidized form of TEMPO, such as sodium hypochlorite, that

require alkaline pH.

References

1. Riva, S., Laccases: blue enzymes for green chemistry. Trends Biotechnol., 2006, 24, 219–226.2. Baratto, L., Candido, A., Marzorati, M., Sagui, M. and Riva, S., Laccase-mediated oxidation of

natural glycosides. J. Mol. Catal. B Enzym., 2006, 39, 3–8.


7.5 Biocatalysed Synthesis of Monoglucuronides of Hydroxytyrosol,Tyrosol, Homovanillic Alcohol and 3-(40-Hydroxyphenyl)propanolUsing Liver Cell Microsomal FractionsOlha Khymenets, Pere Clapes, Teodor Parella, Marıa-Isabel Covas, Rafael de la

Torre, and Jesus Joglar

Liver cell microsomal fractions are a widely used source of natural metabolic

enzymes with a broad spectrum of regio- and stereo-selective activities due to

their combination of tissue- and species-specific isoforms. This unexpensive and

useful source of diverse uridine 50-diphosphoglucuronyl transferase (UDPGT) activ-

ities was successfully used for the synthesis of glucuronic acid conjugated metabo-

lites of phenolic compounds such as hydroxytyrosol (HOTYR), tyrosol (TYR),

homovanillic alcohol (HVAlc) and hydroxyphenylpropanol (HOPhPr).1 The micro-

somal incubation of substrates, in the presence of uridine 50-diphosphoglucuronic

acid (UDPGA), yielded exclusively the mono �-D-conjugated glucuronides with in

vivo equivalent metabolic regioselectivity.1,2 The application of high-performance

liquid chromatography (HPLC) methodology for reaction monitoring, isolation,

separation, purification and quantification of the O-glucuronoconjugate regioisomers

synthesized provided pure products ready for experimental and analytical application

(Figure 7.2).

O

O

OH

OH

OHOH

O

OHO2C OP

OP

O O

HO OH

N

HN

O O

OH OH

HO OH

OH

O

O

OH

OH

OH

+

HO

HO

O

O

OH

OH

OHOH

O

HO

HO

UDPGT from liver cell microsomes

hydroxytyrosol

UDPGA

HOTYR-3´-O-β-glucuronideHOTYR-4´-O-β-glucuronide

Figure 7.2 Synthesis of HOTYR glucuronides using liver cell microsomal fractions

7.5 Biocatalysed Synthesis of Monoglucuronides 245

7.5.1 Procedure 1: Liver Microsomes Preparation3,4


• KCl solution (1.15 %, V¼ 5 v/w of tissue)

• snap-frozen fresh liver tissue (25 g)

• ice bath

• blade homogenizer

• Potter–Elvehjem tissue homogenizer or stirrer with electrically driven glass–teflon

pestle, adapted to work immersed in ice

• centrifuge generating 10 000� g at 4 �C, with fixed-angle 10 mL tube rotor

• 10 mL polyethylene tubes

• Pasteur pipettes

• ultracentrifuge generating 100 000� g with corresponding rotor and tubes

• 1.5 mL microtubes (Eppendorf) and cryobox

• liquid nitrogen

• �80 �C freezer.

7.5.1.2 Procedure

1. About 25 g of frozen liver tissue was thawed at 4–8 �C, cut with a blade mixer while

some ice-cold 1.15 % KCl (up to 3 v/w of tissue) was added. The tissue disruption was

continued until a uniform, consistent ‘liver juice’ was obtained. The homogenization

was carried out by making 10 strokes at 800–850 rpm for 15 s each using a fixed glass

tube immersed in an ice-dish and an electrically driven Teflon pestle (Potter–Elvehjem

tissue homogenizer). The homogenized suspension was transferred to precooled 10 mL

polyethylene tubes and centrifuged at 10 000� g for 20 min at 4 �C. The supernatants

were transferred into precooled ultracentrifuge tubes and ultracentrifuged at

100 000� g for 45 min at 0 �C. The supernatant was discarded.

Note: The supernatant, the raw cytosolic fraction, could be saved for another experi-

ment as a source of liver cytosolic enzymes.

2. Each pellet (‘dirty microsomal fraction’) was resuspended in 2 mL of ice-cold 1.15 %

KCl, collected together and ultracentrifuged again at 100 000� g for 45 min at 0 �C.

The supernatant was discarded and the pellet was rinsed thrice with ice-cold dilution

buffer. The rinsed pellet was resuspended in ice-cold 1.15 % KCl to yield a final

concentration 20 mg mL�1 of microsomal proteins. The microsomal suspension was

immediately aliquoted, frozen and stored at �80 �C.

Note: Microsomes should be thawed immediately before use by rapid rotation in

hands. Once defrosted, they should be used immediately and cannot undergo even short

bench storage or repeated freeze–thawing.

7.5.2 Procedure 2: Biotransformation via Microsomal Glucuronidation.

Product Identification and Reaction Monitoring


• Freshly prepared substrate solution (100 mM in 20 % dimethylsulfoxide (DMSO))

• tris-HCl buffer (1 M, pH 8.0)


• CaCl2 solution (60 mM)

• freshly prepared UDPGA (100 mM)

• dithiothreitol (DTT) (200 mM)

• freshly prepared bovine serum albumin (BSA) (30 % (w/v))

• UDPGT (microsomal fraction with 20 mg protein/mL)

• 1 and 5 mL microcentrifuge tubes

• 33% acetic acid in methanol (pH 3.3)

• ice-cold 100% methanol

• distilled water

• 5–10% of acetonitrile (MeCN) in 5 mM ammonium acetate (pH 5.0)

• reciprocal shaker

• microcentrifuge, generating 13 000 rpm

• Pasteur pipettes


• N2 gas

• equipment for column chromatography coupled to an ultraviolet (UV; 215 nm) and/or

mass spectrometry (MS) detector

• Atlantis� C18, 5 mm, 4.6 mm � 150 mm, pre-column 5 mm guard 2.1 mm � 10 mm.

7.5.2.2 Procedure

1. Tris-HCl (120 mL), CaCl2 (60 mL), BSA (60 mL), DTT (3 mL), water (87 mL) and freshly

thawed microsomal fraction (30 mL) were mixed well and divided into three reaction

tubes: one reaction and two control tubes (1 and 2). 40mL of freshly prepared UDPGA was

added to the reaction tube and control 1 and 40mL of water to the control 2. The tubes were

pre-incubated for 2 min in a water bath shaker at 35 �C. Then, 20 mL of freshly prepared

solutions of substrate in 20 % DMSO was added to the reaction and control 2 tubes, as well

as 20 mL of water to the control 1 tube. Aliquots of 50 mL were withdrawn from all tubes

and were quenched in acidified ice-cold methanol (160 mL per each 50 mL aliquot). The

quenched aliquots collected during the incubation were left on ice for at least 20 min. After

centrifugation for 10 min at 12 000 rpm, the supernatants were collected into glass tubes

and the remaining proteinaceous pellets were washed twice with 50 mL of water and

reprecipitated with 150 mL of ice-cold methanol. The combined supernatants were dried

under N2 gas and reconstituted into 50 mL of appropriate mobile phase.

Note: Control 2 was used for testing substrate stability during microsomal incuba-

tion; it was very helpful in prediction of outcomes for synthesis, especially in the case of

HOTYR, which is easily oxidized under neutral and basic aqueous conditions.

2. The products were detected by HPLC-UV (215 nm) with an Atlantis� C18 5 mm

4.6 mm � 150 mm column (5–10 % MeCN in 5 mM ammonium acetate, pH 5.0), 0.5

mL min�1 or by HPLC-MS detection, 0.35 mL min�1 in the negative mode. The

injection volume was 10 mL. The ion spectra of [M � H]� showed at least one of the

two diagnostic product ions [M � H � Gluc]� and [Gluc � H]�, clearly indicating the

presence of the glucuronide moiety.

Note: MS analysis was carried out in negative ionization mode due to the moderate

acidity of glucuroconjugates. The mobile phase ionic strength was adjusted to 5 mM

ammonium acetate at pH 5.0 in order to facilitate in-liquid ionization under established


chromatographic separation and, subsequently, to support both MS monitoring of

substrates and MS detection of glucuroconjugated products.

3. To identify the regioisoforms of glucuroconjugated metabolites, the NMR spectro-

scopic analysis was carried out with chromatographically isolated products.

7.5.3 Procedure 3: Medium-scale (10 mL) Microsomal Glucuroconjugates

Synthesis


• Freshly prepared stock solutions substrate compounds (1 mL, 100 mM in 20 %

DMSO)

• tris-HCl buffer (2 mL, 1 M, pH 8.0)

• CaCl2 (1 mL, 60 mM)

• freshly prepared UDPGA (2 mL, 100 mM)

• DTT (50 mL, 200 mM)

• BSA [1mL 30%(w/v)]

• UDPGT (as a microsomal fraction) (1.5 mL, 20 mg protein/mL)

• ice-cold 33 % acetic acid in methanol (pH 3.3)

• ice-cold 100 % methanol

• distilled water

• 4–8 % of MeCN in 5 mM ammonium acetate (pH 5.0)

• 25 mL glass tube with rubber seal

• 50 mL Falcon tube

• N2 gas



• 0.2 mm filter (MilliQ)

• Atlantis� C18, 5 mm, 19 mm � 150 mm

• equipment for column preparative chromatography with UV detector (215 nm).

7.5.3.2 Procedure

1. UDPGA (2 mL, 100 mM) and freshly thawed microsomal fraction (1.5 mL, 20 mg

protein/mL) were added to a 25 mL glass tube containing tris-HCl solution(2 mL), 1 mL

CaCl2 (1 mL), BSA (1 mL), DTT (50 microL) and H2O (1.45 mL). The mixture was

mixed well, sealed and pre-incubated in a reciprocal shaker for 2 min at 35 �C. Freshly

prepared 1 mL substrate solution was added to the reaction mixture, which was

subsequently sealed with rubber cap, purged with N2 gas and left in the reciprocal

shaker (130 rpm) for 6–8 h at 35 �C.

2. The reaction was quenched by transferring it into a 50 mL Falcon tube containing

33 mL of ice-cold acidic methanol and left immersed in ice for 20 min. The proteinac-

eous pellet was precipitated by centrifugation at 8000 rpm for 15 min at 4 �C and the

supernatant collected in a glass bottle. The pellet was re-extracted twice by washing it

in MilliQ water (5 mL) and reprecipitated with ice-cold 100% methanol (35 mL) over

20 min in ice and then centrifuged as above. The combined supernatants were evapo-

rated at 35 �C and the residue was reconstituted in 20 mL of mobile phase (4–8 %

MeCN in 5 mM ammonium acetate, pH 5.0).


3. The extract was filtered through a 0.2 mm MilliQ filter and chromatographically

separated using an Atlantis� C18, 5 mm, 19 mm � 150 mm column, 10 mL min�1

(4–8 % of MeCN in 5 mM ammonium acetate, pH 5.0). Glucuroconjugated metabolite

peak fractions were collected and the mobile phase was removed by lyophilization. The

products were weighed and reconstituted either in CD3OD to perform NMR analysis or

in 100 % methanol and stored at �80 �C.

7.5.4 Conclusion

Up to 100 % substrate conversion1 and high product purity were achieved in the synthesis

of HOTYR, TYR, HVAlc and HOPhPr mono-O-�-D-glucuronides (Table 7.6).

Table 7.6 Microsomal synthesis and HPLC-UV isolation of HOTYR, TYR, HVAlcand HOPhPr O-�-monoglucuronides

Substrate, mg Substrateconversion (%)

Products Yield,mg (%)

Purity (%)

HOTYR, 15.4 74 HOTYR-40-O-�-glucuronidea 11.9 (36.1) 99.8HOTYR-30-O-�-glucuronideb 5.1 (15.4) 97.8

TYR, 13.8 100 TYR-40-O-�-glucuronidec 13.1 (41.7) 98.8TYR-1-O-�-glucuronided 1.7 (5.4) 98.5

HVAlc, 16.8 95 HVAlc-40-O-�-glucuronidee 30.3 (88.0) 99.78HVAlc-1-O-�-glucuronidef 2.1 (6.1) 96.4

HOPhPr, 15.2 100 HOPhPr-40-O-�-glucuronideg 18.4 (56.0) 99.8HOPhPr-1-O-�-glucuronideh 5.6 (5.5) 99.0

a 1H NMR (500.13 MHz, CD3OD): � 7.13 (d, J¼ 8.15 Hz, H-50), 6.72 (d, J¼ 1.87 Hz, H-20), 6.63 (dd, J¼8.15, 1.73 Hz,H-60), 4.68 (d, J¼ 6.98 Hz, H-100), 3.68 (t, J¼7.16 Hz, H-1), 3.58–3.43 (m, H-500, H-200, H-300 and H-400), 2.69(t, J¼ 7.12 Hz, H-2).b 1H NMR (500.13 MHz, CD3OD): � 7.09 (d, J¼ 0.78 Hz, H-20), 6.79 (dd, J¼8.19, 1.51 Hz, H-60), 6.76 (d, J¼8.14 Hz,H-50), 4.76 (d, J¼6.56 Hz, H-100), 3.77 (bs, H-500), 3.69 (t, J¼7.09 Hz, H-1), 3.58–3.48 (m, H-200, H-300 and H-400), 2.71(t, J¼ 7.07 Hz, H-2).c 1H NMR(500.13 MHz, CD3OD): � 7.13 (d, J¼ 8.36 Hz, H-20 and H-60), 7.04 (d, J¼8.43 Hz, H-30 and H-50), 4.87(d, J¼ 7.10 Hz, H-100), 3.76 (bs, H-500), 3.70 (t, J¼7.12 Hz, H-1), 3.57–3.45 (m, H-200, H-300 and H-400), 2.76 (t, J¼7.12 Hz,H-2).d 1H NMR(500.13 MHz, CD3OD): � 7.07 (d, J¼ 8.46 Hz, H-20 and H-60), 6.69 (d, J¼ 8.49 Hz, H-30 and H-50), 4.30(d, J¼ 7.78 Hz, H-100), 4.08 (m, H-1), 3.69 (m, H-1), 3.56 (d, J¼9.0 Hz, H-500), 3.45 (t, J¼ 9.12 Hz, H-400), 3.40 (t, J¼8.86Hz, H-300), 3.22 (dd, J¼ 8.83, 8.00 Hz, H-200), 2.84 (t, J¼7.49 Hz, H-2).e 1H NMR (500.13 MHz, CD3OD): � 7.12 (d, J¼ 8.25 Hz, H-50), 6.91 (s, H-20), 6.79 (d, J¼8.23 Hz, H-60), 4.89 (d, J¼7.31Hz, H-100), 3.86 (s, —OMe), 3.74 (t, J¼ 7.03 Hz, H-1), 3.72 (t, J¼ 4.63 Hz, H-500), 3.60–3.51 (m, H-200, H-300 and H-400),2.78 (t, J¼ 7.02 Hz, H-2).f 1H NMR (500.13 MHz, CD3OD): � 6.81 (s, H-20), 6.66–6.60 (m, H-50 and H-60), 4.25 (d, J¼7.79 Hz, H-100), 4.06(m, H-1), 3.77 (s,—OMe), 3.66 (m, H-1), 3.50 (d, J¼9.6 Hz, H-500), 3.38 (t, J¼ 9.6 Hz, H-400), 3.34 (t, J¼ 8.7 Hz, H-300),3.18 (t, J¼ 8.9, 7.8 Hz, H-200), 2.81–2.76 (m, H-2).g 1H NMR(500.13 MHz, CD3OD): � 7.05 (d, J¼ 8.56 Hz, H-20 and H-60), 6.98 (d, J¼8.62 Hz, H-30 and H-50), 4.81(d, J¼ 7.25 Hz, H-100), 3.67 (d, J¼8.97 Hz, H-500), 3.49 (t, J¼6.8 Hz, H-1), 3.46–3.41 (m, H-200, H-300 and H-400), 2.56(t, J¼ 7.5 Hz, H-3), 1.74 (q, J¼ 6 Hz, H-3).h 1H NMR(500.13 MHz, CD3OD): � 7.02 (d, J¼ 8.10 Hz, H-20 and H-60), 6.68 (d, J¼ 8.07 Hz, H-30 and H-50), 4.26(d, J¼ 7.75 Hz, H-100), 3.98 (dd, J¼15.62, 6.46 Hz, H-1), 3.60–3.48 (m, H-1 and H-500), 3.45 (t, J¼8.80 Hz, H-400), 3.40(t, J¼ 8.90 Hz, H-300), 3.23 (t, J¼ 8.27 Hz, H-200), 2.62 (t, J¼7.54 Hz, H-3), 1.90–1.83 (m, H-2).


The procedure is very easy to reproduce and could be easily adapted to a wide range of

substrates due to the favourable combination of different UDPGT isoforms in liver

microsomal fractions. Microsomal glucuroconjugation can also be successfully applied

for the synthesis of N- and C-linked glucuroconjugates. The microsomes could also be

extracted from other UDPGT-rich tissues (such as gut and kidney).

Note: The production of minor products, alcohol O-glucuronides, in given examples was

prevented by saturation with both UDPGA and substrate, which usually does not happen in

vivo.

References

1. Khymenets, O., Joglar, J., Clapes, P., Parella, T., Covas, M.-I. and de la Torre, R., Biocatalyzedsynthesis and structural characterization of monoglucuronides of hydroxytyrosol, tyrosol, homo-vanillic alcohol, and 3-(40-hydroxyphenyl)propanol. Adv. Synth. Catal., 2006, 348, 2155.

2. Lukkanen, L., Kilpelainen, I., Kangas, H., Ottoila, P., Elovaara, E. and Taskinen, J., Enzyme-assisted synthesis and structural characterization of nitrocatechol glucuronides. BioconjugateChem., 1999, 10, 150.

3. Shiraga, T., Niwa, T., Ohno, Y. and Kagayama, A., Interindividual variability in 2-hydroxylation,3-sulfation, and 3-glucuronidation of ethynylestradiol in human liver. Biol. Pharm. Bull., 2004,27, 1900.

4. Bradford, M.M., A rapid and sensitive method for the quantitation of microgram quantities ofprotein utilizing the principle of protein–dye binding. Anal. Biochem., 1976, 72, 248.


7.6 Synthesis of the Acyl Glucuronide of Mycophenolic AcidMatthias Kittelmann, Lukas Oberer, Reiner Aichholz and Oreste Ghisalba

Enzymatic synthesis using uridine diphosphate (UDP)-glucuronosyl transferase has pro-

ven to be an effective method for preparation of drug glucuronides which often are

unstable even under mild conditions. By avoiding harsh reaction conditions and multistep

protecting group chemistry, this method often leads selectively to the desired isomer in one

step.1 The immunosuppressive mycophenolic acid (MPA) is used as a drug against organ

rejection after transplantation.2,3 In many species, including humans, the major metabolite

is the O-glucuronide (biologically inactive), whereas the pharmacologically active acyl-

glucuronide is formed only in traces. Via a screening of 10 liver preparations from nine

vertebrate species for acyl glucuronidation of MPA, horse liver S9 fraction was identified

as the most suitable biocatalyst producing acyl and O-glucuronides in a 1:1 ratio, and a

straightforward method for the synthesis of the acylglucuronide on multi-100 mg scale was

developed (Figure 7.3).3

7.6.1 Procedure 1: Preparation of Horse Liver S9 Fraction


• Distilled water (30 mL)

• NaCl (0.27 g)

• horse liver (30 g)

• scalpel

• ice water bath

• Braun Potter-S tissue homogenizer (B. Braun Biotech Co., Melsungen, Germany)


• two centrifuge flasks �50 mL.

7.6.1.2 Procedure

1. NaCl (0.27 g) was dissolved in 30 mL of distilled water and cooled in an ice-water bath.

Horse liver (30 g) was cut into small pieces using a scalpel, mixed with the ice-cold

NaCl solution and homogenized in the Potter-S tissue homogenizer under cooling in

ice-water.

O

OHO

OOH

O

O

O

O

O

O

OH

OH

OHO

COOHO

O

OH

OH

OHOH

O

OH

O

O

OO

O-GlucuronideMycophenolic acid (Myfortic)

Acylglucuronide

UDP-glucuronic acid UDP

UDP-glucuronosyl transferase(horse liver S9 fraction)

+

Figure 7.3 Glucuronidation of mycophenolic acid under catalysis of horse liver S9preparation.


2. The homogenate was centrifuged at 4 �C first for 5 min at 6400g and subsequently for

10 min at 10 200g. The supernatant served as the biocatalyst and could be stored at

�80 �C for months until use.

7.6.2 Procedure 2: Synthesis the MPA-acyl and O-Glucuronide

O

O

O

O

O

OH

OH

OHOH

O

COOHO

O

OH

OH

OHOH

O

OH

O

O

OO

O-GlucuronideAcylglucuronide


• Distilled water

• MPA (450 mg, 1.31 mmol)

• dimethylsulfoxide (DMSO, 13 mL)

• UDP-glucuronic acid sodium salt (4250 mg, 6.58 mmol)

• 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES, 8.14 g, 34.2 mmol)

• MgCl2 (0.876 g, 9.2 mmol)

• NaOH solution, 1 M in distilled water

• XAD-16 (4.4 g, 1.67 % w/v)

• acetonitrile (�4.5 L)

• formic acid solution 0.1 % in water, HPLC-grade (1.0 L)

• nine polypropylene tubes, each 50 mL volume

• nine magnetic stirrers (e.g. device with multiple stirrers)

• temperature-controlled incubator

• laboratory shaker

• filter funnel

• gauze


• preparative HPLC equipment, column Chromasil C18, 7 mm, 100 A, 50 mm � 500 mm

(Eka Chemicals AB, Bohus, Sweden)

• lyophilizor.

7.6.2.2 Procedure

1. HEPES (8.14 g, 34.2 mmol) and MgCl2 (0.876 g, 9.2 mmol) were dissolved in 197 mL

of distilled water and the pH was adjusted to 7 with 1 M NaOH solution. To the resulting

solution UDP-glucuronic acid sodium salt (4250 mg, 6.58 mmol) and a solution of


MPA (450 mg, 1.2 mmol) in DMSO (13 mL) was added. After adjusting the pH again to

7, 52 mL of horse liver S9 preparation was added. The reaction mixture was distributed

into nine polypropylene vessels in equal portions (�29 mL) and incubated under gentle

magnetic stirring in a temperature-controlled incubator at 25 �C for 20 h.

2. The reaction mixture was acidified to pH 2.5 with formic acid and then 4.4 g of the

adsorber resin XAD-16 was added. After further incubation under gentle shaking in a

laboratory shaker for 30 min the resin was filtered off and extracted three times each

with 300 mL of acetonitrile. The solvent was then removed under reduced pressure at

30 �C bath temperature and the residue subjected to preparative RP18-HPLC under the

following elution conditions: solvent A, 0.1 % HCOOH; solvent B, acetonitrile, HPLC

grade; gradient 5–80 % solvent B in 50 min with flow rate 80 mL min�1; detection at

215 nm. The eluent was removed from the fractions containing the acyl and the

O-glucuronide by lyophilization overnight to afford acylglucuronide (240 mg, 35 %

molar yield, >95 % purity according to NMR) and O-glucuronide (10 mg, 97 % purity

according to HPLC-UV, >90 % NMR).1H NMR (CD3CN, 400 MHz) of aclyglucuronide: �¼ 1.82 (br s, 3H,) 2.15 (s, 3H),

2.31 (br t, 2H), 2.50 (t, 2H), 3.38 (br d, 2H), 3.44 (t, 1H), 3.33 (t, 1H), 3.53 (t, 1H), 3.38

(br d, 2H), 3.91 (d, 1H), 5.47 (d, 1H), 5.25 (s, 2H), 5.26 (m, 1H), 7.7 (b); of

O-glucuronide: �¼ 1.80 (br s, 3H,) 2.22 (s, 3H), 2.26 (br t, 2H), 2.38 (t, 2H), 3.46

(t, 1H), 3.50 (t, 1H), 3.56 (t, 1H), 3.62/3.42 (AB(X), 2H), 3.76 (d, 1H), 3.78 (s, 3H), 5.23

(d, 1H), 5.24 (m, 1H), 5.26 (AB, J¼ 15.5 Hz, 2H).

7.6.3 Conclusion

The method is easy to perform and reliable. Screening for a suitable biocatalyst was

carried out and then optimization of the reaction conditions by lowering the tem-

perature from 37 to 25 �C and the UDP-glucuronic acid concentration from 40 to

25 mM shifted conversion up for the acylglucuronide from 24 % to˜50 %. Whereas

the enzymatic formation of O-glucuronides is often favoured at more alkaline pH

(pH 8–8.75), pH 7 was more productive for the acylglucuronide of MPA. This might

partially be due to the fact that acylglucuronides are generally more stable in the

acidic range.4 Although the acylglucuronide of MPA proved to be sufficiently stable

at pH 2.5 for chromatographic purification and lyophilization, storage at low tem-

perature (�20 or �80 �C) is advisable.

O-Glucuronidation of MPA can be performed in high yield with the same technology

(also pH 7), except using rabbit liver S9 fraction as the catalyst, which produces exclu-

sively the O-glucuronide (>95 % conversion).

References

1. Zaks, A. and Dodds, D.R., Enzymatic glucuronidation of a novel cholesterol absorption inhibitor,Sch 58235. Appl. Biochem. Biotechnol., 1998, 73, 205.

2. Shipkova, M., Armstrong, V.M., Wieland, E., Niedmann, P.D., Schutz, E., Brenner-Weiß, G.,Voihsel, M., Braun, F. and Oellerich, M., Identification of glucoside and carboxyl-linkedglucuronide conjugates of mycophenolic acid in plasma of transplant recipients treated withmycophenolate mofetil. Brit. J. Pharmacol., 1999, 126, 1075.


Section 7.6 reproduced from reference 3, with permission from Wiley-VCH Verlag

GmbH & Co. KGaA.

3. Kittelmann, M., Rheinegger, U., Espigat, A., Oberer, L., Aichholz, R., Francotte, E. and GhisalbaO., Preparative enzymatic synthesis of the acylglucuronide of mycophenolic acid. Adv. Synth.Catal., 2003, 345, 825.

4. Spahn-Langguth, H. and Benet, L.Z., Acyl glucuronides revisited: is the glucuronidation proces atoxification as well as a detoxification mechanism? Drug Metab. Rev., 1992, 24, 5.


8

Synthesis of Cyanohydrins UsingHydroxynitrile Lyases

8.1 Synthesis of (S)-2-Hydroxy-2-methylbutyric Acid by a ChemoenzymaticMethodologyManuela Avi and Herfried Griengl

The hydroxynitrile lyase (HNL)-catalysed cyanohydrin reaction is a useful method to

synthesize enantiopure �-hydroxy nitriles and the corresponding �-hydroxy acids.1,2

However, small ketones, such as 2-butanone, are converted with low selectivities, due to

the poor discrimination between methyl and ethyl.3–5

Recently, the synthesis of (S)-2-hydroxy-2-methylbutyric acid has been reported using

the docking/protecting group concept6 and HNL from Hevea brasiliensis (HbHNL) as

catalyst (Scheme 8.1).5

S

O

S

OH

CN

S

OH

COOH

OH

CH2CH3

CH3

HOOCHCN/HbHNL H+ Raney Ni

80%, 91% ee 80%, 99% ee 70%, 99% ee

Scheme 8.1 Chemoenzymatic synthesis of (S)-2-hydroxy-2-methylbutyric acid by using thedocking protecting group concept



8.1.1 Procedure 1: (S)-3-Hydroxytetrahydrothiophene-3-carbonitrile

S

OH

CN


• 4,5-Dihydro-3(2H)-thiophenone (10.2 g, 100 mmol)

• HbHNL solution (31 mL, 6000 IU mL�1)

• HCN (19.6 mL, 500 mmol)

• tert-butyl methyl ether (TBME, 235 mL)

• K2HPO4/citrate buffer (24 mL, 50 mM, pH 4.0)

• Anhydrous Na2SO4

• 250 mL reactor with flow beakers

• mechanical stirrer

• thermostat

• frit


8.1.1.2 Procedure

1. The aqueous HbHNL solution (31 mL) was diluted with phosphate/citrate buffer (1/2

v/v, 50 mM, pH 4.0) and the pH adjusted to 4.5 with 10 % citric acid.

2. The above solution was added to tetrahydrothiophen-3-one (10.2 g, 100mmol) in

TBME (35 mL) and the resulting mixture was stirred until an emulsion formed.

3. Freshly generated HCN (19.6 mL, 5 equiv) was added and the mixture was stirred at

0 �C and 950 rpm until quantitative conversion.

4. The reaction was diluted with TBME (100 mL) and stirred for 30 min. The phases were

separated and once more TBME (100 mL) was added to the aqueous phase. The mixture was

stirred for 10 min. The organic phases were combined and dried over Na2SO4. Evaporation

of the solvent yielded the crude (S)-3-Hydroxytetrahydrothiophene-3-carbonitrile (10.26 g,

80 %) as slightly yellow oil with 91 % ee measured by chiral gas chromatography (GC) after

TMS-protection (Chirasil-Dex, 120 �C isotherm: 11.9 min (S), 12.1 min (R)).

½��20D ¼�11:0� (c¼ 1.0, CHCl3 for 50 % ee). 1H NMR (500 MHz; CDCl3) d 3.94 (s, 1H,

OH); 3.28 (d, 1H, H2, J¼ 12.23 Hz); 3.12 (dd, 1H, H20, J¼ 11.71 Hz, 0.98 Hz); 3.08–2.98

(m, 2H, H5, H50); 2.48 (ddd, 1H, H4, J¼ 3.90 Hz, 0.46 Hz,); 2.28 (ddd, 1H, H40, J¼ 7.81

Hz, 1.47 Hz); 13C NMR (125 MHz; CDCl3) d 120.11 (CN); 74.31 (C3—OH); 42.85 (C2);

42.71 (C4); 28.30 (C5); elemental analysis calc. (%) for C5H7NOS: C 46.49, H 5.46.

N 10.84, S 24.82; found: C 46.41, H 5.50, N 10.90, S 24.83.

8.1.2 Procedure 2: (S)-3-Hydroxytetrahydrothiophene-3-carboxylic Acid

S

OH

COOH

256 Synthesis of Cyanohydrins Using Hydroxynitrile Lyases


• (S)-3-Hydroxytetrahydrothiophene-3-carbonitrile (6.0 g, 46 mmol)

• HCl concentrated (45 mL)

• NaOH (40 mL, 20 %)

• TBME (160 mL)

• anhydrous Na2SO4

• (S)-phenylethylamine (2 g)

• 50 mL round-bottom flask equipped with magnetic stirrer bar

• reflux condenser



• frit


8.1.2.2 Procedure

1. A solution of (S)-3-hydroxytetrahydrothiophene-3-carbonitrile (6.0 g, 46 mmol) was

stirred with concentrated HCl (15 mL) at 70 �C for 15 h.

2. After cooling, NaOH (40 mL, 20 %) was added and the mixture (pH 10) was washed

with TBME (2� 30 mL). The aqueous phase was acidified again with concentrated

HCl (10 mL, pH 1.0) and extracted with TBME (2� 30 mL). The combined organic

phase was dried over Na2SO4. The filtrate was concentrated in vacuo to give crude

(S)-hydroxy acid (5.47 g, 80 %) as a beige solid.

3. 2.5 g of product was dissolved in 25 mL of hot TBME and (S)-phenylethylamine (2 g)

added. After cooling, the precipitate was filtered off, dissolved in HCl (20 mL, 2 M) and

extracted with TBME (3� 15 mL). After concentration in vacuo, (S)-3-hydroxytetra-

hydrothiophene-3-carboxylic acid was obtained with 99 % ee.

M.p. 74–76 �C; ½��20D ¼�50:3� (c¼ 2.0 in CHCl3) 99 % ee. 1H NMR (500 MHz; CDCl3)

d 3.33 (d, 1H, H2, J¼ 11.23 Hz); 3.09–2.98 (m, 2H, H5, H50, J¼ 10.25 Hz); 2.90 (dd, 1H,

H20, J¼ 11.71 Hz; 1.46 Hz); 2.33–2.21 (m, 2H, H4, H40, J¼ 10.74 Hz); 13C NMR (125

MHz; CDCl3) d 178.45 (COOH); 82.91 (C3—OH); 42.52 (C2); 42.10 (C4); 30.15 (C5);

elemental analysis calc. (%) for C5H8O3S: C 40.53, H 5.44, S 21.64; found: C 41.13,

H 5.38, S 21.64.

8.1.3 Procedure 3: (S)-2-Hydroxy-2-methylbutanoic Acid

OH

CH2CH3

CH3

HOOC


• (S)-3-Hydroxytetrahydrothiophene-3-carboxylic acid (300 mg, 2 mmol)

• Raney nickel (5.5 g moist, 50 % Ni)


• NaOH (1 mL, 2 M)

8.1 Chemoenzymatic Synthesis of (S)-2-Hydroxy-2-methylbutyric Acid 257

• H2SO4 (1 M)

• TBME (40 mL)

• 50 mL round- bottom flask equipped with magnetic stirrer bar

• reflux condenser



• frit


8.1.3.2 Procedure

1. A suspension of (S)-3-hydroxytetrahydrothiophene-3-carboxylic acid (300 mg, 2 mmol)

and Raney nickel (5.5 g moist 50 % Ni) in water (28 mL) and NaOH (1 mL, 2 M) was

stirred at 75 �C for 2 h.

2. After acidification with H2SO4 (1 M) to pH 1, filtration and extraction with TBME, the

solvent was removed in vacuo to afford (S)-2-hydroxy-2-methylbutanoic acid (70 %) as

a white solid.

3. Recrystallization from n-hexane at ambient temperature gave the (S)-2-hydroxy-2-

methylbutanoic acid as a white solid in 99 % ee, detected by chiral GC after derivatiza-

tion to the acetonide via acetone and concentratred H2SO4 (Chirasil-Dex, 100 �C, 3.5

min, 10 �C min�1, 160 �C, 2.5 min).

M.p. 74 �C; NMR data are equal to literature;7 elemental analysis calc. (%) for C5H10O3:

C 50.84, H 8.53; found: C 50.80, H 8.64.

References

1. Gregory, R.J.H., Cyanohydrins in nature and the laboratory: biology, preparations, and syntheticapplications. Chem. Rev., 1999, 99, 3649.

2. Fechter, M.H. and Griengl, H., Enzymatic synthesis of cyanohydrins. In Enzyme Catalysis inOrganic Synthesis: A Comprehensive Handbook, vol. 2, 2nd edn, Drauz, K. and Waldmann, H.(eds). Wiley–VCH, Weinheim, 2002, pp. 974–989.

3. Forster,S., Roos, J., Effenberger, F., Wajant, H. and Sprauer, A. Uber die erste rekombinanteHydroxynitril-Lyase und ihre Anwendung in der Synthese von (S)-Cyanhydrinen. Angew. Chem.,1996, 108, 493.

4. Effenberger, F., Hoersch, B., Weingart, F., Ziegler, T. and Kuehner, S., Enzyme-catalyzedsynthesis of (R)-ketone-cyanohydrins and their hydrolysis to (R)-�-hydroxy-�-methyl-carboxylic acids. Tetrahedron Lett., 1991, 32, 2605.

5. Fechter, M.H., Gruber, K., Avi, M., Skranc, W., Schuster, C., Pochlauer, P., Klepp, K.O. andGriengl, H., Stereoselective biocatalytic synthesis of (S)-2-hydroxy-2-methylbutyric acid viasubstrate engineering by using ‘thio-disguised’ precursors and oxynitrilase catalysis. Chem.Eur. J., 2007, 13, 3369.

6. De Raadt, A., Griengl, H. and Weber, H., Chem. Eur J., 2001, 7, 27.7. Pouchert, C.J. and Behnke, J., The Aldrich Library of 13C and 1H FT NMR Spectra, vol. 1,

Aldrich, Milwaukee, WI, USA, 1993, p. 808A.


8.2 (S)-Selective Cyanohydrin Formation from Aromatic Ketones UsingHydroxynitrile LyasesChris Roberge, Fred Fleitz and Paul Devine

The production of optically active cyanohydrins, with nitrile and alcohol functional groups

that can each be readily derivatized, is an increasingly significant organic synthesis

method. Hydroxynitrile lyase (HNL) enzymes have been shown to be very effective

biocatalysts for the formation of these compounds from a variety of aldehyde and aliphatic

ketone starting materials.1–4 Recent work has also expanded the application of HNLs to the

asymmetric production of cyanohydrins from aromatic ketones.5 In particular,

commercially available preparations of these enzymes have been utilized for high ee

(S)-cyanohydrin synthesis from phenylacetones with a variety of different aromatic sub-

stitutions (Figure 8.1).

8.2.1 Procedure 1: General Procedure for the Preparation of Enantioenriched

Methyl Ketone Cyanohydrins


• Ketone substrate (750 mL)

• 0.1 M citrate buffer pH 4.5 (18 mL)

• diisopropyl ether (2.5 mL)

• HNL enzyme solution (Codexis Inc, 2.5 mL)

• trimethylsilylcyanide (1.25 mL)

• saturated ammonium sulfate solution (2.5 mL)


• nitrogen gas

• fume hood

• cyanide detector

• 50 mL flask

• magnetic stirrer.

8.2.1.2 Procedure

1. Diisopropyl ether (2.5 mL), ketone substrate (750 mL) and trimethylsilylcyanide (1.25

mL) were added to 18 mL of 0.1 M pH 4.5 citrate buffer which was stirring at 5 �C. The

mixture was prepared in a chemical fume hood with the use of a cyanide detector to

measure for any release of cyanide vapor.

R1

R2

OR1

R2

OH

CN

83–99% ee53–77% yield

LuHNL or MeHNL

Figure 8.1 Enzymatic (S)-selective substituted phenylacetone cyanohydrin synthesis

8.2 (S)-Selective Cyanohydrin Formation from Aromatic Ketones 259

2. The reaction was started with the further addition of 2.5 mL of a solution of commercial

HNL isolated from either flax (LuHNL) or cassava (MeHNL). The reaction vessel was

closed and the reaction was aged for 24 h.

3. The cyanohydrin product was extracted into organic solvent with the addition of saturated

ammonium sulfate (2.5 mL) and ethyl acetate (25 mL) while continuing to stir. The organic

layer was transferred to another vessel and evaporated under nitrogen, yielding an oil (Table 1).

Assay of the reaction mixture. The samples were then resuspended in 1.5 mL

isopropanol and assayed to determine both the yield and ee by chiral normal phase

high-performance liquid chromatography (HPLC). A 250 mm� 4.6 mm Chiralpak

AD-H column was used with an eluant of 95:5 heptane/ethanol, a flow rate of 3 mL

min�1, a temperature of 10 �C and a detection wavelength of 210 nm.

8.2.2 Conclusion

The generalized procedure described here has been demonstrated with a large number of

aromatic ketone substrates, including those described in Table 8.1. When the goal is the

production of a particular (S)-cyanohydrin, specialized process improvements to parameters,

such as the operating temperature and pH and the choice and concentration of organic

solvent and cyanide donor, may further increase both the product ee and yield values.

Table 8.1 Results for the generalized procedure demonstrated with various aromatic ketonesubstrates

Ketone

RO

R = 2-BrR = 3-Fa

R = 3-ClR = 3-BrR = 3-CH3

R = 3-CF3b

R = 3-CH3OR = 4-BrR = 4-CH3O

c

MeHNL

Yield (%) Ee (%)

44 9770 7971 9361 9365 8867 9753 9277 9062 47

LuHNL

Yield (%) Ee (%)

20 8324 8937 9940 9719 9631 99

9 9330 84

a2-Hydroxy-3-(3-fluorophenyl)-2-methyl-propanenitrile. 1H NMR (CDCl3; 400 MHz):δ1.68 (s, 3H), 2.97–3.13 (m, 2H), 7.04–7.14 (m, 4H). HPLC: R T(ketone) = 2.5 min, R T[(S )-cyanohydrin] = 4.2min, R T[(R )-cyanohydrin] =12.2min.b2-Hydroxy-3-(3-trifluoromethyl-phenyl)-2-methylpropanenitrile. 1H NMR (CD4O; 400 MHz):δ1.52 (s, 3H), 2.93–3.26 (m, 2H), 7.46–7.63 (m, 4H). HPLC: R T(ketone) = 2.0 min, R T[(S )-cyanohydrin] = 2.6 min R T[(R )-cyanohydrin] =4.6min.c2-Hydroxy-3-(4-methoxyphenyl) -2-methyl-propanenitrile.1H NMR (CDCl3; 400 MHz): δ 1.65 (s, 3H), 2.89–3.08 (m, 2H), 3.82 (s, 3H), 6.88 (d, 2H, J = 8.6Hz), 7.26 (d, 2H, J = 7.8Hz). HPLC: TR(ketone) = 3.6 min, TR [(S )-cyanohydrin]= 6.7min TR [(R )-cyanohydrin] = 13.6 min.

References

1. Fechter, M.H. and Griengl, H., Hydroxynitrile lyases: biological sources and application asbiocatalysts. Food Technol. Biotechnol., 2004, 42, 287.

2. Brussee, J. and van der Gen, A., Biocatalysis in the enatioselective formation of chiral cyanohy-drins, valuable building blocks in organic synthesis. In Stereoselective Biocatalysis, Patel, R.N.(ed.). Dekker: New York, 2000, pp. 289–320.

3. Effenberger, F., Hydroxynitrile lyases in stereoselective synthesis. In StereoselectiveBiocatalysis; Patel, R.N. (ed.). Dekker: New York, 2000; pp. 321–342.


4. Sharma, M., Sharma, N.N. and Bhalla, T.C., Hydroxynitrile lyases: at the interface of biology andchemistry. Enzyme Microb. Technol., 2005, 37, 279.

5. Roberge, C., Fleitz, F., Pollard, D. and Devine, P., Synthesis of optically active cyanohydrinsfrom aromatic ketones: evidence of an increased substrate range and inverted stereoselectivity forthe hydroxynitrile lyase from Linum usitatissimum. Tetrahedron Asymm., 2007, 18, 208.

8.2 (S)-Selective Cyanohydrin Formation from Aromatic Ketones 261

8.3 Hydroxynitrile-lyase-catalysed Synthesis of Enantiopure(S)-Acetophenone CyanohydrinsJan von Langermann, Annett Mell, Eckhard Paetzold and Udo Kragl

Chiral cyanohydrins are versatile intermediates in the synthesis of �-hydroxy acids,

�-amino alcohols, amino nitriles, �-hydroxy ketones and aziridines. For the synthesis

of enantiopure cyanohydrins, the use of hydroxynitrile lyases is currently the most

effective approach.1–4 Application of an organic-solvent-free system allows thermo-

dynamically hindered substrates to be converted with moderate to excellent yields.

With the use of the highly selective hydroxynitrile lyase from Manihot esculenta, the

syntheses of several acetophenone cyanohydrins with excellent enantioselectivities

were developed (Figure 8.2). (S)-Acetophenone cyanohydrin was synthesized on a

preparative scale.5

8.3.1 Procedure 1: Preparation of (S)-Acetophenone Cyanohydrin


• acetophenone (40 mL, 0.34 mol)

• hydroxynitrile lyase from M. esculenta (350 kU) (purchased from Julich Chiral

Solutions, A Codexis Company, Julich, Germany)

• hydroxynitrile lyases from other sources (e.g. Hevea brasiliensis, Prunus amygdalus)

may be also used in a similar procedure

• sodium cyanide (128 g, 2.6 mol)

• sulfuric acid in water (5 M, 320 mL)

• deionized water (320 mL, 17.8 mol)

• citrate buffer pH 4.0 (50 mM, 750 mL)

• diisopropylether (300 mL)

• sodium sulfate (anhydrous)

• distillation equipment (with dropping funnel) for the distillation of

hydrogen cyanide

• electrochemical HCN detector for continuous monitoring

• reaction flask 1 L

• stirrer

• centrifuge.

CH3

O

CH3

HO CN(S)-Hydroxynitrile lyasefrom M. esculenta

+HCN

R R

Figure 8.2 M. esculenta-catalysed synthesis of acetophenone cyanohydrins.


8.3.1.2 Procedure

The synthesis should be performed within a well-ventilated hood.

Safety note. An electrochemical HCN detector (Micro III G203, GfG-Gesellschaft fur

Geratebau mbH, Dortmund, Germany) was placed in the fume hood for continuous

monitoring.

1. The required amount of HCN was freshly distilled in a well-ventilated fume hood.

Sodium cyanide (128 g) was dissolved in deionized water (320 mL) and 5 M sulfuric

acid (320 mL) was added dropwise within the distillation equipment and the resulting

solution was heated up to 75 �C. The hydrogen cyanide was condensed immediately in a

receiving flask cooled to 5 �C. Total yield of hydrogen cyanide: 68 mL.

2. Citrate buffer (pH 4, 750 mL, 50 mM) was placed in a 1 L reaction flask and

thermostatically cooled to 5 �C. Then the freshly distilled hydrogen cyanide (68

mL) and acetophenone (40 mL) were added to the buffer and the mixture was cooled

to 5 �C.

3. The reaction was started with the addition of 350 kU hydroxynitrile lyase from M.

esculenta. The reaction mixture was vigorously stirred and the reaction was monitored

by gas chromatography (GC) until the equilibrium conversion of 22 % was reached

(�1.5 h).

Note. One unit (U) of enzyme activity was defined as the amount of enzyme that

catalysed the cleavage of 1 mmol mandelonitrile per minute under assay conditions.

The enzyme activity was determined by following the cleavage of rac-mandelonitrile

into benzaldehyde and HCN at 25 �C. The formation of benzaldehyde was measured

spectrometrically at 280 nm. The nonenzymatic cleavage reaction was monitored under

identical conditions and subtracted.

Assay conditions. 700 mL citrate–phosphate buffer pH 5.0, 100 mL enzyme solution

(dilution if required) and 200 mL mandelonitrile stock solution (60 mmol L�1 in citrate–

phosphate buffer pH 3.5) were mixed in a cuvette with 1 cm pathlength and the increase

of absorbance at 280 nm was measured for 2 min.

4. The reaction mixture was then extracted twice with diisopropylether (2 � 150 mL). (If

problems with the phase separation of the reaction mixture occur, then it should be

centrifuged.) The combined solutions were dried with sodium sulfate and the organic

solvent was removed under reduced pressure.

5. Distillation under reduced pressure occurred without racemization or decomposition to

afford (S)-acetophenone cyanohydrin (5 g, 10 %) in >95 % purity and 98.5 %

enantiomeric excess.6

The conversion of acetophenone to acetophenone cyanohydrin and enantiomeric excess

were determined by gas chromatographic analysis after product derivatisation as the

trifluoroacetate. GC was performed using a Chiraldex capillary GC column (G-PN –

g-Cyclodextrin, Propionyl) from Astec using a CP3800 (Varian) with a flame ionization

detector. Carrier gas was helium at 2 mL min�1. Temperature gradient: 80 �C for 0.5 min,

raise at 10.8�C min�1 to 130 �C and hold 130 �C for 15 min. The injector and detector

temperatures were set to 250 �C.

8.3 Catalysed Synthesis of Enantiopure (S)-Acetophenone Cyanohydrins 263

A sample of the suspension (100 ml) was extracted with 100 ml of diisoropyl ether.

50 ml from the resulting organic phase (or if available: 1 ml crude product) were added to

a mixture of 500 ml dichloromethane, 50 ml trifluoroacetic anhydride and 50 ml pyridine

for the acetylation procedure. The mixture was then directly injected into the gas

chromatograph.

All waste solutions from the reaction were collected and disposed with hydrogen

peroxide.

8.3.2 Conclusion

The use of organic-solvent-free systems can be applied to the cyanohydrin synthesis of a

wide range of acetophenone derivatives (Table 8.2); electronegative substituents (e.g.

fluorine) facilitate high conversions and enantiomeric excess of the product, whereas

electropositive substituents (e.g. methoxy-) result in low to no conversion into the corre-

sponding cyanohydrins.

Table 8.2 Acetophenone (AP) derivative cyanohydrin formationa

AP DERIVATIVE TIME (H) CONVERSIONb (%) EE (S) (%)

40-F-AP 1.5 14 >9930-F-AP 3 48 >9920-F-AP 3 71 >9920,30,40,50,60-F-AP 6 <1 -40-CL-AP 6 18 9730-CL-AP 6 23 9720-CL-AP 4.5 6 8040-BR-AP SOLID30-BR-AP 6 10 >9920-BR-AP 6 9 6840-I-AP SOLID20-I-AP 6 <1 —40-ME-AP 6 <1 —30-ME-AP 6 <1 —20-ME-AP 6 <1 —40-MEO-AP SOLID30-MEO-AP 6 <1 —20-MEO-AP 6 <1 —40-NO2-AP SOLID30-NO2-AP SOLID20-NO2-AP 1.5 40 >9940-NH2-AP SOLID30-NH2-AP SOLID20-NH2-AP 6 <1 —40-OH-AP SOLID30-OH-AP SOLID20-OH-AP 6 <1 —

aReaction conditions: reaction time, 1.5–6 h; 0.4 mmol AP derivative; 2 mmol hydrogen cyanide; 1 mL citrate buffer pH4.0 or pH 4.8; 5 �C; 450 U mL�1; 400 rpm.bSolid: conversion not determined due to the absence of an organic layer.


References

1. Gregory, R.J.H., Cyanohydrins in nature and the laboratory: biology, preparations, and syntheticapplications. Chem. Rev., 1999, 99, 3649–3682.

2. Johnson, D.V. and Griengl, H., Chiral cyanohydrins: their formation synthetic potential andapplication. Chim. Oggi-Chem. Today, 1997, 15, 9–13.

3. Breuer, M., Ditrich, K., Habicher, T., Hauer, B., Kesseler, M., Sturmer,R. and Zelinski, T.,Industrial methods for the production of optically active intermediates. Angew. Chem. Int. Ed.,2004, 43, 788–824.

4. M. North, Synthesis and applications of non-racemic cyanohydrins. Tetrahedron Asymm. 2003,14, 147–176.

5. Von Langermann, J., Mell, A., Paetzold, E., Daussmann, T. and Kragl, U., Hydroxynitrile lyase inorganic solvent-free systems to overcome thermodynamic limitations. Adv. Synth. Catal., 2007,349, 1418–1424.

6. Gassman, P.G. and Talley, J.J., Cyanohydrins – a general synthesis. Tetrahedron Lett., 1978, 19,3773–3776.

8.3 Catalysed Synthesis of Enantiopure (S)-Acetophenone Cyanohydrins 265

8.4 (R)- and (S)-Cyanohydrin Formation from Pyridine-3-carboxaldehydeUsing CLEATM-immobilized Hydroxynitrile LyasesChris Roberge, Fred Fleitz and Paul Devine

The use of both ligands and enzymes for the asymmetric synthesis of cyanohydrins has

been a topic of much recent research, with advances being made to processes converting a

diverse array of aldehydes and ketones.1–4 Nitrogen-containing compounds have been

shown to be generally poor substrates for both of these classes of catalysts, though,

yielding products with poor to moderate chiral purity.5–8 One significant cause for this is

the presence of an aqueous background nonenzymatic reaction involving the racemic

addition of cyanide. By immobilizing the biocatalyst in the form of a cross-linked enzyme

aggregate (CLEATM) it can be successfully used in essentially water-free environments,

reducing the impact of this background reaction and enhancing the overall enantioselec-

tivity. In this way, commercially available CLEATM preparations of hydroxynitrile lyases

(HNLs) from cassava (MeHNL) and almond (PaHNL) have been used to generate >93%

ee (S)- and (R)-cyanohydrin from 3-pyridinecarboxaldehyde (Figure 8.3).9

8.4.1 Procedure 1: Preparation of Enantioenriched Piperidine-3-carboxaldehyde

Cyanohydrin


• Citrate (1.6 g)

• Dichloromethane (10 mL)

• Potassium cyanide (520 mg)

• 0.1 M citrate buffer pH 2.5 (20 mL)

• 3-pyridinecarboxaldehyde (300 mg)

• HNL-CLEATM enzyme preparation (CLEATM Technologies, 40 mg)

N

O N

OH

CN

4 g . L–1 MeHNL-CLEAHCN

30 g . L–1

85 % yield94 % e.e.

CH2Cl2, 5 °C2 h

CH2Cl2, 5 °C2 h N

OH

CN

4 g . L–1PaHNL-CLEAHCN

65 % yield93 % e.e.

Figure 8.3 Enzymatic enantiocomplementary synthesis of the cyanohydrins from pyridine-3-carboxaldehyde.


• nitrogen gas

• fume hood

• cyanide detector

• 25 mL flask



8.4.1.2 Procedure

1. Citrate (1.6 g) and potassium cyanide (520 mg) were added to a mixture of dichloro-

methane (10 mL) and water (2.5 mL) that was stirring at 0 �C. The solution was allowed

to mix for an additional 15 min and the organic layer which now contained HCN was

removed for use as the reaction medium. The mixture was prepared in a chemical fume

hood with the use of a cyanide detector to measure for any release of cyanide vapor.

2. Citrate buffer (0.1 M, pH 2.5, 20 mL), 3-pyridinecarboxaldehyde (300 mg) and

CLEATM-immobilized HNL (40 mg) were added to 9.7 mL of the dichloromethane–

HCN solution and the reaction was aged for 2–20 h at 5 �C. Commercially available

CLEASTM of almond HNL (PaHNL-CLEATM) and cassava HNL (MeHNL-CLEATM)

were used to catalyze the syntheses of (R)- and (S)-cyanohydrin respectively.

3. The product was isolated by evaporating the dichloromethane and HCN under nitrogen,

yielding a solid.

Assay of the reaction mixture. A 50 mL sample was removed from the reaction and the

dichloromethane component was evaporated under nitrogen for 20 s. The sample was

then resuspended in 600 mL isopropanol and assayed by chiral high-performance liquid

chromatography. A 250 mm � 4.6 mm Chiralpak AD-H column was used with an

eluant of 85:15 heptane/ethanol, a flow rate of 3 mL min�1, a temperature of 10 �C, a

detection wavelength of 245 nm and a sample injection volume of 2 mL.

8.4.2 Conclusion

The use of CLEATM preparations of commercially available HNLs allowed for the

enantiocomplementary production of cyanohydrins from a pyridinecarboxaldehyde at a

much higher chiral purity than had previously been demonstrated with any chemical

catalyst. The key to the success of this process was the use of the CLEATM-immobilized

biocatalysts that allowed reaction conditions to be chosen to minimize the negative effects

of the nonspecific background reaction.

References

1. Fechter, M.H. and Griengl, H., Hydroxynitrile lyases: biological sources and application asbiocatalysts. Food Technol. Biotechnol., 2004, 42, 287.

2. Brussee, J. and van der Gen, A., Biocatalysis in the enatioselective formation of chiral cyanohy-drins, valuable building blocks in organic synthesis. In Stereoselective Biocatalysis, Patel, R.N.(ed.). Dekker: New York, 2000, pp. 289–320.

3. Effenberger, F., Hydroxynitrile lyases in stereoselective synthesis. In StereoselectiveBiocatalysis; Patel, R.N. (ed.). Dekker: New York, 2000; pp. 321–342.

4. Sharma, M., Sharma, N.N. and Bhalla, T.C., Hydroxynitrile lyases: at the interface of biology andchemistry. Enzyme Microb. Technol., 2005, 37, 279.

8.4 Cyanohydrin Formation from Pyridine-3-carboxaldehyde 267

5. Baeza, A., Casa, J., Najera, C., Sansano, J.M. and Saa, J.M., Enantioselective synthesis ofO-methoxycarbonyl cyanohydrins: chiral building blocks generated by bifunctional catalysiswith BINOLAM-AlCl. Eur. J. Org. Chem., 2006, 1949.

6. Schmidt, M., Herve, S., Klempier, N. and Griegl, H., Preparation of optically active cyanohydrinsusing the (S)-hydroxynitrile lyas from Hevea brasiliensis. Tetrahedron, 1996, 52, 7833.

7. Chen, P., Han, S., Lin, G., Huang, H. and Li, Z., A study of asymmetric hydrocyanation ofheteroaryl carboxaldehydes catalyzed by (R)-oxynitrilase under micro-aqueous conditions.Tetrahedron Asymm., 2001, 12, 3273.

8. Nanda, S., Kato, Y. and Asano, Y., A new (R)-hydroxy-nitrile lyase from Prunus mume: asym-metric synthesis of cyanohydrins. Tetrahedron, 2005, 61, 10908.

9. Roberge, C., Fleitz, F., Pollard, D. and Devine, P., Asymmetric synthesis of cyanohydrin derivedfrom pyridine aldehyde with cross-linked aggregates of hydroxynitrile lyases. Tetrahedron Lett.,2007, 48, 1473.


8.5 A New (R)-Hydroxynitrile Lyase from Prunus mume for AsymmetricSynthesis of CyanohydrinsYasuhisa Asano

A new hydroxynitrile lyase (HNL) was isolated from the seed of Japanese apricot

(Prunus mume).1 It accepts benzaldehyde and a large number of unnatural substrates

for the addition of HCN to produce the corresponding (R)-cyanohydrins in excellent

optical and chemical yields. A new high-performance liquid chromatography

(HPLC)-based enantioselective assay technique was developed for the enzyme,

which promotes the addition of KCN to benzaldehyde in a buffered solution (pH

4.0). Asymmetric synthesis of (R)-cyanohydrins by a new HNL is described

(Figure 8.4).2,3

8.5.1 Procedure 1: Activity Measurement and Partial Purification of HNL

from P. mume


• Ripened Ume fruits (P. mume) (obtained from local fruit market in June and stored at

4 �C until use, 1 kg or more)

• potassium cyanide

• benzaldehyde

• (R)- and (S)-mandelonitrile

• (NH4)2SO4

• citrate buffer (pH 4.0)

• potassium phosphate buffer (pH 6.0)

• hexane

• isopropanol


• dialysis tubing

• hammer

• homogenizer

• cheesecloth

• centrifuge (with a cooling apparatus)

• HPLC with a Chiralcel OJ-H column (0.46 cm� 25 cm, Daicel industries)

• ice bucket and ice.

R

N

OH

R H

O

HCN+

(R)-Cyanohydrin

(R)-Hydroxynitrilelyase

Aldehyde or ketone

Figure 8.4 HNL-catalyzed asymmetric synthesis of (R)-cyanohydrin

8.5 A New (R)-Hydroxynitrile Lyase from Prunus mume 269

8.5.1.2 Procedure

1. Ripened Ume fruit (P. mume) was taken and the fleshy cover was removed to obtain the

seeds. The upper layer of the seeds was cracked with a hammer to give the soft kernels

inside. Those kernels were collected and crushed in the process homogenizer at 4 �C, with

10 mM potassium phosphate buffer, pH 6.0, to give a milky suspension. The suspension

was filtered through four layers of cheesecloth to remove the insoluble part. The HNL

activity shown by P. mume extract was 6.9 U mg�1 in the milky suspension.1

2. The suspension was next centrifuged (18, 800� g, 30 min at 4 �C) and the removal of the

residue gave a crude enzyme preparation, which was fractionated with (NH4)2SO4. Proteins

precipitating with 30 % saturation were collected by centrifugation (18, 800 � g, 30 min at

4 �C), dissolved in minimum volume of 10 mM potassium phosphate buffer, pH 6.0, and

dialyzed against the same buffer with three changes. After that the dialyzed solution was

centrifuged and the supernatant was stored at 4 �C and assayed for the HNL activity.

3. The protein content in HNL was measured by the Bradford method using a Bio-Rad

protein assay kit and bovine serum albumin as the standard.4 The protein content of

crude HNL after the fractionation with 30 % (NH4)2SO4 saturation was found to be

roughly 10 mg mL�1 and the activity was 120 U mL�1 (specific activity: 12 U mg�1).

4. The enzyme activity was assayed by measuring the production of optically active

mandelonitrile synthesized from benzaldehyde and cyanide. The standard assay solu-

tion contained 300 mmol citrate buffer (pH 3.5–6.0), 50 mmol of benzaldehyde,

100 mmol potassium cyanide and 100 ml of the enzyme in a final volume of 1.0 mL.

The reaction was started by an addition of 100 ml of the enzyme solution and incubated

at 25 �C for 1–120 min. Aliquots (100 ml) were withdrawn at various reaction times and

the reaction was stopped by the addition of 0.9 mL of organic solvent (9:1 hexane:iso-

propanol by volume). The mandelonitrile formed was extracted and the supernatant,

obtained by centrifugation (15, 000� g, 1.0 min at 4 �C), was assayed by HPLC. A

blank reaction was also performed without enzyme and the amount of mandelonitrile

obtained was deducted from the biocatalyzed reaction product. One unit of the enzyme

is defined as the amount of the enzyme that produces 1 mmol of (R)-mandelonitrile

under reaction conditions in 1 min.

8.5.2 Procedure 2: Synthesis of (R)-Mandelonitrile and Other Chiral

Cyanohydrins


• Aldehyde or ketone (5 g) (4-nitrobenzaldehyde, piperonal, naphthalene-2-carboxalde-

hyde, 2-furan carboxyaldehyde, 2-thiophene carboxyaldehyde, propanal, butanal, piva-

laldehyde or cyclohexanecarboxaldehyde)

• solvent: di-isopropyl ether (DIPE), tert-butyl methyl ether (TBME) or di-n-butyl ether

(DBE) (50 mL)

• EtOAc (50 mL)

• citrate buffer, pH 4 (5 mL)

• acetone cyanohydrin (1.5 equiv)

• NaHCO3




• filter paper

• separatory funnel


• gas chromatograph

• cyclodextrins column as chiral stationary phase (fused-silica capillary column,

30 m� 0.25 mm� 0.25 mm thickness, �-Dex-120 and �-Dex-325 from Supelco, USA).

8.5.2.2 Procedure

1. Carbonyl compounds (5 g) were taken in appropriate solvent (DIPE, TBME or DBE,

50 mL) saturated with 5 mL of citrate buffer (pH 4.0). A solution of partially purified

(R)-HNL from kernels of P. mume (50 U mmol�1 of substrate) followed by acetone-

cyanohydrin (1.5 equiv) was added to the reaction mixture.

2. The reaction was vigorously stirred at room temperature for several hours to a few days

until the desired conversion was achieved (by TLC).

3. The reaction mixture was extracted three times with 50 mL ethyl acetate. The organic

layer was dried over anhydrous Na2SO4 and concentrated by evaporation in vacuo.

After the residue had been dried, optically active cyanohydrin was obtained, as shown

in Tables 8.3 and 8.4. More examples are available.2,3

Table 8.3 HNL-catalyzed asymmetric synthesis of cyanohydrins with aromatic aldehydes

Entry Aldehyde Product

Yield (%) Ee (%)

1 Benzaldehyde 65 952 4-Nitrobenzaldehyde 93 713 Piperonal 88 974 Naphthalene-2-carboxaldehyde 78 965 2-Furan carboxyaldehyde 65 966 2-Thiophene carboxyaldehyde 82 88

Table 8.4 HNL-catalyzed asymmetric synthesis of cyanohydrins with aliphatic aldehydes andmethyl ketones

Entry Aldehyde or ketone Product

Yield (%) Ee (%)

1 Propanal 68 942 Butanal 58 903 Pivalaldehyde 52 964 Cyclohexanecaroboxaldehyde 72 935 2-Pentanone 60 726 4-Methyl-2-pentanone 56 887 5-Methyl-2-hexanone 49 65

8.5 A New (R)-Hydroxynitrile Lyase from Prunus mume 271

All aldehydes used in the experiment were freshly distilled or washed with aqueous

NaHCO3 solution to minimize the amount of free acid. Chiral HPLC was performed using

a chiral OJ-H column (0.46 cm� 25 cm, Daicel industries) with a water 717 auto sampler

and a UV–vis detector (254 nm). The eluting solvent used was different ratios of hexane

and 2-propanol. Chiral gas chromatography analysis was performed in a Shimadzu auto

sampler with cyclodextrins columns as chiral stationary phase (fused-silica capillary

column, 30 m� 0.25 mm� 0.25 mm thickness, �-Dex-120 and �-Dex-325 from Supelco,

USA) using He as a carrier gas (detector temperature 230 �C and injection temperature

220 �C).

8.5.3 Conclusion

We have found a new (R)-hydroxynitrile lyase from Japanese apricot (P. mume). The new

enzyme accepts a broad array of substrates, ranging from aromatic, heteroaromatic,

bicyclic to aliphatic carbonyl compounds, and yields the corresponding cyanohydrins

with excellent enantioselection.

References

1. Asano, Y., Tamura, K., Doi, N., Ueatrongchit, T., H-Kittikun,A. and Ohmiya, T., Screening fornew hydroxynitrilases from plants. Biosci. Biotech. Biochem., 2005, 69, 2349.

2. Nanda, A., Kato, Y. and Asano, Y., A new (R)-hydroxy-nitrile lyase from Prunus mume:asymmetric synthesis of cyanohydrins. Tetrahedron Lett., 2005, 61, 10908.

3. Nanda, A., Kato, Y. and Asano, Y., PmHNL catalyzed synthesis of (R)-cyanohydrins derivedfrom aliphatic aldehydes. Tetrahedron: Asymm., 2006, 17, 735.

4. Bradford, M.M., A rapid and sensitive method for the quantitation of microgram quantities ofprotein utilizing the principle of protein–dye binding. Anal. Biochem. 1976, 72, 248.


9

Synthesis of Chiral sec-Alcoholsby Ketone Reduction

9.1 Asymmetric Synthesis of (S)-Bis(trifluoromethyl)phenylethanol byBiocatalytic Reduction of Bis(trifluoromethyl)acetophenoneDavid Pollard, Matthew Truppo and Jeffrey Moore

The chiral compounds (R)- and (S)-bis(trifluoromethyl)phenylethanol are particularly

useful synthetic intermediates for the pharmaceutical industry, as the alcohol functionality

can be easily transformed without a loss of stereospecificity and biological activity, and the

trifluoromethyl functionalities slow the degradation of the compound by human metabo-

lism. A very efficient process was recently demonstrated for the production of the

(S)-enantiomer at >99% ee through ketone reduction catalyzed by the commercially

available isolated alcohol dehydrogenase enzyme from Rhodococcus erythropolis

(Figure 9.1).1 The (R)-enantiomer could be generated at>99% ee as well using the isolated

ketone reductase enzyme KRED-101.

9.1.1 Procedure 1: Preparation of (S)-Bis(trifluoromethyl)phenylethanol


• Nicotinamide adenine dinucleotide (NADþ, 40 g)

• glucose (60 g)

• alcohol dehydrogenase from R. erythropolis ADH-RE (Codexis Inc, 3.4 g)

• glucose dehydrogenase GDH-103 (Codexis Inc, 3.1 g)



• 50 mM potassium phosphate buffer pH 7.2 (10 L)

• 3,5-bistrifluoromethylphenyl ketone (1 kg)

• reaction vessel with temperature and pH control

• 2 M sodium hydroxide

• heptane (10 L)

9.1.1.2 Procedure

1. NADþ (40 g), glucose (60 g), alcohol dehydrogenase ADH-RE (3.4 g) and glucose

dehydrogenase GDH-103 (3.1 g) were added to 50 mM potassium phosphate buffer

pH 7.2 (10 L) that was stirring at 45 �C. The reaction was started with the addition of

ketone substrate (1 kg) and aged for 24 h while maintaining a pH of 6.5 through the

addition of 2 M sodium hydroxide.

2. The reaction was extracted twice with 5 L heptane. The organic layers were then

combined, washed with 2.5 L water and evaporated by distillation until the alcohol

product concentration was 200 g L�1. The solution was cooled to 35 �C and seeded with

1 g alcohol product prior to aging for 1 h and then cooling again to�10 �C. The alcohol

crystallized into a solid with >99% purity.

1H NMR: � 7.85 (s, 2H), 7.80 (s, 1H), 5.05 (qd, J¼ 6.5, 3.3, 1H), 2.04 (d, J¼ 3.3, 1H),

1.56 (d, J¼ 6.5, 3H). 13C NMR: � 148.44, 131.99 (q, J¼ 33.2), 125.87 (br q, J¼ 2.8),

123.58 (q, J¼ 272.6), 121.53 (septet, J¼ 3.9), 69.31, 25.79.

Chiral analysis for ee determination was by normal-phase high-performance liquid

chromatography with a Chiralcel OD-H column using 98 % hexanes/2 % 2-propanol at

1 mL min�1, 25 �C and monitoring at 265 nm.

9.1.2 Conclusion

This novel biocatalytic method for the production of (S)-bis(trifluoromethyl)phenyletha-

nol was easily and reproducibly demonstrated up to pilot plant scale in reactions generat-

ing 25 kg of >99% ee material. Substrate concentrations were increased up to 580 mM,

ADH-RE

GDH-103glucose

F3C

CF3

O

F3C

CF3

OH

gluconic acid

NADH NAD+ >99% ee>98% yield

580 mM

Figure 9.1 Production of the (S)-bis-(trifluoromethyl)phenylethanol using alcohol dehydro-genase enzyme from R. erythropolis

274 Synthesis of Chiral sec-Alcohols by Ketone Reduction

resulting in a space–time yield of 260 g L�1 day�1. Additionally, enantiocomplementary

results were obtained by using an identical procedure with the commercially available

isolated ketoreductase KRED-101 (Biocatalytics) in place of alcohol dehydrogenase

ADH-RE.

Reference

1. Pollard, D., Truppo, M., Pollard, J., Chen, C. and Moore, J. , Effective synthesis of (S)-3,5-bistrifluoromethylphenyl ethanol by asymmetric enzymatic reduction. Tetrahedron Asymm.,2006, 17, 554–559.

9.1 Asymmetric Synthesis of (S)-Bis(trifluoromethyl)phenylethanol 275

9.2 Enantioselective and Diastereoselective Enzyme-catalyzed DynamicKinetic Resolution of an Unsaturated KetoneBirgit Kosjek, David Tellers and Jeffrey Moore

Whole-cell cultures and isolated enzymes have been shown to be very useful in catalyzing

highly chemoselective reductions of �,�-unsaturated ketones. The presence of an addi-

tional racemic centre in a ketone substrate for this reduction has a strong potential for

decreasing the overall yield of the reaction by introducing a competing directing effect on

the enzyme. To reduce such a compound effectively and efficiently, a biocatalytic process

was developed that incorporates a racemization step to increase the theoretical yield of

enantiomerically pure product to 100 %.1 This process was used to generate allylic alcohol

with an enantioselectivity of 95 % ee and a diastereoselectivity of 99 % de (Figure 9.2).

9.2.1 Procedure 1: Ketoreductase Reduction of Ketone 1


• Ketoreductase KRED-104 (Codexis Inc, 120 mg)

• ketoreductase KRED-108 (Codexis Inc, 30 mg)

• nicotinamide adenine dinucleotide phosphate (NADPþ, 30 mg)

• 0.5 M potassium phosphate buffer pH 6.5 (9.5 mL)

• ketone substrate (100 mg)

• isopropanol (0.5 mL)

• ethyl acetate (10 mL).

9.2.1.2 Procedure

1. Ketoreductase enzymes KRED-104 (120 mg) and KRED-108 (30 mg) and NADPþ

(30 mg) were added to 0.5 M potassium phosphate buffer pH 6.5 (9.5 mL) that was

stirring at 35 �C. The reaction was started with the addition of a solution of ketone

substrate (100 mg) in isopropanol (0.5 mL) and aged for 12 h.

O

CO2Me

CO2Me

OH

CO2Me

CO2Me

OHO

1 2NADPH NADP+

99% de95% ee94% yield

KRED-108

KRED-104

Figure 9.2 Enantioselective and diastereoselective reduction of a,�-unsaturated ketones


2. The reaction was extracted with an equal volume of ethyl acetate and the organic layer

was evaporated, resulting in isolation of the product 2 at a yield of 94 %, a de of 99% cis,

and an ee of 95 % in favour of the S-enantiomer.

1H NMR (399.9 MHz; acetonitrile-d3, 27 �C) � 6.95 (s, 1H), 4.30 (m, 1H), 3.65 (s, 3H),

3.42 (m, 1H), 2.45 (m, 2H), 2.35 (m, 2H), 2.1 (m, 2H) 1.98 (m, 1H). 13C NMR (125 MHz,

tetrahydrofuran-d8, 27 �C) �¼ 24.8, 28.9, 41.0, 52.7, 52.8, 66.3, 130.0, 144.4, 167.6, 175.1

ppm.

Conversion and diastereomeric excess were determined on an Agilent HPLC system

using a Zorbax eclipse XDB C18 column (4.6 mm � 150 mm) at a gradient from 35/65

MeCN/water (0.1 % H3PO4) to 95/5 over 14 min at 1 mL min�1, room temperature,

210 nm.

Enantiomeric excess was determined with a Berger SFC system employing a tandem

Chiralpak OD (250 mm � 4.6 mm)–Chiralpak OB (250 mm � 4.6 mm), isocratic 3 %

2-propanol/CO2 at 2 mL min�1, 200 bar, 35 �C, 30 min. Alternatively, product enantio-

meric excess could be measured by chiral gas chromatography: Agilent GC system, Varian

Chiralsil-Dex Cb (25 m � 0.32 mm, 0.25 mm film thickness) ramp from 70 �C to 190 �C at

2 �C min�1, ramp to 200 �C at 1 �C min�1, hold for 10 min, average velocity 39 cm s�1.

9.2.2 Conclusion

Whereas the use of chemical catalysts to reduce this unsaturated ketone does not afford any

diastereoselective discrimination, the biocatalytic method described here generates pro-

duct that is almost exclusively the cis diastereomer at a very high ee of 95 %. Important

process improvements included optimizing the ester moiety of the starting material, after it

was found to have a significant impact on the observed enantioselectivity, and the inclu-

sion of isopropanol in the reaction mixture to serve both as the hydrogen source for the

recycling of the cofactor NADPH and as a cosolvent for increasing the solubility of the

ketone substrate.

Reference

1. Kosjek, B., Tellers, D.M., Biba, M., Farr, R. and Moore, J.C., Biocatalytic and chemocatalyticapproaches to the highly stereoselective 1,2-reduction of an �,�-unsaturated ketone.Tetrahedron Asymm., 2006, 17, 2798–2803.

9.2 Enzyme-catalyzed Dynamic Kinetic Resolution of an Unsaturated Ketone 277

9.3 Enzyme-catalysed Synthesis of a-Alkyl-b-hydroxy Ketones and Estersby Isolated KetoreductasesIoulia Smonou and Dimitris Kalaitzakis

Usingisolatedenzymesascatalysts fororganic reactions isbecomingamorestandardizedand

practical tool in the hands of organic chemists.1 The biocatalytic reduction of �-alkyl-1,3-

diketones and�-alkyl-�-keto esters employing commercially available reduced nicotinamide

adeninedinucleotidephosphate (NADPH)-dependentketoreductases (KREDs)proved tobea

highlyefficientmethodfor thepreparationofopticallypureketoalcohols,1,3-diolsorhydroxy

esters.2,3 These enzymatic reactions provide a simple, highly stereoselective and quantitative

methodfor the synthesis ofdifferent stereoisomersofvaluable chiral synthons fromnonchiral,

easily accessible 1,3-diketones or keto esters (Figure 9.3). Chiral keto alcohols and diols

represent very useful synthons in organic synthesis and have been used as precursors in the

synthesis of various biologically active compounds4,5 and pharmaceuticals.

9.3.1 Procedure 1: Synthesis of (3R,4S)-3-Allyl-4-hydroxy-2-pentanone

OOH


• 200 mM phosphate buffer solution, pH 6.9 (100 mL)

• 3-allyl-2,4-pentanedione (700 mg, 50 mmol)

• glucose (2.16 g, 120 mmol)

• NADPH (45 mg)

• glucose dehydrogenase (50 mg)

• KRED-102 (50 mg) (Codexis Inc.)

• NaOH solution (2 M)


12

34

O O

KRED-102

NADPH

KRED-A1B

NADPH

KRED-108

NADPH

OH O

3R,4S > 99%ee, > 99%de

12

34

OH O

3S,4R > 99%ee, > 98%de

12

34

OH O

3S,4S > 99%ee, > 98%de

Figure 9.3 Enzyme-catalysed stereoselective reduction of 3-allyl-2,4-pentanedione


• saturated NaCl solution (70 mL)


• pH meter

• one-necked reaction flask equipped with magnetic stirring bar, 250 mL

• magnetic stirring plate



9.3.1.2 Procedure

1. A 200 mM phosphate-buffered solution, pH 6.9 (100 mL), containing 3-allyl-

2, 4-pentanedione (700 mg, 50 mmol), glucose (2.16 g, 120 mmol), NADPH

(45 mg), glucose dehydrogenase (50 mg) and KRED-102 (50 mg) was stirred at

room temperature for 24 h, until gas chromatography (GC) analysis of the crude

extracts showed complete reaction. Periodically, the pH was readjusted to 6.9 with

NaOH (2 M).

2. The product (3R,4S)-3-allyl-4-hydroxy-2-pentanone was isolated by extracting the

crude reaction mixture with EtOAc (2 � 100 mL). The combined organic layers were

then extracted with saturated NaCl solution (70 mL), dried over MgSO4 and evapo-

rated to dryness to afford optically active (3R,4S)-3-allyl-4-hydroxy-2-pentanone

(617 mg, 87 %).

1H NMR (CDCl3; 500 MHz) � 5.74–5.82 (m, 1H), 5.01–5.11 (m, 2H), 4.01–4.07

(m, 1H), 2.64–2.68 (m, 1H), 2.39–2.42 (m, 2H), 2.18 (s, 3H), 1.18 (d, J¼ 6.5 Hz, 3H).

The optical purity was determined by chiral GC, using a 20 % permethylated cyclodex-

trin column, after esterification of the pure product with (CF3CO)2O in dry CH2Cl2 (65 �C

isothermal; carrier gas: N2, pressure 70 kPa). TR¼ 26.305 min (>99%, (3R,4S)-3-allyl-4-

hydroxy-2-pentanone). The enantiomeric purity was estimated to be >99 % and the

diastereomeric purity >99 %.

9.3.2 Procedure 2: Synthesis of (3S,4R)-3-Allyl-4-hydroxy-2-pentanone

OOH





• NADPH (45 mg)


• KRED-A1B (50 mg) (Codexis Inc.)



9.3 Synthesis of a-Alkyl-�-hydroxy Ketones and Esters by Isolated KREDS 279



• pH meter





9.3.2.2 Procedure

1. A 200 mM phosphate-buffered solution, pH 6.9 (100 mL), containing 3-allyl-2,

4-pentanedione (700 mg, 50 mmol), glucose (2.16 g, 120 mmol), NADPH (45 mg), glucose

dehydrogenase (50 mg) and KRED-A1B (50 mg) was stirred at room temperature for 8 h,

until GC analysis of crude extracts showed complete reaction. Periodically, the pH was

readjusted to 6.9 with NaOH (2 M).

2. Product (3S,4R)-3-allyl-4-hydroxy-2-pentanone was isolated by extracting the

crude reaction mixture with EtOAc (2� 100 mL). The combined organic layers

were then extracted with saturated NaCl solution (70 mL), dried over MgSO4 and

evaporated to dryness to afford optically active (3S,4R)-3-allyl-4-hydroxy-2-pen-

tanone (604 mg, 85 %).


(m, 1H), 2.64–2.68 (m, 1H), 2.39–2.43 (m, 2H), 2.18 (s, 3H), 1.18 (d, J¼ 6.5 Hz, 3H).

The optical purity was determined by chiral GC, using a 20 % permethylated

cyclodextrin column, after esterification of the pure product with (CF3CO)2O in dry

CH2Cl2 (65 �C isothermal; carrier gas: N2, pressure 70 kPa). TR¼ 27.604 min

(99 %, (3R,4S)-3-allyl-4-hydroxy-2-pentanone), TR¼ 28.776 min (1 %, (3R,4R)-3-

allyl-4-hydroxy-2-pentanone). The enantiomeric purity was estimated to be >99 %

and the diastereomeric purity 98 %.

9.3.3 Procedure 3: Synthesis of (3S,4S)-3-Allyl-4-hydroxy-2-pentanone

OOH





• NADPH (45 mg)


• KRED-108 (70 mg) (Codexis Inc.)






• pH meter



• filter paper



9.3.3.2 Procedure

1. A 200 mM phosphate-buffered solution, pH 6.9 (100 mL), containing 3-allyl-2, 4-pen-

tanedione (700 mg, 50 mmol), glucose (2.16 g, 120 mmol), NADPH (45 mg), glucose

dehydrogenase (50 mg) and KRED-108 (70 mg) was stirred at room temperature for

24 h, until GC analysis of crude extracts showed complete reaction. Periodically, the

pH was readjusted to 6.9 with NaOH (2 M).

2. Product (3S,4S)-3-allyl-4-hydroxy-2-pentanone was isolated by extracting the crude

reaction mixture with EtOAc (2 � 100 mL). The combined organic layers were then

extracted with saturated NaCl solution (70 mL), dried over MgSO4 and evaporated to

dryness to afford optically active (3S,4S)-3-allyl-4-hydroxy-2-pentanone (614 mg,

86 %).


(m, 1H), 2.60–2.65 (m, 1H), 2.34–2.37 (m, 2H), 2.19 (s, 3H), 1.23 (d, J¼ 6 Hz, 3H).

The optical purity was determined by chiral GC, using a 20 % permethylated cyclodex-

trin column, after esterification of the pure product with (CF3CO)2O in dry CH2Cl2 (65 �C

isothermal; carrier gas: N2, pressure 70 kPa). TR¼ 26.739 min (1 %, (3R,4S)-3-allyl-4-

hydroxy-2-pentanone), TR¼ 29.864 min (99 %, (3S,4S)-3-allyl-4-hydroxy-2-pentanone). The

enantiomeric purity was estimated to be >99 % and the diastereomeric purity 98 %.

R1 R2

O O

R3 R4

INADPH

NADP+ II

R1 R2

OH

R3 R 4

O

D-GlucoseGlucono-lactone

GDH

KRED

**

Figure 9.4 Enzymatic reduction of a-alkyl-1,3-diketones with NADPH-dependent KREDs


9.3.4 Conclusion

The enzymatic transformation of a large number of �-monoalkyl and dialkyl symmetrical

and nonsymmetrical 1,3-diketones and keto esters (Figure 9.4) shows excellent chemical

and optical yield and can be tailored to afford most of the four possible diastereomers

from the same starting substrate at will, depending on the chosen enzyme. Besides being

regio- and stereo-selective, these enzymes exhibited high chemoselectivity by giving a

keto alcohol or hydroxy ester and not the diol.

Table 9.1 shows some examples of the different substrates that can be reduced to various

single diastereomers of the same compound, as the keto alcohols and hydroxy esters, by

choosing the use of different enzymes. The chemoenzymatic syntheses of the aggregation

pheromones (þ)-Sitophilure and Sitophilate by the use of the above isolated, NADPH-

dependent KREDs were successfully accomplished by our group4,5 with high chemical

and optical purities (98 % de, >99 % ee).

Table 9.1 Enzyme-catalysed stereoselective reduction of diketones/keto esters to ketoalcohols/hydroxy esters

Entry R1 R2 R3 R4 KRED Product yield (%) Conversion(%) [time]

AR-Alkyl,S-OH

BS-Alkyl,S-OH

CS-Alkyl,R-OH

DR-Alkyl,R-OH

1 Me Me Me H 102 >993R,4S

— — — >99 [24 h]

2 Me Me Me H 127 3 943S,4S

— 3 90 [24 h]

3 Me Me Et H 102 >993R,4S

— — — >99 [12 h]

4 Me Me Et H A1B — — 953S,4R

5 >99 [1 h]

5 Me Me Et H 118 — >983S,4S

— — >99 [24 h]

6 Et Et Me H A1B — 974S,5R

3 >99 [40 min]

7 Et Et Me H 119 <1 >994S,5S

— — >99 [12 h]

8 Me Et Me H 102 >992S,3R

— — — >99 [24 h]

9 Me Et Me H 127 3 962S,3S

— 1 92 [24 h]

10 Me Me Me Allyl 101 983R,4S

— 2 — >99 [1 h]

11 Me Me Me Allyl A1B — — 10 903R,4R

>99 [24 h]

12 Me Me Me Allyl 118 — >993S,4S

— — >99 [6 h]

13 Me OEt Me H 102 >992R,3S

— — — >99 [24 h]

14 Me OEt Me H 107 — 15 — 852S,3S

>99 [6 h]


References

1. Faber, K., Biotransformations in Organic Chemistry. 1997, Springer-Verlag, Berlin, pp. 160–206.

2. Kalaitzakis, D., Rozzell, J.D., Kambourakis, and S. Smonou, I., Highly stereoselective reduc-tions of �-alkyl-1,3-diketones and �-alkyl-�-keto esters catalyzed by isolated NADPH-depen-dent ketoreductases. Org. Lett., 2005, 7, 4799–4801.

3. Kalaitzakis, D., Rozzell, J.D., Kambourakis, S. and Smonou, I., Synthesis of valuable chiralintermediates by isolated ketoreductases: application in the synthesis of -alkyl--hydroxy ketonesand 1,3-diols. Adv. Synth. Catal., 2006, 348, 1958–1969.

4. Kalaitzakis, D., Rozzell, J.D., Kambourakis, S. and Smonou, I., A two-step chemoenzymaticsynthesis of the natural pheromone (þ)-Sitophilure utilizing isolated, NADPH-dependent ketor-eductases. Eur. J. Org. Chem., 2006, 2309–2313.

5. Kalaitzakis, D., Kambourakis, S., Rozzell, J.D. and Smonou, I., Stereoselective chemoenzy-matic synthesis of sitophilate: a natural pheromone. Tetrahedron Asymm. 2007, 18, 2418–2426.


9.4 Asymmetric Reduction of Phenyl Ring-containing Ketones UsingXerogel-encapsulated W110A Secondary Alcohol Dehydrogenase fromThermoanaerobacter ethanolicusMusa M. Musa, Karla I. Ziegelmann-Fjeld, Claire Vieille, J. Gregory Zeikus and

Robert S. Phillipsa

There has been a growing interest in using enzymes for asymmetric transformations

of unnatural organic compounds in organic solvents.1 Recently, we have used xerogel-

immobilized W110A mutant secondary alcohol dehydrogenase from

Thermoanaerobacter ethanolicus (W110A TeSADH) to reduce a series of phenyl ring-

containing ketones to the corresponding (S)-alcohols in good yields and high optical

purities in organic solvents (Figure 9.5).2 The resulting alcohols have (S)-configuration,

in agreement with Prelog’s rules, in which the reduced nicotinamide adenine dinu-

cleotide phosphate (NADPH) cofactor transfers its pro-R hydride to the re face of the

ketone.

9.4.1 Procedure 1: Preparation of Xerogel-encapsulated W110A TeSADH


• Tetramethyl orthosilicate (TMOS, 2.10 g)

• distilled water (0.47 g)

• HCl (0.04 m, three drops)

• W110A TeSADH (0.43 mg)

• NADPþ (3.0 mg, 3.6 mmol)

• tris-HCl buffer (50 mM, pH 7.0, 1.0 mL)

• one 10 mL round-bottomed flask

• sonifier.

9.4.1.2 Procedure

1. The silica sol was prepared by mixing TMOS (2.10 g), distilled water (0.47 g) and HCl

(0.04 M, three drops). The mixture was then sonicated until one layer was formed.

R

O

O

NADPH NADP+

OH

O

OH

R

OH

na (S )-nb

xerogel W110A TeSADH

6a

(S)-6b

Hexane

xerogel W110A TeSADH

R = Ph(CH2)2, PhOCH2, p-MeOC6H4(CH2)2, PhCH2, p-MeOC6H4CH2

Figure 9.5 Reduction of ketones with W110A secondary alcohol dehydrogenase


2. The gels were prepared by mixing the above sol (1.0 mL) with enzyme stock (1.0 mL)

in a 10 mL round-bottomed flask. The enzyme stock was prepared in 50 mM tris-HCl

buffer (pH 8.0) such that the concentration of the enzyme, expressed and purified as

described previously,3 was 0.43 mg mL�1, and that of NADPþ was 3.0 mg mL�1. The

sol–gel was then left in the same flask sealed with Parafilm at room temperature for 48 h

to allow gel to age.

3. The hydrogel was dried at room temperature in air for 24 h to give hydrated silica,

SiO2�nH2O, the so-called xerogel.

9.4.2 Procedure 2: Asymmetric Reduction Using Xerogel-encapsulated W110A

TeSADH in Organic Solvents


• Ketone substrate (0.34 mmol)

• 2-propanol (600 mL)

• hexane (2 mL)



• pyridine

• acetic anhydride

• silica gel (60 A, 32–63 mm)

• one 10 mL round-bottomed flask equipped with a magnetic stirrer

• hot and magnetic stirrer plate

• filter paper



9.4.2.2 Procedure

1. All reactions were performed using W110A TeSADH (0.43 mg) and NADPþ (3.0 mg,

3.6 mmol) encapsulated in sol–gel, substrate (0.34 mmol), 2-propanol (600 mL), and

2.0 mL of hexane in a 10 mL round-bottomed flask equipped with a magnetic stirrer.

The reaction mixture was stirred at 50 �C for 12 h.

2. The sol–gel was then removed by filtration and washed with ethyl acetate (2 � 2 mL).

The combined organic filtrates were dried with Na2SO4 and then concentrated under

vacuum.

3. The remaining residue was analyzed by gas chromatography (GC) to determine the

yield, then purified by silica-gel column chromatography (eluent: ethyl acetate: hexane,

15:85).

The product alcohol was then converted to the corresponding acetate derivative.4 The ee

was determined by GC equipped with a flame-ionization detector and a Supelco �-Dex

120 chiral column (30 m, 0.25 mm (internal diameter), 0.25 mm film thickness) by using

He as the carrier gas. The injector temperature was 250 �C and the detector temperature

was 300 �C. The flow rate was 19.0 psi. The column was programmed between 120 �C and

170 �C.

9.4 Asymmetric Reduction of Phenyl Ring-containing Ketones 285

9.4.3 Conclusion

This method allows the asymmetric reduction of hydrophobic ketones in high yields and

enantioselectivities (Table 9.2). It is a facile method, not only for making the enzyme

accessible to a wide variety of water-insoluble substrates by switching the traditional

aqueous medium to organic media, but also for reusing the enzyme. This method allows for

the use of high concentrations of substrate and catalytic quantities of cofactor, both of

which are crucial for large-scale synthetic applications. Reusable catalysts for chemo-,

regio-, and enantio-selective asymmetric reduction may be of industrial interest.

Table 9.2 Asymmetric reduction of phenyl ring-containing ketones by TeSADH usingProcedure 2

Entry R Product

Yield (%) Ee (%)

1a Ph(CH2)2 74 972b PhOCH2 >99 >993c p-MeOC6H4(CH2)2 61 944d PhCH2 80 695e p-MeOC6H4CH2 67 >996f 2-Tetralol (see Figure 9.1) 94 76

a(S)-4-Phenyl-2-butanol: [�]D20¼þ16.5 (c¼ 1.81, CHCl3), >99 % ee, lit.5 [�]20

D¼þ17.4 c¼ 1.80, CHCl3), 99 % ee.Spectral data were consistent with that reported previously.6b(S)-Phenoxy-2-propanol: [�]D

20¼þ30.7 (c¼1.32, CHCl3), >99 % ee, lit.7 [�]D20¼þ28.9 c¼1.10, CHCl3), 99 % ee.

Spectral data were consistent with that reported previously.8c(S)-4-(4-Methoxyphenyl)-2-butanol: [�]D

20¼þ12.8 (c¼ 2.41, CHCl3), >91 % ee, lit.9 [�]D20¼þ30.9 c¼1.0, CHCl3), 94

% ee. Spectral data were consistent with that reported previously.9d(S)-1-Phenyl-2-propanol: [�]D

20¼þ14.5 (c¼1.04, CHCl3), >37 % ee, lit.10 [�]D25¼þ42.2 c¼ 1.0, CHCl3), >99 % ee.

Spectral data were consistent with that reported previously.11

e(S)-4-(4-Methoxyphenyl)-2-propanol: [�]D20¼þ16.3 (c¼ 1.86, CHCl3), >99 % ee, lit.12 [�]D

20¼þ27.0 c¼ 4.40, CHCl3),95 % ee. Spectral data were consistent with that reported previously.10

f(S)-2-Tetralol: [�]D20¼�43.77 (c¼ 0.911, CHCl3),>71 % ee, lit.13 [�]D

20¼�29.6 c¼ 0.50, CHCl3), 85 % ee. Spectral datawere consistent with that reported previously.14

References

1. Faber, K., Biotransformations in Organic Chemistry, 5th edn. Springer: Heidelberg, 2004.2. Musa, M., Ziegelman-Fjeld, K., Vieille, C., Zeikus, J. and Phillips, R., Xerogel-encapsulated

W110A secondary alcohol dehydrogenase from Thermoanaerobacter ethanolicus performsasymmetric reduction of hydrophobic ketones in organic solvents. Angew. Chem. Int. Ed.,2007, 46, 3091–3094.

3. Ziegelman-Fjeld, K., Musa, M., Phillips, R., Zeikus, J. and Vieille, C., A Thermoanaerobacterethanolicus secondary alcohol dehydrogenase mutant derivative highly active and stereoselec-tive on phenylacetone and benzylacetone. Protein Eng. Des. Sel., 2007, 20, 47–55.

4. Ghanem, A. and Schuring, V., Lipase-catalyzed access to enantiomerically pure (R)- and(S)-trans-4-phenyl-3-butene-2-ol. Tetrahedron Asymm., 2003, 14, 57–62.

5. Nakamura, K., Inoue, Y., Matsuda, T. and Misawa, I., Stereoselective oxidation and reduction byimmobilized Geotrichum candidum in an organic solvent. J. Chem. Soc. Perkin Trans. 1, 1999,2397–2402.


6. Kuwano, R., Uemura, T., Saitoh, M. and Ito, Y., A trans-chelating bisphosphine possessing onlyplanar chirality and its application to catalytic asymmetric reactions. Tetrahedron Asymm.,2004,15, 2263–2271.

7. Nakamura, K., Takenaka, K., Fujii, M. and Ida, Y., Asymmetric synthesis of both enantiomers ofsecondary alcohols by reduction with a single microbe. Tetrahedron Lett., 2002, 43, 3629–3631.

8. Dragovich, P. S., Prins, T. J. and Zhou, R., Palladium catalyzed, regioselective reduction of1,2-epoxides by ammonium formate. J. Org. Chem., 1995, 60, 4922–4924.

9. Donzelli, F., Fuganti, C., Mendozza, M., Pedrocchi-Fantoni, G., Servi, S. and Zucchi, G., On thestereochemistry of the Baeyer–Villiger degradation of arylalkylketones structurally related toraspberry ketone by Beauveria bassiana. Tetrahedron Asymm., 1996, 7, 3129–3134.

10. Erdelyi, B., Szabo, A., Seres, G., Birincsik, L., Ivanics, J., Szatzker, G. and Poppe, L.,Stereoselective production of (S)-1-aralkyl- and 1-arylethanols by freshly harvested and lyophi-lized yeast cells. Tetrahedron Asymm., 2006, 17, 268–274.

11. Ley, S. V., Mitchell, C., Pears, D., Ramarao, C., Yu, J. and Zhou, W., Recyclable polyurea-microencapsulated Pd(0) nanoparticles: an efficient catalyst for hydrogenolysis of epoxides.Org. Lett., 2003, 5, 4665–4668.

12. Ferraboschi, P., Grisenti, P., Manzocchi, A. and Santaniello, E., Baker’s yeast-mediated pre-paration of optically active aryl alcohols and diols for the synthesis of chiral hydroxy acids.J. Chem. Soc. Perkin Trans. 1, 1990, 2469–2474.

13. Stampfer, W., Kosjek, B., Faber, K. and Kroutil, W., Biocatalytic asymmetric hydrogen transferemploying Rhodococcus ruber DSM 44541. J. Org. Chem., 2003, 68, 402–406.

14. Orsini, F., Sello, G., Travaini, E. and Di Gennaro, P., A chemoenzymatic synthesis of (2R)-8-substituted-2-aminotetralins. Tetrahedron Asymm., 2002, 13, 253–259.

9.4 Asymmetric Reduction of Phenyl Ring-containing Ketones 287

9.5 (R)- and (S)-Enantioselective Diaryl Methanol Synthesis UsingEnzymatic Reduction of Diaryl KetonesMatthew Truppo, Krista Morley, David Pollard and Paul Devine

The asymmetric formation of industrially useful diaryl methanols can be realized through

either the addition of aryl nucleophiles to aromatic aldehydes or the reduction of diaryl

ketones.1 The latter route is frequently the more desirable, as the starting materials are often

inexpensive and readily available and nonselective background reactions are not as com-

mon. For good enantioselectivity, chemical catalysts of diaryl ketone reductions require

large steric or electronic differentiation between the two aryl components of the substrate

and, as a result, have substantially limited applicability.2,3 In contrast, recent work has shown

commercially available ketoreductase enzymes to have excellent results with a much

broader range of substrates in reactions that are very easy to operate (Figure 9.6).4

9.5.1 Procedure 1: General Procedure for the Ketoreductase Reduction of Diaryl

Ketones

9.5.1.1 Materials and equipment

• Ketone substrate (1 g)

• glucose (800 mg)


• 0.1 M potassium phosphate buffer pH 7 (36 mL)

• tetrahydrofuran (THF, 4 mL)

• ketoreductase enzyme (Codexis Inc, 80 mg)

• glucose dehydrogenase enzyme (Codexis Inc, 80 mg)

• 2-butanone (80 mL)

• nitrogen gas

• flask (100 mL)



9.5.1.2 Procedure

1. Glucose (800 mg), NADPþ (40 mg), ketone substrate example (1 g) and THF (4 mL)

were added to 0.1 M potassium phosphate buffer pH 7 (36 mL) that was stirring at 30 �C.

O

Ar1 Ar2

ketoreductase OH

Ar1 Ar2*

NADPH NADP+

glucosedehydrogenase

glucosegluconolactone

O

Ar1 Ar2

ketoreductase OH

Ar1 Ar2*

NADPH NADP+

ketoreductase

OHO

Figure 9.6 Asymmetric reduction of diaryl ketones with ketoreductases


The reaction was started with the addition of ketoreductase (80 mg) and glucose

dehydrogenase (80 mg) enzymes.

2. The reaction was extracted with 2-butanone (80 mL) that was then washed twice with 5 mL

water. The organic layer was evaporated under nitrogen, yielding the alcohol product.

Table 9.3 Reduction of various diaryl ketones with ketoreductases.

Ketone (R)-Alcohol (S)-Alcohol

Ee (%) Ketoreductasea Ee (%) Ketoreductasea

R1¼ o-CH3 98 121 95 119R1¼m-CH3 99 CDX P1H10 92 CDX P2C12R1¼ p-CH3 99 CDX P1H10 9 119R1¼m-NO2 34 111 99 108R1¼ p-NO2 99 CDX P1H10 97 119R1¼ o-OH 84 111R1¼m-OH 82 CDX P1H10 13 119R1¼ p-OH 96 CDX P1H10 55 117R1¼ o-NH2 91 101 64 114R1¼ p-OMe 60 111 51 119R1¼ p-NO2 70 101 64 119R1¼ o-Cl 64 121 99 118R1¼m-Cl 97 CDX P1H10 99 108R1¼ p-Cl 99 CDX P1H10R1¼m-CNR2¼ p-Cl

84 112 90 108

R1¼m-CO2MeR2¼ p-Cl

33 115 99 108

97 101 77 119

82 101 38 120

44 Lactobacillus kefir 99 119

94 124 60 119

a Ketoreductases identified by numbers refer to enzymes commercially available from Codexis (Pasadena, CA – formerlyBiocatalytics) and those with CDX prefixes refer to enzymes obtained under license from Codexis (Redwood City, CA –CodexTM KRED Panel v 1.0).

O

R1 R2

NO

N

O

N

O

N

O

Cl

9.5 (R)- and (S)-Enantioselective Diaryl Methanol Synthesis 289

9.5.2 Procedure 2: Screening of Ketoreductase Enzymes


• Ketone

• THF

• NADPþ

• 0.1 M potassium phosphate buffer, pH 7

• aqueous glucose solution

• ketoreductase enzyme (Codexis Inc)

• glucose dehydrogenase enzyme (Codexis Inc)

• methyl tert-butyl ether (MTBE)

• 96-well plate

• chiral analytical method.

9.5.2.2 Procedure

1. To rapidly screen libraries of ketoreductase enzymes in parallel against ketone starting

materials of interest, a substrate solution containing 20 mg mL�1 ketone in THF, a

cofactor solution containing 5 mg mL�1 NADPþ in 0.1 M potassium phosphate buffer

pH 7, and a glucose solution containing 20 mg mL�1 glucose in water were prepared.

2. For each Biocatalytics (Codexis Pasadena) ketoreductase enzyme, 50 mL each of the

substrate, cofactor and glucose solutions were added to 350 mL 0.1 M potassium

phosphate buffer pH 7, 1 mg ketoreductase and 1 Mg glucose dehydrogenase enzymes

in one location of a 96-well plate.

3. For each Codexis ketoreductase enzyme, 50 mL each of the substrate and cofactor

solutions were added to 300 mL isopropanol, 100 mL 0.1 M potassium phosphate buffer

pH 7 and 1 mg ketoreductase in one location of a 96-well plate.

4. After aging the reactions for 24 h at 30 �C, they were each extracted with 1 mL MTBE

for analysis using a suitable chiral method.

9.5.3 Conclusion

A large number of diaryl ketone substrates, including those listed in Table 9.3, have been

reduced with high enantioselectivity with the protocol described here. Unlike analogous

chemical catalysts, the commercially available biocatalysts displayed no dependence on

ortho substitutions or electronic dissymmetry, and produced diaryl methanols with good to

excellent ee values in nearly all cases.

References

1. Devaux-Basseguy, R., Bergel, A. and Comtat, M., Potential applications of NAD(P)-dependentoxidoreductases in synthesis: a survey. Enzyme Microb. Technol., 1997, 20, 248.

2. Welch, C. J., Grau, B., Moore, J. and Mathre, D., Use of chiral HPLC–MS for rapid evaluation ofthe yeast-mediated enantioselective bioreduction of a diaryl ketone. J. Org. Chem., 2001, 66,6836.

3. Itsuno, S. Enantioselective reduction of ketones. In Organic Reactions, vol. 52, Paquette, L.A.(ed.), John Wiley & Sons, Inc.: New York, 1998, pp. 395–576.

4. Truppo, M. D., Pollard, D. and Devine, P., Enzyme-catalyzed enantioselective diaryl ketonereductions. Org. Lett., 2007, 9, 335.


9.6 Highly Enantioselective and Efficient Synthesis of Methyl (R)-o-Chloromandelate, Key Intermediate for Clopidogrel Synthesis, withRecombinant Escherichia coliTadashi Ema, Nobuyasu Okita, Sayaka Ide and Takashi Sakai

Clopidogrel is a platelet aggregation inhibitor widely administered to atherosclerotic

patients with the risk of a heart attack or stroke that is caused by the formation of a clot

in the blood. Worldwide sales of Plavix (clopidogrel bisulfate) amounted to $6.4 billion

per year (data for the 12 months ending June 2006), which ranks second to Lipitor for

sales.1 We have recently found that methyl (R)-o-chloromandelate ((R)-1), which is a key

intermediate for clopidogrel synthesis, can be obtained in >99 % ee by the asymmetric

reduction of methyl o-chlorobenzoylformate (2) (up to 1.0 M) with recombinant

Escherichia coli overproducing a versatile carbonyl reductase called SCR

(Saccharomyces cerevisiae carbonyl reductase) together with a glucose dehydrogenase

(GDH).2 A remarkable temperature effect on productivity was observed in the whole-cell

reduction of 2, and the optimum productivity as high as 178 g L�1 was attained at 20 �C

(Scheme 9.1).

9.6.1 Procedure 1: Cultivation of Recombinant E. coli


• Ampicillin (250 mg)

• chloramphenicol (85 mg)

• E. coli BL21(DE3) cells harboring pESCR and pABGD

• Luria–Bertani (LB) medium: tryptone (25 g), yeast extract (13 g), NaCl (25 g)

• isopropyl-�-D-thiogalactopyranoside (IPTG, 60 mg)

• Milli-Q water (2.5 L)

Cl

Cl

CO2Me

NADPH

gluconolactone

gluconic acid

glucoseGDH

SCR

recombinant E. coli

NADP+

Cl

(R)-12

CO2Me

O

clopidogrel

OH

N

S

CO2Me

Scheme 9.1

9.6 Enantioselective and Efficient Synthesis of Methyl (R)-o-Chloromandelate 291

• 0.1 M phosphate buffer (200 mL)

• test tube (� 8)

• rotary shaker

• sterile toothpicks

• 1 L Erlenmeyer flask (� 8)

• measuring cylinder

• cotton plug

• autoclave


• centrifuge.

9.6.1.2 Procedure

1. E. coli BL21(DE3) cells harboring pESCR and pABGD, previously constructed,3 were

grown in LB medium (3 mL � 8) containing ampicillin (100 mg mL�1) and chloram-

phenicol (34 mg mL�1) at 37 �C for 15 h with shaking at 230 rpm.

2. The culture (3 mL � 8) was transferred to the same medium (300 mL � 8) in a 1 L

Erlenmeyer flask and shaken at 200 rpm at 37 �C.

3. IPTG (0.1 mM) was added when optical density at 600 nm reached 0.6–0.8. The cells

were further incubated at 28 �C for 18 h with shaking at 200 rpm and then harvested by

centrifugation (7000 rpm, 4 �C, 10 min) into four portions.

4. Each of the four portions was washed with 0.1 M phosphate buffer (pH 7.0, 50 mL).

5. The wet cell pellet (7–8 g) obtained was stored at �20 �C until it was used for

asymmetric reduction.

9.6.2 Procedure 2: Synthesis of Methyl (R)-o-Chloromandelate ((R)-1)

Cl Cl OH

CO2Me

(R)-1

O

CO2Me

NADP+, glucose, buffer

recombinant E. coli

2


• Cells prepared above (2 g)

• cell pellet of recombinant E. coli (2.0 g)

• methyl o-chlorobenzoylformate (2) (1.98 g)

• D-glucose (3.6 g)


• 0.1 M phosphate buffer (pH 7.0, 10 mL)

• 2 M NaOH (5 mL)

• NaCl (5.5 g)

• MgSO4 (1 g)

• silica gel

• hexane


• ethyl acetate

• 100 mL test tube (2.7 cm diameter)

• water bath with a thermostat


• vortex mixer

• pH indicator paper

• centrifuge




• 30 mL round-bottom flask.

9.6.2.2 Procedure

1. To a mixture of glucose (3.60 g, 20.0 mmol), NADPþ (10 mg, 12 mmol), and E. coli

BL21(DE3) cells harboring pESCR and pABGD (2.0 g) in 0.1 M phosphate buffer (pH

7.0, 10 mL) in a 100 mL test tube was added methyl o-chlorobenzoylformate (2)

(1.98 g, 10.0 mmol).

2. The mixture was stirred in a water bath at 20 �C for 24 h, during which 2 M NaOH was

added to maintain pH 7 by neutralizing the acid formed in the progress of the

reaction.

3. Solid NaCl (5.5 g) was added and the product was extracted with EtOAc (25 mL

� 3). Phase separation was effected by centrifugation (3200 rpm, 10 min). The

combined organic layers were dried over MgSO4, filtered and concentrated under

reduced pressure. Purification by silica-gel column chromatography (hexane/

EtOAc (10:1)) gave methyl (R)-o-chloromandelate ((R)-1) as a colorless oil

(1.78 g, 89 %).

[�] 19D¼�178.3 (c¼ 1.3, CHCl3), >99 % ee, (R). High-performance liquid chromato-

graphy (HPLC): Chiralpak AD-H (Daicel Chemical Industries, Ltd), hexane/i-PrOH (9:1),

flow rate 0.5 mL min�1, detection 254 nm, (S) 20.3 min, (R) 22.7 min. 1H NMR (CDCl3,

600 MHz) � 3.56 (d, J¼ 5.4 Hz, 1H), 3.78 (s, 3H), 5.57 (d, J¼ 5.4 Hz, 1H), 7.28–7.29 (m,

2H), 7.39–7.40 (m, 2H). 13C NMR (CDCl3, 150 MHz) � 53.2, 70.3, 127.2, 128.8, 129.8,

130.0, 133.5, 135.9, 173.7. IR (film) 3454, 3003, 2955, 1744, 1441, 1223, 1090, 756 cm�1.

9.6.3 Conclusion

An efficient and green chemoenzymatic method for methyl (R)-o-chloromandelate ((R)-1)

has been developed. The asymmetric reduction of methyl o-chlorobenzoylformate (2)

with recombinant E. coli overproducing a versatile carbonyl reductase, SCR, gave (R)-1

with >99% ee. This is the first example of the direct asymmetric synthesis of (R)-1 with

>99 % ee. A remarkable temperature effect on productivity was observed in the whole-cell

reduction of 2, and the optimum productivity as high as 178 g L�1 was attained at 20 �C

(Table 9.4). The bioreduction of 2 is a green process, because the hydride source is glucose,

which is a cheap biomass-derived reagent, and because the E. coli catalyst can be multiplied

easily and inexpensively. Moreover, the bioreduction is performed in an aqueous solution

under air.

9.6 Enantioselective and Efficient Synthesis of Methyl (R)-o-Chloromandelate 293

References

1. Grimley, J., Pharma challenged. Chem. Eng. News, 2006, Dec. 4, 17–28.2. Ema, T., Okita, N., Ide, S., Sakai, T., Highly enantioselective and efficient synthesis of methyl

(R)-o-chloromandelate with recombinant E. coli: toward practical and green access to clopido-grel. Org. Biomol. Chem., 2007, 5, 1175–1176.

3. Ema, T., Yagasaki, H., Okita, N., Takeda, M., Sakai, T., Asymmetric reduction of ketones usingrecombinant E. coli cells that produce a versatile carbonyl reductase with high enantioselectivityand broad substrate specificity. Tetrahedron, 2006, 62, 6143–6149.

Table 9.4 Asymmetric reduction of 2 with recombinant E. coli.a

Entry [2] (M) [2] (g L�1) T (�C) C (%)b Yield (%)c Ee (%)d

1 0.3 60 30 92 76 >992 0.3 60 25 >99 88 >993 0.6 120 25 94 88 >994 1.0 198 25 90 85 >995 1.0 198 20 99 89 >996 1.0 198 15 86 82 >99

a Conditions: 2 (0.60–1.98 g, 3.0–10.0 mmol), wet cells of E. coli BL21(DE3) harboring pESCR and pABGD (2.0 g), glucose(2 equiv), NADPþ (10 mg, 12 mmol), 0.1 M phosphate buffer (pH 7.0, 10 mL).

b Conversion determined by 1H NMR.c Isolated yield of (R)-1.d Determined by HPLC (Chiralpak AD-H, hexane/i-PrOH (9:1)).


10

Reduction of Functional Groups

10.1 Reduction of Carboxylic Acids by Carboxylic AcidReductase Heterologously Expressed in Escherichia coliAndrew S. Lamm, Arshdeep Khare and John P.N. Rosazza*

The biocatalytic reduction of carboxylic acids to their respective aldehydes or alcohols is a

relatively new biocatalytic process with the potential to replace conventional chemical

processes that use toxic metal catalysts and noxious reagents. An enzyme known as

carboxylic acid reductase (Car) from Nocardia sp. NRRL 5646 was cloned into

Escherichia coli BL21(DE3).1–7 This E. coli based biocatalyst grows faster, expresses

Car, and produces fewer side products than Nocardia. Although the enzyme itself can be

used in small-scale reactions, whole E. coli cells containing Car and the natural cofactors

ATP and NADPH, Hþ are easily used to reduce a wide range of carboxylic acids,

conceivably at any scale. The biocatalytic reduction of vanillic acid to the commercially

valuable product vanillin is used to illustrate the ease and efficiency of the recombinant

Car E. coli reduction system.4 A comprehensive overview is given in Reference 6, and

experimental details below are taken primarily from Reference 7.

10.1.1 Biocatalytic Synthesis of Vanillin

OH

OMe

OH

OH

OMe

HOO

OH

OMe

HO

Vanillic acidVanillin

VanillylAlcohol

Car, ATP, NADPH Aldehyde reductase

Mg+2




• E. coli BL21(DE3) harboring plasmid pPV2.85 (frozen glycerol stocks)

• Luria–Bertani (LB) broth powder (20 g L�1)

• LB agar powder (15 g L�1)

• ampicillin (100 mg mL�1 stock solution in water, filter sterilized)

• high-performance liquid chromatography (HPLC)-grade acetonitrile (800 mL)

• HPLC-grade water (200 mL)

• HPLC-grade formic acid (1 mL)

• sodium dihydrogen phosphate (12g.L-1)

• sodium hydrogen carbonate (5 g)

• sodium vanillate stock solution (50 mg mL�1)

• vanillyl alcohol (1 g as HPLC or thin-layer chromatography (TLC) standard)

• vanillin (1 g as HPLC or TLC standard)

• 0.22 mm polyvinylidene difluoride syringe filters

• 10 mL and 1 mL syringes

• sterile loop

• Petri dish

• two 125 ml and two 1 L stainless-steel-capped DeLong flasks

• rotary shakers at 37 �C

• centrifuge capable of reaching 5000g while holding 4 �C

• HPLC system and UV detection

• Econosil HPLC column (C18, 5 mm, 150 mm � 3.2 mm; Alltech).

Optional

• Silica gel TLC plates (silica gel 60 F254, Merck)

• 30 % w/v phosphomolybdic acid in ethanol (100 mL)

• reagent spray bottle

• heat gun

• UV lamp/viewing box

• benzoic acid (50 mg mL�1 in water)

• 3-chlorobenzoic acid (50 mg mL�1 in water)

• 4-chlorobenzoic acid (50 mg mL�1 in water)

• 3-(4-hydroxy-3-methoxyphenyl)-propenoic acid (50 mg mL�1 in water).

10.1.1.2 Procedure

Initial culture

1. Crystals from a frozen glycerol stock of E. coli BL21(DE3)/pPV2.85 were streaked

onto LB agar plates with ampicillin (100 mg mL�1) to obtain single colonies.

2. Single colonies were inoculated into 20 mL of LB medium (containing 100 mg mL�1

ampicillin) in 125 mL stainless-steel-capped DeLong flasks.

3. Cultures were incubated with shaking at 250 rpm on a rotary shaker at 37 �C. A 1 %

inoculum derived from 8 h stage I cultures was used to initiate fresh LB cultures (200

mL) with antibiotics in a 1 L DeLong flask. These cultures were incubated at 37 �C for

16 h with shaking at 250 rpm.

4. Car-containing E. coli cells were pelleted by centrifugation at 5000g for 6 min at

4 �C.

296 Reduction of Functional Groups

Whole-cell Carboxylic Acid Reduction

1. Car-containing E. coli cells were resuspended in 200 mL of 0.9 % (w/v) NaCl and

pelleted once again by centrifugation at 5000g for 6 min at 4 �C.

2. A sodium vanillate stock solution (50 mg mL�1) was prepared by dissolving equimolar

amounts of vanillic acid and NaHCO3 in 0.1 M Na2HPO4 (pH 7). Reaction mixtures of

50 mL contained 0.4 % glucose, 1.5 g of wet E. coli cells, 200 mg of sodium vanillate in

pH 7, 0.1 M Na2HPO4.

3. Reactions were incubated at 28 �C with shaking at 220 rpm and 1 mL samples were

withdrawn at various time intervals for analysis. Samples are treated as described in

Section 10.1.1.3.

10.1.1.3 Analytical Methods

Standard solutions were prepared by dissolving weighed amounts of compounds in a 1:1

(v/v) mixture of pH 7, 0.1 M Na2HPO4/acetonitrile. Aliquots of 0.5 mL of biotransforma-

tion samples were mixed with 0.5 mL of acetonitrile and mixtures were vortexed for 30 s.

After standing at room temperature for 30 min, samples were microcentrifuged at 20 000g

for 3 min, the supernatants filtered through 0.22 mm polyvinylidene difluoride syringe

filters and 1–2 ml injected for HPLC analysis.

The HPLC system used a mobile phase consisting of CH3CN/H2O/HCOOH (20:80:1, v/

v/v). Quantitation of standards and samples was achieved by isocratic elution over a C18, 5

mm, Econosil HPLC column at a flow rate of 0.4 mL min�1. HPLC retention volumes and

detection wavelengths for standards were as follows: vanillyl alcohol, 1.7 mL and 277 nm;

vanillic acid, 2.5 mL and 260 nm; and vanillin, 4.1 mL and 284 nm.

TLC analysis of samples was conducted on silica-gel plates carefully spotted with 10–

20 mg of standard compounds, and 30 mL of bioconversion reaction samples. Plates

developed with 75:25:1 (v/v/v) CH2Cl2/CH3CN/HCOOH solvent may be visualized

with a 254 nm UV lamp and/or by spraying with a 30 % w/v phosphomolybdic acid/

95% ethanol spray reagent followed by gentle heating. Rf values of standards are: vanillyl

alcohol, 0.8; vanillic acid, 0.5; and vanillin 0.4.

Other suitable alternate aromatic carboxylic substrates include 3-chlorobenzoic acid, 4-

chlorobenzoic and 3-(4-hydroxy-3-methoxyphenyl)-propenoic acid.

10.1.2 Conclusion

Most of the vanillic acid was reduced by E. coli containing Car in 2 h to vanillin (80 %) and

vanillyl alcohol (20 %). Car does not reduce aldehydes to alcohols. However, E. coli’s

endogenous aldehyde reductase/dehydrogenase reduces vanillin to vanillyl alcohol. The

broad substrate specificity of Car enables the wide application of this biocatalyst to other

important applications, such as enantiomeric resolution of isomers such as ibuprofen1 and

the reductions of many other natural and synthetic carboxylic acids.

References

1. Chen, Y. and Rosazza, J.P.N., Microbial transformation of ibuprofen by a Nocardia species. Appl.Environ. Microbiol., 1994, 60, 1292–1296.

2. Li, T. and Rosazza, J.P.N. Purification, characterization, and properties of an aryl aldehydeoxidoreductase from Nocardia sp. strain NRRL 5646. J. Bacteriol., 1997, 179, 3482–3487.

10.1 Reduction of Carboxylic Acids by Carboxylic Acid Reductase 297

3. Li, T. and Rosazza, J.P.N., NMR Identification of an acyl-adenylate intermediate in the aryl-aldehyde oxidoreductase catalyzed reaction. J. Biol. Chem., 1998, 273, 34230–34233.

4. Li, T. and Rosazza, J.P.N., Biocatalytic synthesis of vanillin. Appl. Environ. Microbiol., 2000, 66,684–687.

5. He, A., Li, T., Daniels, L., Fotheringham, I. and Rosazza, J.P.N. Nocardia sp. carboxylic acidreductase: cloning, expression, and characterization of a new aldehyde oxidoreductase family.Appl. Environ. Microbiol., 1994, 70, 1874–1881.

6. Venkitasubramanian, P., Daniels, L. and Rosazza, J.P.N., Biocatalytic reduction of carboxylicacids: mechanism and application. In Biocatalysis in the Pharmaceutical and BiotechnologyIndustries, Patel, R. (ed). CRC Press LLC: Boca Raton, FL, 2006, pp. 425–440.

7. Venkitasubramanian, P., Daniels, L. and Rosazza, J.P.N., Reduction of carboxylic acids byNocardia aldehyde oxidoreductase requires a phosphopantetheinylated enzyme. J. Biol. Chem.,2007, 282, 478–485.


10.2 Light-driven Stereoselective Biocatalytic Oxidations and ReductionsAndreas Taglieber, Frank Schulz, Frank Hollmann, Monika Rusek

and Manfred T. Reetz*

Recently, the use of visible light to promote the direct reductive regeneration for flavin-

dependent enzymes has been proposed.1 The feasibility of this concept has been success-

fully demonstrated for stereoselective Baeyer–Villiger oxidations using an engineered

variant of phenylacetone monooxygenase (PAMO-P31–3) from Thermobifida fusca and for

the enantioselective reduction of ketoisophorone catalyzed by YqjM from Bacillus sub-

tilis.4 The light-driven regeneration system is based on the use of flavin cocatalysts that are

activated by white light and react from their excited state with a sacrificial electron donor

(ethylenediaminetetraacetic acid (EDTA)). In this reaction, the flavin cocatalyst is con-

verted into a reduced species which transfers the reducing equivalents to the enzyme-

bound flavin, thereby regenerating the reduced enzyme (Scheme 10.1).

By means of this reaction, the use of the costly and unstable natural redox cofactor

reduced nicotinamide adenine dinucleotide phosphate (NADPH) was circumvented and

the reactions were carried out in a straightforward procedure in a chemical laboratory

(Scheme 10.2, Table 10.1).

flavinred

EDTA

decompositionproducts

A

B flavinox

light YqjM

E-FMNred

E-FMNox

flavinred

EDTA

decompositionproducts

flavinox

light BVMO

E-FADred

E-FADox

R R'

O

R O

O

H2O

O2

R'

O

O

R1

R4R3

R2

R1H

R2

HR3 R4

Scheme 10.1 Light-driven regeneration of (A) a Baeyer–Villiger monooxygenase (BVMO)and (B) for the flavin-dependent reductase YqjM

O O

O O6 (R)-7

light, YqjM, FMN, EDTA

4 h, 30 °C100% conv.83% ee

Scheme 10.2 Light-driven YqjM-catalyzed reduction of ketoisophorone


10.2.1 Procedure 1: Expression and Purification of PAMO-P3


• Terrific broth5

• carbenicillin

• L-arabinose

• 5 L fermenter (Labfors HT, Infors)

• centrifuge

• sonicator

• flavine adenine dinucleotide (FAD)

• lysozyme (from chicken egg-white)

• Fractogel His-Bind resin (Novagen)

• 500 mL glass column (with glass frit at the bottom)

• imidazole

• centrifuge filters (Amicon, molecular weight cutoff (MWCO) 10 000)

• PD-10 desalting columns (GE Healthcare)

• Escherichia coli TOP10 [pPAMO-P3].3

10.2.1.2 Procedure

1. For the expression of PAMO-P3, 100 mL of an overnight preculture of E. coli TOP10

[pPAMO-P3] in terrific broth medium supplemented with 100 mg L�1 of carbenicillin

was used to inoculate 5 L terrific broth medium supplemented with 100 mg L�1 of

carbenicillin and 0.1 % L-arabinose in a 5 L fermenter. The expression was carried out

Table 10.1 Light-driven PAMO-P3-catalyzed Baeyer–Villiger oxidations.

rac-1

O

R O

O

R

O

R

(R)-2a R = C6H5b R = CH2C6H5

(S)-1

+

O

O

OO

+ O

rac -3 (–)-4 (–)-5

light, PAMO-P3FAD, EDTA

30 °C

light, PAMO-P3FAD, EDTA

30 °C

Ee(%) of productConversion(%)Substrate1a 79841b 79033 93 4: 92;5: ≥95


over the course of 7 h at 37 �C and constant stirrer revolution (800 rpm). The cells were

harvested by centrifugation (10 000g, 4 �C, 15 min) and frozen at �80 �C.

2. After thawing, the complete cell paste was suspended in 250 mL of 20 mM potassium

phosphate buffer (pH 7.4) containing 10 mM of FAD and 0.2 mg mL�1 of lysozyme and

incubated at 4 �C for 30 min.

3. Cells were disrupted by sonication and the lysate clarified by centrifugation (10 000g,

4 �C, 60 min). The supernatant was incubated at 50 �C for 1 h in a water bath and

subsequently centrifuged.

4. The supernatant was supplemented with NaCl to a concentration of 0.5 M NaCl and

mixed with 35 mL of Novagen Fractogel His-Bind resin (pre-equilibrated and loaded

with Ni2þ as recommended by the manufacturer). The suspension was gently mixed for

30 min and then manually loaded into a 500 mL glass column and packed under 1.5 bar

Ar pressure.

5. The material eluted was loaded onto the column once more; subsequently, the column

was washed with 350 mL of 20 mM KH2PO4 (pH 7.4) and then with 350 mL of 20 mM

potassium phosphate buffer (pH 7.4) supplemented with 1 mM imidazole. PAMO-P3

was eluted with 100 mL of 50 mM tris-HCl (pH 7.4) containing 200 mM imidazole. In

total, 25 mL of yellow eluate was collected and concentrated to 12.5 mL by centrifuge

filters (Amicon, MWCO 10 000).

6. The final purification step in each case was desalting of the eluates via PD-10 columns

(Amersham, 8.3 mL Sephadex G-25 medium) according to the recommendations of the

column manufacturer, using 50 mM tris-HCl (pH 7.4) as equilibration and elution

buffer. The concentration of purified enzyme in 50 mM tris-HCl (pH 7.4) was deter-

mined by the UV–vis absorbance at 441 nm ("441 nm¼ 12.4 mM�1 cm�1).6

10.2.2 Procedure 2: Light-driven PAMO-P3-catalyzed Baeyer–Villiger Oxidations


• PAMO-P3, purified enzyme

• NADPþ

• substrate (2-phenylcyclohexanone, 2-benzylcyclohexanone or bicyclo[3.2.0]hept-2-en-

6-one)

• EDTA

• flavin cocatalyst (FAD, flavin mononucleotide (FMN) or riboflavin)

• 100 W white-light bulb.

10.2.2.2 Procedure

1. A final reaction volume of 250 mL, containing 10 mM PAMO-P3, 25 mM EDTA, 100 mM

FAD, 250 mM NADPþ, 1 mM or 2 mM substrate and 50 mM tris-HCl (pH 7.4), was

incubated under aerobic conditions at 30 �C in a water bath exposed to light from a

100 W Osram� white-light bulb. The light was filtered through 1 cm of water and�0.5

cm of DURANTM glass. The approximate distance between light source and the

reaction vessels was 6 cm. The reaction mixture was extracted with 275 mL of ethyl

acetate each and analyzed by gas chromatography (GC). The reaction took around 6 h

and was followed by GC analysis using authentic standards.


For all experiments, PAMO-P3 was produced and purified as described above to

allow for accurate quantification of the results. However, the reaction also works using

crude enzyme as obtained after bacterial lysis.2

10.2.2.3 Characterization of the Products

For all products, synthetic standards were prepared by meta-chloroperbenzoic acid-

mediated Baeyer–Villiger oxidation.7 The crude products were purified by silica-gel

chromatography and the NMR spectra matched the data reported in the literature (see

below). Stereochemical configurations were assigned based on analogous conversions

carried out using cyclohexanone monooxygenase. All GC results were confirmed by GC–

mass spectrometry (MS) analysis (instrument: Finnigan SSQ7000; GC–electron impact

(EI), achiral GC methods described below).

10.2.2.4 GC Analyses

Compound 2a8

Achiral method. Instrument: Agilent Technologies 6890N; carrier gas: 0.6 bar H2; column:

15 m ZB1 (100 % dimethylpolysiloxane, 0.25 mm inner diameter, 0.5 mm film); injector

T¼ 220 �C, detector T¼ 350 �C; program: ramp 80 �C to 195 �C with 8 �C min�1, then

20 �C min�1 to 340 �C; retention times: 7.96 min (1a), 10.85 min (2a). GC-factor

correction was performed versus n-C16 standard; correction factor: 1.15.

Chiral method. Instrument: Agilent Technologies 6890N; carrier gas: 0.6 bar H2;

column: 30 m BGB-176 (20 % 2,3-dimethyl-6-tert-butyldimethylsilyl-�-cyclodextrin dis-

solved in BGB-15, 0.25 mm inner diameter, 0.1 mm film); injector T¼ 220 �C, detector

T¼ 350 �C; program: 150 �C (iso) 10.5 min, ramp to 160 �C with 50 �C min�1, 160 �C (iso)

16 min; retention times: (S)-1a, 9.56 min; (R)-1a, 9.77 min; (S)-2a, 21.20 min; (R)-2a,

21.51 min.

Compound 2b8




320 �C (iso) 1 min; retention times: 9.59 min (1b), 12.46 min (2b).

Chiral method. Instrument: Agilent Technologies 6890N; carrier gas: 0.6 bar H2; column:

30 m BGB-176 (20 % 2,3-dimethyl-6-tert-butyldimethylsilyl-�-cyclodextrin dissolved in

BGB-15, 0.25 mm inner diameter, 0.1 mm film); injector T¼ 220 �C, detector T¼ 350 �C,

program: 150 �C (iso) 12.5 min, ramp to 200 �C with 100 �C min�1, 200 �C (iso) 5 min;

retention times: (R)-1b, 11.97 min; (S)-1b, 11.72 min; (R)-2b, 17.12 min; (S)-2b, 16.86 min.

Compounds 4 and 59,10


30 m RTX-5 (5 % diphenylpolysiloxane, 95 % dimethylpolysiloxane, 0.25 mm inner

diameter, 0.25 mm film); injector T¼ 220 �C, detector T¼ 350 �C; program: ramp 60 �C

to 330 �C with 6 �C min�1, then 350 �C (iso) 10 min, retention times: 4.60 min (3), 9.79

min (4), 9.91 min (5).


Chiral method. Agilent Technologies 6890N; carrier gas: 0.7 bar H2; column: 30 m

BGB-178 (20 % 2,3-diethyl-6-tert-butyldimethylsilyl-�-cyclodextrin dissolved in BGB-15,

0.25 mm inner diameter, 0.25 mm film); injector T¼ 220 �C, detector T¼ 350 �C; program:

125 �C (iso) 14.5 min, ramp to 230 �C with 10 �C min-1, 230 �C (iso) 5 min; retention times:

10.61 min (�4), 11.30 min (þ5), 11.98 min (�5), 12.50 min (þ4).

10.2.3 Procedure 3: Expression of YqjM


• IPTG

• E. coli Rosetta (DE3) [pET21a-YqjM]11

• terrific broth medium5

• FMN

• carbenicillin

• 5 L fermenter (Labfors HT, Infors)

• centrifuge

• sonicator

• PD-10 desalting columns (GE Healthcare).

10.2.3.2 Procedure

Expression of YqjM was carried out in a 5 L fermenter using terrific broth medium

supplemented with 100 mg L�1 carbenicillin. As an inoculum, 100 mL of preculture was

used. Temperature and stirrer speed were kept constant (800 rpm, 37 �C). At an optical

density at 600 nm of 0.67, 500 mL of 1 M IPTG was added and the temperature adjusted to

30 �C. After 4 h of expression, the cells were harvested by centrifugation and the

resulting cell paste frozen at �80 �C overnight. After thawing, the complete cell paste

was suspended in 100 mL of 50 mM tris-HCl buffer (pH 7.4). Cell lysis was achieved by

sonication. DNAse I (0.1 mg mL�1) was added and the crude lysate incubated at 4 �C for

30 min. The lysate was clarified by centrifugation (10 000g, 4 �C, 60 min). After the

reconstitution of the enzyme with FMN (1 h incubation at 4 �C in the presence of 100 mM

FMN) followed by removal of excess FMN using PD-10 desalting columns (GE

Healthcare) according to the recommendations of the column manufacturer with

50 mM tris-HCl (pH 7.4) as equilibration and elution buffer, the protein was analyzed

by sodium dodecyl sulfate polyacrylamide gel electrophoresis using a 12.5 % gel. Using

the densitograph function of BioDocAnalyze (Biometra), the YqjM content was deter-

mined to be 32 % of the total protein content of the protein preparation. The total protein

content was determined using a Bradford assay reagent (Bio-Rad) with bovine serum

albumin as standard.12

10.2.4 Procedure 4: Light-driven YqjM-catalyzed Reduction of Ketoisophorone


• YqjM

• ketoisphorone


• EDTA

• FMN

• 100 W white-light bulb

10.2.4.2 Procedure

1. A final reaction volume of 250 mL, containing 2.6 mM YqjM (corresponds to approxi-

mately 0.1 mg mL�1 total protein content), 25 mM EDTA, 100 mM FMN, 1 mM ketoi-

sophorone and 50 mM tris-HCl (pH 7.4), was incubated under anaerobic conditions

(closed reaction vessel with small head space) at 30 �C in a water bath exposed to the

light of a 100 W Osram� white-light bulb for 4 h. The experimental setup was the same

as described in Procedure 2 (Section 10.2.2).

2. Reactions were followed and analyzed by GC. For GC analysis, samples were extracted

with an equal volume of ethyl acetate and analyzed. Peak assignment was performed

with an authentic standard of the product and confirmed by GC–MS (instrument:

Finnigan SSQ7000, GC–EI, achiral GC method described below).

10.2.4.3 GC-Analyses




20 �C min-1 to 340 �C; retention times: 2.83 min (6), 3.05 min (7).

Chiral method. Agilent Technologies 6890N; carrier gas: 0.6 bar H2; column: 30 m

BGB176 (20 % 2,3-dimethyl-6-tert-butyldimethylsilyl-�-cyclodextrin dissolved in BGB-

15, 0.25 mm inner diameter, 0.1 mm film); injector T¼ 220 �C, detector T¼ 350 �C;

program: 100 �C (iso) 15 min; retention times: 12.11 min (6); 12.35 min ((R)-7) and 13.90

min ((S)-7).

10.2.5 Conclusion

The straightforward concept for the direct light-driven regeneration of flavin-dependent

enzymes has been successfully applied for two representative classes of such enzymes: a

reductase and a monooxygenase. Therefore, it can be expected that this concept can also be

applied to other flavin-dependent enzymes, potentially leading to additional practical

catalyst systems for applications in synthetic organic chemistry.

References

1. Hollmann, F., Taglieber, A., Schulz, F. and Reetz, M.T., A light-driven stereoselective bioca-talytic oxidation. Angew. Chem. Int. Ed., 2007, 46, 2903.

2. Schulz, F., Leca, F., Hollmann, F. and Reetz, M.T., Towards practical biocatalytic Baeyer–Villiger reactions: applying a thermostable enzyme in the gram-scale synthesis of optically-active lactones in a two-liquid-phase system. Beilstein J. Org. Chem., 2005, 1, 10.

3. Bocola, M., Schulz, F., Leca, F., Vogel, A., Fraaije, M.W. and Reetz, M.T., Convertingphenylacetone monooxygenase into phenylcyclohexanone monooxygenase by rational design:towards practical Baeyer–Villiger monooxygenases. Adv. Synth. Catal., 2005, 347, 979.


4. Taglieber, A., Schulz, F., Hollmann, F., Rusek, M. and Reetz, M.T., Light-driven biocatalyticoxidation and reduction reactions: scope and limitations. ChemBioChem, 2008, 9, 565.

5. Sambrook, J. and Russel, D., Molecular Cloning: A Laboratory Manual, 3rd edn. Cold SpringHarbor Laboratory Press, New York, 2000.

6. Fraaije, M.W., Wu, J., Heuts, D.P.H.M., van Hellemond, E.W., Spelberg, J.H.L. and Janssen,D.B., Discovery of a thermostable Baeyer–Villiger monooxygenase by genome mining. Appl.Microbiol. Biotechnol. 2005, 66, 393.

7. Krow, G.R., The Baeyer–Villiger oxidation of ketones and aldehydes. In Organic Reactions,vol. 43, Paquette, L.A. (ed.). John Wiley & Sons, Inc.: New York, 1993, pp. 251–798.

8. Alphand, V., Furstoss, R., Pedragosa-Moreau, S., Roberts, S.M. and Willetts, A.J., Comparisonof microbiologically and enzymatically mediated Baeyer–Villiger oxidations: synthesis ofoptically active caprolactones. J. Chem. Soc. Perkin Trans. 1, 1996, 1867.

9. Hudlicky, T., Reddy, D.B., Govindan, S.V., Kulp, T., Still, B. and Sheth, J.P., Intramolecularcyclopentene annulation. 3. Synthesis and carbon-13 nuclear magnetic resonance spectroscopyof bicyclic cyclopentene lactones as potential perhydroazulene and/or monoterpene synthons. J.Org. Chem., 1983, 48, 3422.

10. Grieco, P.A., Cyclopentenones. Efficient synthesis of cis-jasmone. J. Org. Chem., 1972, 37,2363.

11. Fitzpatrick, T.B., Amrhein, N. and Macheroux, P., Characterization of YqjM, an old yellowenzyme homolog from Bacillus subtilis involved in the oxidative stress response. J. Biol. Chem.,2003, 278, 19891.

12. Bradford, M.M., A rapid and sensitive method for the quantitation of microgram quantities ofprotein utilizing the principle of protein–dye binding. Anal. Biochem., 1976, 72, 248.


10.3 Unnatural Amino Acids by Enzymatic Transamination: Synthesis ofGlutamic Acid Analogues with Aspartate AminotransferaseThierry Gefflaut*, Emmanuelle Sagot and Jean Bolte

Aminotransferases (ATs) catalyse the stereoselective transfer of the amino group from a

donor substrate to an acceptor prochiral carbonyl derivative. ATs are very common

enzymes with high specific activities and relaxed substrate specificity. The development

of equilibrium shifted transamination processes allowed the preparation of a variety of

biologically active compounds, including unnatural L- and D-�-amino acids1,2 as well as �-

aminoacids or simple amines.3–6 Aspartate aminotransferase (AspAT) offers the opportu-

nity to shift the transamination equilibrium through the use of cysteine sulfinic acid (CSA)

as the amino donor substrate: CSA, which is a close analogue of aspartic acid, is converted

into the very unstable pyruvyl sulfinic acid, which spontaneously decomposes into pyruvic

acid and sulfur dioxide, thus providing the equilibrium shift. Recently, AspAT has proven

useful for the stereoselective preparation of a variety of neuroactive glutamic acid deri-

vatives.7–10 This methodology is exemplified below with the preparation of (2S,4R)-4-

methyl Glu (2), a potent selective ligand for kainate receptors: AspAT gives exclusively

the L-amino acid and allows the kinetic resolution of the racemic �-keto acid substrate 1

readily prepared by conventional chemical methods.7 This catalyst thus offers the stereo-

control of two asymmetric centres.

10.3.1 Procedure 1: Synthesis of (2S,4R)-4-Methyl Glutamic Acid

HO

O

O

O

OH HO

O O

OHNH2

HO

O

OHO S

O

NH2

OH

O

HO

O

O

O

OH

AspAT

CSA

SO2

+

+

(2S,4R)-2rac-1

H2O, pH7.6


• 4-Methyl-2-oxoglutaric acid (1)7 (0.5 g, 2.9 mmol)

• CSA (0.45 g, 2.9 mmol)

• acetaldehyde (128 mg, 2.9 mmol)

• AspAT (from pig heart, Sigma) (1 mg)

• KH2PO4 (1.36 g, 10 mmol)

• KOH

• lactate dehydrogenase (from rabbit muscle, Sigma) (0.1 mg mL�1)

• reduced nicotinamide adenine dinucleotide (NADH, 10 mg mL�1)

• Dowex 50WX8 (200–400 mesh, Hþ form) (20 g)

• Dowex 1X8 (200–400 mesh, AcO� form) (20 g)

• 1 M NH4OH (100 mL)


• 1 M AcOH (200 mL)

• ninhydrin (0.2 g in 100 mL EtOH) (for thin-layer chromatography (TLC) dip)

• propan-1-ol (50 mL) (for TLC elution)

• pH meter

• 250 mL flask


• 1 mL adjustable volume pipette

• 20 mL adjustable volume pipette

• 1.5 mL microcentrifuge tubes (for enzyme solutions)

• centrifuge (for microtubes)

• 1.5 mL disposable cuvettes (for spectrophotometry at 340 nm)

• UV or visible spectrophotometer

• equipment for column chromatography (column: 2 cm � 20 cm, tubes 5–10 mL)

• TLC plates (silica gel 60F254, Merck)

• rotatory evaporator.

10.3.1.2 Procedure

1. In a 250 mL flask equipped with a stir bar were introduced 4-methyl-2-oxoglutaric

acid 1 (0.5 g, 2.9 mmol), CSA (0.45 g, 2.9 mmol), water (145 mL) and acetaldehyde

(128 mg, 2.9 mmol).11 The pH of the solution was adjusted to 7.6 with 1 M KOH

before the addition of pig heart AspAT (1 mg). The commercial enzyme suspension in

3 M (NH4)2SO4 was centrifuged (5 min at 10 000 rpm), the supernatant eliminated and

the enzyme pellet dissolved in the reaction mixture. The reaction was stirred slowly at

room temperature and monitored by titration of pyruvic acid formed from CSA.

2. The solutions needed for pyruvic acid titration were all prepared in 0.1 M potassium

phosphate buffer pH 7.6. In a disposable 1.5 mL cuvette were introduced a 10 mg mL-1

solution of NADH (20ml), a 0.1 mg mL�1 solution of rabbit muscle lactate dehydrogenase

(10mL, 1.2 unit) and phosphate buffer (965mL). The initial optical density (ODi £ 1.5) was

measured at 340 nm. An aliquot of the reaction mixture (5mL) was then added and the final

stable OD (ODf) was measured. Pyruvic acid concentration in the reaction mixture was

calculated using "NADH¼ 6220 M�1 cm�1: [Pyruvate]¼ (ODi�ODf) � 200/6220.

3. When a conversion of 40 % was reached (8 mM pyruvate formed in 2–3 h), the reaction

mixture was rapidly passed through a column of Dowex 50WX8 resin (Hþ form,

2 cm � 10 cm). The column was then washed with water (100 mL) until complete elution

of residual substrate 1, pyruvicacid and CSA. Itwas thenelutedwith1 M NH4OH(100mL).

The fractions (5mL) were analysedbyTLC (eluent n-PrOH/H2O,7:3,v/v). The ninhydrin-

positive fractions were combined and concentrated to dryness under reduced pressure. The

purity of the product was further increased by anion-exchange chromatography:

4. The residue was diluted in water (5 mL) and, if necessary, the pH adjusted to 7.0 with

1 M KOH before adsorption of the product on a column of Dowex 1X8 resin (200–400

mesh, AcO� form, 2 cm � 10 cm). The column was washed with water (50 mL) and

then eluted with AcOH aqueous solutions (50 mL of 0.1 M, 50 mL of 0.2 M and 50 mL of

0.5 M AcOH). The ninhydrin-positive fractions were combined and dried under reduced

pressure to afford (2S,4R)-4-methyl glutamic acid 2 isolated as a white solid (192 mg,

41 %) and with a high purity (>98 %).

10.3 Unnatural Amino Acids by Enzymatic Transamination 307

M.p. 178 �C; ½��25D ¼þ24:0� (c¼ 1.3, 6 M HCl). 1H NMR (400 MHz, D2O) � 3.83 (1H,

dd, J¼ 4.5 and 8.5 Hz), 2.58 (1H, m, J¼ 5.0, 7.0, 8.5 Hz), 2.23 (1H, ddd, J¼ 5.0, 8.5 and

14.0 Hz), 1.96 (1H, ddd, J¼ 5.0, 8.5, 13.5 Hz), 1.25 (3H, d, J¼ 7.0 Hz); 13C NMR (100

MHz, D2O) � 185.3, 178.5, 53.9, 39.5, 36.9, 17.8. Anal. (C6H11NO4) C, H, N: calc., 44.72,

6.88, 8.69; found, 44.60, 6.94, 8.64.

10.3.2 Conclusion

AspAT has been shown to display a broad substrate spectrum. This chemoenzymatic

procedure is, therefore, a very convenient way to prepare a variety of L-2,4-syn Glu

analogues substituted at the 4-position by alkyl7 or functionalized substituents.12

Moreover, this catalyst has been used for the preparation of 4,4-disubstituted10 and

(2S,3R)-3-methyl9 Glu derivatives, as well as the cyclobutane analogues LCBG

II–IV.8 The different Glu analogues prepared to date using this methodology are reported

in Figure 10.1.


1. Hwang, B.-Y., Cho, B.-K., Yun, H., Koteshwar, K. and Kim, B.-G., Revisit of aminotransferasein the genomic era and its application to biocatalysis. J. Mol. Catal. B: Enzym., 2005, 37, 47–55.

2. Ager, D.J., Li, T., Pantaleone, D.P., Senkpeil, R.F., Taylor, P.P. and Fotheringham, I.G., Novelbiosynthetic routes to non-proteinogenic amino acids as chiral pharmaceutical intermediates.J. Mol. Catal. B: Enzym., 2001, 11, 199–205.

3. Yun, H., Cho, B.-K. and Kim, B.-G., Kinetic resolution of (R,S)-sec-butylamine using omega-transaminase from Vibrio fluvialis JS17 under reduced pressure. Biotechnol. Bioeng., 2004, 87,772–778.

4. Shin, J.S. and Kim, B.G., Comparison of the o-transaminases from different microorganismsand application to production of chiral amines. Biosci. Biotechnol. Biochem., 2001, 65,1782–1788.

H2NHO

OO

HO

* *

HO

O

H2NOH

O

HO

O

H2NOH

O

Alk

HO

O

H2NOH

O

OHHO

O

H2NOH

O

OR HO

O

H2NOH

O

O

X

HO

O

H2NOH

O

OH

(3R,4R ) : L-CBG-II(3S,4R ) : L-CBG-III(3R,4S ) : L-CBG-IV

Alk = Me, Et, Pr, Bu, Pn iPr, iBu, iPn, Bn

n = 1,2X = OR, NHR

n

HO

O

H2NOH

O

Figure 10.1 Glu analogues prepared by AspAT-catalysed transamination


5. Iwasaki, A., Yamada, Y., Ikenaka, Y. and Hasegawa, J., Microbial synthesis of (R)- and (S)-3,4-dimethoxyamphetamines through stereoselective transamination. Biotechnol. Lett., 2003, 25,1843–1846.

6. Yun, H., Lim, S., Cho, B.-K. and Kim, B.-G., o-Amino acid:pyruvate transaminase fromAlcaligenes denitrificans Y2k-2: a new catalyst for kinetic resolution of �-amino acids andamines. Appl. Environ. Microbiol., 2004, 70, 2529–2534.

7. 1 was prepared in three steps from commercially available methyl 3-hydroxy-2-methylenebu-tyrate and triethyl orthoacetate (Aldrich): Alaux, S., Kusk, M., Sagot, E., Bolte, J., Jensen, A.A.,Brauner-Osborne, H., Gefflaut, T. and Bunch, L., Chemoenzymatic synthesis of a series of 4-substituted glutamate analogues and pharmacological characterization at human glutamatetransporters subtypes 1�3. J. Med. Chem., 2005, 48, 7980–7992.

8. Faure, S., Jensen, A.A., Maurat, V., Gu, X., Sagot, E., Aitken, D.J., Bolte, J., Gefflaut, T. andBunch, L., Stereoselective chemoenzymatic synthesis of the four stereoisomers of L-2-(2-carboxycyclobutyl)glycine and pharmacological characterization at human excitatory aminoacid transporter subtypes 1, 2, and 3. J. Med. Chem., 2006, 49, 6532–6538.

9. Xian, M., Alaux, S., Sagot, E. and Gefflaut, T., Chemoenzymatic synthesis of glutamic acidanalogues: substrate specificity and synthetic applications of branched chain aminotransferasefrom Escherichia coli. J. Org. Chem., 2007, 72, 7560–7566.

10. Helaine, V., Rossi, J., Gefflaut, T., Alaux, S. and Bolte, J., Synthesis of 4,4-disubstituted L-glutamic acids by enzymatic transamination. Adv. Synth. Catal., 2001, 343, 692–697.

11. Acetaldehyde is used to limit enzyme inhibition by trapping SO2 produced from CSA. It can beomitted if Escherichia coli AspAT is used instead of pig heart enzyme, the bacterial enzymebeing less sensitive to inhibition by SO2.

12. (a) Sagot, E., Jensen, A.A., Pickering, D., Stensbol, T.B., Nielsen, B., Assaf, Z., Aboab, B.,Bolte, J., Gefflaut, T. and Bunch L., Chemo-enzymatic synthesis of (2S,4R)-2-amino-4-(3-(2,2-diphenylethylamino)-3-oxopropyl)pentanedioic acid: a novel selective inhibitor of humanexcitatory amino acid transporter subtype 2. J. Med. Chem., 2008, 51, 4085–4092. (b)Sagot, E., Jensen, A.A., Pickering, D., Stensbol, T.B., Nielsen, B., Chapelet, M., Bolte, J.,Gefflaut, T. and Bunch L., Chemo-enzymatic synthesis of a series of 2,4-syn-functionalized (S)-glutamate analogues: new insight into the structure–activity relation of ionotropic glutamatereceptor subtypes 5, 6, and 7. J. Med. Chem., 2008, 51, 4093–4103.

10.3 Unnatural Amino Acids by Enzymatic Transamination 309

10.4 Synthesis of L-Pipecolic Acid with D1-Piperidine-2-carboxylateReductase from Pseudomonas putidaHisaaki Mihara and Nobuyoshi Esaki

L-Pipecolic acid, a key component of many antibiotic and anticancer biomolecules,1 serves

as an important chiral pharmaceutical intermediate. We have developed an enzyme-

coupled system consisting of D1-piperidine-2-carboxylate reductase (Pip2C) from

Pseudomonas putida, glucose dehydrogenase (GDH) from Bacillus subtilis, and L-lysine

�-oxidase from Trichoderma viride, affording L-pipecolic acid from L-lysine in high yield

with an excellent enantioselectivity (Figure 10.2).2

10.4.1 Procedure 1: Preparation of Enzymatic Crude Extract


• Bacto-tryptone (1 g)

• bacto-yeast extract (0.5 g)

• NaCl (1 g)

• NaOH (5 M)

• deionized water

• ampicillin (100 mg mL�1, sterilized through a 0.22 mm filter)

• chloramphenicol (25 mg mL�1 in ethanol)

• isopropyl-�-D-thiogalactopyranoside (1 M, sterilized through a 0.22 mm filter)

• tris-HCl buffer (20 mM, pH 7.0)

• membrane disc filters with 0.22 mm pore size

• autoclave

• one 500 mL Sakaguchi flask with a poromeric silicone plug

• one 20 mL test tube with a poromeric silicone plug

• clean bench

• incubation shaker

• sonicator

• centrifuge.

H2N

COOH

NH2

N COOH

NH

COOHNADP+

NADPH

Glucose

Gluconolactone

GDHPip2C reductase

L-Lysine α-oxidase

90%, >99.7% ee

Figure 10.2 Enzyme-coupled system for synthesis of L-pipecolic acid from L-lysine


10.4.1.2 Procedure

1. Bacto-tryptone (1 g), bacto-yeast extract (0.5 g) and NaCl (1 g) were dissolved in 95 mL of

deionized water and the pH was adjusted to 7.0 with 5 M NaOH. The volume was adjusted to

100mLwithdeionizedwater.Aportionoftheresultingsolution(5mL)wasplacedina20mL

test tube with a poromeric silicone plug and the rest was placed in a 500 mL Sakaguchi flask

with a poromeric silicone plug. The media were sterilized by autoclaving (121 �C, 20 min).

2. The 5 mL medium in a test tube was charged with ampicillin at 100 mg mL�1 and

chloramphenicol at 25 mg mL�1 and inoculated with a single colony of recombinant

Escherichia coli BL21(DE3) cells harbouring both pDPKA,3 which carries a gene for

Pip2C reductase, and pSTVbsGDH,3 which has a B. subtilis GDH4 gene between the

EcoRI and PstI sites of pSTV28.

3. The cells were shaken at 245 rpm for 20 h at 37 �C. The preculture (100 mL) was

transferred to the 100 mL medium containing 100 mg mL�1 ampicillin and 25 mg mL�1

chloramphenicol in a Sakaguchi flask and shaken at 245 rpm for 16 h at 37 �C.

4. The culture was charged with 1 mM isopropyl-�-D-thiogalactopyranoside to induce

gene expression and cultivated another 3 h.

5. The cells were harvested by centrifugation at 5000 rpm for 10 min at 4 �C and washed

twice with 20 mM tris-HCl (pH 7.0). The washed cells were disrupted by sonication for

1 min on ice. The supernatant was collected by centrifugation at 7500 rpm for 30 min at

4 �C to obtain the cell-free crude extract.

10.4.2 Procedure 2: Synthesis of L-Pipecolic Acid

NH

COOH


• L-Lysine monohydrochloride (502 mg, 2.75 mmol)

• glucose (990 mg, 5.5 mmol)

• �-nicotinamide adenine dinucleotide phosphate (NADPþ) sodium salt (4.87 mg,

2 nmol)

• flavin adenine dinucleotide (FAD) disodium salt hydrate (8.66 mg, 10 nmol)

• 100 mM tris-HCl buffer pH 7.5

• L-lysine �-oxidase from T. viride (Seikagaku Corporation, 30 units)

• catalase from bovine liver (Sigma–Aldrich, 500 units)

• NaOH (10 M)


• one 100 mL flask with a poromeric silicone plug.

10.4.2.2 Procedure

1. To a 100 mL flask containing L-lysine (102 mg, 0.6 mmol), glucose (990 mg, 5.5 mmol),

NADPþ (4.87 mg, 2.0 mmol), FAD (8.66 mg, 10.0 mmol), L-lysine �-oxidase

10.4 Synthesis of L-Pipecolic Acid with D1-Piperidine-2-carboxylate Reductase 311

(30 units), catalase from bovine liver (500 units) in 100 mM tris-HCl buffer solution pH

7.5 (10 mL) was added a crude extract of the recombinant E. coli BL21(DE3) cells with

pDPKA and pSTVbsGDH (30 mg protein – as prepared above). The flask was stoppered

with a poromeric silicone plug and shaken at 30 �C for 26 h.

2. L-Lysine (100 mg portions) was added to the reaction mixture at 3, 6, 11, and 17 h

intervals. To prevent a decrease in pH, the reaction mixture was adjusted to pH 7.5 with

10 M NaOH throughout the reaction course.

3. After 17 h, a titre of 210 mM (27 g L�1) L-pipecolic acid was achieved with satisfactorily

high optical purity (>99.7 % ee). The molar yield of L-pipecolic acid relative to L-lysine

was 90 %.

4. L-Pipecolic acid can be isolated from the resultant reaction solution by commonly used

methods, such as ion-exchange chromatography and crystallization, as described

previously.5

Enantiomeric excess was determined by high-performance liquid chromatography with

a Chiralpak WE column (4.6 mm � 250 mm, Daicel Chemical Industries, Tokyo), 2 mM

CuSO4, 0.75 mL min�1, 50 �C, monitored at 254 nm; L-pipecolic acid rt¼ 14.7 min,

D-pipecolic acid rt¼ 18.0 min.

10.4.3 Conclusion

The procedure can provide a higher amount of L-pipecolic acid in a shorter reaction time

than the previously reported system,6 indicating that it is applicable in industrial produc-

tion of L-pipecolic acid. A similar system was successfully employed in the enzymatic

synthesis of several cyclic amino acids by our group.7

References

1. (a) Germann, U.A., Shlyakhter, D., Mason, V.S., Zelle, R.E., Duffy, J.P., Galullo, V., Armistead,D.M., Saunders, J.O., Boger, J. and Harding, M.W., Cellular and biochemical characterization ofVX-710 as a chemosensitizer: reversal of P-glycoprotein-mediated multidrug resistance in vitro.Anti-Cancer Drugs, 1997, 8, 125. (b) Vezina, C., Kudelski, A. and Sehgal, S.N., Taxonomy of theproducing streptomycete and isolation of the active principle. J. Antibiot. (Tokyo), 1975, 28, 721.(c) Lehmann, J., Hutchison, A.J., McPherson, S.E., Mondadori, C., Schmutz, M., Sinton, C.M.,Tsai, C., Murphy, D.E., Steel, D.J., Williams, M., Cheney, D.L. and Wood, P.L., CGS 19755, aselective and competitive N-methyl-D-aspartate-type excitatory amino acid receptor antagonist.J. Pharmacol. Exp. Ther., 1988, 246, 65. (d) Boger, D.L., Chen, J.H. and Saionz, K.W., (�)-Sandramycin: total synthesis and characterization of DNA binding properties. J. Am. Chem. Soc.,1996, 118, 1629. (e) Hirota, A., Suzuki, A., Aizawa, K. and Tamura, S., Structure of Cyl-2, anovel cyclotetrapeptide from Cylindrocladium scoparium. Agric. Biol. Chem., 1973, 37, 955.

2. Muramatsu, H., Mihara, H., Yasuda, M., Ueda, M., Kurihara, T. and Esaki, N., Enzymaticsynthesis of L-pipecolic acid by D1-piperidine-2-carboxylate reductase from Pseudomonasputida. Biosci. Biotechnol. Biochem., 2006, 70, 2296.

3. (a) Muramatsu, H., Mihara, H., Kakutani, R., Yasuda, M., Ueda, M., Kurihara, T. and Esaki, N.,The putative malate/lactate dehydrogenase from Pseudomonas putida is an NADPH-dependentD1-piperidine-2-carboxylate/D1-pyrroline-2-carboxylate reductase involved in the catabolism ofD-lysine and D-proline. J. Biol. Chem., 2005, 280, 5329. (b) Mihara, H., Muramatsu, H., Kakutani,R., Yasuda, M., Ueda, M., Kurihara, T. and Esaki, N., N-Methyl-L-amino acid dehydrogenasefrom Pseudomonas putida. FEBS J., 2005, 272, 1117.


4. Lampel, K.A., Uratani, B., Chaudhry, G.R., Ramaley, R.F. and Rudikoff, S., Characterization ofthe developmentally regulated Bacillus subtilis glucose dehydrogenase gene. J. Bacteriol., 1986,166, 238.

5. Rodwell, V.W., Pipecolic acid. Methods Enzymol. Pt 2, 1971, 17, 174.6. (a) Fujii, T., Mukaihara, M., Agematu, H. and Tsunekawa, H., Biotransformation of L-lysine to L-

pipecolic acid catalyzed by L-lysine 6-aminotransferase and pyrroline-5-carboxylate reductase.Biosci. Biotechnol. Biochem., 2002, 66, 622. (b) Fujii, T., Aritoku, Y., Agematu, H. andTsunekawa, H., Increase in the rate of L-pipecolic acid production using lat-expressingEscherichia coli by lysP and yeiE amplification. Biosci. Biotechnol. Biochem. 2002, 66, 1981.

7. Yasuda, M., Ueda, M., Muramatsu, H., Mihara, H. and Esaki, N., Enzymatic synthesis of cyclicamino acids by N-methyl-L-amino acid dehydrogenase from Pseudomonas putida. TetrahedronAsymmetry. 2006, 17, 1775.

10.4 Synthesis of L-Pipecolic Acid with D1-Piperidine-2-carboxylate Reductase 313

10.5 Synthesis of Substituted Derivatives of L-Phenylalanine and of otherNon-natural L-Amino Acids Using Engineered Mutants ofPhenylalanine DehydrogenasePhilip Conway, Francesca Paradisi and Paul Engel

We have used a series of biocatalysts produced by site-directed mutations at the active site

of L-phenylalanine dehydrogenase (PheDH) of Bacillus sphaericus,1,2 which expand the

substrate specificity range beyond that of the wild-type enzyme, to catalyse oxidoreduc-

tions involving various non-natural L-amino acids. These may be produced by enantiose-

lective enzyme-catalysed reductive amination of the corresponding 2-oxoacid.3,4 Since the

reaction is reversible, these biocatalysts may also be used to effect a kinetic resolution of a

D,L racemic mixture.5

Here, we describe, as a representative example, an efficient chemical synthesis of 4-

fluorophenylpyruvic acid (Procedure 1, Section 10.5.1) followed by its biocatalytic con-

version to L-4-fluorophenylalanine catalysed by the N145V mutant2–4 of PheDH

(Procedure 2, Section 10.5.2).

10.5.1 Procedure 1: Preparation of 4-Fluorophenylpyruvic Acid

F

O

HNNH

O

O

NH

HNNH

O

OF F

O

O

OH+130 °C

NaOH 5M

100°C

%37dleiy)g4.8(dleiyedurc%78


• Hydantoin (5.2 g, 52 mmol)

• 4-fluorobenzaldehyde (5 mL, 47 mmol)

• piperidine (9.9 mL, 100 mmol)

• cyclohexane

• ethyl acetate


• nitrogen gas

• HCl 12 M (�20 mL)

• NaOH 5 M aqueous (21 mL)


• Whatman pH indicator paper type CF (1–14 range)

• two-necked reaction flask equipped with a magnetic stirrer bar, 100 mL

• magnetic stirrer and heating plate

• equipment for reflux condenser

• oil bath

• filter paper

• glass pipettes



• flask


10.5.1.2 Procedure

1. Hydantoin (5.2 g, 52 mmol) was added to piperidine (9.9 mL, 100 mmol) in a two-

necked reaction flask equipped with a magnetic stirrer bar and heated to 130 �C under

nitrogen flux. 4-Fluorobenzaldehyde (5 mL, 47 mmol) was added dropwise to the

stirring mixture. The reaction was monitored by TLC (eluent: ethyl acetate/cyclohex-

ane, 1:4) and reached completion in 30 min.

Attention. At room temperature hydantoin is insoluble in piperidine, but it will

dissolve at approximately 80 �C. Nitrogen is required to remove any traces of oxygen,

but the reaction does not need to be moisture-free.

2. The reaction mixture was cooled to 60 �C and water (200 mL) was added and stirred

vigorously for 30 min. A yellow, tarry side product precipitated and was removed by

filtration on filter paper.

3. The filtrate was acidified with HCl 12 M (approximately 20 mL) to pH 2.0, monitored

with Whatman pH indicator paper. The resulting precipitate was collected on filter

paper and dried under vacuum to afford 4-fluorobenzyl hydantoin (8.4 g, 87 %).1H NMR (500 MHz; DMSO) � 6.41 (s, 1H, olefin), 11.43–10.32 (m, 2H, NHs), 7.22

(t, J¼ 8.76 Hz, 2H, Aromatic), 7.66 (dd, J¼ 7.38, 5.99 Hz, 2H, Aromatic).

4. A fraction of the crude 4-fluorobenzyl-hydantoin (1 g, 4.9 mmol) was mixed with a

solution of NaOH 5 M (21 mL) in a two-necked reaction flask equipped with a magnetic

stirrer bar. The mixture was refluxed at 100 �C for 2.5 h. The reaction showed a strong

colour change as it progressed. As the starting material was added to the base, the mixture

turned a bright orange colour, which then lightened as the final product was formed.

5. The reaction was allowed to cool to room temperature and HCl 12 M was added

dropwise to generate the free acid. The 4-fluorophenylpyruvic acid was extracted

with EtOAc and the organic layer was dried with MgSO4 (which was then removed

by passing through filter paper) and evaporated in vacuo. 4-Fluorophenylpyruvic acid

was obtained as a yellow solid (0.7 g, 73 %).1HNMR(500MHz;CDCl3)�1.37(3H,t,J¼ 7.1Hz,CH3),4.35(2H,q,J¼ 7.1Hz,CH2),

6.48 (1H, s, H-3), 6.64 (1H, bs, OH), 7.01–7.07 (2H, m, Ar-H), 7.71–7.78 (2H, m, Ar-H).

10.5.2 Procedure 2: Enzymatic Synthesis of L-4-Fluorophenylalanine

O

O

ONa

NADH NAD+

PheDH mutant

EtOHCH3CHO

YADH

NH3 H2O

yield 0.19 g (86%)>99% ee

F

(S)

O

OH

FNH2

10.5 Synthesis of Substituted Derivatives of L-Phenylalanine 315


• 4-Fluorophenylpyruvate sodium salt (204 mg, 1 mmol)

• nicotinamide adenine dinucleotide (NADþ, 12 mg, 20 mmol)

• KCl (76 mg, 1 mmol)

• HCOONH4 (252 mg, 4 mmol)

• ethylenediaminetetraacetic acid (EDTA, 4 mg, 10 mmol)

• EtOH (0.5 mL)

• tris buffer 50 mM, pH 8.5 (10 mL)

• yeast alcohol dehydrogenase (ADH, 1 mg, 633 U mg�1)

• PheDH N145V mutant

• HCl 6 M (2 mL)

• NH4OH aqueous as needed

• HCl 1 M (150 mL)

• ninhydrin

• 15 mL plastic tubes with sealable cap

• orbital shaker incubator

• centrifuge

• high-performance liquid chromatograph fitted with chiral column such as

CHIROBIOTIC T

• Dowex� monosphere 550A (OH) anion-exchange resin (60 mL)

• equipment for ion-exchange chromatography


10.5.2.2 Procedure

1. 4-Fluorophenylpyruvate sodium salt (204 mg, 1 mmol) was mixed with NADþ (12 mg,

20 mmol), KCl (76 mg, 1 mmol), HCOONH4 (252 mg, 4 mmol) and EDTA (4 mg,

10 mmol). EtOH (0.5 mL) and tris buffer 50 mM pH 8.5 (10 mL) were used to dissolve

the reagents.

Attention. Despite the use of the oxo acid in the salt form, and the addition of EtOH,

the non-natural oxoacid substrates are not fully soluble at 0.1 M. Greater dilution

resulted in lower conversion rates, however.

2. Reaction was initiated with 1 mg yeast ADH (663 U mg�1) plus 10 mg N145V and the

mixture was held at 25 �C in an orbital shaker incubator.

3. Amino acid formation was monitored over 24 h by chiral high-performance liquid

chromatography (CHIROBIOTIC T column) with samples diluted 10-fold (in H2O) to a

suitable concentration.

4. The reaction was quenched by adding HCl 6 M (2 mL). After centrifugation, the crude

reaction mixture (precipitate and supernatant) was ready for purification.

5. The crude reaction mixture was brought to pH 12 with aqueous NH3 and applied to a

60 mL column of Dowex� monosphere 550A (OH) anion exchanger. Inorganic salts

were eluted with 300 mL water. Subsequently, the amino acid was eluted with 1 M HCl

(150 mL) and detected with ninhydrin. The eluate was concentrated in vacuo to give the

pure amino acid as a white solid (0.19 g, 86 %).


10.5.3 Conclusion

The hydantoin route for synthesis of the 2-oxoacid6 has been performed with a variety of

starting aldehydes and appears to work more reliably than the method described earlier.3 In

spite of the restricted solubility of the intermediate oxoacid substrate for the biocatalytic

step, the reaction proceeds to a good final overall yield with more substrate dissolving to

replace that consumed in the reaction. The combined procedure appears to be quite

versatile.

References

1. Seah, S.Y.K., Britton, K.L., Rice, D.W., Asano, Y. and Engel, P.C., Single amino acid substitu-tion in Bacillus sphaericus phenylalanine dehydrogenase dramatically increases its discrimina-tion between phenylalanine and tyrosine substrates. Biochemistry, 2002, 41, 11390.

2. Seah, S.Y.K., Britton, K.L., Rice, D.W., Asano, Y. and Engel, P.C., Kinetic analysis of pheny-lalanine dehydrogenase mutants designed for aliphatic amino acid dehydrogenase activity withguidance from homology-based modelling. Eur. J. Biochem., 2003, 270, 4628.

3. Busca, P., Paradisi, F., Moynihan, E., Maguire, A.R. and Engel, P.C., Enantioselective synthesisof non-natural amino acids using phenylalanine dehydrogenases modified by site-directedmutagenesis. Org. Biomol. Chem., 2004, 2, 2684.

4. Paradisi, F., Collins, S., Maguire, A.R. and Engel, P.C., Phenylalanine dehydrogenase mutants:efficient biocatalysts for synthesis of non-natural phenylalanine derivatives. J. Biotechnol., 2007,128, 408.

5. Paradisi, F., Conway, P.A., Maguire, A.R. and Engel, P.C., Engineered dehydrogenase biocata-lysts for non-natural amino acids: efficient isolation of the D-enantiomer from racemic mixtures.Org. Biomol. Chem., 2008, 6, 3611.

6. Billek, G., p-Hydroxyphenylpyruvic acid. Org. Synth. Collect. Vol., 1973, 5, 627.

10.5 Synthesis of Substituted Derivatives of L-Phenylalanine 317

11

Enzymatic Oxidation Chemistry

11.1 Monoamine Oxidase-catalysed Reactions: Application Towardsthe Chemo-enzymatic Deracemization of the Alkaloid (–)-Crispine AAndrew J. Ellis, Renate Reiss, Timothy J. Snape and Nicholas J. Turner

Previously, we reported a general method for the chemo-enzymatic deracemization of pri-

mary,1 secondary2 and tertiary3 amines in high yield and enantiomeric excess. The deracemi-

zation process involves a two-step, one-pot reaction and employs an enantioselective amine

oxidase (MAO-N) in combination with a nonselective chemical reducing agent (Figure 11.1).

We have further demonstrated the utility of this variant enzyme by way of the asymmetric

synthesis of the natural product (þ)-crispine A in 97 % ee.4 The previously reported MAO-

N-D5 variant, which contains five important mutations (Ile246Met/Asn336Ser/Met348Lys/

Thr384Asn/Asp385Ser) was used; its preparation has been described previously.5,6

R3 R4

N

R3 R4

NR3 R4

N+

NH3:BH3

enantio selective amineoxidase

R1 R2

R1 R2

R1 R2(S)

(R)

Figure 11.1 Enzymatic deracemization of racemic amines via a two-step, one-pot processutilizing an enantioselective amine oxidase in combination with ammonia–borane.



11.1.1 Procedure 1: Preparation of the Biocatalyst


• aqueous suspension of Escherichia coli BL21 competent cells (50 mL – Invitrogen)

• plasmid pET16b (Novagen) containing the variant MAO-N-D5 gene (1 mL)5,6

• Luria–Bertani (LB) broth containing 100 mg mL�1 ampicillin

• LB agar containing 100 mg mL�1 ampicillin (contained in Petri dishes)

• potassium phosphate buffer (K2HPO4–KH2PO4) (50 mM, pH 7.6)

• Erlenmeyer flask (250 mL) with foam bung

• ice-bath

• centrifuge

• shaker/incubator

• static incubator

• Falcon tube (50 mL).

11.1.1.2 Procedure

1. The plasmid was transformed into E. coli BL21 competent cells as per the manufac-

turer’s instructions (Invitrogen).

2. The transformed cell suspension (50 mL) was spread onto an LB-ampicillin agar plate

and incubated at 37 �C for 16 h.

3. A single colony was used to inoculate 5 mL of LB-ampicillin broth contained in a 50

mL Falcon tube. This was incubated at 37 �C, 250 rpm for 3 h.

4. The cell suspension (130 mL) was used to inoculate 25 mL of fresh LB-ampicillin broth

contained in a 250 mL Erlenmeyer flask with a foam bung. This culture was incubated

at 30 �C, 250 rpm for 24 h.

5. Cells were harvested by centrifugation at 3000 Gav for 30 min.

6. Cell pellets were washed by resuspension in potassium phosphate buffer (10 mL) and

were subjected to further centrifugation at 3000 Gav.

7. Wet cell pellets were then used directly in Procedure 2 (Section 11.1.2).

11.1.2 Procedure 2: Deracemization of (–)-Crispine A

N N

H

MeO

MeO

MeO

MeO

MAO-N-D5

BH3.NH3

43%, 97% ee

buffer (pH7.6)H


• whole wet cells expressing the MAO-N-D5 variant protein (�440 mg)

• potassium phosphate buffer (K2HPO4–KH2PO4) (2.46 mL, 0.1 M, pH 7.6)

• racemic crispine A (6.0 mg, 0.03 mmol)

320 Enzymatic Oxidation Chemistry

• ammonia–borane complex (3.5 mg, 0.11 mmol)

• dichloromethane

• anhydrous MgSO4

• glass vial (10 mL) sealed with a stopper

• shaker/incubator (set to 30 �C, 250 rpm)

• microcentrifuge

• 0.2 mm in-line syringe filter


11.1.2.2 Procedure

8. Whole wet cells (�440 mg) expressing the MAO-N-D5 variant enzyme were sus-

pended in 0.1 M potassium phosphate buffer pH 7.6 (2.46 mL). Racemic crispine

A (6.0 mg, 0.03 mmol) was added to this suspension followed by ammonia–borane

(3.5 mg, 0.11 mmol). The vial was sealed with the stopper and the mixture was

incubated at 30 �C, 250 rpm and samples (0.5 mL) taken periodically for analysis.

9. For high-performance liquid chromatography (HPLC) analysis: samples (0.5 mL)

were clarified by centrifugation at 14 000 Gav for 5 min and the supernatant was

decanted, filtered through a 0.2 mm in-line syringe filter and analysed directly by

chiral HPLC (see below).

10. For isolated material: the reaction mixture was clarified by centrifugation at 14 000

Gav for 5 min and the supernatant decanted and extracted with dichloromethane. The

dichloromethane phase was dried (MgSO4) and concentrated in vacuo to yield the title

compound as a colourless oil (4 mg, 43 %).1H NMR (CDCl3): d 1.67–1.76 (1H, m), 1.83–1.95 (2H, m), 2.29–2.37 (1H, m), 2.64

(1H, q, J 8.5), 2.66–2.71 (1H, m), 2.73 (1H, br. dt, J 16.1, 3.8), 2.93–3.00 (1H, m), 3.02–3.07

(1H, m), 3.12–3.17 (1H, m), 3.51 (1H, br. t, J 6.0), 3.82 (6H, s), 6.54 (1H, s), 6.58 (1H, s).13C NMR (CDCl3): d 22.1, 27.6, 30.5, 48.1, 53.0 (CH2), 55.8, 55.9 (CH3), 62.6,

108.7, 111.1 (CH), 125.8, 130.2, 147.2, 147.3 (C).

Electrospray ionization mass spectrometry (þve): found m/z 234.1 (MHþ, 100%).

[�]D ¼ þ88.4 (c ¼ 1.0, CHCl3).

Enantiomeric excess was determined by HPLC with an OD-H column (90 %

isohexane in isopropanol), 1.0 mL min�1, 210 nm; major enantiomer Rt ¼ 18.6 min,

97 % ee.

11.1.3 Conclusion

The procedure is very easy to perform and is highly reproducible and may be applied to a

wide range of substrates; see below for selected examples:

NH

95%, 99% ee

HN

80%, 98% ee

N

75%, 99% ee

11.1 Monoamine Oxidase-catalysed Reactions 321

References

1. Alexeeva, M., Enright, A., Dawson, M.J., Mahmoudian, M. and Turner, N.J., Deracemization of�-methylbenzylamine using an enzyme obtained by in vitro evolution. Angew. Chem. Int. Ed.,2002, 41, 3177.

2. Carr, R., Alexeeva, M., Dawson, M.J., Gotor-Fernandez, V., Humphrey, C.E., Turner, N.J.,Directed evolution of an amine oxidase for the preparative deracemisation of cyclic secondaryamines. ChemBioChem, 2005, 6, 637.

3. Dunsmore, C.J., Carr, R., Fleming, T. and Turner, N.J., A chemo-enzymatic route to enantio-merically pure cyclic tertiary amines. J. Am. Chem. Soc., 2006, 128, 2224.

4. Bailey, K.R., Ellis, A.J., Reiss, R., Snape, T.J. and Turner, N.J., A template-based mnemonic formonoamine oxidase (MAO-N) catalyzed reactions and its application to the chemo-enzymaticderacemisation of the alkaloid (–)-crispine A. Chem. Commun., 2007, 3640.

5. Alexeeva, M., Carr, R. and Turner, N.J., Directed evolution of enzymes: new biocatalysts forasymmetric synthesis. Org. Biomol. Chem., 2003, 1, 4133.

6. Carr, R., Alexeeva, M., Enright, A., Eve, T.S.C., Dawson, M.J. and Turner, N.J., Directedevolution of an amine oxidase possessing both broad substrate specificity and high enantios-electivity. Angew. Chem. Int. Ed., 2003, 42, 4807.


11.2 Glucose Oxidase-catalysed Synthesis of Aldonic AcidsFabio Pezzotti, Helene Therisod and Michel Therisod

The classical chemical synthesis of aldonic acids makes use of stoichiometric amounts of

bromine, copper or silver hydroxides, or mercuric acetate in totally nonecologically

acceptable processes.1,2 Glucose oxidase (EC 1.1.3.4) has the reputation of being extre-

mely specific for D-glucose (a characteristic commonly used in analytical biochemistry).

However, by extending the reaction time and the amount of enzyme, we were able to

prepare gram-scale quantities of xylonic, galactonic, mannonic, 2-deoxygluconic and

2-aminogluconic acids.3,4 This was made possible by the availability of glucose oxidase,

produced on an industrial scale and at low cost by Novozymes. The reaction takes place in

water under atmospheric oxygen (as an oxidant). The products are isolated in pure form

either by precipitation (2-aminogluconic acid) or by filtration through an ion-exchange

resin.

11.2.1 Procedure: Synthesis of Xylonic, Galactonic, Mannonic, 2-Deoxygluconic

Acid and Synthesis of 2-Amino-2-deoxy-gluconic Acid (Glucosaminic Acid)

O

OH

HO

Glucose oxidase

O2 2, H O H2O2

Catalase

O

O

HO

Aldose

COOH

OH

HO

H2O/NaOH (pH-stat)

Dowex 1 (AcO –)elution HCl

dicacinodlAenotcalonodlA

O

OH

Glucose oxidase

O2, H2O H2O2

Catalase

O

O

Glucosamine

COO –

OHH2O/NaOH

Glucosaminic acid(precipitates)

NH2

HO

HO

OH

NH2

HO

HO

OH OH

(pH-stat)HO

HONH3

+


• Aldose (D-xylose, D-galactose, D-mannose, D-2-deoxyglucose, D-glucosamine hydro-

chloride) (11.1 mmol)

• glucose oxidase (Gluczyme�, from Novozymes) (200 mg, 400 U)

• catalase* (Catazyme�, from Novozymes) (1 mL, 25 kU)

• 1 M sodium hydroxide (11.1 mmol)

• 1 M hydrochloric acid

• Dowex 1 (acetate form)

• pH-stat


*The use of catalase is not essential, as long as large amounts of glucose oxidase are

used, since this last enzyme apparently is quite resistant to high concentration of hydrogen


peroxide. Moreover, Gluczyme� contains some catalase activity. Catalase may be more

useful if another source of purified glucose oxidase is used.

11.2.1.2 Procedure

11. The aldose (11.1 mmol) was dissolved in water to a final concentration of 0.5 M and

subjected to oxidation by addition of glucose oxidase (200 mg, 400 U) and a large

excess of catalase (1 mL, 25 kU). The mixture was vigorously stirred under air and the

pH was kept constant at 7.5 by means of a pH-stat adding continuously 1 M NaOH. The

conversion degree was directly calculated considering the volume of added 1 M

NaOH, since 1 mol of NaOH neutralizes 1 mol of aldonic acid formed.

12. The reaction mixture was then filtered through a Dowex 1 (AcO�) column to

eliminate the enzymes and any residual substrate.

13. The aldonic acid (as a mixture of lactones) was then recovered by elution with 1 M

aqueous HCl and evaporation in vacuo.

14. For the synthesis of 2-aminogluconic acid, the starting glucosamine hydrochloride was

first neutralized by 1 M NaOH before starting the enzymatic oxidation. The insoluble

product was recovered by concentration of the reaction medium and filtration.

15. The identity and purity of each product were confirmed by 1H and 13C NMR spectro-

scopy (D2O–Na2CO3).

11.2.1.3 2-Deoxy-D-gluconic Acid

Yield 92 %.1H NMR (360 MHz, D2O): d 2.52 (dd, 1H, J 15, J 5.8, H2), 2.55 (dd, 1H, J 15, J 8.3,

H20), 3.51 (dd, 1H, J 8.3, J 2, H4), 3.76 (dd,1H, J 12, J 6.3, H6), 3.8 (m, 1H, H5), 3.9 (dd,

1H, J 11.5, J 2.9, H60), 4.3 (ddd, 1H, J 8.3, J 5.4, J 1.8, H3).13C NMR (62 MHz, D2O): d 41.61 C2, 63.07 C6, 67.89 C3, 71.28 C5, 72.87 C4, 180.33

C1 (in accordance with literature data5).

[�]D ¼ þ5.05 (c ¼ 2.18, H2O) (lit. 6 [�]D ¼ þ4.2).

11.2.1.4 D-Galactonic Acid

Yield 77 %.1H NMR (360 MHz, D2O): d 3.56 (dd, 1H, J 9.6, J 1.5, H4), 3.62 (d, 2H, J 6.4, H6–H60),

3.88 (dd, 1H, J 9.5, J 1.5, H5), 3.90 (dd, 1H, J 9.5, J 1.5, H3), 4.2 (d, 1H, J 1.5, H2).13C NMR (62 MHz, D2O): d 63.35 C6, 69.82 C4, 70.16 C5, 71.42 C3, 71.57 C2, 179.64


[�]D ¼ þ1.6 (c ¼ 10, H2O) (lit.8 [�]D ¼ þ0.4).

11.2.1.5 D-Mannonic Acid

Yield 70 %.1H NMR (360 MHz, D2O): d 3.6 (dd, 1H, J 11.5, J 2.7, H6), 3.67 (bs, 2H, H5–H4), 3.75

(d, 1H, J 11.5, H60), 3.92 (d, 1H, J 5.6, H3), 4.06 (d, 1H, J 5.6, H2).13C NMR (62 MHz, D2O): d 62.95 C6, 70.42 C3, 70.49 C5, 70.86 C4, 73.84 C2, 179.25


[�]D ¼ �8.9 (c ¼ 10, H2O) (lit.8 [�]D ¼ �8.8).


11.2.1.6 D-Xylonic Acid

Yield 90 %.1H NMR (360 MHz, D2O): d 3.50 (dd, 1H, J 11.8, J 5.4, H5), 3.62 (dd, 1H, J 11.7, J 3.9,

H50), 3.71 (m, 1H, H4), 3.76 (dd, 1H, J 5.9, J 2.5, H3), 3.99 (d, 1H, J 2.55, H2), 3.84 (d, 1H,

J 2.4).13C NMR (62 MHz, D2O): d 62.72 C5, 72.54 C3, 73.09 C2, 73.19 C4, 179.19 C1 (in

accordance with literature data10).

[�]D ¼ þ7.05 (c ¼ 10, H2O) (lit. 11 [�]D ¼ þ7.4).

11.2.1.7 D-Glucosaminic Acid

Yield 76 %.1H NMR (360 MHz, D2O): d 3.05 (d, 1H, J 5.4, H2), 3.28 (dd, 1H, J 10.4, J 4.7, H6),

3.3–3.4 (m, 2H, H4, H5), 3.42 (dd, 1H, J 10.4, J 4.3, H60), 3.57 (dd, 1H, J 5.4, J 1.8, H3).13C NMR (62 MHz, D2O): d 59.54 C2 , 62.92 C6, 72.18 C3, 72.97 C5, 73.8 C4, 181.11


[�]D ¼ �15 (c ¼ 4, 2.5 % aqueous HCl) (lit.13 [�]D ¼ �14)

11.2.2 Conclusion

We have devised a very simple procedure for the preparative synthesis of various aldonic

acids from the corresponding aldoses. This ‘green chemistry’ process takes advantage of

the availability of cheap, robust industrial enzymes.

References

1. (a) Varela, O., Oxidative reactions and degradations of sugars and polysaccharides. Adv.Carbohydr. Chem. Biochem., 2003, 58, 307. (b) DeLederkremer, R.M. and Marino, C., Acidsand other products of oxidation of sugars. Adv. Carbohydr. Chem. Biochem., 2003, 58, 199.

2. Pringsheim, H. and Ruschman, G., Preparation of glucosaminic acid. Ber. Dtsch. Chem. Ges.1915, 48, 680.

3. Pezzotti, F., Therisod, H. and Therisod, M., Enzymatic synthesis of D-glucosaminic acid fromD-glucosamine. Carbohydr. Res., 2005, 340, 139.

4. Pezzotti, F. and Therisod, M., Enzymatic synthesis of aldonic acids. Carbohydr. Res., 2006, 341,2290.

5. Freimund, S., Huwig, A., Giffhorn, F. and Kopper, S., Rare keto-aldoses from enzymaticoxidation: substrates and oxidation products of pyranose 2-oxidase. Chem. Eur. J., 1998, 4,2442.

6. Horton, D. and Philips, K.D., Diazo derivatives of sugars. Synthesis of methyl 2-deoxy-2-diazo-D-arabino-hexonate, its behaviour on photolysis and thermolysis, and conversion into a pyrazolederivative. Carbohydr. Res., 1972, 22, 151.

7. Ramos, M.L., Caldeira, M.M. and Gil, V., NMR study of the complexation of D-galactonic acidwith tungsten (VI) and molybdenum (VI). Carbohydr. Res., 1997, 297, 191.

8. Levene, P.A., The specific rotations of hexonic and 2-amino-hexonic acids and of their sodiumsalts. J. Biol. Chem., 1924, 59, 123.

9. Horton, D., Walaszek, Z. and Ekiel, I., Conformations of D-gluconic, D-mannonic, and D-galactonic acids in solution, as determined by n.m.r. spectroscopy. Carbohydr. Res., 1983,119, 263.

10. Serianni, A.S., Nunez, H.A. and Barker, R., Cyanohydrin synthesis: studies with carbon-13-labeled cyanide. J. Org. Chem., 1980, 45, 3329.


11. Browne, C.A. and Tollens, B., Uber die Bestandtheile des Mais-Marks und des Hollander-Marks und das gleichzeitige Vorkommen von Araban und Xylan in den Pflanzen. Berl. Dtsch.Chem. Ges., 1902, 35, 1457.

12. Horton, D., Thomson, J.K., Varela, O., Nin, A. and Lederkremer, R.M., Confirmation of thestructures of the products obtained on acylation of 2-amino-2-deoxy-D-gluconic acid.Carbohydr. Res., 1989, 193, 49.

13. Hope, D.B. and Kent, P.W., Ester and lactone linkages in acidic polysaccharides. Part II.Lactones of D-glucosaminic acid. J. Chem. Soc. Abstr., 1955, 1831.


11.3 Oxidation and Halo-hydroxylation of Monoterpenes withChloroperoxidase from Leptoxyphium fumagoBjoern-Arne Kaup, Umberto Piantini, Matthias Wust and Jens Schrader

Chloroperoxidase (CPO) from Leptoxyphium fumago (formerly Caldariomyces fumago) is

a unique enzyme showing broad substrate specificity and featuring industrially relevant

reactions, including halogenation and oxidation reactions. Recently, monoterpenes were

discovered to be substrates for both oxidation (Figure 11.2) and halo-hydroxylation

(Figure 11.3) by CPO in the absence and presence of halide ions respectively.1 In the

case of halo-hydroxylation of (1S)-(þ)-3-carene, excellent ee values of >99 % were

detected. The introduction of two stereogenic centres in one step makes this reaction

very interesting given the fact that 3-carene has been investigated as starting material for

the synthesis of different valuable target compounds, such as fragrances and �-lactam

antibiotics.2,3

11.3.1 Procedure 1: Halo-hydroxylation of (1S)-(þ)-3-Carene by CPO


• (1S)-(þ)-3-Carene (>99 %, 10 mM)

• sodium chloride, bromide or iodide (p.a., 10 mM)

• citric acid buffer (100 mM, pH 3.5)

• tert-butanol (>99 %)

CH2OH

CPO

+H2O2

CHO

GeranialGeraniol

Figure 11.2 Oxidation of the monoterpene alcohol geraniol to geranial by CPO in thepresence of hydrogen peroxide and absence of halide ions.

CPO

+H2O2 + X–

XHO

(1S)-(+)-3-Carene (1S,3R,4R,6R )-4-Halo-3,7,7-trimethyl-bicyclo[4.1.0]-

heptane-3-ol

Figure 11.3 Stereoselective halo-hydroxylation of the monoterpene hydrocarbon (1S)-(þ)-3-carene by CPO in the presence of hydrogen peroxide and halide ions (X� ¼ Cl�, Br� or I�).

11.3 Oxidation and Halo-hydroxylation of Monoterpenes 327

• CPO (0.045 mg mL�1, �1.2 mM)

• Hydrogen peroxide (1 M)

• n-hexane (>99 %)

• sodium sulfate (p.a.)

• one 50 mL vessel with screw cap

• magnetic stir bar


11.3.1.2 Procedure

16. For the conversion of (1S)-(þ)-3-carene approximately 0.045 mg mL�1 (�1.2 mM)

CPO was incubated in 100 mM citric acid buffer, pH 3.5 with 25 % (v/v) tert-butanol

containing 10 mM (1S)-(þ)-3-carene (final assay concentration) and 10 mM sodium

chloride, sodium bromide or sodium iodide (final assay concentrations) in a 50 mL

vessel on a magnetic stirrer (300 rpm) at room temperature. Hydrogen peroxide was

added to a total concentration of 10 mM over a reaction time of 60 min at a rate of 165

mM min�1 (165 mM portions every minute).

17. Samples were extracted with n-hexane, dried over sodium sulfate and stored at�20 �C

until analysed by coupled gas chromatography–mass spectrometry (GC-MS).

GC–MS: Shimadzu GC-17A/GCMS-OP5050 system; column: Valco Bond VB-S

(30 m � 0.25 mm � 0.25 mm); injection: split 50:1 at 230 �C, 1 mL; carrier gas: helium,

1.3 mL min�1; interface temperature: 250 �C; oven program: starting temperature 50 �C, rate

5 �C min�1 for 30 min to 200 �C.

Product identification was carried out by NMR analysis using 1H, 13C and 1H/1H

correlation spectroscopy techniques.1 Under the given conditions, molar conversion yields

were �90 % after 10 min, �90 % after 20 min and �60 % after 30 min for iodo-, bromo-

and chloro-halohydrin formation respectively.

11.3.2 Procedure 2: Oxidation of Geraniol by CPO


• Geraniol (96 %, 2 mM)

• citric acid buffer (100 mM, pH 3.5)

• tert-butanol (>99 %)

• CPO (0.2 mg mL�1, �5 mM)

• hydrogen peroxide (1 M)

• n-hexane (>99 %)

• sodium sulfate (p.a.)

• one 50 mL vessel with screw cap

• magnetic stir bar


11.3.2.2 Procedure

18. For conversion of geraniol approximately 0.2 mg mL�1 (�5 mM) CPO was incubated

in 100 mM citric acid buffer, pH 3.5, with 25 % (v/v) tert-butanol containing 2 mM


geraniol (final assay concentration) in a 50 mL vessel on a magnetic stirrer (300 rpm)

at room temperature. Hydrogen peroxide was added to a total concentration of 2 mM

over a reaction time of 60 min at a rate of 33 mM min�1 (100 mM portions every 3 min).

19. Samples were extracted with n-hexane, dried over sodium sulfate and stored at�20 �C

until GC–MS analysis.

GC–MS: Shimadzu GC-17A/GCMS-OP5050 system; column: Valco Bond VB-S

(30 m� 0.25 mm� 0.25 mm); injection: split 50:1 at 230 �C, 1 mL; carrier gas: helium,

1.3 mL min�1; interface temperature: 250 �C; oven program: starting temperature 50 �C,

rate 5 �C min�1 for 30 min to 200 �C.

The product was identified by comparison of its retention time and mass spectrum with

those of a commercial standard substance. Under the given conditions, the final molar

conversion yield was 37.5 %.

11.3.3 Conclusion

The procedures are very easy to reproduce and to scale up. Reaction products are isolated

by evaporation of the extraction solvent (e.g. hexane, pentane). In the case of the carene

halohydrin, further product purification is not necessary if reaction is allowed to proceed

until total substrate conversion due to the high selectivity of product formation.

References

1. Kaup, B.A., Piantini, U., Wust, M. and Schrader, J., Monoterpenes as novel substrates foroxidation and halo-hydroxylation with chloroperoxidase from Caldariomyces fumago. Appl.Microbiol. Biotechnol., 2007, 73, 1087–1096.

2. Bhawal, B.M., Joshi, S.N., Krishnaswamy, D. and Deshmukh, A.R., (þ)-3-Carene, an efficientchiral pool for the diastereoselective synthesis of �-lactams. J. Indian Inst. Sci., 2001, 81,265–276.

3. Lochynski, S., Kowalska, K. and Wawrzenczyk, C., Synthesis and odour characteristics of newderivatives from the carane system. Flavour Fragr. J., 2002, 3, 181–186.

11.3 Oxidation and Halo-hydroxylation of Monoterpenes 329

11.4 Chloroperoxidase-catalyzed Oxidation of Phenyl Methylsulfide inIonic LiquidsCinzia Chiappe

The heme enzyme chloroperoxidase (CPO), produced by the marine fungus

Caldariomyces fumago, is a versatile enzyme which exhibits a broad spectrum of chemical

reactivities and it is recognized as a most promising biocatalyst for synthetic applications.1

Recently, pure (R)-phenyl methylsulfoxide (ee > 99 %) was prepared by chemo- and

stereo-selective oxidation of phenyl methylsulfide with CPO in citrate buffer–ionic liquid

mixtures.2

11.4.1 Procedure: Synthesis of (R)-Phenyl Methylsulfoxide

SCH3

SCH3

OS

CH3

O O

+H2O2

Ionic Liquid/ citrate buffer 95–100% 5–0%

e.e. >99%

CPO


• Thioanisole (6.2 mg, 50 mmol)

• CPO (67.4 U)

• hydrogen peroxide solution 7 wt% in water (50 mmol þ 25 mmol)

• sodium citrate buffer solution (0.1 M, pH 5) (1 mL)

• ionic liquid: cholinium citrate ([N1112OH][Citr], 1 mL) or 1,3-dimethylimidazolium

methylsulfate ([mmim][MeSO4], 1 mL) or cholinium acetate ([N1112OH][OAc], 0.5 mL)

• sodium thiosulfate


• diethyl ether

• filter paper

• one 10 mL test tube with a screw cap and equipped with a magnetic stirrer bar




11.4.1.2 Procedure

20. Thioanisole (6.2 mg, 50 mmol) and CPO (67.4 U) were magnetically stirred at room

temperature in a 10 mL test tube for 5 min in 2 mL of a mixture of ionic liquid–sodium

citrate buffer solution (0.1 M, pH 5). A 1:1 mixture was used in the case of

[N1112OH][Citr] and [mmim][CH3SO4], whereas a 0.6:1.4 mixture was employed in

the case of [N1112OH][OAc]. Hydrogen peroxide solution (7 wt%) was added in two

portions: initially 50 mmol and then an additional 25 mmol after 2 h. After 4 h the

reaction was quenched by the addition of excess Na2S2O3.


21. The reaction mixture was extracted with diethyl ether. The organic portion was

collected, dried over anhydrous magnesium sulfate and concentrated using a rotary

evaporator.

The crude product was analysed by gas chromatography on a chiral 30 m Chiraldex G-

TA column (helium flow 50 kPa, evaporator and detector set at 200 �C, column tempera-

ture 90 �C for 1 min, 8 �C min�1 to 170 �C) after addition of anisole as an internal standard.

The reaction performed in the 1:1 mixture of [mmim][CH3SO4]/sodium citrate buffer

has been scaled up 50 times without any change in chemo- and enantio-selectivity. The

crude reaction mixture was purified by chromatography on silica gel (hexane/ethyl acetate

6:1–1:1) to give pure phenyl methylsulfoxide (yield 70 %, ee 99 %).

11.4.2 Conclusions

Ionic liquids can be used as co-solvents for CPO-catalysed sulfoxidation. Table 11.1 gives

details about different ionic liquids. The procedure is very easy to reproduce and the

oxidation of thioanisole proceeds with high chemo- and stereo-selectivity.

Table 11.1 Oxidation of thioanisole with hydrogen peroxide and CPO at room temperaturein a 1:1 ionic liquid/citrate buffer

Ionic liquid Amount (v/v) of IL in citratebuffer (%)

Conversion(%)

Products

Sulfoxide/sulfone

Ee (%)

[N1112OH][Citr] 50 48 95:5 >99 (R)[mmim][CH3SO4] 50 76 98:2 >99 (R)[N1112OH][OAc] 30 42 100:0 >99 (R)

References

1. Dembitsky, M.V., Oxidation, epoxidation and sulfoxidation reactions catalysed by haloperox-idases. Tetrahedron, 2003, 59, 4701.

2. Chiappe C., Neri, L. and Pieraccini, D., Application of hydrophilic ionic liquids as co-solvents inchloroperoxidase catalyzed oxidations. Tetrahedron Lett., 2006, 47, 5089.

11.4 Chloroperoxidase-catalyzed Oxidation of Phenyl Methylsulfide 331

11.5 Stereoselective Synthesis of b-Hydroxy Sulfoxides Catalyzed byCyclohexanone MonooxygenaseStefano Colonna, Nicoletta Gaggero, Sara Pellegrino and Francesca Zambianchi

Chiral �-hydroxy sulfoxides are well known for their usefulness as chiral auxiliaries for

the preparation of a great variety of compounds, such as biaryl sulfoxides,1a cyclic

sulfides,1b benzoxathiepines,1c benzothiazepines,1d allylic alcohols,1e macrolides1f and

leucotrienes.1g The most straightforward method for their synthesis is the selective oxida-

tion of the parent sulfides. Although a variety of catalytic systems2 have been introduced

for this kind of oxidation, many of these methodologies are associated with several

disadvantages, such as environmentally unfriendly catalysts, volatile organic solvents,

harsh reaction conditions and low stereoselectivities. Therefore, the development of an

environmentally benign, high yielding and clean approach for the synthesis of �-hydroxy

sulfoxides is needed.

We have found that these goals can be achieved through the direct oxidation of

different �-hydroxy sulfides using cyclohexanone monooxygenase from

Acinetobacter calcoaceticus (CHMO) as catalyst.3 CHMO is a bacterial Baeyer–

Villigerase that has been used for a chemoenzymatic synthesis of a variety of key chiral

products. This enzyme catalyzes the Baeyer–Villiger oxidation of cyclohexanone with

formation of the corresponding E-caprolactone; the only reagents consumed are O2 and

reduced nicotinamide adenine dinucleotide phosphate (NADPH). Apart from cyclohex-

anone and other ketones and aldehydes, CHMO can oxidize a wide series of organic

compounds containing electron-rich heteroatoms, i.e. sulfides are converted to sulf-

oxides,4 sulfites to sulfates,5 selenides to selenoxides,6 tertiary amines to N-oxides7 and

phospines to phospinoxides.8 In many cases these reactions are highly enantioselective9

(Figure 11.4).

S

OH

CHMO/G6PDH

G6P/NADP/O2

S

OH

S

OH

O O

S

OH

S

OH

O O

1R,2R,RS 1S,2S,SS

1R,2R,SS 1S,2S,RS

n

n

n n

n

(±)-trans -1-3

1-3amajor diasteroisomer

1-3bminor diasteroisomer

Figure 11.4 Oxidation of �-hydroxy sulfides to �-hydroxy sulfoxides catalyzed by CHMO


11.5.1 Procedure 1: Preparation and Purification of CHMO


• Buffer A: 0.02 M potassium phosphate buffer with 0.01 M dithiothreitol, pH 7.2

• (NH4)2SO4

• NaCl

• NADPþ

• sonicator

• centrifuge

• Fractogel EMD DEAE, 10 cm � 2 cm ion-exchange column

• Matrex gel red A, agarose 5 %; 10 cm � 2 cm affinity column

• lyophilizor.

11.5.1.2 Procedure

1. A. calcoaceticus was grown as described by Trudgill10 and the transformed micro-

organism was cultivated essentially as described previously by Doig et al.11

2. All steps of purification were carried out at 4 �C using buffer A.

3. The cells obtained from 1 L of culture medium were harvested, disrupted by sonication

and cell debris removed by centrifugation.

4. The supernatant was subjected to fractionation with (NH4)2SO4 and the fraction which

precipitated between 40 and 85 % saturation was recovered by centrifugation at 6000g

for 30 min.

5. The pellet was redissolved in buffer A, dialysed overnight against the same buffer and

loaded on an anion-exchange column (Fractogel EMD DEAE, 10 cm � 2 cm) which

was previously equilibrated with buffer A. The enzyme was eluted with a linear

gradient from 0 to 0.15 M NaCl for 30 min in the same buffer, at a flow rate of 2 mL

min�1.

6. Active fractions were collected and loaded on an affinity column (Matrex gel red A,

agarose 5 %; 10 cm � 2 cm) previously equilibrated with buffer A and eluted with the

same buffer but containing NADPþ 0.05 M (flow rate of 1 mL min�1). Active fractions

were collected, dialysed overnight against buffer A and lyophilized.

11.5.2 Procedure 2: Oxidation of b-Hydroxy Sulfides to b-Hydroxy Sulfoxides

Catalyzed by CHMO


• Sulfides (–)-1–3 (1 mg)

• CHMO (1 U) is not that commercially available12 (see Procedure 1, Section 11.5.1)

• glucose-6-phosphate dehydrogenase from Leuconostoc mesenteroides (G6PDH) (18 U

mL�1)

• glucose-6-phosphate (G6P) (50 mM, 15 mg)

• NADPþ (0.5 mM, 1 mg)

• pH 8.6 tris-HCl buffer solution 50 mM (1 mL)

• diethyl ether (1 mL)

• propan-2-ol (1 mL)

11.5 Stereoselective Synthesis of �-Hydroxy Sulfoxides 333

• n-hexane

• ethyl acetate

• silica gel 60 (230-400 mesh)

• reaction flask equipped with a magnetic stirrer bar




• high-performance liquid chromatography (HPLC) equipment

• Chiracel OD column (Daicel, Illkirch, France)


11.5.2.2 Procedure

1. Sulfide (–)-1 or (–)-2 or (–)-3 (1 mg) was added to 50 mM tris-HCl buffer pH 8.6 (1 mL)

containing NADPþ (1 mg), G6P (15 mg), partially purified CHMO (1 U) and G6PDH

(18 U). The reaction mixture was gently stirred at 25 �C (see Table 11.2 for reaction times).

Table 11.2 Oxidation of racemic �-hydroxy sulfides catalyzed by CHMO

Sulfide Time (h) C (%)aee sulfidesa

(%) [E]dr major/minora

eea (%)

Major Minor

(–)-1b,c

n ¼ 024 97 69 [1.6] 83:17 53 91

(–)-2d,e

n ¼ 11 36 47 [261] >99:1 �98 -

(–)-2d,e

n ¼ 15 47 87 [299] 99:1

�98(1S,2S,SS)

�95(1R,2R,SS)

(–)-3f,g

n ¼ 21 52 50 [4.3] 89:11 63 56

(–)-3f,g

n ¼ 23 78 79 [3.3] 82:18 45 78

aDetermined by HPLC analysis on Chiralcel OD column.btrans-2-(Phenylsulfinyl)cyclopentan-1-ol (1a). Major diastereoisomer. M.p. 102 �C. IR (KBr): � 3390, 3058, 1651, 1085,1028 cm�1. 1H NMR (CDCl3, 200 MHz): d 1.62–1.83 (m, 5 H, 30-H, 4-H, 5-H), 2.07 (m, 1 H, 300-H), 3.02 (m, 2 H, 2-H,OH), 4.64 (m, 1 H, 1-H), 7.50 (m, 3 H, Ar-H), 7.77 (m, 2 H, Ar-H) ppm. Electron-ionization mass spectrometry (EI-MS): m/z(%) ¼ 210 (100) [M]þ.ctrans-2-(Phenylsulfinyl)cyclopentan-1-ol (1b). Minor diastereoisomer. M.p. 97 �C. IR (KBr): � 3295, 3058, 1637, 1085,1012 cm�1. 1H NMR (CDCl3, 200 MHz): d 1.53–1.71 (m, 4 H, 4-H, 5-H), 1.84–2.06 (m, 3 H, 3-H, OH), 3.03 (m, 1 H, 2-H),4.56 (m, 1 H, 1-H), 7.85 (m, 5 H, Ar-H) ppm. EI-MS: m/z (%) ¼ 210 (100) [M]þ.dtrans-2-(Phenylsulfinyl)cyclohexan-1-ol (2a). Major diastereoisomer (1S,2S,SS). M.p. 156–157 �C. IR (KBr): � 3444, 2931,1628, 1077, 1002 cm–1. 1H NMR (CDCl3, 300 MHz): d 1.11–1.40 (m, 6 H, 4-H, 5-H, 6-H), 1.73 (m, 2 H, 3-H), 2.75 (m,1 H, 2-H), 3.03 (m, 1 H, 1-H), 4.16 (br s, 1 H, OH), 7.55 (m, 3 H, Ar-H), 7.74 (m, 2 H, Ar-H) ppm. EI-MS: m/z (%) ¼ 224(100) [M]þ. ½��20

D ¼þ134 (c¼ 1.5, CHCl3).etrans-2-(Phenylsulfinyl)cyclohexan-1-ol (2b). Minor diastereoisomer (1R,2R,SS). M.p. 138–139 �C. IR (KBr): � 3450,2931, 1648, 1077, 1012 cm�1. 1H NMR (CDCl3, 300 MHz): d 1.09–1.71 (m, 7 H, 30-H, 4-H, 5-H, 6-H), 2.11 (m, 1 H, 300-H), 2.66 (m, 1 H, 2-H), 3.93 (m, 1 H, 1-H), 4.01 (br s, 1 H, OH), 7.58 (m, 5 H, Ar-H) ppm. EI-MS: m/z (%)¼ 224 (100) [M]þ.½��20

D ¼þ100:9 (c ¼ 1.5, CHCl3).ftrans-2-(Phenylsulfinyl)cycloheptan-1-ol (3a). Major diastereoisomer. M.p. 148 �C. IR (KBr): � 3353, 3059, 1579, 1050,1014 cm�1. 1H NMR (CDCl3, 200 MHz): d 1.19–1.90 (m, 10 H, 3-H, 4-H, 5-H, 6-H, 7-H), 2.85 (br s, 1 H, OH), 2.94 (m, 1H, 2-H), 4.32 (m, 1 H, 1-H), 7.56 (m, 3 H, Ar-H), 7.76 (m, 2 H,Ar-H) ppm. EI-MS: m/z (%) ¼ 238 (100) [M]þ.gtrans-2-(Phenylsulfinyl)cycloheptan-1-ol (3b). Minor diastereoisomer. M.p. 125 �C. IR (KBr): � 3358, 3057, 1638, 1050,1014 cm�1. 1H NMR (CDCl3, 200 MHz): d 1.24–1.89 (m, 10 H, 3-H, 4-H, 5-H, 6-H, 7-H), 2.16 (br s, 1 H, OH), 2.99 (m, 1H, 2-H), 4.16 (m, 1 H, 1-H), 7.53 (m, 3 H, Ar-H), 7.88 (m, 2 H, Ar-H) ppm. EI-MS: m/z (%) ¼ 238 (100) [M]þ.


2. The reaction medium was extracted with an equal volume of diethyl ether (1 mL),

evaporated, diluted to original volume with propan-2-ol (1 mL) and analysed by chiral

HPLC (Chiralcel OD column: rate flow 1 mL min�1, l 254 nm; for (–)-1: n-hexane/

propan-2-ol, 90/10; for (–)-2: n-hexane/propan-2-ol, 95/5; for (–)-3: n-hexane/propan-

2-ol, 90/10) in order to evaluate the degree of oxidation and the enantiomeric excess.

3. The crude reaction mixture was purified by column chromatography (n-hexane/ethyl

acetate: for (–)-1, 2/8; for (–)-2, 4/6; for (–)-3, 4/6) affording pure diastereomers 1a–3a

and 1b–3b. Enantiomeric excess (ee), conversion C and diastereomeric ratio (dr) are

reported in Table 11.2.

11.5.3 Conclusion

The kinetic resolution of �-hydroxy sulfides mediated by CHMO provides an excellent

result in the case of sulfide (–)-2 and moderate results with (–)-1 and (–)-3. Indeed, the

enzyme-catalysed oxidation to sulfoxide 2a showed remarkable enantio- and diastereo-

selectivity with an enantiomeric ratio E of 299 and with an ee � 98 % (C ¼ 47 %).

References

1. (a) Broutin, P.E. and Colobert, F., Enantiopure �-hydroxysulfoxide derivatives as novel chiralauxiliaries in asymmetric biaryl Suzuki reactions. Org. Lett., 2003, 5, 3281. (b) Eames, J. andWarren, S., Synthesis of cyclic sulfides and allylic sulfides by phenylsulfanyl (PhS-) migrationof �-hydroxy sulfides. J. Chem. Soc. Perkin Trans. 1, 1999, 2783. (c) Gelebe, A.C. and Kaye,P.T., Benzodiazepine analogues. Part 15. Synthesis of benzoxathiepine derivatives. Synth.Commun., 1996, 26, 4459. (d) Schwatz, A., Madan, P.B., Mohacsi, E., O’Brien, J.P., Todaro,L.J. and Coffen, D.L., Enantioselective synthesis of calcium channel blockers of the diltiazemgroup. J. Org. Chem., 1992, 57, 851. (e) Kesavan, V., Bonnet-Delpon, D. and Begue, J.P.,Fluoro alcohol as reaction medium: one-pot synthesis of �-hydroxy sulfoxides from epoxides.Tetrahedron Lett., 2000, 41, 2895. (f) Solladie, G., Almario, A. and Dominguez, C.,Asymmetric synthesis of natural products monitored by chiral sulfoxides. Pure Appl. Chem.,1994, 66, 2159. (g) Corey, E.J., Clark, D.A., Goto, G., Marfat, A., Mioskowski, C., Samuelsson,B. and Hammarstrom, S., Stereospecific total synthesis of a ‘slow reacting substance’ ofanaphylaxis, leukotriene C-1. J. Am. Chem. Soc., 1980, 102, 1436.

2. (a) Conte, V., Di Furia, F., Licini, G., Modena, G., Sbampato, G. and Valle, G., Enantioselectiveoxidation of �-hydroxythioethers. Synthesis of optically active alcohols and epoxides.Tetrahedron Asymm., 1991, 2, 257. (b) Pitchen, P. and Kagan, H.B., An efficient asymmetricoxidation of sulfides to sulfoxides. Tetrahedron Lett., 1984, 25, 1049.

3. Colonna, S., Pironti, V., Zambianchi, F., Ottolina, G., Gaggero, N. and Celentano, G.,Diastereoselective synthesis of -hydroxy sulfoxides: enzymatic and biomimetic approaches.Eur. J. Org. Chem., 2007, 363.

4. Colonna, S., Gaggero, N., Pasta, P., Ottolina G., Enantioselective oxidation of sulfides tosulfoxides catalysed by bacterial cyclohexanone monooxygenases. Chem. Commun., 1996,2303.

5. Colonna, S., Gaggero, N., Carrea, G. and Pasta, P., Oxidation of organic cyclic sulfites tosulfates: a new reaction catalyzed by cyclohexanone monooxygenase. Chem. Commun., 1998,415.

6. Branchaud, P. and Walsh C.T., Functional group diversity in enzymatic oxygenation reactionscatalyzed by bacterial flavin-containing cyclohexanone oxygenase J. Am. Chem. Soc., 1995,107, 2153, and references cited therein.

7. Ottolina, G., Bianchi, S., Belloni, B., Carrea, G. and Danieli, B., First asymmetric oxidation oftertiary amines by cyclohexanone monooxygenase. Tetrahedron Lett., 1999, 40, 8483.

11.5 Stereoselective Synthesis of �-Hydroxy Sulfoxides 335

8. Alphand, V., Archelas, A. and Furstoss, R., Microbial transformations 16. One-step synthesis ofa pivotal prostaglandin chiral synthon via a highly enantioselective microbiological Baeyer–Villiger-type reaction. Tetrahedron Lett., 1989, 30, 3663.

9. Colonna, S., Pironti, V., Carrea, G., Pasta, P. and Zambianchi, F., Oxidation of secondaryamines by molecular oxygen and cyclohexanone monooxygenase. Tetrahedron, 2004, 60, 569.

10. Trudgill, P.W., Cyclohexanone 1,2-monooxygenase from Acinetobacter NCIMB 9871.Methods Enzymol., 1990, 188, 70.

11. Doig, S.D., O’Sullivan, L.M., Patel, S., Ward, J.M. and Woodley, J.M., Large scale productionof cyclohexanone monooxygenase from Escherichia coli TOP10 pQR239. Enzyme Microb.Technol., 2001, 28, 265.

12. Secundo, F., Zambianchi, F., Crippa, G., Carrea, G. and Tedeschi, G., Comparative study of theproperties of wild type and recombinant cyclohexanone monooxygenase, an enzyme of syn-thetic interest. J. Mol. Catal. B Enzym., 2005, 34, 1.


11.6 Enantioselective Kinetic Resolution of Racemic 3-Phenylbutan-2-one Using a Baeyer–Villiger MonooxygenaseAnett Kirschner and Uwe T. Bornscheuer*

Baeyer–Villiger monooxygenases (BVMOs) mimic the chemical Baeyer–Villiger oxi-

dation and belong to the class of oxidoreductases. Using molecular oxygen, they can

convert ketones into esters or lactones.1 Most stereoselective Baeyer–Villiger oxida-

tions were described for mono- and bi-cyclic ketones.2 Recently, we have shown that

aliphatic acyclic3 and arylaliphatic4 ketones are also enantioselectively converted by a

BVMO from Pseudomonas fluorescens DSM 50106, which was recombinantly

expressed in Escherichia coli.5 Using whole cells of E. coli JM109 pGro7

pJOE4072.6 expressing this BVMO, preparative kinetic resolution of racemic 3-phe-

nylbutan-2-one and subsequent hydrolysis of the ester product was performed giving

(R)-3-phenylbutan-2-one in 45 % yield with 80 % ee and (S)-1-phenylethanol in 35 %

yield and 93 % ee.

11.6.1 Procedure 1: Recombinant Expression of the BVMO from P. fluorescens

DSM 50106 in E. coli


• Tryptone (5 g)

• yeast extract (2.5 g)

• NaCl (5 g)

• distilled water

• ampicillin stock solution (100 mg mL�1)

• chloramphenicol stock solution (50 mg mL�1)

• L-rhamnose solution (20 % w/v)

• L-arabinose solution (50 mg mL�1)

• stored culture of E. coli JM109 harboring the chaperone plasmid pGro7 and the BVMO-

expression plasmid pJOE4072.6

• phosphate buffer solution (50 mM, pH 7.5)

• one 100 mL shake flask with a cotton plug

• one 1 L shake flask with a cotton plug

• shaker

• photometer

• centrifuge.

11.6.1.2 Procedure

1. Tryptone (5 g), yeast extract (2.5 g) and NaCl (5 g) were dissolved in distilled water, the

volume was adjusted to 500 mL and then autoclaved (20 min, 120 �C). A small portion

of this Luria–Bertani (LB) medium (10 mL) was placed into a sterile 100 mL shake

flask and ampicillin and chloramphenicol solutions were added (LBampþcm) to final

concentrations of 100 mg mL�1 and 20 mg mL�1 respectively. The solution was

inoculated with E. coli JM109 pGro7 pJOE4072.6 and shaken overnight at 37 �C and

200 rpm. This overnight culture (2 mL) was used to inoculate 200 mL LBampþcm in a

11.6 Enantioselective Kinetic Resolution of Racemic 3-Phenylbutan-2-one 337

1 L shake flask supplemented with 0.5 mg mL�1L-arabinose. The culture was incu-

bated at 37 �C and 200 rpm to an optical density (OD) at 600 nm of 0.6, where

expression of the recombinant BVMO was induced by the addition of 0.2 % (w/v)

L-rhamnose. Expression was performed for 4 h at 30 �C and 200 rpm.

2. Cells were then harvested by centrifugation for 20 min at 4400g and 4 �C. The medium

was removed and the cell pellet was washed once with 50 mL phosphate buffer solution

and centrifuged again. The cells can be stored in the fridge for a few days or used

directly for biotransformation.

11.6.2 Procedure 2: Kinetic Resolution of Racemic 3-Phenylbutan-2-one

O O+

O

O

O2

NADPH+H+

H2ONADP+

BVMO


• Phosphate buffer solution (50 mM, pH 7.5)

• racemic 3-phenylbutan-2-one (0.15 g, 1 mmol)

• �-cyclodextrin (0.07 g, 0.5 mmol)

• glucose solution (1 M, 4 mL)

• ethyl acetate


• one 1 L shake flask with a cotton plug

• shaker

• photometer

• centrifuge

• one separatory funnel


11.6.2.2 Procedure

1. The cell pellet of E. coli JM109 pGro7 pJOE4072.6 was resuspended in phosphate

buffer to a final OD at 600 nm of around 20. To 100 mL of this suspension in a 1 L

shake flask racemic 3-phenylbutan-2-one (0.15 g, 1 mmol), �-cyclodextrin (0.07 g,

0.5 mmol) and 1 M glucose solution (2 mL) were added. The reaction mixture was

incubated at 30 �C and 220 rpm. After 4 h, further 1 M glucose solution (2 mL) was

added.

2. After 6 h, the mixture was centrifuged to remove cells from the solution. The pellet was

washed and the reaction solution was extracted several times with ethyl acetate. The

combined organic layers were dried over anhydrous sodium sulfate and concentrated

using a rotary evaporator.


3. The crude product was analyzed by chiral GC (Hydrodex�-�-3P column)4 revealing 46

% conversion with 80 % and 94 % enantiomeric excess of substrate and product

respectively, corresponding to an E-value of 82.

11.6.3 Procedure 3: Enzymatic Hydrolysis of (S)-1-Phenylethyl Acetate

O

+O

O

O

+

OHCAL-A

H2O


• Phosphate buffer solution (50 mM, pH 7.5), 50 mL

• hexane, 10 mL

• Candida antarctica lipase A (CAL-A, Chirazyme L-5, C2), 100 mg

• ethyl acetate


• silica gel

• thin-layer chromatography plates (silica gel 60 F254)

• reaction flask, 250 mL

• water bath





11.6.3.2 Procedure

1. The crude product after kinetic resolution of racemic phenylbutan-2-one was dissolved

in 10 mL hexane and transferred to a reaction flask containing 100 mg CAL-A in 50 mL

phosphate buffer. The reaction mixture was stirred for 24 h at 30 �C.

2. The reaction mixture was extracted several times with ethyl acetate. The combined

organic layers were dried over anhydrous sodium sulfate and concentrated using a

rotary evaporator after filtration.

3. Purification by silica-gel column chromatography (eluent: hexane:ethyl acetate, 5:1)

gave 43 mg (S)-1-phenylethanol (35 % yield, 93 % ee) and 67 mg (R)-3-phenylbutan-

2-one (45 % yield, 80 % ee).

11.6.4 Conclusion

Using the procedure described herein, a racemic arylaliphatic ketone could be efficiently

resolved using a BVMO. Similar arylaliphatic substrates were also shown to be enantio-

selectively converted on an analytical scale by a phenylacetone monooxygenase from

Thermobifida fusca and a 4-hydroxyacetophenone monooxygenase from P. fluorescens

ACB with good to high enantioselectivities.

11.6 Enantioselective Kinetic Resolution of Racemic 3-Phenylbutan-2-one 339

References

1. (a) Walsh, C.T. and Chen, Y.C.J., Enzymic Baeyer–Villiger oxidations by flavin-dependentmonooxygenases. Angew. Chem. Int. Ed. Engl., 1988, 27, 333. (b) Mihovilovic, M.D., Muller, B.and Stanetty, P., Monooxygenase-mediated Baeyer–Villiger oxidations. Eur. J. Org. Chem.2002, 3711. (c) Mihovilovic, M.D., Enzyme mediated Baeyer–Villiger oxidations. Curr. Org.Chem., 2006, 10, 1265. (d) Kamerbeek, N.M., Janssen, D.B., van Berkel, W.J.H. and Fraaije,M.W., Baeyer–Villiger monooxygenases, an emerging family of flavin-dependent biocatalysts.Adv. Synth. Catal. 2003, 345, 667.

2. (a) Mihovilovic, M.D., Rudroff, F., Grotzl, B., Kapitan, P., Snajdrova, R., Rydz, J. and Mach, R.,Family clustering of Baeyer–Villiger monooxygenases based on protein sequence and stereo-preference. Angew. Chem. Int. Ed., 2005, 44, 3609. (b) Taschner, M.J., Black, D.J. and Chen,Q.Z., The enzymatic Baeyer-Villiger oxidation : a study of 4-substituted cyclohexanones.Tetrahedron Asymm., 1993, 4, 1387. (c) Fraaije, M.W., Wu, J., Heuts, D.P.H.M., vanHellemond, E.W., Spelberg, J.H.L. and Janssen, D.B., Discovery of a thermostable Baeyer–Villiger monooxygenase by genome mining. Appl. Microbiol. Biotechnol., 2005, 66, 393.

3. Kirschner, A. and Bornscheuer, U.T., Kinetic resolution of 4-hydroxy-2-ketones catalyzed by aBaeyer–Villiger monooxygenase. Angew. Chem. Int. Ed., 2006, 45, 7004.

4. Geitner, K., Kirschner, A., Rehdorf, J., Schmidt, M., Mihovilovic, M.D. and Bornscheuer, U.T.,Enantioselective kinetic resolution of 3-phenyl-2-ketones using Baeyer–Villiger monooxy-genases. Tetrahedron Asymm., 2007, 18, 892.

5. Kirschner, A., Altenbuchner, J. and Bornscheuer, U.T., Cloning, expression, and characteriza-tion of a Baeyer–Villiger monooxygenase from Pseudomonas fluorescens DSM 50106 in E. coli.Appl. Microbiol. Biotechnol. 2007, 73, 1065.

6. Rodrıguez, C., de Gonzalo, G., Fraaije, M.W. and Gotor, V., Enzymatic kinetic resolution ofracemic ketones catalyzed by Baeyer–Villiger monooxygenases. Tetrahedron: Asymm., 2007,18, 1338.


11.7 Desymmetrization of 1-Methylbicyclo[3.3.0]octane-2,8-dione by theRetro-claisenase 6-Oxo Camphor HydrolaseGideon Grogan* and Cheryl Hill

A range of symmetrical bicyclic �-diketones can be converted to 2,3-disubstituted

cycloalkanones in high yield with high diastereomeric and enantiomeric excess using a

cell-free preparation of a retro-Claisenase enzyme, or �-diketone hydrolase, the gene for

which has been heterologously expressed in Escherichia coli.1

11.7.1 Procedure 1: Preparation of the Crude Enzyme


• Plasmid pGG3

• E. coli BL21 (DE3)

• Luria–Bertani (LB) agar

• stock solution of kanamycin (1 mL, 30 mg mL�1)

• stock solution of isopropylthio-�-galactopyranoside (IPTG, 2 mL, 1 M)

• phosphate buffer pH 7.0 (1 L of 50 mM)

• sterile plastic Petri dishes

• 30 mL sterile plastic culture Sterilin bottles (or 50 mL Falcon tubes)

• orbital shaker with controlled temperature (37 �C)

• 2 L baffled Erlenmeyer flasks

• centrifuge with capacity to centrifuge several hundred millilitres

• ultrasonicator

• liquid nitrogen for snap freezing.

11.7.1.2 Procedure

1. Plasmid pGG3 (a Novagen pET-26b vector into which had been ligated the gene

encoding 6-oxo camphor hydrolase (OCH))1 was transformed into E. coli BL21

(DE3) and the recombinant strain maintained on LB agar plates containing 30 mg

mL�1 kanamycin.

2. A single colony was used to inoculate a 5 mL starter culture in LB medium with 30 mg

mL�1 kanamycin, which was grown overnight at 37 �C.

3. The turbid culture was then used to inoculate 500 mL of LB medium containing 30 mg

mL�1 kanamycin in a 2 L flask. The organism was grown at 37 �C until an optical

density A600 ¼ 0.5.

4. In order to induce expression of the OCH gene, 500 mL of a 1 M solution of IPTG was

then added and the organism incubated at 37 �C for 3 h.

5. The culture was then centrifuged and the cell pellet resuspended in 50 mL 50 mM

phosphate buffer pH 7.

6. The cell suspension was disrupted by ultrasonication and the cell debris removed by

centrifugation.

7. The supernatant was then snap-frozen in liquid nitrogen in aliquots for use directly as

the biocatalyst, which had a specific activity of approximately 9 U mL�1.

11.7 Desymmetrization of 1-Methylbicyclo(3.3.0)octane-2,8-dione 341

11.7.2 Procedure 2: Desymmetrization of 1-Methylbicyclo[3.3.0]octane-2,

8-dione2

O O ORR

CO2HR = Me, Et, allyl, propargyl

6-axo camphor hydrolasephosphate buffer pH 7.0


• Phosphate buffer pH 7.0, 75 mL

• crude OCH preparation (225 U – approximately 25 mL of the preparation described

above)

• diketone substrate3 (100 mg)

• 2 M hydrochloric acid (few drops)


• anhydrous MgSO4 for drying


• 250 mL round-bottomed flask with a magnetic stirrer bar


• filter paper



11.7.2.2 Procedure

1. Transfer 75 mL of the phosphate buffer into a 250 mL round-bottomed flask. To this

add the enzyme solution (25 mL) and stir for 10 min at room temperature.

2. Make up a solution of 100 mg of the diketone substrates in ethanol (2 mL) and add this

dropwise to the stirred buffer. Stir the reaction at room temperature overnight.

3. Analyse the reaction by TLC in a solvent system consisting of 1:1 ethyl acetate/hexane.

The substrate has an Rf of approximately 0.55, and the keto acid product appears at or

just above the baseline. If substrate is still present, then add a further 5 mL of the enzyme

preparation (45 U approximately) and continue stirring for 2 h at room temperature.

4. When TLC shows that the reaction is complete, acidify the mixture to pH 3.0 using a

few drops of 2 M HCl. The enzyme will precipitate. At this stage, the extraction of the

product is facilitated if the precipitated protein is removed by centrifugation.

5. Extract the clear supernatant with ethyl acetate (3� 50 mL) and dry the combined

organic fractions with anhydrous magnesium sulfate. Filter off the drying agent and

remove the solvent in vacuo to yield the crude keto-acid product.

For purposes of analysis, the crude keto-acid was then treated with trimethylsilyl

diazomethane converted to afford 3-(2-methyl-3-oxo-cyclopentyl)-propionic acid methyl

ester, Rf 1:1 petrol/ethyl acetate (0.55).


1H NMR (400 MHz; CDCl3) d 3.66 (3 H, s, OCH3), 2.48–231 (3 H, m), 2.20–2.02 (4

H, m), 1.72–1.58 (2 H, m), 1.37–1.32 (1 H, m) and 1.06 (3 H, d, J 7.0, CH3). 13C NMR (400

MHz; CDCl3) d 220.5 (C¼O), 173.9 (CO2Me), 51.8 (OCH3), 50.4 (CH), 44.2 (CH), 37.3

(CH2), 31.9 (CH2) and 29.6 (CH2), 26.9 (CH2) and 12.6 (CH3).

m/z (chemical ionization; NH3) 202 [100 %, (M þ NH4)þ]. [Found: (M þ NH4)þ,

202.1439 C10H16O3 requires M þ NH4, 202.1443].

The diastereomeric excess (de) and enantiomeric excess (ee) were determined by first

converting the methyl ester to the diastereomeric acetal by acid-catalysed reaction with

(2R,3R)-2,3-butanediol. The acetals were then analysed on a capillary GC HP5 column

(30 m� 0.32 mm� 0.25 mm): injector 250 �C; 320 �C; column 130 �C isothermal. The de

was calculated to be 82 % and the ee >95 %.2

11.7.3 Conclusion

The desymmetrization of 1-alkylbicyclo[3.3.0]octane-2,8-diones can be achieved in a

facile coenzyme-independent enzymatic reaction in buffer. Alkyl chains in the 1-position

of up to at least five carbon atoms are tolerated.2 The yields of the crude keto-acids are

essentially quantitative, and the enantiotopic discrimination by the enzyme is usually

excellent.4 Work remains to be done on the optimization of this biocatalyst with respect

to protein stability and reaction engineering, but it remains a unique and intriguing

possibility for the generation of interesting intermediates bearing multiple chiral centres.


1. Whittingham, J.L., Turkenburg, J.P., Verma, C.S., Walsh, M.A. and Grogan, G., The 2-A crystalstructure of 6-oxo camphor hydrolase: new structural diversity in the crotonase superfamily. J.Biol. Chem., 2003, 278, 1744.

2. Hill, C.L., Verma, C.S. and Grogan, G., Desymmetrisations of 1-alkylbicyclo[3.3.0]octane-2,8-diones by enzymatic retro-Claisen reaction yield optically enriched 2,3-substituted cyclopenta-nones. Adv. Synth. Catal. 2007, 349, 916.

3. A synthetic method for the preparation of the diketone substrates has been presented: Hill, C.L.,McGrath, M., Hunt, T. and Grogan, G., A generic and reproducible route to homo- andheteroannular bicyclic �-diketones via Knochel-type 1,4-conjugate additions to �,�-unsaturatedcycloalkenones. Synlett, 2006, 309.

4. A proposed mechanism for the reaction, and the molecular basis for enantiotopic discrimination,based on X-ray crystallographic studies of the enzyme, has been reported: Leonard, P.M. andGrogan, G., Structure of 6-oxo camphor hydrolase H122A mutant bound to its natural product,(2S,4S)-�-campholinic acid: mutant structure suggests an atypical mode of transition statebinding for a crotonase homolog. J. Biol. Chem., 2004, 279, 31312.

Table 11.3 Desymmetrization of 1-alkylbicyclo[3.3.0]octane-2,8-diones by OCH

R De (%) Ee (%)

Me 82 >95Et 81 >95Allyl 86 >95Propargyl 78 91

11.7 Desymmetrization of 1-Methylbicyclo(3.3.0)octane-2,8-dione 343

11.8 Synthesis of Optically Pure Chiral Lactones by CyclopentadecanoneMonooxygenase-catalyzed Baeyer–Villiger OxidationsShaozhao Wang, Jianzhong Yang and Peter C.K. Lau*

Baeyer–Villiger monooxygenases (BVMOs), typified by cyclohexanone or cyclopentanone

monooxygenases derived from Acinetobacter sp. NCIMB 9871 and Comamonas (formerly

Pseudomonas) sp. NCIMB 9872 respectively, have been shown to be useful reagents for the

preparation of optically active lactones with high enantiomeric excess (ee) and yield.1–4 Bio-

oxidation using BVMOs is among the 12 recommended green chemistry research areas in the

pharmaceutical industry, avoiding such hazardous reagents as organic peracids, chlorinated

solvents or metals that are otherwise used in the chemical Baeyer–Villiger reactions.5 We

recently introduced a new recombinant BVMO, called cyclopentadecanone monooxygenase

(CPDMO) of Pseudomonas origin, that is capable of lactone formation from a broad spectrum

of cyclic ketones ranging in size from substituted C6 to C15 ring compounds. In many cases,

excellent enantioselectivity for the preparation of optically pure chiral lactones was demon-

strated in whole-cell biotransformation experiments.6 The following section describes the

biotransformations of 4-substituted cyclohexanones, 4-t-butyl cyclohexanone in particular,

and several prochiral substrates to the corresponding lactones in good yield and excellent ee

by whole-cell CPDMO desymmetrization, with a simple solvent (ethyl acetate) extraction for

product recovery (Figure 11.5). CPDMO was also found to have an excellent enantioselec-

tivity (E > 200) as well as 99 % (S)-selectivity toward 2-methyl-cyclohexanone for the

production of 7-methyl-2-oxepanone, a potentially valuable chiral building block (Figure

11.6). In the latter case, scale-up synthesis in a 3 L fermenter was demonstrated.

11.8.1 Procedure 1: Propagation of Engineered Escherichia coli Strain

BL21(DE3)[pCD201]


• Luria–Bertani (LB) medium (tryptone peptone 10 g L�1, yeast extract, 5 g L�1, NaCl 5 g

L�1)

• LB-ampicillin (100 mg mL�1) plates

• LB-ampicillin media

• isopropyl-�-D-thiogalacto-pyranoside (IPTG)

• 30 % v/v sterile glycerol, 3 mL


• 10 (2.5 mL) Eppendorf tubes

• shaker.

O

R

O

O

R

E. coli / CPDMO

R = CH3, CH2CH3, C(CH3)3

Figure 11.5 Lactone formation from 4-substituted cyclohexanone catalyzed by E. coli wholecells expressing CPDMO


11.8.1.2 Procedure

1. The E. coli strain BL21(DE3)[pCD201] expressing an IPTG-inducible CPDMO activity

was streaked from a frozen stock on LB-ampicillin plates and incubated at 30 �C until

colonies were 1–2 mm in size. Refer to Sambrook et al.7 concerning media preparation, etc.

2. One colony was used to inoculate 10 mL of a LB-ampicillin medium in a 50 mL

Erlenmeyer flask and incubated at 30 �C, 200 rpm overnight.

3. Sterile glycerol (30 % v/v) was added and the mixture was divided into 1.0 mL aliquots

and stored in a �80 �C freezer.

4. The control carrier strain BL21(DE3)(pSD80) containing the plasmid pSD80 vector

only was propagated using the same protocol except that no ampicillin was used in the

plates or medium.

11.8.2 Procedure 2: Synthesis of (S)-5-t-Butyl-2-oxepanone

O

O

H3C

CH3H3C


• LB-ampicillin medium (90 mL)

• 20% glucose solution (10 mL)

• 4-t-butyl cyclohexanone (50 mg, 0.32 mmol)

• 100 mM IPTG stock solution (100 mL)

• �-cyclodextrin (0.25 g)

• ethyl acetate (� 100 mL)


• hexane

• ethyl acetate

• 500 mL baffled Erlenmeyer flask



• shaker.

11.8.2.2 Procedure

1. One tube of stock culture (1 mL) was thawed in a warm hand and used to inoculate an

LB-ampicillin medium (90 mL) supplemented with 20 % glucose solution (10 mL) in a

500 mL baffled Erlenmeyer flask.

2. The culture was incubated at 30 �C, 200 rpm until the optical density at 600 nm (OD600)

was approximately 0.5–0.7 (around 1.5 h).

11.8 Synthesis of Optically Pure Chiral Lactones by Cyclopentadecanone 345

3. 100 mM IPTG stock solution was added (1 mL per milliliter of medium, final concen-

tration 0.1 mM) followed by the substrate 4-t-butyl cyclohexanone (50 mg, 0.32 mmol)

and �-cyclodextrin (0.25 g).

4. The mixture solution was agitated at 30 �C at 200 rpm for 26 h until the reaction was

finished.

5. The culture solution was extracted with ethyl acetate (3� 100 mL). Combined extracts

were washed once with brine and dried with anhydrous Na2SO4. The solvent was

removed on a rotary evaporator and the residue was purified by flash chromatography

over silica gel to afford the title compound as white crystals (36 mg, 65 % isolated yield).

Chiral-phase gas chromatography (GC) showed >99 % ee, [�]D ¼ �36 (c ¼ 1.7,

CHCl3).

Electron impact mass spectrometry (EI-MS) (m/e): 171 (1 %, Mþ þ 1), 155 (3 %), 114

(100 %), 86 (90 %).1H NMR (CDCl3, 500 MHz) d 4.35 (1H, dd, J1 ¼ 7.8 Hz , J2 ¼ 5.6 Hz), 4.16 (1H, dd,

J1 ¼ 12.6 Hz, J2 ¼ 10.7 Hz), 2.72 (1H, dd, J1 ¼ 14.7 Hz, J2 ¼ 13.1 Hz), 2.58 (1H, t, J ¼11.7 Hz), 2.05 (2H, m), 1.50 (1H, m), 1.32 (2H, m), 0.89 (9H, s) ppm. 13C NMR (CDCl3,

500 MHz) d 23.7, 27.3, 27.4, 27.5, 30.3, 32.9, 33.4, 50.7, 68.6, 176.2 ppm.

11.8.3 Procedure 3: Synthesis of Both Enantiomers of 7-Methyl-2-oxepanone

See Figure 11.6.


• Racemic 2-methyl cyclohexanone (100 mg, 0.89 mmol)

• LB-ampicillin medium (90 mL)

• 20% glucose solution (10 mL)

• 100 mM IPTG stock solution (10 mL)

• ethyl acetate (3 � 100 mL)

• hexane

• ethyl acetate

• anhydrous Na2SO4 (3 g)

O

O

O

kinetic resolution with CPDMO reaction stoppedat 50% conv, &separation

CH3

CH3

O

CH3

36%

19%

99% ee

m -CPBA/TFACH2Cl2

O

O

CH3

99% ee99% ee

Figure 11.6 Kinetic resolution of 2-methycyclohexanone by CPDMO-catalyzed oxidation toyield both enantiomers of 7-methyl-2-oxepanone with high ee values.


• m-chloroperoxybenzoic acid (50 mg)

• trifluoroacetic acid (TFA, 0.2 mL)

• silica gel 60, 200–425 mesh, Fisher Scientific (15 g)

• thin-layer chromatography plates (silica gel 60 F254, Merck)

• flask equipped with a magnetic stirrer bar, 50 mL


• chiral-phase GC, �-Dex 225 column (Supelco Inc.)

• 50 mL and 500 mL separatory funnels



11.8.3.2 Procedure

1. One tube of stock culture (1 mL) was thawed in a warm hand and was used to inoculate

an LB-ampicillin medium (90 mL) plus 20 % glucose solution (10 mL) in a 500 mL

baffled Erlenmeyer flask. The culture was incubated at 30 �C, 200 rpm until OD600 was

approximately 0.5–0.7 (around 1.5 h).

2. 100 mM IPTG stock solution (10 mL) was added followed by the substrate 2-methyl

cyclohexanone (100 mg, 0.89 mmol).

3. The mixture was agitated at 200 rpm at 30 �C (to monitor the reaction, aliquots were

extracted with ethyl acetate and the organic layer analyzed by chiral-phase GC).

4. The kinetic resolution of racemic substrate with CPDMO reaction was stopped at 50 %

conversion and immediately extracted with ethyl acetate. Combined extracts were

washed once with brine and dried with anhydrous Na2SO4.

5. The mixture of optically pure lactone and ketone solution was evaporated by rotary

evaporator to dryness. The residue mixture was separated by flash chromatography

over silica gel (hexane:ethyl acetate 5:1), eluted first as colorless oil (S)-lactone (41 mg,

36 % yield, 99 % ee, [�]D ¼ �16, c ¼ 10, in CH2Cl2), followed by (R)-ketone as

colorless oil (19 mg, 19 % yield, 99 % ee).

6. An analytical sample of (R)-ketone was chemically oxidized with m-chloroperoxyben-

zoic acid (TFA, CH2Cl2) to give (R)-lactone with 99 % ee on chiral-phase GC without

losing any optical purity.

1H NMR (250 MHz; CDCl3) d 1.36(d, J ¼ 6.5 Hz), 1.62(m, 4H), 1.93(m, 2H), 2.65(m,

2H), 4.28(m, 1H) ppm.13C NMR (63 MHz; CDCl3) d 22.5, 22.9, 28.2, 35.0, 36.2, 76.8, 175.6 ppm.

EI-MS: 128 (1 %, Mþ), 113 (2 %), 84 (95 %), 55 (100 %).

11.8.4 Procedure 4: Scale-up Synthesis of (S)-7-Methyl-2-oxepanone in a 3 L

Fermenter

O

O

CH3



• LB medium (100 mL)

• sugar solution (200 g L�1)

• supplemented M9 medium (1 L) containing: 4.0 g Na2HPO4, 2.0 g KH2PO4, 3.0 g

(NH4)2SO4, 0.5 g NaCl, 1.0 g casamino acid, 0.12 g MgSO4, 58.0 mg CaCl2�2H2O, 50.0

mg thiamine, 50.0 mg ampicillin, 6.0 mg FeSO4�7H2O, 20 g glucose, and 4.5 mL US

trace element solution as described.8

• racemic 2-methyl cyclohexanone (20 g)

• phosphate buffer (0.05 M, pH 7.2)

• KOH solution (2 M)

• antifoam (Mazu DF 204, BASF)

• 100 mM IPTG (1 mL)

• dodecane (0.2 mL)

• ethyl acetate


• micro-centrifuge tubes, 1.5 mL

• cuvette, 1 mL

• fermenter (3 L, Biobundles, Applikon Inc., US)

• ReactIRTM 4000 spectrometer (Mettler Toledo, ASI Applied Systems, USA) (optional)

• chiral-GC �-Dex 225 column (Supelco Inc.)

• high-performance liquid chromatograph (Hewlett Packard, Hp 1047A)

• spectrophotometer (Hitachi Model U3210)

• orbital incubator shaker (30 �C, New Brunswick Scientific Innova�43)

• Erlenmeyer flasks, 500 mL

• refrigerators (4�C and �80 �C)




11.8.4.2 Procedure

1. Preculture. Frozen stock cell (1 mL, E. coli BL21(DE3)[pCD201] stored at �80 �C)

was thawed and precultured in 100 mL of LB medium in a 500 mL Erlenmeyer flask.

The rotation speed of the incubator shaker was controlled at 250 rpm and the culture

incubated overnight at 30 �C.

2. Culture and bioconversion. The precultured cells were recovered by centrifugation at

4 �C and the cell pellet was inoculated into a 3 L fermenter (stirred tank with two Rushton

turbine impellers and four baffles) containing 1.0 L of supplemented M9 medium.

3. The cell culture was carried out under the following conditions: temperature 30 �C; pH

7.0 controlled by the addition of 2 M KOH. The fermenter was aerated at 1 vvm via a

submerged sparger and the agitation rate was controlled between 600 and 1000 rpm in

order to maintain the dissolved oxygen concentration above 20 % air saturation.

Foaming was controlled by addition of antifoam (Mazu DF 204, BASF).

4. The dissolved oxygen tension (DOT), feed rate and KOH consumption were monitored.

When the cell density reached an OD600¼ 1, IPTG (1 mL, 100 mM) was added to induce

the CPDMO expression.


5. After about 1 h of induction, 2-methyl cyclohexanone (20.0 g) was dispersed into the

cell medium for the biotransformation.

6. Samples of 10 mL were withdrawn from the fermenter during the course of biotrans-

formation. Sample (1 mL) was immediately extracted with an equal volume of ethyl

acetate. The extracted sample solutions were analyzed by GC. Sample (1 mL) was also

extracted with dodecane (0.2 mL) for fast determination of the biotransformation using

a ReactIR 4000. The cell density was measured using a spectrophotometer and the

residual glucose in the aqueous phase was monitored using high-performance liquid

chromatography.

7. Downstream extraction. The culture broth was diluted with ethyl acetate and the

aqueous phase separated using a separation funnel. The organic layer was collected

and dried over anhydrous sodium sulfate. Removal of the solvent by rotary evaporator

gave (S)-7-methyl-2-oxepanone as a light yellow oil (6.5 g, 38 % yield). Chiral-phase

GC showed 99 % ee and >97 % purity. EI-MS and NMR confirmed the product. Note:

the unconverted (R)-2-methyl cyclohexanone evaporated completely under the aeration

conditions used during the overnight incubation.

11.8.5 Conclusion

CPDMO is a new bioreagent for the synthesis of optically pure lactones with excellent

enantioselectivity. CPDMO is not only effective in desymmetrization of meso and prochiral

compounds (Procedure 2, Section 11.8.2), but excellent in carrying out the kinetic resolution of

racemates (Procedure 3, Section 11.8.3). Additional examples of optically pure lactones that

can be obtained are summarized in Table 11.4. In the fermenter work (Procedure 4, Section

11.8.4), (R)–2-methyl cyclohexanone was not converted, but evaporated under aeration con-

dition (1 vvm). This led to the expected product (S)-7-methyl oxepanone at the end of the

experiment. The optically pure lactone could be recovered without silica-gel chromatography

separation. However, the production yield may be improved by using a better condenser.

Table 11.4 Baeyer–Villiger oxidation by recombinant CPDMO using Procedure 2

Substrate Yield (%) Enantioselectivity Product

4-Methylcyclohexanone 54 99% ee (S) 5-Methyl-2-oxepanone4-Ethylcyclohexanone 74 99% ee (S) 5-Ethyl-2-oxepanonecis-2,6-Dimethylcyclohexanone

74 99% ee (?) cis-3,7-Dimethyl-2-oxepanone

References

1. Stewart, J.D., Cyclohexanone monooxygenase: a useful reagent for asymmetric Baeyer–Villigerreactions. Curr. Org. Chem., 1998, 2, 195–216.

2. Iwaki, H., Hasegawa, Y., Wang, S., Kayser, M.M. and Lau, P.C.K., Cloning and characterizationof a gene cluster involved in cyclopentanol metabolism in Comamonas sp. strain NCIMB 9872and biotransformations effected by Escherichia coli-expressed cyclopentanone 1,2-monooxy-genase. Appl. Environ. Microbiol., 2002, 68, 5671-5684.


3. Mihovilovic, M.D., Rudroff, F. and Grotzl, B., Enantioselective Baeyer–Villiger oxidations.Curr. Org. Chem., 2004, 8, 1057–1069.

4. Ten Brink, G.-J., Arends, I.W.C.E. and Sheldon, R.A., The Baeyer–Villiger reaction: towardsgreener procedures. Chem Rev., 2004, 104, 4105–4123.

5. Constable, D.J.C., Dunn, P.J., Hayler, J.D., Humphrey, G.R., Leazer, Jr, J.L., Linderman, R.J.,Lorenz, K., Manley, J., Pearlman, B.A., Wells, A., Zaks, A. and Zhang, T.Y., Key greenchemistry research areas – a perspective from pharmaceutical manufacturers. Green Chem.,2007, 9, 411–420.

6. Iwaki, H., Wang, S., Grosse, S, Bergeron, H., Nagahashi, A., Lertvorachon, J., Yang, J., Konishi,Y., Hasegawa, Y. and Lau, P.C.K., Pseudomonad cyclopentadecanone monooxygenase display-ing an uncommon spectrum of Baeyer-Villiger oxidations of cyclic ketones. Appl. Environ.Microbiol., 2006, 72, 2707–2720.

7. Sambrook, J.E., Fritsch E.F. and Maniatis, T. Molecular Cloning: A Laboratory Manual, 2ndedn. Cold Spring Harbor Laboratory Press: Cold Spring Harbor, NY, 1989.

8. Panke, S., Held, M., Wubbolts, M.G., Witholt, B. and Schmid, A., Pilot-scale production of (S) -styrene oxide from styrene by recombinant Escherichia coli synthesizing styrene monooxygen-ase. Biotechnol. Bioeng., 2002, 80, 33–41.


12

Whole-cell Oxidations andDehalogenations

12.1 Biotransf ormations of Naphthalene to 4-Hydroxy-1-tetralone byStreptomyces griseus NRRL 8090Arshdeep Khare, Andrew S. Lamm and John P.N. Rosazza

Streptomyces griseus NRRL 8090 catalyzes a series of biotransformations of

naphthalene and 2-methyl-1,4-naphthaquinone to their corresponding racemic and

diastereomeric 4-hydroxy-1-tetralones (Figure 12.1). The yields of 4-hydroxy-1-

tetralone obtained with S. griseus are much higher than those produced by various

fungi that oxidize naphthalene.1

12.1.1 Procedure 1: Cultivation of S. griseus NRRL 8090


• Glycerol (20 g)

• soybean flour (30 g)

• sterile water (5 mL)

• culture of S. griseus NRRL 8090 stored on Sabouraud maltose agar slant at 4 �C

• distilled water

• sterile loop

• two 125 mL DeLong culture flasks with stainless steel cap

• rotary shaker.



12.1.1.2 Procedure

1. Cultures were grown in a two-stage procedure in 25 mL of soybean flour and glycerol

medium (30 g soybean flour and 20 g glycerol in 1 L distilled water) held in stainless-

steel capped, 125 mL DeLong culture flasks. The flasks containing the medium were

autoclaved at 15 psi at 121 �C for 15 min. The surface growth from slants was

suspended in 5 mL of sterile water with a sterile loop and used to inoculate 25 mL

sterile medium (Stage I culture). Cultures were incubated for 72 h, at 29 �C, with

shaking at 200 rpm. A 10 % inoculum derived from the 72-h-old Stage I culture was

used to inoculate sterile medium (Stage II culture), which was incubated for 24 h before

adding naphthalene substrate for biotransformation.

12.1.2 Procedure 2: Synthesis of 4-Hydroxy-1-tetralone

OH

O


• Distilled water

• naphthalene (150 mg)

• N,N-dimethylformamide (DMF, 30–40 mL)

• ethyl acetate

• hexanes

OH O O

OHNaphthalene 1-Naphthol 1-Tetralone 4-Hydroxy-1-Tetralone

O

OH

CH3

O

O

CH3

2-Methyl-1,4-Naphthoquinone 2-Methyl-4-Hydroxy-1-Tetralone

Figure 12.1 S. griseus-catalyzed oxidation of naphthalene and 2-methyl-1,4-naphthaquinone

352 Whole-cell Oxidations and Dehalogenations

• silica gel (GF254) plates, 0.25 mM

• thin-layer chromatography (TLC) solvent system: ethyl acetate:hexanes (50:50 v/v)

• 254 nm UV lamp for TLC plate visualization


• ten 125 mL DeLong flasks with stainless-steel caps

• Shimadzu GC-17A series

• RTX-5 column, 15 m (length), 0.25 mm (i.d.) and 0.15 m film thickness.


• desktop centrifuge.

12.1.2.2 Procedure

2. For analytical purposes, Stage II cultures of S. griseus were prepared in 125 mL

DeLong culture flasks as described. The cultures were shaken at 200 rpm at 29 �C for

24 h and then 10 mg of substrate naphthalene in 30–40 mL of DMF was added to each

25 mL volume of culture and incubations were continued with shaking.

3. Samples (3 mL) of substrate containing cultures were taken at 24, 48 and 120 h after

substrate addition and extracted with equal volumes of ethyl acetate. The organic

phases were separated by centrifugation for 3 min in a desktop centrifuge and used

for TLC analysis. 30–40 mL of sample extracts were spotted onto TLC plates that were

developed with ethyl acetate:hexane (v/v). Visualization of TLC plates was done by

fluorescence quenching under 254 nm UV light. Rf values were: naphthalene, 0.90;

1-naphthol, 0.85; 1-tetralone, 0.8; 4-hydroxy-1-tetralone, 0.32; menadione, 0.86; and

2-methyl-4-hydroxy-1-tetralone, 0.43.

4. Preparative-scale biotransformation of 150 mg naphthalene was conducted using ten

125 mL DeLong flasks, each containing 25 mL of 24-h-old Stage II cultures and 15 mg

of naphthalene in DMF (30–40 mL). After 120 h, contents of all flasks were combined,

centrifuged at 7000g for 20 min. The supernatant was extracted three times with 150 mL

ethyl acetate and cells washed twice with 20 mL of ethyl acetate each time. Organic

extracts were combined, washed with distilled H2O, dried over anhydrous Na2SO4 and

concentrated in vacuo. The residue was dissolved in a minimum amount of ethyl acetate,

applied to a 2� 22 cm silica-gel column and eluted with a hexane ethyl acetate gradient

ranging from 100:3 to 75:25. 4-Hydroxy-1-tetralone was obtained in 43 % yield.

1H NMR (CDCl3, 400 MHz) � 2.17 (1H, m, H-3), 2.41 (1H, m, H-3), 2.58 (1H,

ddd, J¼ 17.8, 9.6, 4.8 Hz, H-2), 2.92 (1H, ddd, J¼ 17.8, 7.5, 4.6 Hz, H-2), 4.98

(1H, dd, J¼ 8.1, 3.9 Hz, H-4), 7.41 (1H, m, H-7), 7.60 (2H, m, H-5 and H-6), 8.03

(1H, d, J¼ 7.7 Hz, H-8).

12.1.3 Procedure 3: Synthesis of 2-Methyl-4-hydroxy-1-tetralone

OH

Me

O

12.1 Biotransformations of Naphthalene to 4-Hydroxy-1-tetralone 353

1. The same method described in Procedure 2 (Section 12.1.2) was used for pre-

parative-scale biotransformation of 150 mg of 2-methyl-1,4-naphthoquinone,

except that reactions were incubated for only 72 h before being combined, cen-

trifuged, extracted and chromatographically purified to give 50 % yield (92 mg) of

product.

1H NMR (CDCl3, 400 MHz) � 1.15 (CH3, 3H, d, J¼ 6.8 Hz), � 1.30 (CH3, 3H, d,

J¼ 6.6 Hz), � 1.42 (CH3, 3H, d, J¼ 6.6 Hz), � 2.51 (1H, m, H-3), � 2.60 (1H, m,

H-3), � 2.8 (1H, m, H-2), � 5.04 (1H, dd, J¼ 11.1, 4.8 Hz, H-4), � 7.38 (1H, m,

H-7), � 7.58 (1H, m, H-6), � 7.7 (1H, d, J¼ 7.52 Hz, H-5), � 8.02 (1H, m, H-8);13C NMR (CDCl3, 100 MHz), � 16.31, � 16.43, and � 17.21 (3-CH3); and signals for

three � 199.20, � 200.67, � 201.50 (3-C¼O).


1. Gopishetty, S.R., Heinemann, J., Deshpande, M. and Rosazza, J.P.N., Aromatic oxidations byStreptomyces griseus: biotransformations of naphthalene to 4-hydroxy-1-tetralone. EnzymeMicrobiol Technol., 2007, 40, 1622.

2. For gas chromatography analysis, samples were spiked with 2-methyl-naphthalene as an internalstandard. Samples were analyzed using a Shimadzu GC-17A series gas chromatograph equippedwith RTX-5 column, 15 m (length) 0.25 mm (i.d.) and 0.25 mm (film thickness). The initialcolumn temperature was 70 �C and temperature was increased at 20 �C min�1 300 �C, andcolumn temperature was held for 13 min. Retention times Rt: naphthalene, 3.2 min; 2-methyl-naphthalene, internal standard, 4.09 min; 1-tetralone, 4.7 min; menadione, 5.68 min; 1-naphthol,5.7 min; 4-hydroxy-1-tetralone, 6.1 min; and 2-methyl-4-hydroxy-1-tetralone, 6.18, 6.27, 6.3 and6.4 min.


12.2 Hydroxylation of Imidacloprid for the Synthesis of OlefinImidacloprid by Stenotrophomonas maltophilia CGMCC 1.1788Sheng Yuan and Yi-jun Dai

Resting cells of bacterium Stenotrophomonas maltophilia CGMCC 1.1788 catalyze the

stereoselective hydroxylation at position C12 of imidacloprid (IMI) in the imidazolidine

ring to form (R)-5-hydroxy IMI. Under acidic conditions, 5-hydroxy IMI is converted into

olefin IMI (Figure 12.2), which exhibits about 19 times more insecticidal efficacy than IMI

against horsebean aphid imago.

12.2.1 Procedure 1: Cultivation of S. maltophilia CGMCC 1.1788


• Luria–Bertani (LB) broth

• tryptone (35 g)


• NaCl (35 g)

• distilled water (3.5 L)

• stored culture of S. maltophilia CGMCC 1.1788

• one plate, 11 cm

• flask with a poromeric silicone plug, 1 L

• fermentor, 5 L

• shaker

• high-speed freeze centrifuge.

12.2.1.2 Procedure

1. A single colony of bacterium S. maltophilia CGMCC 1.1788 strain on LB agar

plate is inoculated to a 1 L flask containing 300 mL of LB broth and cultivated

in a rotary shaker at 220 rpm at 30 �C for 13 h. Then, the culture broth is

poured into the fermentor containing 3.2 L LB broth for cultivation. During

cultivation, the fermentor is constantly aerated and stirred at 500 rpm at 30 �C.

After cultivation for 10 h, the fermentation broth is centrifuged at 6000g for 20

min to obtain the cells of S. maltophilia CGMCC 1.1788 (about 60 g wet

weight).

Figure 12.2 Chemical structure of IMI and its transformation products

12.2 Hydroxylation of Imidacloprid for the Synthesis of Olefin Imidacloprid 355

12.2.2 Procedure 2: Synthesis of 5-Hydroxy IMI


• KH2PO4 (1.3609 g)

• Na2HPO4�12H2O (68.0466 g)

• IMI (3.0 g)

• sucrose (150 g)

• distilled water (3 L)

• anhydrous sodium sulfate (30 g)

• dichloromethane (6.1 L)

• ethyl acetate (3 L)

• acetonitrile (10 mL)

• ultrafiltration membranes, 0.22 mm pore size

• one flask, 5 L

• beaker, 50 mL

• one separatory funnel, 12 L

• fermentor, 5 L


• vacuum pump

12.2.2.2 Procedure

1. Fresh harvested cells were suspended in 3.0 L of 87 mmol L�1 phosphate buffer

(pH 8.0) with 3 g IMI and 150 g sucrose in a 5 L fermentor for transformation.

During transformation, the fermentor was constantly aerated and stirred at 500 rpm at

30 �C for 72 h. At the end of transformation, cells were removed by centrifugation at

6000g for 20 min and the supernatant is collected.

2. The supernatant was first extracted with dichloromethane (2� 3 L) to eliminate the

remaining IMI. The aqueous fraction was then extracted with ethyl acetate (3 L). The

ethyl acetate extract, containing 5-hydroxy IMI, wais dried with 30 g anhydrous sodium

sulfate and concentrated to about 1/20th of the original volume in a vacuum rotary

evaporator and then filtered with 0.22 mm pore size ultrafiltration membranes. The

filtered solution was evaporated again until white crystals were produced. The crystals

were filtered, washed twice with dichloromethane and then dissolved in 10 mL

acetonitrile by heating. At 4 �C, the 5-hydroxy IMI crystallized from the above solution

and was filtered and dried under vacuum. A total of 413 mg of 5-hydroxy IMI was

obtained.


1H NMR (dimethylsulfoxide (DMSO); 400 MHz) � 9.17 (s, 1H, H-10), 8.38 (d, J¼ 2.4

Hz, 1H, H-2), 7.82 (dd, J¼ 8.2, 2.4 Hz, 1H, H-4), 7.51 (d, J¼ 8.2 Hz, 1H, H-5), 6.82

(d, J¼ 7.5 Hz, 1H, 12-OH), 5.25 (ddd, J¼ 7.5, 7.5, 2.5 Hz, 1H, H-12), 4.58 (d, J¼ 16.1 Hz,

1H, H-7), 4.42 (d, J¼ 16.1 Hz, 1H, H-7), 3.84 (dd, J¼ 12.0, 7.6 Hz, 1H, H-11), 3.37

(J¼ 12.0, 2.4 Hz, 1H, H-11). 13C NMR (DMSO; 100 MHz) � 158.9 (C9), 149.7 (C2), 149.6

(C6), 139.7 (C4), 132.9 (C3), 124.5 (C5), 80.3 (C12), 50.6 (C11), 41.8 (C7).

12.2.3 Procedure 3: Synthesis of Olefin IMI


• Distilled water (350 mL)

• hydrochloric acid (5 mL)

• 5-hydroxyl IMI (0.3 g)



• one beaker, 1 L

• one separatory funnel, 1 L

• vacuum rotary evaporator

• water bath.

12.2.3.2 Procedure

1. 5-Hydroxyl IMI (0.3 g) was added to 350 mL distilled water and heated to 80 �C to

obtain a solution. Hydrochloric acid (5 mL) was added and the solution heated at 80 �C

for 35 min. After cooling to room temperature, the reaction solution was extracted with

ethyl acetate (350 mL). The extracts were dried with 5 g anhydrous sodium sulfate and

concentrated in a vacuum rotary evaporator until the product appeared as white needle

crystals. The crystals were collected and dried in air (0.1 g).

1H NMR (DMSO; 400 MHz) � 12.79 (s, 1H, H-10), 8.41 (s, 1H, H-2), 7.77 (d, J¼ 8.0

Hz, 1H, H-4), 7.53(d, J¼ 8.0 Hz, 1H, H-5), 7.38 (s, 1H, H-12), 7.07(s, 1H, H-11), 5.13 (s,

2H, H-7). 13C NMR (DMSO; 100 MHz) � 150.3 (C9), 149.8 (C2), 146.2 (C6), 139.8 (C4),

131.8 (C3), 124.9 (C5), 117.3 (C12), 114.3 (C11), 45.1 (C7).

12.2.4 Conclusion

The procedure is very easy to reproduce and the stereoselective hydroxylation of IMI with

S. maltophilia CGMCC 1.1788 may be applied to some other neonicotinoid insecticides,

such as thiacloprid (Table 12.1).

12.2 Hydroxylation of Imidacloprid for the Synthesis of Olefin Imidacloprid 357

Table 12.1 Transformation of substrates by S. maltophiliaCGMCC 1.1788

Substrates Products Transformationyield (%)

26

28

23

References

1. Dai, Y.J., Yuan, S., Ge, F., Chen, T., Xu, S.C. and Ni, J.P., Microbial hydroxylation ofimidacloprid for the synthesis of highly insecticidal olefin imidacloprid. Appl. Microbiol.Biotechnol., 2006, 71, 927–934.

2. Dai, Y.J., Chen, T., Ge, F., Huan, Y., Yuan, S. and Zhu, F.F., Enhanced hydroxylation ofimidacloprid by Stenotrophomonas maltophilia upon addition of sucrose. Appl. Microbiol.Biotechnol., 2007, 74, 995–1000.


12.3 Biocatalytic Synthesis of 6-Hydroxy Fluvastatin using Mortierellarammaniana DSM 62752 in Shake Flask Culture and on Multi-gramScale using a Wave BioreactorMatthias Kittelmann, Maria Serrano Correia, Anton Kuhn, Serge Parel, Jurgen

Kuhnol, Reiner Aichholz, Monique Ponelle and Oreste Ghisalba

Fluvastatin is a serum cholesterol-lowering drug belonging to the class of ‘statins’, which

acts through inhibition of 3-hydroxy-3-methyl-glutaryl coenzyme A (HMG-CoA) reduc-

tase, the rate-limiting enzyme in cholesterol biosynthesis.1 The 5- and 6-hydroxy and the

N-de-isopropyl derivative represent the major human metabolites of this drug.2 The

synthesis of oxidized drug metabolites via microbial biotransformation has broadly been

discussed in the literature in recent years.3–5 We evaluated the biotransformation of

fluvastatin using different bacterial and fungal wild-type strains as an alternative to

chemical synthesis. With Mortierella (M.) rammaniana DSM 62752 6-hydroxy fluvastatin

was produced (Figure 12.3) in multi-hundred milligram amounts via shake flask culture

and in gram amounts using a BioWave bioreactor. 5-Hydroxy fluvastatin was synthesized

with Streptomyces violascens ATCC 31560 on multi-milligram scale, though with much

lower yield, so that this method will not be outlined in detail.

12.3.1 Procedure 1: Reactivation of M. rammaniana DSM 62752 from a Frozen

Culture on Agar Plates


• Malt extract (2.25 g)

• casitone (0.375 g)

• agar (1.13 g)


• culture of M. rammaniana DSM 62752 frozen at �80 �C

• glass bottle, 200 mL, screw capped

• three Petri dishes

• inoculation loop, sterile

• steam-sterilizator

• water bath, temperature controlled

• incubator, temperature controlled.

12.3.1.2 Procedure

1. Malt extract (2.25 g), casitone (0.375 g) and agar (1.13 g) were dissolved in 75 mL of

distilled water in, for example, a 200 mL screw-capped glass bottle. The screw cap was

not completely closed and the mixture together with a magnetic bar sterilized in the

steam-sterilizer for 20 min at 121 �C.

2. The hot liquid agar medium was mixed by magnetic stirring and cooled to 45 �C in a

temperature-controlled water bath. Then the agar was poured into three Petri dishes and

solidified by cooling to room temperature.

3. The agar plates were inoculated on the whole surface from a culture of M. rammaniana

frozen at �80 �C using a sterile inoculation loop and incubated for 4 days at 28 �C.

12.3 Biocatalytic Synthesis of 6-Hydroxy Fluvastatin 359

12.3.2 Procedure 2: Preculture and Main Culture of M. rammaniana and Synthesis

of 6-Hydroxyfluvastatin

N

OH OH

F

OH

O

OH


• Distilled water (16.5 L)

• glucose (281 g)

• Lab-Lemco (Oxoid) (42 g)

• peptone from casein (52.5 g)


• casitone (Becton Dickinson) (31.5 g)

• NaCl (15.75 g)

• 3-morpholino propane sulfonic acid (MOPS) (220.5 g)

• NaOH solution, 4 M

• fluvastatin-Na (1 g, 2.23 mmol)


• XAD-16 adsorber resin (16 g) (Rohm and Haas France S.A.S., Lauterbourg, France)

• isopropanol (2 L)

• ethyl acetate (4.4 L)


• NaCl (100 g)

N

OH OH

F

O

O

N

OH OH

F

O

O

OH

N

OH OH

F

O

O

OH

6

5

5-Hydroxy fluvastatin-Na

6-Hydroxy fluvastatin-Na

Mortierella rammaniana

Fluvastatin-Na

DSM 62572

Streptomyces violascensATCC 31560

Na+

Na+

Na+

Figure 12.3 Synthesis of 5- and 6-hydroxy fluvastatin by microbial biotransformation


• MgSO4, anhydrous

• RP18 silica gel (30 g) (Lichroprep RP18 40–60 mm, Merck KGaA, Darmstadt,

Germany)

• acetonitrile, high-performance liquid chromatography (HPLC) gradient grade (280 mL)

• KH2PO4 (0.134 g, 0.98 mmol)

• Na2SO4, anhydrous

• 25 Erlenmeyer flasks, 2 L, four baffles

• 5 Erlenmeyer flasks, 500 mL, one baffle

• 1 Erlenmeyer flask, 1 L

• cotton

• gauze

• inoculation loop, sterile

• steam sterilizer

• laboratory shaker, 5 cm agitation radius

• 20–30 pipettes, 25 mL, sterile

• 10–15 pipettes, 1 mL, sterile

• polypropylene tube, 50 mL, screw capped, presterilized (e.g. Falcon tubes, Becton

Dickinson Labware, Franklin Lakes, NJ, USA)

• filter funnel

• sinter glass filter funnel

• separatory funnel for 1 L extraction

• filter paper


• high-vacuum pump.

12.3.2.2 Procedure

Growth of M. rammaniana and Biotransformation of Fluvastatin

1. Glucose (281 g), Lab-Lemco (Oxoid) (42 g), peptone from casein (52.5 g), yeast

extract (52.5 g), casitone (31.5 g), NaCl (15.75 g), and MOPS (220.5 g) were

dissolved in 10.5 L of distilled water and the pH was adjusted to 6.5 with 4 M

NaOH.

2. The resulting solution was filled in 400 mL portions into 25 Erlenmeyer flasks

with 2 L total volume equipped with four baffles and in 100 mL portions in five

Erlenmeyer flasks with 500 mL total volume equipped with one baffle. The flasks

were closed by cotton plugs wrapped in gauze and autoclaved at 121 �C for

20 min.

3. The five flasks with each 100 mL of medium (precultures) were inoculated with

mycelium of M. rammaniana from the agar plates using a sterile inoculation loop and

incubated on a laboratory shaker with 5 cm agitation radius at 28 �C and 220 rpm for

3 days.

4. Each of the 2 L Erlenmeyer flasks (main cultures) was then inoculated with 20 mL of

preculture and incubated at 28 �C and 180 rpm.

5. 1 g of fluvastatin-Na was dissolved in 10 mL of methanol (fluvastatin solution) in, for

example, a presterilized, screw-capped 50 mL polypropylene tube. Glucose (50 g) was

dissolved 0.5 L of distilled water and sterilized at 121 �C for 20 min.


6. After 48 h of incubation, 0.4 mL of the methanolic fluvastatin solution and

20 mL of the sterile glucose solution were added to each of the 2 L shake flasks

under sterile conditions. Incubation under shaking was continued for another 42 h.

The degree of conversion was measured by analytical RP18-HPLC with diode

array detection.6

Purification of 6-Hydroxy Fluvastatin

1. To each of the flasks, 16 g of the adsorber resin XAD-16 was added and the flasks were

shaken for a further 3 h. The resin was collected by filtering off over gauze in a filter

funnel and washed with 4 L of distilled water. Then it was eluted four times with

portions of 500 mL of isopropanol by gentle shaking in a 1 L Erlenmeyer flask for

30 min and filtering off the resin. The solvent was removed under reduced pressure at

30 �C bath temperature. The residue was dissolved in 400 mL of ethyl acetate and

washed twice with 400 mL of saturated NaCl solution. The organic phase was dried

over anhydrous MgSO4 and the solvent removed under reduced pressure at 30 �C bath

temperature. Further purification was performed via a second solid-phase extraction on

RP18 silica gel.

2. The crude extract (3 g) was dissolved in acetonitrile (30 mL) and mixed with dry solid

RP18-phase (30 g) and 270 mL of potassium phosphate buffer 0.7 mM pH 7 (¼Kpi-

buffer, preparation: KH2PO4 (0.134 g, 0.98 mmol) dissolved in 1400 mL distilled

water, pH adjusted to 7 with 0.1 M KOH). The mixture was filtered in a sinter-glass

filter funnel and the RP18 silica gel was washed with a 10 % (v/v) solution of

acetonitrile in Kpi-buffer (300 mL). Subsequently, the RP18 silica gel was eluted

twice with 500 mL of a 25 % (v/v) solution of acetonitrile in Kpi-buffer. To each of

the two resulting fractions, �50 g of NaCl was added and they were extracted twice

with 500 mL of ethyl acetate. The two organic phases were dried over anhydrous

Na2SO4, filtered over filter paper and the solvent was removed under reduced pressure

at 20 �C and finally under high vacuum for 2 h. (Fraction 1: 480 mg, light brown resin,

62 % purity RP18 HPLC-UV205 nm, 57 % RP18 HPLC–mass spectrometry (MS),

>27 % molar yield, structure identification in comparison with chemically synthesized

6-hydroxy fluvastatin).

1H NMR (400 MHz, dimethylsulfoxide) �¼ 1.19 (6H, dd, appears as t), 1.29 (1H, m),

4.83 (1H, hept J¼ 8 Hz), 5.66 (1H, dd, J¼ 4 and 16 Hz), 6.50–6.53 (2H, m), 7.20–7.29

(4H, m), 7.40–7.45 (2H, m).

12.3.3 Procedure 3: Synthesis of 6-Hydroxy Fluvastatin with M. rammaniana DSM

62752 in a BioWave Bioreactor on 22 L Scale


• Distilled water 23 L

• Tween 80 (70 mg)

• glycerol (14 g)

• cellulose powder (4 g)

• oat meal (2 g)


• tomato paste (2 g), low salt, no preservatives, in the original recipe the brand is Hunt’s

Tomato Paste

• KH2PO4 (11.3 g)

• MgSO4 (0.2 g)

• agar (4 g)

• glucose (44 g)

• malt extract (220 g)

• yeast extract (88 g)

• NH4Cl (11 g)

• 2-morpholino ethane sulfonic acid monohydrate (MES, 429 g)

• antifoam 204 (Sigma)

• antifoam Y-30 solution (Sigma)

• glass bottle, screw capped, 100 mL

• glass bottle, screw capped, 500 mL

• magnetic bar

• steam sterilizator

• water bath, temperature controlled

• eight Petri dishes

• 10 pipettes, 10 ml, sterile

• 10 Eppendorf tips, truncated, sterile

• 10 L-shaped spreaders, plastic, sterile (e.g. VWRI612-1560, VWR International)

• two polypropylene tube, 50 mL, screw-capped, presterilized (e.g. Falcon tubes, Becton

Dickinson Labware, Franklin Lakes, NJ, USA)

• BioWave 50SPS bioreactor (Wave Biotech AG, Tagelswangen, Switzerland) (since

recently distributed by Sartorius BBI Systems GmbH, Melsungen, Germany, as

‘Biostat Cultibag RM 50’)

• Wavebag 50 L total volume, exhaust gas line 0.5 inch

• peristaltic pump (Heidolph Pumpdrive 5006)

• sterile microfiltration capsule, pore size 0.45þ 0.2 mm (Sartobran 300, 5231307-H5–00,

Sartorius Biotech GmbH, Gottingen, Germany)

• six sterile disposable syringes, 50 mL

• 50–60 sterile disposable syringes, 10 mL (for sampling)

• membrane pump KNF Labport type N86KN.18 (KNF Neuberger, Freiburg, Germany)

• thermal mass flowmeter for air, type GCA-B5SA-BA20 and

• thermal mass flowmeter for oxygen, type GCR-A9SA-BA15 (Thermal Mass Flow Co.,

USA)

• sterilized glass bottle, 1 L, screw capped (as foam trap)

• spectrophotometer.

12.3.3.2 Procedure

Preparation of Spore Suspension

1. Tween–glycerol solution. Tween 80 (70 mg) and glycerol (14 g) were dissolved in

distilled water (59 mL) and sterilized at 121 �C for 20 min.


2. Preparation of sporulation agar plates. Cellulose powder (4 g), oat meal (2 g), tomato

paste (2 g), KH2PO4 (0.3 g), MgSO4 (0.2 g) and agar (4 g) were dissolved in 200 mL of

distilled water in a 500 mL screw-capped glass bottle; the screw cap was not completely

closed, and the mixture together with a magnetic bar was sterilized in the steam-

sterilizator for 20 min at 121 �C; the hot liquid agar medium was mixed by magnetic

stirring and cooled to 45 �C in a temperature-controlled water bath; then the agar was

filled into eight Petri dishes and solidified by cooling to room temperature.

3. 7 mL of Tween–glycerol solution was mixed with M. rammaniana mycelium grown

densely on a malt extract agar plate (see Procedure 1, Section 12.3.1) using a sterile

plastic L-shaped spreader.

4. Portions (100 mL) of mycelium suspension were transferred to eight sporulation agar

plates with a sterile, truncated Eppendorf tip and spread with a sterile plastic L-shaped

spreader.

5. The plates were incubated for 8 days at 28 �C. Then Tween–glycerol solution (7 mL) was

filled onto each plate and the spores/mycelium were suspended by rigorous scraping of

the agar surface again with a sterile plastic L-shaped spreader. The spore/mycelium

suspension was stored in two sterile 50 mL polypropylene tubes at �80 �C until use.

Growth of M. rammaniana and Biotransformation

In BioWave bioreactors, a disposable polyethylene bag (Wavebag) serves as the cell

containment which is rocked on a temperature-controlled table for mixing and gas

exchange. Oxygen is supplied via a stream of sterile air/oxygen mixture through the head-

space of the bag. For the cultivation of highly oxygen-demanding microorganisms the

required gas flow exceeds the capacity of the built-in pump, so that an external membrane

pump had to be employed. Furthermore, supplementation with pure oxygen was necessary.

1. Preparation of concentrated medium. Glucose (44 g), malt extract (220 g), yeast extract

(88 g), KH2PO4 (11 g), NH4Cl (11 g) and MES (429 g) was dissolved in 4.4 L of

distilled water, the pH was adjusted to 5.9 with concentrated NaOH and the liquid was

sterilized at 121 �C for 30 min in a steam sterilizer.

2. The Wavebag (50 L volume) was placed on the temperature-controlled tray and

completely inflated with air using a membrane pump. The airflow was adjusted to

2250 mL using a thermal mass flow meter.

3. The concentrated sterilized medium was pumped into the Wavebag with a peristaltic

pump through autoclaved silicone tubes. Then distilled water (17.6 L) was pumped in

through a disposable, sterile microfiltration capsule. Rocking was started at an angle of

10.5� at 36–37 rpm as well as the temperature regulation, set-point 28 �C.

4. When the temperature was equilibrated, spore suspension (28 mL) was added with a

sterile disposable 50 mL syringe through the sampling port, followed by 10 mL of a

heat-sterilized antifoam 204 emulsion (1 mL antifoam 204 mixed with 9 mL of distilled

water). Then, the supply of pure oxygen was started and adjusted to 250 mL min�1

using a second thermal mass flow meter. The mixing of air and oxygen in the desired

ratio was effected by joining the line for pure oxygen and the gas inlet of the pump

(aspiration port) with a ‘T’ connector, allowing both gases to be taken in (Figure 12.4).

In the gas inline tubing between the air flow meter and the Wavebag, a hydrostatic

pressure relief valve was built in, providing a maximum backpressure of 20 cm water


column. In the exhaust gas line between the Wavebag and the sterile filter, a sterile 1 L

screw-capped glass bottle containing a few millilitres of antifoam Y-30 emulsion was

included as a foam trap.

5. During the day, samples were taken around every 2 h and analysed without dilution for

pH. Growth was estimated by measuring the optical density at 600 nm (OD600) against

distilled water in samples diluted to OD600 £ 0.3 with distilled water using a

spectrophotometer.

6. After 30 h from inoculation, NaOH solution (4 m, 40 ml) was injected into the Wavebag

with a sterile disposable syringe. 8.8 g of fluvastatin-Na was dissolved in 88 mL of

methanol. At 48, 74, 120, and 192 h after inoculation with spores, fluvastatin solution

(22 mL) was supplemented to the culture via the sampling port and a sterile disposable

50 mL syringe. The degree of conversion was measured by analytical RP18-HPLC with

diode array detection.6

7. After 10–12 days, the maximum product concentration was achieved and fluvastatin

completely consumed. The 6-hydroxy-fluvastatin-containing culture liquid was stored

in the Wavebag at �20 �C for later purification. In other cases, the culture liquid was

conveniently and safely harvested from the Wavebag without aerosol formation by

sucking into glass bottles in a vacuum line with a sterile filter installed between the

collecting vessel and the vacuum pump.

12.3.4 Conclusion

The synthesis of 6-hydroxy fluvastatin with M. rammaniana DSM 62752 gave high con-

version (>95 %) in shake flask culture on 400 mL scale with 0.1 g L�1 of fluvastatin as well

as on 22 L scale in a Wave bioreactor-fed batch process at a final substrate concentration of

0.4 g L�1. Instead of the partial purification by a second solid-phase extraction described

above, 6-hydroxy fluvastatin can be obtained in high purity (�95 %) by, for example,

preparative medium-pressure liquid chromatography (MPLC) on RP18 silica gel.7

5-Hydroxy fluvastatin could be prepared analogously via biotransformation in shake

flask culture with Streptomyces violascens ATCC 31560. Different media and minor

variations of the process schedule had to be applied.8 Before supplementation of

F

Exhaust airF

Oxygen

Air

Membrane pump

Air flow meter

Oxygenflow meter

Wavebag

Sterileinlet filter

Foam trap

Sterile filter

Overpressurerelease via 20 cm water-column

1/2" exhaust tubing

Figure 12.4 Gas flow in the BioWave reactor under oxygen supplementation


fluvastatin it was important that glucose had been completely consumed (check with

urine–glucose test sticks). Furthermore, the pH had to be stabilized in the culture by

addition of CaCO3 at the time of fluvastatin addition.9 Since the organism produced

both 5- and 6-hydroxy fluvastatin, purification via RP18-LC was needed.7


1. Christians, U., Jacobsen, W. and Floren, L.C., Metabolism and drug interactions of 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors in transplant patients: are the statins mechan-istically similar? Pharmacol. Ther., 1998, 80, 1–34.

2. Fischer, V., Johanson, L., Heitz, F., Tullmann, R., Graham, E., Baldeck, J.P. and Robinson, W.T.,The 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitor fluvastatin: effect on humancytochrome P-450 and implications for metabolic drug interactions. Drug Metab. Dispos., 1999,27, 410–416.

3. Azerad, R., Microbial models for drug metabolism. In Biotransformations, Faber, K. andScheper, T. (eds), Advances in Biochemical Engineering/Biotechnology, vol. 63, Springer,1999, pp. 169–218.

4. Venisetty, R.K. and Ciddi, V., Application of microbial biotransformation for the new drugdiscovery using natural drugs as substrates. Curr. Pharm. Biotechnol., 2003, 4, 153–167.

5. Ghisalba, O. and Kittelmann, M., Preparation of drug metabolites using fungal and bacterialstrains. In Modern Biooxidation – Enzymes, Reactions and Applications, Schmid, R.D. andUrlacher, V.B. (eds). Wiley–VCH Verlag, Weinheim, 2007, pp. 211–232.

6. Sample preparation. Culture liquid (0.5 mL) was mixed with isopropanol (0.5 mL), kept at roomtemperature for 10 min and centrifuged at 25 000g and 20 �C in a refrigerated Eppendorfcentrifuge. The supernatant was subjected to HPLC-analysis. HPLC system Agilent 1100;column: Chromolith Performance RP-18e 100 mm � 4.6 mm, pre-column Chromolith GuardCartridge RP-18e 5 mm � 4.6 mm (Merck KgaA, Darmstatt, Germany); elution: flow rate 2 mLmin�1, eluent A¼ 3 mM H3PO4, eluent B¼ acetonitrile (gradient grade), gradient 5–100 % B in4.75 min; injection volume 10 mL; diode array detection 190–400 nm.

7. Preparative MPLC. Solid phase Lichroprep RP18 40–60 mm (Merck KGaA, Darmstadt,Germany), first gradient 5–45 % acetonitrile against 1 mM ammonium formate (six columnvolumes), second gradient 5–50 % acetonitrile with 1 mM formic acid as the aqueous phase.

8. Variations for 5-hydroxy fluvastatin. Medium for growth on agar plates: Plate Count Agar (Fluka/Sigma Aldrich, Buchs, Switzerland); medium for preculture and main culture: glucose 20 g L�1,soytone (Becton Dickinson) 15 g L�1, yeast extract 10 g L�1, pH adjusted to 6.5 with NaOH.Incubation time of main culture before fluvastatin addition 3 days; biotransformation period 24 h.

9. Each flask (400 mL culture) was supplemented with a steam-sterilized (121 �C) suspension ofCaCO3 (3 g) in distilled water (30 mL) immediately after fluvastatin addition.


12.4 Synthesis of 1-Adamantanol from Adamantane through RegioselectiveHydroxylation by Streptomyces griseoplanus CellsKoichi Mitsukura,* Yoshinori Kondo, Toyokazu Yoshida and Toru Nagasawa

Microbial hydroxylation, which introduces regioselectively a hydroxyl group at a non-

activated carbon atom of alicyclic compounds, is an attractive and promising method for

the synthesis of useful fine chemicals. Recently, we found that Streptomyces griseoplanus

AC122 catalyzed highly regioselective hydroxylation of adamantane.1 Through hydro-

xylation of adamantane by S. griseoplanus AC122 cells, 1-adamantanol was synthesized in

the culture broth (Figure 12.5).

12.4.1 Procedure 1: Cultivation of S. griseoplanus AC122


• Glucose (0.4 g)



• tap water 100 mL



• magnesium sulfate heptahydrate (0.012 g)

• tap water 100 mL

• one 50 mL test tube with a poromeric silicone plug

• one shaking-flask with a cotton plug, 500 mL

• reciprocal shaker.

12.4.1.2 Procedure

1. Glucose (0.4 g), malt extract (1 g) and yeast extract (0.4 g) were dissolved with water

and the volume was adjusted to 100 mL with tap water. The seed culture medium

(4 mL) was placed in a 50 mL test tube. Seed culture of S. griseoplanus AC122 was

carried out for 48 h at 28 �C with reciprocal shaking (115 strokes per minute).

2. Malt extract (1 g), yeast extract (1 g) and magnesium sulfate heptahydrate (0.012 g)

were dissolved with water and the volume was adjusted to 100 mL with tap water. The

culture medium (30 mL) was placed in a 500 mL shaking-flask with a cotton plug and

sterilized (121 �C, 20 min). The seed culture broth was transferred to a 500 mL shaking-

flask containing 30 mL of the culture medium. Cultivation was carried out for 48 h at 28

�C and 115 strokes per minute.

OH

S. griseoplanus AC122

32% yield

Figure 12.5 Hydroxylation of adamantane using Streptomyces cells

12.4 Synthesis of 1-Adamantanol from Adamantane 367

12.4.2 Procedure 2: Synthesis of 1-Adamantanol

OH


• Adamantane (41 mg, 0.3 mmol)

• Tween 60 (900 mg)

• ethyl acetate


• n-hexane and ethyl acetate

• filter paper

• silica gel (Wakogel C300 45–75 mm), 15 g



12.4.2.1 Procedure

1. Adamantane (40.9 mg, 0.3 mmol) and Tween 60 (900 mg) were added to 30 mL of

culture medium after 48 h of cultivation. The conversion of adamantane was carried out

for 72 h with reciprocal shaking (115 strokes per minute) at 28 �C.

2. The cells were removed from the culture broth by centrifugation (12 000 rpm, 30 min)

and the supernatant was extracted with ethyl acetate. The organic layer was collected,

dried over anhydrous magnesium sulfate and concentrated using a rotary evaporator.

The crude product was purified by a silica-gel column chromatography using eluents

(the mixture of n-hexane and ethyl acetate was used stepwise at a ratio (v/v) of 1:0, 8:1,

4:1 and 2:1) to give 1-adamantanol (13 mg, 32 % yield).

1H NMR (500 MHz, CDCl3) � 1.47 (1H, s), 1.62 (6H, dd, J 24.3, 12.3 Hz) 1.71 (6H, d, J

2.3 Hz), 2.14 (3H, s); 13C NMR (125 MHz, CDCl3) � 30.7, 36.0, 45.3, 68.2; electron impact

mass spectrometry m/z (%) 152 (Mþ, 29.4), 95 (58.9).

12.4.3 Conclusion

The procedure is an approach for the synthesis of 1-adamantanol from adamantane by a

green bioprocess.

Reference and Note

1. Mitsukura, K., Kondo, Y. Yoshida T. and Nagasawa, T., Regioselective hydroxylation ofadamantane by Streptomyces griseoplanus cells. Appl. Microbiol. Biotechnol., 2006, 71, 502.Adamantane-hydroxylating strains such as Dothiora phaeosperma and Streptomyces griseusIFO3237 can be utilized for 1-adamantanol production.


12.5 Enantioselective Benzylic Microbial Hydroxylation of Indan andTetralinRenata P. Limberger, Cleber V. Ursini, Paulo J.S. Moran and J. Augusto

R. Rodrigues

Benzylic microbial hydroxylations of hydrocarbons have been found to be an important

tool in organic chemistry. Their efficiency, specificity and environmentally benign con-

ditions make this approach superior to many chemical-based methods,1 especially in view

of unsolved chemical problems, such as a certain lack of control and predictability of the

product structures and the expense of oxidizing reagents.2 Recently, we have described a

screening of 15 strains of bacteria and fungi targeted at the production of specific hydro-

xylated benzylic derivatives of indan and tetralin.3 Among the cultures screened,

Mortierella isabellina CCT3498, Mortierella ramanniana CCT4428 and Beauveria bassi-

ana CCT3161 (from ATCC 7159) were shown to mediate the respective conversions of the

hydrocarbons into 1-indanol and 1-tetralol. The most satisfactory results were achieved

with M. isabellina, which afforded (R)-1-indanol (78 % conversion, 64 % yield, 86 % ee)

after a 2-day incubation and (R)-1-tetralol (50 % conversion, 38 % yield, 92 % ee) in a

4-day incubation. Overoxidation of 1-indanol and 1-tetralol during the reactions resulted in

the formation of 1-indanone and 1-tetralone respectively.

12.5.1 Procedure 1: Synthesis of (R)-1-Indanol

indian (R)-1-indanol 1-indanone64% yield

OH O

M. isabellina


• Indan (95 %, Acros) (100 mg, 0.846 mmol, 2.0 mL of ethanolic solution at 50 mg mL�1)

• M. isabellina CCT3498 strain (Tropical Culture Collection, Andre Tosello Research

Foundation)4

• potato–dextrose–carrot broth (PDCB, 200 mL)5

• pH 6.0 potassium phosphate buffer (18.4 g KH2PO4 and 4.025 g Na2HPO4 in 1 L)

(200 mL)

• ethyl acetate (500 mL, Merck)

• hexane (700 mL, Merck)

• sodium chloride, saturated aqueous solution

• silica gel (400–200 mesh, Aldrich) (25 cm height in glass column)


• two 500 mL Erlenmeyer flasks

12.5 Enantioselective Benzylic Microbial Hydroxylation of Indan and Tetralin 369

• orbital shaker

• centrifuge

• 500 mL filter flask

• Buchner funnel

• filter paper


• 250 mL separating funnel


• equipment for continuous liquid–liquid extraction

• equipment for flash column chromatography using a glass column (2.5 cm� 25 cm)


• gas chromatograph–mass spectrometer (GC–MS)

• J&W Scientific HP-5 (30 m� 0.25 mm� 0.25 mm) or Supelco Simplicity 1

(30 m� 0.25 mm� 0.25 mm) fused-silica capillary column

• Marcherey 212117/91 Hydrodex-� 3P (25 m� 0.25 mm� 0.25 mm) fused silica capil-

lary column

• polarimeter

• nuclear magnetic resonance spectrometer.

12.5.1.2 Procedure

1. A culture of M. isabellina CCT3498 was aseptically transferred into conical

Erlenmeyer flasks (500 mL) containing 200 mL of sterile PDCB5 and kept on a rotary

shaker (150 rpm) at 30 �C for 3 days to acquire biomass.

2. Microbial biomass was harvested by centrifugation (5000 rpm, 10 min) and 30 g (wet

weight) was transferred to clean 500 mL conical flasks containing 200 mL of 0.2 ionic

strength potassium phosphate buffer solution at pH 6.0.

Attention: to avoid the lack of efficiency, new cultures of M. isabellina should be

used and a careful control of incubation conditions (temperature, pH and medium) is

necessary.

3. 100 mg of indan (2.0 mL of ethanolic solution at 50 mg mL�1) was added.

4. The flask was returned to the shaker (150 rpm) at 30 �C for 2 days.

5. After 2 days, three 5.0 mL portions of the incubation mixture were harvested by

vigorous shaking and extraction with 5.0 mL of ethyl acetate. If necessary, a cen-

trifugation procedure (3 min at 3000 rpm) was used to break the emulsion.

6. The organic fraction was collected and 1 mL of the solution was submitted to the

GC–MS for qualitative and chiral analysis to certify conversion to 1-indanol.

7. For qualitative analyses, the GC system was equipped with a J&W Scientific HP-5 or a

Supelco Simplicity 1 fused-silica capillary column. Injector and detector temperatures

were set at 220 �C and 240 �C respectively; the oven temperature was programmed

from 60 to 230 �C at 40 �C min�1. Helium was employed as carrier gas (1 mL min�1).

Compound identification was based on a comparison of mass spectra with those of

synthetic racemic and enantiomeric-enriched samples. The retention times for indan,

1-indanol and 1-indanone were 4.7 min, 5.9 min and 6.2 min respectively.

8. For chiral analyses the GC system was equipped with a Marcherey 212117/91

Hydrodex-� 3P fused-silica capillary column. The oven temperature was


programmed from 100 to 210 �C at 10 �C min�1. Injector and detector temperatures

were set at 200 �C and 240 �C respectively. Hydrogen was employed as carrier gas

(1 mL min�1). Under these conditions, the retention times obtained for (S)-1-indanol

and (R)-1-indanol were 8.07 min and 8.11 min respectively.

9. After the incubation time, the culture was harvested and filtered. The filtrate was

saturated with a sodium chloride saturated aqueous solution, stirred for 3–4 h at room

temperature and then extracted six times with ethyl acetate (20 mL).

10. The organic phase was saved and the sodium chloride saturated aqueous solution

remaining was extracted again by a continuous liquid–liquid process at�50 �C for 24 h.

11. The combined organic extracts were dried over anhydrous sodium sulfate and con-

centrated in vacuum after filtration.

12. The crude residue was purified by flash column chromatography on silica gel using

300 mL portions of hexane:ethyl acetate (90:10 and 80:20); 10 mL fractions were

collected, giving indanone with a 9:1 ratio and 1-indanol (64 mg, 0.541 mmol) with

8:2 ratio.

13. The isolated 1-indanol was collected and 1 mL of the solution was submitted for

GC–MS analysis, as described above, and the compound identity was confirmed by

nuclear magnetic resonance spectrometry.7,8

(R)-1-Indanol was isolated as a white solid in 64 % yield. M.p. 67–68.0 �C;

[�]23D ¼�25� (c¼ 0.41, CHCl3), 88 % ee. Lit.:9,10 m.p. 72 �C, [�]22

D ¼þ34� (c¼ 1.895,

CHCl3) for (S) enantiomer. 1H NMR (CDCl3, 300 MHz): � 7.46–7.42 (m, 1H, Ph),

7.24–7.20 (m, 2H, Ph), 7.14–7.10 (m, 1H, Ph), 5.27 (br t, J¼ 5.9 Hz, 1H, CHOH), 3.07

(ddd, J¼ 16.2 Hz, J¼ 8.4 Hz, J¼ 4.8 Hz, 1H, CHHCH2CH(OH)), 2.84 (br dd, J¼ 16.2 Hz,

J¼ 7.0 Hz, 1H, CHHCH2C(OH)H), 2.51 (m, 1H, CHHC(OH)H), 1.97 (m, 1H,

CHHC(OH)H), 1.75 (br s, 1H, OH). MS (electron impact (EI)): m/z (relative intensity)

134 (Mþ, 51), 133 (100), 117 (12), 116 (14), 115 (28), 105 (30), 103 (12), 91 (32), 89 (9), 79

(25), 78 (15), 77 (45), 74 (2), 66 (16), 65 (22), 63 (25), 57 (25), 55 (32), 53 (9), 52 (13), 51

(57), 50 (29).

Racemic standards of 1-indanol, to be used for chiral GC analysis, can be prepared by

treatment of indanone with NaBH4 in MeOH, as described by Aina et al.6

Enantiomerically enriched samples of 1-indanol, used to determine the enantiomer

specificity by GC, can be prepared by hydrogen transfer ruthenium-catalysed reduction,

as described by Ursini et al.7

12.5.2 Procedure 2: Synthesis of (R)-1-Tetralol

tetralin (R)-1-tetralol 1-tetralone38% yield92% ee

OH O



• �-Tetralin (99%, Sigma–Aldrich) (100 mg, 0.756 mmol, 2.0 mL of ethanolic solution at

50 mg mL�1)

• M. isabellina CCT3498 strain (Tropical Culture Collection, Andre Tosello Research

Foundation)4

• PDCB (200 mL)5

• two 500 mL Erlenmeyer flasks

• pH 7.0 potassium phosphate buffer (4.71 g KH2PO4 and 7.85 g Na2HPO4 in 1 L)

(200 mL)

• ethyl acetate (500 mL, Merck)

• hexane (700 mL, Merck)

• sodium chloride, saturated aqueous solution


• silica gel (400–200 mesh, Aldrich) (23 cm height in glass column)

• orbital shaker

• centrifuge

• 500 mL Kitasato

• Buchner funnel

• filter paper


• 250 mL Separating funnel


• equipment for continuous liquid–liquid extraction

• equipment for flash column chromatography using a glass column (2.5 cm� 25 cm)


• GC–MS

• J&W Scientific HP-5 (30 m� 0.25 mm� 0.25 mm) or Supelco Simplicity 1

(30 m� 0.25 mm� 0.25 mm) fused-silica capillary column

• Marcherey 212117/91 Hydrodex-� 3P (25 m� 0.25 mm� 0.25 mm) fused-silica capil-

lary column

• polarimeter

• nuclear magnetic resonance spectrometer.

12.5.2.2 Procedure

1. A culture of M. isabellina CCT3498 was aseptically transferred to conical Erlenmeyer

flasks (500 mL) containing 200 mL of sterile PDCB5 and kept on a rotary shaker (150

rpm) at 30 �C for 3 days to acquire biomass.

2. Microbial biomass was harvested by centrifugation (5000 rpm, 10 min) and 30 g (wet

weight) was transferred to clean 500 mL conical flasks containing 200 mL of 0.2 ionic

strength potassium phosphate buffer solutions at pH 7.0.

Attention: to avoid the lack of efficiency, new cultures of M. isabellina should be

used and a careful control of incubation conditions (temperature, pH and medium) is

necessary.

3. 100 mg of tetralin (2.0 mL of ethanolic solution at 50 mg mL�1) was added.

4. The flask was returned to the shaker (150 rpm) at 30 �C for 4 days.


5. After 4 days, three 5.0 mL portions of the incubation mixture were harvested

and submitted to vigorous shaking and extraction with 5.0 mL of ethyl acetate.

If necessary, a centrifugation procedure (3 min at 3000 rpm) was used to break

the emulsion.

6. The organic fraction was collected and 1 mL of the solution was submitted to the

GC–MS for qualitative and chiral analysis to certify the conversion to 1-tetralol.

7. For qualitative analyses, the GC system was equipped with a J&W Scientific HP-5 or

a Supelco Simplicity 1 fused-silica capillary column. Injector and detector tempera-

tures were set at 220 �C and 240 �C respectively; the oven temperature was pro-

grammed from 60 to 230 �C at 40 �C min�1. Helium was employed as carrier gas

(1 mL min�1). Compound identification was based on a comparison of mass spectra

with those of synthetic racemic and enantiomeric-enriched samples. The retention

times for tetralin, 1-tetralol and 1-tetralone were 5.6 min, 6.5 min and 6.6 min

respectively.

8. For chiral analyses the GC system was equipped with a Marcherey 212117/91

Hydrodex-� 3P fused-silica capillary column. The oven temperature was pro-

grammed from 100 to 210 �C at 10 �C min�1. Injector and detector temperatures

were set at 200 �C and 240 �C respectively. Hydrogen was employed as carrier gas

(1 mL min�1). Under these conditions, the retention times obtained for (S)-1-tetralol

and (R)-1-tetralol were 9.6 min and 9.7 min respectively.

9. After the incubation time, the culture was harvested and filtered. The filtrate was

saturated with a sodium chloride saturated aqueous solution, stirred for 3–4 h at room

temperature and extracted six times with ethyl acetate (20 mL).

10. The organic phase was saved and the sodium chloride saturated aqueous solution

remaining was extracted again by continuous liquid–liquid process at�50 �C for 24 h.

11. The combined organic extracts were dried over anhydrous sodium sulfate and con-

centrated in vacuum after filtration.

12. The crude residue was purified by flash column chromatography on silica gel using

300 mL portions of hexane:ethyl acetate (90:10 and 80:20); 10 mL fractions were

collected, giving 1-tetralone with 9:1 ratio and 1-tetralol (38 mg, 0.287 mmol) with

8:2 ratio.

13. The isolated 1-tetralol was collected and 1 mL of the solution was submitted for GC–

MS analysis, as described above; compound identity was confirmed by nuclear

magnetic resonance spectrometry.7,8

(R)-1-Tetralol was isolated as a colourless oil in 38 % yield. [�]22D ¼�34.0�

(c¼ 2.12, CHCl3), 92 % ee. Lit.:11,12 [�]25D ¼þ34.4� (c¼ 1.01, CHCl3) for S enan-

tiomer. 1H NMR (CDCl3, 300 MHz): � 7.46–7.42 (m, 1H, Ph), 7.24–7.20 (m, 2H,

Ph), 7.14–7.10 (m, 1H, Ph), 4.79 (apparent t, J¼ 4.4 Hz, 1H, CHOH), 2.85–2.65

(m, 2H, CH2), 2.05–1.75 (m, 5H, CH2, CH2, OH). MS (EI): m/z (relative intensity)

148 (Mþ, 18), 147 (25), 131 (18), 129 (43), 128 (20), 127 (13), 121 (8), 120 (80),

119 (67), 115 (28), 105 (47), 104 (15), 92 (20), 91 (100), 90 (15), 89 (15), 79 (10),

78 (30), 77 (34), 66 (10), 65 (47), 64 (30), 63 (41), 62 (10), 60 (10), 57 (11), 55

(10), 53 (14), 52 (16), 51 (69), 50 (28), 43 (23), 41 (31), 40 (12).

Racemic standards of 1-tetralol, to be used for chiral GC analysis, can be prepared by

treatment of tetralone with NaBH4 in MeOH, as described by Aina et al.6


Enantiomerically enriched samples of 1-tetralol, used to determine the enantiomer

specificity by GC, can be prepared by hydrogen transfer ruthenium-catalysed reduction,

as described by Ursini et al.7

12.5.3 Conclusion

The enantioselective benzylic hydroxylation of indan and tetralin can be achieved with

M. isabellina, affording 78 % conversion to 1-indanol (64 % yield, 86 % (1R)- ee) in a 2-

day incubation and 52 % conversion to 1-tetralol (38 % yield, 92 % (1R)- ee) in a 4-day

incubation. The good yields and ee allow their use in future scaling-up processes; however,

to avoid the lack of efficiency, careful control of the temperature, pH and medium is

necessary, since the reactions are strongly dependent on the incubation and reaction

conditions. Tables 12.2 and 12.3 give details of some of the different incubation condi-

tions/results and time-course analysis found in the benzylic hydroxylation of indan and

tetralin mediated by M. isabellina CCT3498.

Table 12.2 Benzylic hydroxylation of indan and tetralin mediated by M. isabellinaCCT3498 a

Entry Parameter Indan to 1-indanol Tetralin to 1-tetralol

1 pH 6.0 7.02 Incubation time (days) 2 43 Relative conversion to alcohol (%)b 78 504 Alcohol yield (%)c 64 385 Ee (% R)d 86 92

a The microorganism was grown in PDCB at 30 �C. The reactions were performed in buffer solutions, also at 30 �C.b Determined by GC on an HP-5 or a Supelco Simplicity 1 fused-silica capillary column. The percentage compositionswere obtained from electronic integration measurements, without taking into account relative response factors.c The yields quoted are those of isolated, purified material.d Determined by chiral GC on Marcherey 212117/91 Hydrodex-� 3P fused-silica capillary column.

Table 12.3 Time-course analysis obtained in the benzylic hydroxylation of indan and tetralin(30 mg) by M. isabellina (3 g fresh weight) a

Time Conversion (%) of substrate and products

Indan 1-Indanol 1-Indanone Tetralin 1-Tetralol 1-Tetralone

Day 1 75 24 1 89 10 1Day 2 18 78 3 65 28 7Day 3 18 64 18 22 52 28Day 4 18 59 23 14 50 36Day 5 13 36 51 0 40 60

a The experiments were carried out in triplicate and analysed by GC on an HP-5 or a Supelco Simplicity 1 fused-silicacapillary column. The percentage compositions were obtained from electronic integration measurements, without takinginto account relative response factors.



1. Van Berkel, W.J.H., Kamerbeek, N.M. and Fraaije, M.W., Flavoprotein monooxygenases, adiverse class of oxidative biocatalysts. J. Biotechnol., 2006, 124, 670.

2. Burton, S.G., Oxidizing enzymes as biocatalysts. Trends Biotechnol., 2003, 21, 543.

3. Limberger, R.P., Ursini, C.V., Moran, P.J.S. and Rodrigues, J.A.R., Enantioselective benzylicmicrobial hydroxylation of indan and tetralin. J. Mol. Catal. B: Enzym. 2007, 46, 37.

4. Fundacao Andre Tosello de Pesquisa e Tecnologia, Rua Latino Celho 1301, 13087-1001Campinas-SP, Brazil, http://www.fat.org.br.

5. PDCB was prepared by suspending cut-up unpeeled potatoes (100 g L�1) and carrots (10 g L�1)in purified water and heated to boiling in a microwave oven for 5–10 min. Dextrose (D-(þ)-glucose) was added (30 g L�1). The medium was sterilized by autoclaving for 20 min at 121 �Cand then decanting off the broth. The broth is clear to slightly opalescent and yellowish in colour.No pH adjustment was made.

6. Aina, G., Nasini, G. and Pava, O.V., Asymmetric bioreduction of racemic 5,6,7,8-tetrahydro-8-methyl-1,3-dimethoxynaphthalen-6-one to the corresponding chiral �-tetralols. J. Mol. Catal.B: Enzym., 2001, 11, 367.

7. Ursini, C.V., Dias, G.H.M. and Rodrigues, J.A.R., Ruthenium-catalyzed reduction of racemictricarbonyl(�6-aryl ketone)chromium complexes using transfer hydrogenation: a simple alter-native to the resolution of planar chiral organometallics. J. Organomet. Chem., 2005, 690, 3176.

8. Boyd, D.R., Sharma, N.D., Boyle, R., Evans, T.A., Malone, J.F., McCombe, K.M., Dalton, H.and Chima, J., Chemical and enzyme-catalysed syntheses of enantiopure epoxide and diolderivatives of chromene, 2,2-dimethylchromene, and 7-methoxy-2,2-dimethylchromene (pre-cocene-1). J. Chem. Soc. Perkin Trans. 1, 1996, 1757.

9. Brand, J.M., Cruden, D.L., Zylstra, G.J. and Gibson, D.T., Stereospecific hydroxylation of indanby Escherichia coli containing the cloned toluene dioxygenase genes from Pseudomonas putidaF1. Appl. Environ. Microbiol., 1992, 58, 3407.

10. Jaouen, G. and Meyer, A., Facile syntheses of optically active 2-substituted indanones, indanols,tetralones, and tetralols via their chromium tricarbonyl complexes. J. Am. Chem. Soc., 1975, 97,4667.

11. Boyd, D.R., McMordie, R.A S., Sharma, N.D., Dalton, H., Williams, P. and Jenkins, R.O.,Stereospecific benzylic hydroxylation of bicyclic alkenes by Pseudomonas putida: isolation of(þ)-R-1-hydroxy-1,2-dihydronaphthalene, an arene hydrate of naphthalene from metabolism of1,2-dihydronaphthalene. J. Chem. Soc. Chem. Commun., 1989, 339.

12. Palmer, M.J., Kenny, J.A., Walsgrove, T., Kawamoto, A.M. and Wills, M., Asymmetric transferhydrogenation of ketones using amino alcohol and monotosylated diamine derivatives of indane.J. Chem. Soc. Perkin Trans. 1, 2002, 416.


12.6 Stereospecific Biotransformation of (R,S)-Linalool by Corynesporacassiicola DSM 62475 into Linalool OxidesMarco Antonio Mirata and Jens Schrader

The biotransformation of (R,S)-linalool by fungi is a useful method for the preparation of

natural linalool oxides.1 The stereospecific conversion of (R,S)-linalool by Corynespora

cassiicola DSM 62475 led to 5R-configured furanoid linalool oxides and 5S-configured

pyranoid linalool oxides, both via 6S-configured epoxylinalool as postulated intermediate

(Figure 12.6). The biotransformation protocol affords an almost total conversion of the

substrate with high enantioselectivities and a molar conversion yield close to 100 %

(Table 12.4). Pure linalool oxides are of interest for lavender notes in perfumery.1

12.6.1 Procedure 1: Preparation of Spores Suspension of C. cassiicola DSM 62475


• Malt extract (30 g)

• glucose (10 g)

• peptone (4 g)

• agar (17 g)

O

HO

OHO

OHO5 2

O

HO5

2

H O

HO

O

HO

H O

O

(3S )-linalool

(3R )-linalool

(3S, 6S )-epoxylinalool

(3R, 6S )-epoxylinalool

furanoid cis-(2S,5R ) linalool oxide

pyranoid cis-(2S, 5S ) linalool oxide

furanoid trans-(2R,5R ) linalool oxide

pyranoid trans-(2R,5S ) linalool oxide

C. cassiicola DSM 62475

C. cassiicola DSM 62475

52

5 2

OO

OO5 2

Figure 12.6 Stereospecific biotransformation of (R,S)-linalool by C. cassiicola via the(6S)-configured epoxylinalools as postulated intermediates



• NaCl (8.5 g)

• Tween 80 (1 g)

• stock culture of C. cassiicola DSM 62475

• distilled water 2 L

• acetic acid (AcOH, >99.8 %)

• one Petri dish

• one Falcon tube 50 mL

• incubator

• Drigalski spatula.

12.6.1.2 Procedure

1. C. cassiicola DSM 62475 was taken from stock culture and incubated for 15 days

at 25 �C on malt extract agar plate (consisting of 30 g malt extract, 3 g peptone

and 17 g agar in 1 L distilled water adjusted to pH 5.6 with AcOH and sterilized

at 121 �C for 20 min).

2. A spore suspension was prepared by resuspending the spores from the agar plate in 15

mL physiological aqueous solution (8.5 g NaCl, 1 g peptone, and 1 g Tween 80 in 1 L

distilled water, sterilized at 121 �C for 20 min) with a Drigalski spatula. The spore

suspension was diluted to a concentration of approximately 2.5� 107 spores/mL and

stored in a 50 mL Falcon tube at 4 �C.

12.6.2 Procedure 2: Fed-batch Biotransformation of (R,S)-Linalool by C. cassiicola

DSM 62475


• Malt extract (30 g)

• glucose (10 g)

• peptone (10 g)


• spore suspension

• (R,S)-linalool stock solution (3 % w/v in ethanol 99 %)

• glucose aqueous solution 1.1 kg L�1

• acetic acid (AcOH, >99.8 %)

• one Erlenmeyer flask 300 mL

• one Erlenmeyer flask 2 L

• one bottle 1 L

• shaking incubator.

12.6.2.2 Procedure

1. Malt extract (30 g), glucose (10 g), peptone (10 g) and yeast extract (30 g) were

dissolved with water and the volume was adjusted to 1.0 L with distilled water; the

pH was adjusted to 6.4 with AcOH. The resulting solution (MYB medium) was

sterilized (121 �C, 20 min) and stored at 4 �C.

12.6 Stereospecific Biotransformation of (R,S)-Linalool 377

2. For the preparation of the preculture, 50 mL of MYB medium was placed in a sterile

300 mL Erlenmeyer flask and inoculated with 500 mL spore supension of C. cassiicola

DSM 62475. The Erlenmeyer flask was incubated and shaken for 24 h at 25 �C and 130

rpm.

3. A sterile 2 L Erlenmeyer flask was filled with 450 mL MYB medium containing 2.5 mL

of (R,S)-linalool stock solution (150 mg L�1 linalool) and inoculated with the resulting

preculture (step 2) of C. cassiicola DSM 62475. The Erlenmeyer flask was incubated

and shaken for 72 h at 25 �C and 130 rpm. A volume of 1.7 mL (R,S)-linalool stock

solution (100 mg L�1 linalool) and 2 mL glucose aqueous solution (5 g L�1) were fed

after 24 and 48 h cultivation.

12.6.3 Conclusion

The procedure is very easy to reproduce and to scale up. Bioconversion products can be

easily isolated by evaporation of the extraction solvent (e.g. tert-butyl methyl ether).

Table 12.4 summarizes the product concentrations, molecular conversion yields and

enantioselectivities obtained during linalool biotransformation with C. cassiicola

DSM 62475.

Table 12.4 Fed-batch biotransformation of (R,S)-linalool by C. cassiicola DSM 62475 usingProcedures 1 and 2. Of 340 mg L�1 (R,S)-linalool added, 96 % was consumed

Products from (R, S)-linalool Molecular yield (%) Concentration (mg/L) Ee (%)

Furanoid trans-(2R,5R)-linalool oxide 43.0 153 >95Furanoid cis-(2S,5R)-linalool oxide 41.7 148 >99Pyranoid trans-(2R,5S)-linalool oxide 4.7 17 >80Pyranoid cis-(2S,5S)-linalool oxide 9.2 33 >97

Reference

1. Schrader, J. and Berger, R.G., Biotechnological production of terpenoid flavor and fragrancecompounds. In Biotechnology, vol. 10, 2nd edn, Rehm, H.-J. and Reed, G. (eds). Wiley-VCH:Weinheim, 2001, pp. 377–383.


12.7 The Biocatalytic Synthesis of 4-Fluorocatechol from FluorobenzeneLouise C. Nolan and Kevin E. O’Connor*

The microbial synthesis of organic compounds is a useful method for the prepara-

tion of valuable compounds such as substituted catechols. Here, we describe two

approaches to the biological formation of 4-fluorocatechol from fluorobenzene.

First, we describe the biotransformation of fluorobenzene to 4-fluorocatechol

using whole cells of Pseudomonas mendocina KR1 expressing toluene-4-monoox-

ygenase (T4MO). Second, we use whole cells expressing T4MO in tandem with the

enzyme tyrosinase sourced commercially from mushrooms to further improve cate-

chol formation.

F F

OH

F

OH

OH

F

O

O

Fluorobenzene 4-Fluorophenol 4-Fluorocatechol 4-Fluoroquinone

T4MO T4MOAscorbic

acid

Tyrosinase

12.7.1 Procedure 1: Growth Medium and Buffers


• Sodium ammonium phosphate tetrahydrate (NaNH4HPO4�4H2O (35.0 g))

• potassium phosphate dibasic trihydrate (K2HPO4�3H2O (75.0 g))

• potassium phosphate monobasic (KH2PO4 (37.0 g))

• magnesium sulfate heptahydrate (MgSO4�7H2O (4.93 g))

• ferrous sulfate heptahydrate (FeSO4�7H2O (2.78 g))

• manganese chloride tetrahydrate (MnCl2�4H2O (1.98 g))

• cobalt(II) sulfate heptahydrate (CoSO4�7H2O (2.81 g))

• calcium chloride dihydrate (CaCl2�2H2O (1.47 g))

• copper(II) chloride dehydrate (CuCl2�2H2O (0.17 g))

• zinc sulfate heptahydrate (ZnSO4�7H2O (0.29 g))

• hydrochloric acid (HCl) (37 %)

• biotin (20 mg)

• folic acid (20 mg)

• pyrodoxine hydrochloride (100 mg)

• riboflavin (50 mg)

• thiamine hydrochloride (50 mg)

• nicotinic acid (50 mg)

• pantothenic acid (50 mg)

• vitamin B12 (1 mg)

• 4-aminobenzoic acid (50 mg)

• DL-6,8-thioctic acid (50 mg)

• K2HPO4.3H2O (11.423 g)


• KH2PO4 (6.805 g)

• glycerol

• deionized water.

12.7.1.2 Procedure

1. Stock solution 1. 50� stock solution of E2 mineral salts medium was prepared as

follows: NaNH4HPO4�4H2O (35.0 g) was dissolved in 100 mL deionized water using

magnetic stirring. K2HPO4�3H2O (75.0 g) and KH2PO4 (37.0 g) were added to the

solution and the volume adjusted to 200 mL with deionized water. The pH was adjusted

to 7.0. This solution was stored unautoclaved on the bench.

2. Stock solution 2. MgSO4�7H2O (4.93 g) was dissolved in 20 mL of deionized water. This

solution was sterilized by autoclaving and stored as a 1 M stock solution on the bench.

3. 100 mL of 1 M HCl was prepared by adding 8.35 mL of 37 % HCl to 91.65 mL of

deionized water.

4. Stock solution 3. 100� stock solution of trace elements was prepared by dissolving

FeSO4�7H2O (2.78 g), MnCl2�4H2O (1.98 g), CoSO4�7H2O (2.81 g), CaCl2�2H2O

(1.47 g), CuCl2�2H2O (0.17 g) and ZnSO4�7H2O (0.29 g) in 1 M HCl. The final volume

was adjusted to 1.0 L. This stock solution was stored at 4 �C.

5. Stock solution 4. 100� stock solution of vitamins was prepared by dissolving biotin

(20 mg), folic acid (20 mg), pyrodoxine hydrochloride (100 mg), riboflavin (50 mg),

thiamine hydrochloride (50 mg), nicotinic acid (50 mg), pantothenic acid (50 mg),

vitamin B12 (1 mg), 4-aminobenzoic acid (50 mg) and thioctic acid (50 mg) in

deionized water. The volume was adjusted to 1.0 L. The solution was filtered, sterilized

and stored as 10 mL aliquots at �20 �C.

6. Stock solution 5. 1 M stock solution of potassium phosphate buffer was prepared by

dissolving K2HPO4�3H2O (11.423 g) and KH2PO4 (6.805 g) in deionized water to a

final volume of 100 mL. The pH was adjusted to 7.0. This 1 M stock solution was diluted

to the desired concentration of 50 mM with deionized water. Buffers were stored at 0–4

�C.

7. Stock solution 6. 60 % glycerol was prepared by mixing 12 mL glycerol with 8 mL

deionized water. This solution was autoclaved and stored on the bench.

12.7.2 Procedure 2: Storage, Cultivation and Harvesting of P. mendocina KR1


• Stock solution 1 (9 mL)

• deionized water (439 mL)

• stock solution 2 (450 mL)



• toluene (800 mL)

• 1� 250 mL centre column Erlenmeyer flask with cotton wool plug

• 2� 2 L centre column Erlenmeyer flask with cotton wool plugs

• New Brunswich Scientific C25 incubator shaker (Classic Series)



• 15–20 1.5 mL sterile polypropylene tubes

• Unicam (UV–vis) Helios � thermo-spectrophotometer

• Du Pont RC5C-plus fixed-angle centrifuge

12.7.2.2 Procedure

1. Overnight starter cultures of P. mendocina KR1 were grown in a 250 mL glass centre

column (fused to the base of the flask) Erlenmeyer flask. Each flask contained stock

solution 1 (1 mL) and deionized water (49 mL). This solution was autoclaved and

cooled to room temperature before adding 50 mL of stock solution 2, 50 mL stock

solution 3 and 50 mL stock solution 4. Toluene (200 mL) was added to the glass centre

column as the sole source of carbon and energy supplied in the vapour phase. The flask

was then inoculated with 100 mL of a freezer stock of P. mendocina KR1. Cultures were

grown on a shaker table incubator at 30 �C for 18 h at 200 rpm.

2. 750 mL of the above culture was added to stock solution 6 (250 mL) in 1.5 ml sterile

tubes. The cultures were mixed gently before storing them at �80 �C. These culture

tubes were used as stocks for future inoculations.

3. Batch cultivation of P. mendocina KR1 was carried out in 2� 2 L centre column

Erlenmeyer flasks. Each flask contained stock solution 1 (8 mL) and deionized water

(390 mL). This solution was autoclaved and cooled before adding 400 mL stock solution

2, 400 mL stock solution 3 and 400 mL stock solution 4. Toluene (600 mL) was added to

the centre column. Overnight starter cultures were used to inoculate (2 % v/v) the

growth medium. Cultures were incubated at 30 �C shaking at 200 rpm.

4. Cells were harvested at an optical density (OD) (540 nm) of 0.7–0.8. During the harvest

process, cells were kept on ice where possible. Cells were centrifuged at 16 200g for

10 min. Cell pellets were washed with 800 mL of ice-cold stock solution 5 and

centrifuged as above. Cell pellets were combined and concentrated by resuspending

them in a final volume of 50 mL of 50 mM stock solution 5. The OD (540 nm) was

adjusted to 5 (1.5 mg cells dry weight (CDW)/mL).

12.7.3 Procedure 3: Biotransformation of Fluorobenzene by P. mendocina KR1


• Fluorobenzene (4.68 mL)

• ascorbic acid (176.12 mg)

• deionized water

• 1 % (w/v) D-glucose (0.5 g)

• 1� 250 mL Erlenmeyer flask with cotton wool plug

• magnetic stirrer plate and magnetic bar

• 1 M HCl

• Eppendorf centrifuge 5810 R

• nylon filters (0.2 mm)

• high performance liquid chromatography (HPLC) vials and caps

• Hewlett Packard HP1100 instrument equipped with an Agilent 1100 series diode array

detector

• C18 Hypersil ODS 5 m HPLC column (125 mm� 3 mm).


12.7.3.2 Procedure

1. 50 mL of the washed cell suspension of P. mendocina KR1 (1.5 mg CDW/mL) was

transferred to a 250 mL Erlenmeyer flask and the contents brought to room temperature.

2. Ascorbic acid (176.12 mg) was dissolved in 1 mL of deionized water (1 M stock

concentration).

3. 0.5 mL of 1 M ascorbic acid (10 mM final concentration) and D-glucose (0.5 g) were

added to the cell suspension and the flask was placed on a magnetic stirrer. The contents

were stirred magnetically at 200 rpm throughout the biotransformation.

4. 4.68 mL fluorobenzene (1 mM final concentration) was added to the flask and the flask

was plugged with cotton wool.

5. The biotransformation was monitored by analysing samples taken periodically by

HPLC. Samples (450 mL) were withdrawn from the biotransformation medium, acid-

ified with 1 M HCL (50 ml) to stop the reaction and stored on ice for 30 min. All samples

were centrifuged at 23 000g for 10 min at 4 �C to remove the cell debris and the

supernatant filtered into HPLC vials using nylon filters (0.2 mm).

6. Biotransformation samples were analysed by HPLC using a C18 Hypersil ODS 5mcolumn (125 mm� 3 mm) and a Hewlett Packard HP1100 instrument equipped with an

Agilent 1100 series diode array detector. The samples were isocratically eluted using an

aqueous phosphoric acid (0.1 % v/v)/methanol mix (70:30 (v/v)) at a flow rate of 0.5

mL min�1.

7. After 120 min, whole cells of P. mendocina KR1 expressing T4MO activity trans-

formed 1 mM fluorobenzene to 0.8 mM 4-fluorocatechol as a single product via

4-fluorophenol.

12.7.4 Procedure 4: Biotransformation of Fluorobenzene by Whole Cells of

P. mendocina KR1 Expressing T4MO in Tandem with a Cell-free

Preparation of Tyrosinase from Mushroom


• Fluorobenzene (23.4 mL)

• 1 % (w/v) D-glucose (0.5 g)

• 1� 250 mL Erlenmeyer flask with cotton wool plug

• magnetic stirrer plate and magnetic bar

• Du Pont RC5C-plus fixed-angle centrifuge

• heated (30 �C) oxygen electrode chamber

• mushroom (commercial) tyrosinase (10 mg)

• ascorbic acid (1 M stock).

12.7.4.2 Procedure

1. In a 250 mL Erlenmeyer flask, 50 mL of the washed cell suspension of P. mendocina

KR1 (1.5 mg CDW/mL) and 0.5 g D-glucose were added and the contents brought to

room temperature.

2. 23.4 mL of fluorobenzene (5 mM final concentration) was added to the flask and the

flask was plugged with cotton wool. The contents were magnetically stirred at 200 rpm

throughout the biotransformation.


3. After 105 min, the biotransformation contents were transferred to a centrifuge

bucket and the cells spun out at 16 200g for 10 min at 4 �C. 15 mL of the

supernatant was transferred to an oxygen electrode chamber and brought to 30 �C.

Then, 15 mL 1 M ascorbic acid (1 mM final concentration) was added to the

supernatant.

4. 10 mg mushroom tyrosinase was dissolved in stock solution 5 (1 mL) and kept on

ice. 45 ml of mushroom tyrosinase (0.03 mg mL�1 final concentration) was added

to the supernatant to start the tyrosinase reaction and 0.03 mg mL�1 was added

every 15 min thereafter. In addition, 1 mM ascorbic acid was added every 5 min

or until a colour change was observed. The reaction was stirred magnetically

throughout.

5. The biotransformation was monitored by analysing samples by HPLC using the same

sample preparation and HPLC analysis methods as described above (Procedure 3,

Section 12.7.3).

6. After 120 min, tyrosinase transformed 1.8 mM 4-fluorophenol (produced by whole cells

of P. mendocina KR1 expressing T4MO) to 1.3 mM 4-fluorocatechol.

12.7.5 Conclusion

The biotransformation of low levels of fluorobenzene (1 mM final concentration) to

4-fluorocatechol by whole cells of P. mendocina KR1 (1.5 mg CDW/mL) is easy to

reproduce. Under these conditions, 4-fluorocatechol is formed as a single product in

the biotransformation after 120 min (Table 12.5). Biotransformations with

P. mendocina KR1 (1.5 mg CDW/mL) and higher concentrations of fluorobenzene

(5 mM final concentration) result in the formation of 4-fluorophenol (1.8 mM) as a

major product. In addition, minor products, namely 2-fluorophenol, 3-fluorophenol,

4-fluorocatechol and 3-fluorocatechol, are also formed. In the presence of ascorbic

acid, tyrosinase has the ability to convert 4-fluorophenol (1.8 mM) to 4-fluorocatechol

(1.3 mM). While this is a reproducible procedure, the 4-fluorocatechol does not

accumulate as a single product (Table 12.5).

Table 12.5 Product formation in biotransformations with whole cells of P. mendocina KR1expressing T4MO alone and in tandem with mushroom tyrosinase a

Biotransformationconditions

2-Fluorophenol

3-Fluorophenol

4-Fluorophenol

3-Fluorocatechol

4-Fluorocatechol

P. mendocina KR1(1.5 mg CDW/mL)1 mm fluorobenzene

ND ND ND ND 0.8 mM

P. mendocina KR1(1.5 mg CDW/mL),5 mM fluorobenzeneand mushroomtyrosinase in tandem

0.09 mM 0.08 mM 0.5 mM 0.03 mM 1.3 mM

a ND: not detected.


References

1. Nolan, L.C. and O’Connor, K.E., Use of Pseudomonas mendocina, or recombinant Escherichiacoli cells expressing toluene-4-monooxygenase, and a cell-free tyrosinase for the synthesis of4-fluorocatechol from fluorobenzene. Biotechnol. Lett., 2007, 29, 1045.

2. Brooks, S.J., Doyle, E.M., Hewage, C., Malthouse, J.P.G. and O’Connor, K.E.,Biotransformation of halophenols using crude cell extracts of Pseudomonas putida F6. Appl.Microbiol. Biotechnol., 2004, 64, 486.


12.8 Synthesis of Enantiopure (S)-Styrene Oxide by Selective Oxidation ofStyrene by Recombinant Escherichia coli JM101 (pSPZ10)Katja Buehler and Andreas Schmid

Selective oxidation of hydrocarbons is one of the most useful biotransformations for

synthetic applications. Chemical counterparts often do not exist or lack the required

regio- and enantioselectivity. In the respective reaction, recombinant Escherichia coli

JM101 (pSPZ10) carrying and expressing the styAB genes encoding for the two-compo-

nent styrene–monooxygenase from Pseudomonas sp. strain VLB120 is applied in a two-

liquid-phase process for the highly enantioselective production of (S)-styrene oxide from

toxic styrene (Figure 12.7).1,2

12.8.1 Procedure 1: Cultivation of the Seed Culture of Recombinant E. coli JM101

(pSPZ10)


• Luria–Bertani (LB) broth3 containing:

– glucose 1 % (w/v)

– kanamycin (50 mg L�1)

– deionized water

– stored culture of E. coli JM101 (pSPZ10)

• one 10 mL test tube with cap

• sterile filters 0.2 mm pore size

• shaker.

12.8.1.2 Procedure

1. The LB medium and the test tube were sterilized by autoclaving (121 �C, 20 min), while

the glucose and the kanamycin were dissolved separately in water and sterilized by

filtration through a 0.2 mM filter. After allowing the LB medium to cool to room

temperature, glucose and kanamycin solution were added in the appropriate amounts.

2. 5 mL of the thus-prepared medium were transferred to a sterile 10 mL test tube. The

solution was inoculated with one colony of E. coli JM101 (pSPZ10) and left for

incubation on a shaker at 250 rpm and 30 �C overnight.

O

E. coli JM101 (pSPZ10)

ee > 99% Yield: > 76 %

Figure 12.7 Enantioselective epoxidation of styrene to (S)-styrene oxide utilizing recombinantE. coli JM101 (pSPZ10) as biocatalyst

12.8 Synthesis of Enantiopure (S)-Styrene Oxide by Selective Oxidation of Styrene 385

12.8.2 Procedure 2: Cultivation of the Preculture of Recombinant E. coli JM101

(pSPZ10)


• M9* mineral salt medium:

– disodium hydrogen phosphate dihydrate (25.5 g)

– potassium dihydrogen phosphate (9.0 g)

– ammonium chloride (1.0 g)

– sodium chloride (0.5 g)

– deionized water (1 L)

– magnesium sulfate (240.5 mg)

– kanamycin (50 mg)

– thiamine (10 mg)

– glucose (5 g)

• US* trace element solution (� 1000):

– hydrochloric acid (1 M)

– manganese chloride tetrahydrate (1.5 g)

– zinc sulfate (1.05 g)

– boric acid (0.3 g)

– sodium molybdate dihydrate (0.25 g)

– copper(II) chloride dihydrate (0.15 g)

– sodium ethylenediaminetetraacetic acid dihydrate (0.84 g)

– calcium chloride dihydrate (4.12 g)

– ferrous sulfate heptahydrate (4.87 g)

• one 1000 mL shake flask with baffles

• sterile filters 0.2 mm pore size

• shaker.

12.8.2.2 Procedure

1. Disodium hydrogen phosphate dihydrate (25.5 g), potassium dihydrogen phosphate

(9.0 g), ammonium chloride (1.0 g) and sodium chloride (0.5 g) were dissolved in water

and the volume adjusted to 900 mL. The solution was sterilized by autoclaving (121 �C,

20 min) and allowed to cool to room temperature. Magnesium sulfate (240.5 mg),

kanamycin (50 mg), thiamine (10 mg) and glucose (5 g) were dissolved in water and

adjusted to 100 mL volume. This mixture was sterilized by filtration through a 0.2 mm

filter and added to the salt solution.

2. For the US* trace element solution (� 1000) all compounds were subsequently

dissolved in 1 L of 1 M hydrochloric acid. The solution was sterilized by filtration

through a 0.2 mm filter. 1 mL of this solution was added to 1 L of M9* medium prior

to usage.

3. 100 mL of the ready-to-use M9* medium was transferred to a sterile 1000 mL shake

flask with baffles. This solution was inoculated with 1 mL of freshly grown seed culture

of E. coli JM101 (pSPZ10) (see Procedure 1, Section 12.8.1) and incubated overnight at

250 rpm and 30 �C.


12.8.3 Procedure 3: Batch and Fed-batch Cultivation of Recombinant E. coli

JM101 (pSPZ10)2

12.8.3.1 Materials and Equipment4

• Dipotassium hydrogen phosphate trihydrate (15.9 g)

• potassium dihydrogen phosphate (4.0 g)

• disodium hydrogen phosphate dodecahydrate (7 g)

• ammonium sulfate (1.2 g)

• ammonium chloride (0.2 g)

• magnesium sulfate heptahydrate (1 g)

• yeast extract (5 g )

• L-leucine (0.6 g)

• L-proline (0.6 g)

• deionized water (1 L)

• kanamycin (50 mg)

• thiamine (10 mg)

• glucose (5 g)

• US*trace element solution (1 mL)

• glucose feed medium:

– glucose (450 g L�1)

– magnesium sulfate heptahydrate (9 g L�1)

– dissolved in deionized water

• bioreactor specifications:

– lab-scale fermenter with a working capacity of 2.6 L made out of stainless steel, glass

and Viton sealing

– baffles and two six-bladed impellers allowing a stirrer speed of up to 3000 rpm

– temperature control accomplished by using a Testoterm type II sensor and connecting

the fermenter to a heating/cooling system

– pH control connected to a rotary peristaltic pump to feed the titrants

– in situ autoclavable amperometric probe (Pt/Ag) equipped with a fluoroethylene

propylene (25 mm) membrane for dissolved oxygen tension (DOT) control

– sterile filters (0.2 mm) for the air supply

– thermostatted bubble column (i.d. 70 mm; h¼ 350 mm)

– foam probe

– computer-controlled peristaltic pump and a microcomputer-connected balance to

control the feed

– LabTech Notebook software to control process parameters (pH, oxygen tension, air

flow rate, glucose feed)

– polypropylene glycol (20 % v/v), PP-G200

– ammonia (25 %)

– phosphoric acid (25 %).

12.8.3.2 Procedure

1. Dipotassium hydrogen phosphate trihydrate (15.9 g), potassium dihydrogen phos-

phate (4.0g), disodium hydrogenphosphate dodecahydrate (7 g), ammonium sulfate


(1.2 g), ammonium chloride (0.2 g), yeast extract (5 g), L-leucine (0.6 g) and

L-proline (0.6 g) were dissolved in water and the volume adjusted to 900 mL.

2. The fermenter was assembled and filled with 900 mL of the medium prepared above

(see step 1). This setup was then sterilized by autoclaving (121 �C; 20 min) and allowed

to cool to room temperature. Then, the fermenter was properly connected to air supply,

pH titrants (ammonia solution and phosphoric acid) and anti-foam PP-G200, which was

sterilized prior to usage.

3. Magnesium sulfate (1 g), kanamycin (50 mg), thiamine (10 mg) and glucose (5 g) were

dissolved in water and adjusted to 100 mL volume. This mixture was sterilized by

filtration through a 0.2 mm filter and was added together with 1 mL of US* trace

element solution (see Procedure 2, Section 12.8.2) to the fermenter.

4. Batch growth was started by inoculating the fermenter with 100 mL of a freshly grown

preculture of E. coli JM101 (pSPZ10) (see Procedure 2, Section 12.8.2). The pH was

kept at 7.1, the aeration rate was set to 2 L min�1 and the stirrer speed and temperature

were 1300 rpm and 30 �C respectively.

5. The glucose feed was started as soon as the DOT increased significantly, indicating that

the carbon source in the culture medium was consumed. The stirrer rate was increased

to 2400 rpm. All other parameters were kept constant. Feed rate was maintained at 4.5 g

glucose/Laq/h throughout the fed-batch.

12.8.4 Procedure 4: Biotransformation of Styrene into (S)-Styrene Oxide by

recombinant E. coli JM101 (pSPZ10)


• Bis(2-ethylhexyl)phthalate (BEHP, 1 L)

• n-octane (99 %)

• styrene (99 %).

12.8.4.2 Procedure

1. The BEHP was supplemented with 1 % (v/v) octane for induction of the styAB genes

and 4 % (v/v) styrene as substrate.

2. 1 h after the fed-batch was started (Procedure 3, Section 12.8.3) the organic phase was

added to the bioreactor. The biotransformation was left running for 12 h, maintaining

constant conditions as described in Procedure 3 (Section 12.8.3).

3. The organic phase was separated from the aqueous phase containing biomass by centri-

fugation. Epoxide products were recovered from the organic phase by vacuum distillation.

4. Ee was determined by gas chromatography (GC) on a Supelco Beta-DEX 120 column

(fused-silica capillary column, 30 m, 0.25 mm inner diameter, 0.25 mm film thickness;

Supelco, Buchs, Switzerland) with split injection (20:1) and an isothermal oven tem-

perature profile at 90 �C for separation of styrene oxide enantiomers.

12.8.5 Conclusion

During the overall biotransformation, the product formation rate reached a maximum of

61 U g�1 cells dry weight (CDW) and decreased to 27 U g�1 CDW towards the end of the

process. This resulted in a final product concentration of 306 mM (S)-styrene oxide in the


organic phase. By applying the two-liquid phase concept, inhibition by substrate and

product toxicity could be circumvented.

Table 12.6 gives an overview of the different substrates, which are epoxidized with high

enantiomeric excess by this biocatalyst.

Table 12.6 Substrates and products with the corresponding yields and ee–values forthe biocatalyst E. coli JM101 (pSPZ19) 5

Substratea

1

2

3

4

Cl

5

6

7

Productb

O

1a

O

2a

O

3a

O

4a

Cl O

5a

O

6a

O

7a

Yield (%)

76.3

46.5

74.8

87.2

87.3

53.0

47.9

Ee (%)

99.5

99.9

96.7

99.8

99.4

98.5

98.0

a(1) styrene, (2) 4-methylstyrene, (3) �-methylstyrene, (4) trans-�-methylstyrene, (5) 3-chlorostyrene, (6) 1,2-dihydro-naphthalene, (7) indene served as substrates.b(1a) (S)-styrene oxide, (2a) 4-(S)-methyl-styrene oxide, (3a) (S)-�-methylstyrene oxide, (4a) (S)-trans-�-methylstyrene oxide,(5a) (S)-3-chlorostyrene oxide, (6a) (S)-1,2-dihydronaphthalene oxide, (7a) (1S,2R)-indene oxide were the correspondingepoxides synthesized biocatalytically.


References

1. Panke, S., Wubbolts, M.G., Schmid, A. and Witholt, B., Production of enantiopure styrene oxideby recombinant Escherichia coli synthesizing a two-component styrene monooxygenase.Biotechnol. Bioeng., 2000, 69, 91–100.

2. Park, J.B., Buehler, B., Habicher, T., Hauer, B., Panke, S., Witholt, B. and Schmid, A., Theefficiency of recombinant Escherichia coli as biocatalyst for stereospecific epoxidation.Biotechnol. Bioeng., 2006, 95, 501–512.

3. Sambrook, J. and Russell, D.W., Molecular Cloning. A Laboratory Manual, 3rd edn, Nolan, C.(ed.). Cold Spring Harbor Laboratory Press: New York, 2001.

4. Wubbolts, M.G., Favre-Bulle, O. and Witholt, B., Biosynthesis of synthons in two-liquid-phasemedia. Biotechnol. Bioeng., 1996, 52, 301–308.

5. Schmid, A., Hofstetter, K., Feiten, H.J., Hollmann, F., and Witholt, B., Integrated biocatalyticsynthesis on gram scale: the highly enantioselective preparation of chiral oxiranes with styrenemonooxygenase. Adv. Synth. Catal., 2001, 343, 732–737.


12.9 Biotransformation of a-Bromo and a, a0-Dibromo Alkanone intoa-Hydroxyketone and a-Diketone by Spirulina platensisTakamitsu Utsukihara and C. Akira Horiuchi

�-Hydroxy ketones are important as intermediates in organic synthesis.1,2 In a previous

paper we found that a novel reaction of �-bromo ketone under microwave irradiation gives

the corresponding �-hydroxy ketone in good yields.3 Biotransformation of �-bromo and

�,�0-dibromo alkanones was investigated with alga of Spirulina platensis.

Biotransformation of �-bromo ketone with S. platensis gave the corresponding �-hydroxy

ketone in good yields (80–95 %). It was found that �,�0-dibromo ketone is biocatalytically

transformed into the �-diketone and then is reduced into the �-hydroxy ketone. In the case

of 2,6-dibromo menthone, diosphenol (58 %), 1-hydroxy-3-methyl-6-isopropylcyclohex-

ane-1,2-dione (15 %) and 2-hydroxy menthone (4 %) were obtained. This reaction affords a

new, eco-friendly and convenient method for the synthesis of �-hydroxy ketones.

O O

n n

Br OH

1 : n = 02 : n = 13 : n = 24 : n = 3

1– 4 1a – 4a

Spirulina platensis

Spirulina platensis

Spirulina platensis

Spirulina platensis

OBr

5 : R = H

R

6 : R = Me

OOH

R

7 : R = F8 : R = Cl9 : R = Br

O O

n n

Br OH

2' : n = 13' : n = 2

2' – 3' 2a – 3a

BrO

n

O

2b – 3b

+

Br

BrO

OH

O

OH

OHO

OH

O

+ +

10 10a 10b 10c

Figure 12.8 Biotransformation of �-bromo- and �,�0-dibromo alkanone by S. platensis?

12.9 Biotransformation of a-Bromo and a,a0-Dibromo Alkanone 391

12.9.1 Procedure 1: Cultivation of S. platensis

12.9.1.1 Material and Equipment

• SOT medium:1

– NaHCO3 (16.8 g)

– K2HPO4 (0.5 g)

– NaNO3 (2.5 g)

– K2SO4 (1 g)

– NaCl (1 g)

– MgSO4�7H2O (0.2 g)

– CaCl2�2H2O (0.04 g)

– FeSO4�7H2O (0.01 g)

– Na2EDTA (0.08 g)

– A5 solution (1 mL)

– culture of S. platensis

– distilled water 1000 mL

• A5 solution:

– H3BO3 (286 mg)

– MnSO4�7H2O (250 mg)

– ZnSO4�7H2O (22.2 mg)

– CuSO4�5H2O (7.9 mg)

– Na2MoO4�2H2O (2.1 mg)

– distilled water 100 mL

• filter paper

• one 200 mL Erlenmeyer flask

• fluorescent lamp

• air pump.

12.9.1.2 Procedure

1. SOT medium was prepared by mixing NaHCO3 (16.8 g), K2HPO4 (0.5 g), NaNO3 (2.5 g),

K2SO4 (1 g), NaCl (1 g), MgSO4�7H2O (0.2 g), CaCl2�2H2O (0.04 g), FeSO4�7H2O (0.01

g), Na2EDTA (0.08 g) and A5 solution (1 mL) in distilled H2O (1000 mL).

2. A5 solution was H3BO3 (286 mg), MnSO4�7H2O (250 mg), ZnSO4�7H2O (22.2 mg),

CuSO4�5H2O (7.9 mg) and Na2MoO4�2H2O (2.1 mg) dissolved in distilled H2O (100 mL).

S. platensis was grown in SOT medium (pH 10–11) under continuous illumination

provided by fluorescent lamps (2000 lx) with air bubbling at 25 �C for 2 weeks.

3. The mixture was filtered to obtain the alga of S. platensis (yielded about 1 g L�1 dry

weight).

12.9.2 Procedure 2: Biotransformation of 2-Bromoacetophenone

OBr

OOH

Spirulina platensis



• S. platensis in SOT medium (100 mL)

• 2-bromoacetophenone (100 mg, 0.50 mmol)

• ethyl acetate

• ether

• hexane


• filter paper


• silica gel (Kieselgel 60 40–63 mm), 15 g


• shaker


12.9.2.2 Procedure

1. 2-Bromoacetophenone (100 mg, 0.50 mmol) was added to a suspended culture of

S. platensis (adjusted pH 7.0, 1 g L�1 as dry weight) in SOT medium (100 mL). The

mixture was treated with a shaker (120 rpm) for 3 days at 25 �C in the light (2000 lx).

2. At the end of the reaction, S. platensis was filtered off and the resulting mixture was

extracted with EtOAc/Et2O (1:1). The organic layer was collected, dried over anhy-

drous sodium sulfate and concentrated using a rotary evaporator. The resulting oil was

chromatographed on silica gel. Elution with hexane/ether (3:1) gave 2-hydroxyaceto-

phenone (26 mg, 0.1 mmol). All the products were analysed by infrared (IR), 1H NMR

and gas chromatography–mass spectrometry (GC–MS) analyses.

2-Hydroxyacetophenone. M.p. 85–86 �C; 1H NMR (400 MHz, CDCl3): � 3.54

(brs, 1H), 4.88 (s, 2H), 7.49 (t, 2H, J¼ 7.7 Hz), 7.61 (t, 1H, J¼ 7.4 Hz), 7.92 (d, 2H,

J¼ 7.6 Hz); 13C NMR (CDCl3): � 65.4, 127.6, 128.9, 133.3, 134.2, 198.4. IR (KBr):

3428, 1687 cm�1. MS (electron impact (EI)): m/z 136 (Mþ), 105, 77, 51.

12.9.3 Procedure 3: Biotransformation of 2,6-Dibromo Cyclohexanone

O O

Br OHSpirulina platensisBrO

O

+



• 2,6-dibromo cyclohexanone (100 mg, 0.39 mmol)

• ethyl acetate

• ether

• hexane



• filter paper




• shaker


12.9.3.2 Procedure

1. 2,6-Dibromo cyclohexanone (100 mg, 0.39 mmol) was added to suspended culture

of S. platensis (adjusted pH 7.0, 1 g L�1 as dry weight) in SOT medium

(100 mL). The mixture was treated with a shaker (120 rpm) for 3 days at

25 �C in the light (2000 lx).



drous sodium sulfate and concentrated using a rotary evaporator.

3. The resulting oil was chromatographed on silica gel. Elution with hexane/ether

(3:1) gave 2-hydroxycyclohexanone (24 mg, 0.21 mmol) and 1,2-cyclohexane-

dione (0.8 mg, 0.007 mmol). All the products were analysed by IR, 1H NMR and

GC–MS analyses.

2-Hydroxycyclohexanone. 1H NMR (400 MHz, CDCl3): � 1.50–2.15 (m, 6H), 2.30–

2.50 (m, 2H), 3.66 (brs, 1H), 4.15 (ddd, 1H, J¼ 1.6, 4.6, 8.8 Hz); 13C NMR (CDCl3): �23.4, 27.5, 36.7, 39.5, 75.3, 211.4. IR (neat): 3473, 1714 cm�1. MS (EI): m/z 114 (Mþ), 96,

85, 70, 57, 44.

12.9.4 Procedure 4: Biotransformation of 2,6-Dibromo Menthone

Br

BrO

Spirulina platensisOH

O

OH

OHO

OH

O

+ +



• 2,6-dibromo menthone (100 mg, 0.32 mmol)

• ethyl acetate

• ether

• hexane


• filter paper





• shaker


12.9.4.2 Procedure

1. 2,6-Dibromo menthone (100 mg, 0.32 mmol) was added to suspended culture of

S. platensis (adjusted pH 7.0, 1 g L�1 as dry weight) in SOT medium (100 mL). The

mixture was treated with a shaker (120 rpm) for 3 days at 25 �C in light (2000 lx).



drous sodium sulfate and concentrated using a rotary evaporator.

3. The resulting oil was chromatographed on silica gel. Elution with hexane/ether (3:1)

gave diosphenol (15 mg, 0.09 mmol), 6-hydroxy-3-methyl-6-isopropylcyclohexane-

1,2-dione (3.7 mg, 0.02 mmol) and 2-hydroxy menthone (1.1 mg, 0.006 mmol). All the

products were analysed by IR, 1H NMR and GC–MS.

Diosphenol. 1H NMR (400 MHz, CDCl3): � 1.00 (d, 6H), 1.12 (d, 3H), 2.36 (m, 1H), 6.34

(s, 1H); 13C NMR (CDCl3): � 15.3, 19.5, 19.8, 22.3, 27.9, 30.9, 39.9, 138.1, 141.9, 197.4. IR

(neat): 3425, 1670, 1620, 1160 cm�1; MS (EI): 168 (Mþ), 153, 139, 126, 125, 108.

12.9.5 Conclusion

This is the first time that the biotransformation of �-bromo and �,�0-dibromo ketone using

S. platensis has been successfully accomplished. Although enantioselective �-hydroxy

ketones were not obtained, it was found that the hydroxylative biotransformation of

�-bromo and �,�0-dibromo alkanones using S. platensis affords a new synthetic method,

which is more convenient, cleaner, and of lower energy than the chemical method used

heretofore (see Tables 12.7 and 12.8).2–4 Biotransformation for �-hydroxy ketone from

�-bromo ketone is no doubt attributable to the special properties of S. platensis system.

Table 12.7 Biotransformation of �-bromo compounds by S. platensis

Entry Substrate Day Product (yield, %)a

1 1 1 1a (92)2 2 1 2a (89)3 3 1 3a (95)4 4 5 4a (88)5 5 3 5a (55)6 6 3 6a (35)7 7 3 7a (80)8 8 3 8a (11)9 9 3 9a (6)

a Yield was determined by GC–MS peak area.


Table 12.8 Biotransformation of �,�0-dibromo cycloalkanones byS. platensis

Entry Substrate Day Product (yield, %)a

1 20 3 2a (92) 2b (3)2 30 5 3a (42) 3b (1)3 10 3 10a (4) 10b (58) 10c (15)

a Yield was determined by GC–MS peak area.

References

1. SOT: Spirulina–Ogawa–Terui. Ogawa, T. and Terui, G., Studies on the growth of Spindinaplatensis I. On the pure culture of Spindina platensis. J. Ferment. Technol., 1970, 48, 361.

2. Horiuchi, C.A., Takeda, A., Chai, W., Ohwada, K., Ji, S.-J. and Takahashi, T.T., A novel synthesisof �-hydroxy- and �,�0-dihydroxyketone from �-iodo and �,�0-diiodo ketone using photoirra-diation. Tetrahedron Lett., 2003, 44, 9307.

3. Chai, W., Takeda, A., Hara, M., Ji, S.-J. and Horiuchi, C.A., Photo-irradiation of �-halo carbonylcompounds: a novel synthesis of �-hydroxy- and �,�0-dihydroxyketones. Tetrahedron, 2005, 61,2453.

4. Utsukihara, T., Nakamura, H., Watanabe, M. and Horiuchi, C.A., Microwave-assisted synthesisof �-hydroxy ketone and �-diketone and pyrazine derivatives from �-halo and �,�0-dibromoketone. Tetrahedron Lett., 2006, 47, 9359.


Index

Note: Page references to figures are given in italic type; reference to tables are given in

bold type.

Abacavir 40–1

Acetylation 8

N-Acetyl-D-mannosamine (NAM) 33

N-Acetyl-D-neuraminic acid

(NANA) 33

Acinetobacter calcoaceticus 332–5

Acremonium chrysogenum

Acylation, 25–6, 36, 96, 367–8,

Agrobacterium radiobacter 28

Alcalase 165–9

Alcohol dehydrogenases (ADH), see

Ketoreductases 4–5, 48–52,

284–6

Alcohol reductases 8

Alcohols 288–90

esterification 36, 137–9

reduction 273–5

Aldehydes 271Aldolases 52–4

Aldonic acids 323–4

Amano PS30 43

Amberlite XAD-1180 49

American Type Culture Collection 87

Amines

dynamic kinetic resolution 148–52

free radical-mediated

racemization 153–4

Amino acids 96–7, 314–17

7-aminocephalosporic acid

(7-ACA) 19–22, 65

Aminocyclitols 206–10

7-aminodesacetoxycephalosporanic acid

(7-ADCA) 19, 22

Aminoshikimic acid 84

Aminotransferases 306–8

Amygdalin 242–3

Androgen receptor antagonists 37

Antibiotics 19–23

Aprepitant 51–2, 52

Arabidopsis thaliana 17

Arabinonucleosides 31

Archaea 90, 92, 101

Aspartate aminotransferase

(AspAT) 306–8

Asymmetrization 35

Atorvastatin 28–9, 49–50

Azalactones, ethanolysis 162

Azides 232–4

1-azido disaccharides

232–5

Bacillus licheniformis 165

Bacillus sphaericus 314

Bacillus subtilis 190–7, 299

Bacillus subtilis protease 55–7

Bacteria 92, 112



Baeyer-Villiger monooxygenase

(BVMO) 299, 337–9

Baeyer-Villiger reactions 300, 301–4

Baker’s yeast 48

Basic Local Alignment Search Tool

(BLAST) 89

Belgian Coordinated Collections of

Microorganisms 87Bioinformatics 88–90

Biotechnology 83–5

Biotransformation

definition 3

BioWave reactor 362–5

Biphasic biocatalysis 59–61

Bisfuran alcohol 36

Brecanavir 36, 36–7

BRENDA 88

Buchner, Eduard 84

Candida antarctica lipase B (CALB) 24,

92, 133, 148, 170–2, 208–9

Candida rugosa lipase 110–11, 129–31

Carbamoylation 8

Carboxylic acid reductase 295–7

Cassette mutagenesis 107

CASTing 110

Centralbureau voor Schimmelcultures 87

Cephalexin 23Cephalosporins 19–22

Chiral drug candidates 4–5

ChiroCLEC-PC 37

Chloroperoxidase (CPO) 327–9, 330–1

Cloning 91, 98–103

Clonostachys compactiuscula 25

Clopidogrel 43–4, 59–60, 291

Codons 96

Cofactors 86

see also Nicotinamide adenine

dinucleotide (NADH)

Combinatorial active site saturation test

(CAST) 110

Complementary DNA (cDNA) 101

Corrin 69

Corynespora casiicola 376–8

Cosmids 99

Covalent enzyme attachment 62–3

Crispine A 319–21

Crosslinked enzyme aggregate

(CLEA) 63, 266–7

Crosslinked enzyme crystals

(CLEC) 63

Culture collections 87–8, 87, 93–4

Cultures, see Microbial cultures

Cyanation 29

Cyanohydrin formation 255–8, 259–60,

266–7, 270–2

Cyclohexanone monooxygenase

(CHMO) 332–5

Cyclopentadecanone monooxygenase

(CPDMO) 344–9

CYP, see Cytochrome P450

Cytidine deaminase 39–40, 39

Cytochrome P450 9–11

microbial 12–13

Cytosine 96

Dealkylation 8

Deamination 8

Dean-Stark distillation 177

Dehydrated enzymes 56–7

Deoxyribonucleotide triphosphates

(dNTP) 103

2-deoxyribose-5-phosphate aldolase

(DERA) 52–3

Deracemization 320–1

Desymmetrization 41, 45–8, 125–8,

186–9, 341–3

Deutsche Sammlung von Mikroorganismen

und Zellkulturen 87

Diasteroselectivity 30–4

Directed evolution 105–6, 106

DKR, see Dynamic kinetic resolution

DNA 96–8

complementary (cDNA) 101

databases 89–90

noncoding 97–8

transcription 94–5

ligases 96, 100

polymerases 94–5, 103

sampling 90

shuffling 107–8

templates 98–103

398 Index

Drug metabolites 6–18

Dynamic kinetic resolution 42–4, 137–9,

141–164, 276–7

E-factor 64–5, 65–6

Enterobacter aerogenes 32

Entrapment (enzyme

immobilization) 63–4

Environmental health and safety

(EHS) 65

Environmental sampling 90–2

Enzyme activity databases 88–9

Enzyme induction 93

Epoxidation 8

Epoxide hydrolase 190–7

Error-prone PCR 106–7

Escherichia coli 27, 28

BL21(DE3) 291–2, 344–5

BVMO expression 337–9

carboxylic acid reductase 295–6

cytidine deaminase 39

enzyme induction 93

as expression host 111–12

JM101 385–9

Esterification 24–5, 25, 58, 160–4, 171–4

Eukaryotes 101

Eupergit C 63

ExPASy 90

Extremophiles 92, 93

Fagomine 212–17

Federal Drug Administration (FDA) 2

Fluvastatin 12, 29, 359–65

Fondaparinux 17

Free radicals 153–4

D-fructose-6-phosphate aldolase

(FSA) 212–14

Fructose-1,6-bisphosphate aldolase

(RAMA) 206–8

Fruit seed meal 236–9, 237, 269–71

Gene cloning 94–5, 102

testing 101–3

Gene identification, PCR 103–5

Gene synthesis 110

Generic drugs 1

Genetic code 96–8

Genetically modified microorganisms

(GMMs) 5–6

Geotrichum candidum 48–9

Glucose isomerase 223–5

Glucose oxidase 323–5

Glucuronidation 246–9

Glycorandomization 18

Glycosidases 227

Glycosyl azides 232–4

Glycosylation 8, 16–18, 232–4

Glycosynthases 18, 227–9

Gordonae terrae 182–5

Green chemistry 63–6, 66

Halo-hydroxylation 327–9

Halohydrin dehydrogenase 199–200

Housekeeping genes 93

Humicola sp. lipase 125–7

Hydrolases 4, 35–6, 190–7, 341–2

see also Hydrolysis

Hydrolysis 8, 23, 24, 48, 117–20, 135–6,

186–9, 339, 356–7, 359–65, 391–5

Hydroxylation 8, 9–10, 12, 206–8,

355–7, 359–65, 367–71

Hydroxynitrile lyase (HNL) 52–3, 255–7,

259–64

(S)-ibuprofen 157–61

Imidacloprid 355–8

Immobilization 61, 158

covalent attachment 62–3

noncovalent attachment 61–2

entrapment 63–4

hydroxynitrile lyase 266–7

T. versicolor laccase 243–4

Indels 109

Ionic liquids 39, 56

Irbesartan 9, 10

Isopropyl-�-D-thiogalactopyranoside

(IPTG) 93

Kazlaukas rules 46–7

Ketones 259–60, 278–82, 284–6

cyanohydrin synthesis 271

desymmetrization 125–8

Index 399

Ketones (Continued)

reduction 288–90

see also Alcohol dehydrogenases (ADH)

Ketoreductases 276–7, 278–82, 288–90,

289, 290

see also Alcohol dehydrogenases (ADH)

Kinetic resolution 34–44, 117–20, 121–4,

129–31, 337–8

Laccases 15–16, 86–7

glycoside oxidation 240–4

�-lactams 18–19

see also Cephalosporins

Lactones 344–9

Lamivudine 39–40

LCA, see Life cycle analysis

Leloir glycosyltransferases 17

Leptoxyphium fumago 327

Life cycle analysis (LCA) 65

Lipases 129–31, 134–6, 158–60, 173–80

Candida antarctica B 133, 148,

170–2, 208–9

immobilized 62

Pseudomonas fluorescens 41, 125

Lipolase 36

Liver cell microsomal fractions 11–12

horse 251–3

pig 245–9

Lobucavir 24–5, 25

Lotrafiban 38–9

Lovastatin 25, 47

Mandelic acid derivatives 43–4

Membrane reactors 64

Meso-trick 35

Metagenomics 90–2, 91

Microbacterium campoquemadoensis 51

Microbial cultures 9, 111–12

collections 87–8, 87, 93–4

growth conditions 92–4

history 83–5

hosts 112–13

see also Culture collections

Molecular biology 92–4

central dogma 94–5

enzyme tools 95–6

Molecular cloning 98

Monoamine oxidase 319–21

Monophasic biocatalysis 55–9

Monoterpenes 327–9

Montelukast 51, 52

Mortierella species 369–71

Motierella rammaniana 360–5

Mutagenesis 105–10

cassette mutagenesis 107

combinatorial methods 108–9

DNA shuffling 107–8

error-prone PCR 106–7

indels 109

neutral drift 109

rational enzyme design

109–10

Mutator strains 107

Mycophenolic acid 14, 14,

251–2

NADH 49, 86, 273–5

NAM 33

NANA aldolase 33

Napthalene 351–4

National Centre for Biotechnology

Information (NCBI) 90

National Collection of Industrial

Bacteria 87

Nelarabine 31–3, 31

Neutral drift 109

Nicotinamide adenine dinucleotide

(NADH) 49, 86, 273–5

Nitriles 186–9

Noncoding DNA 97–8

Novozym 435 36, 37, 38, 45, 137–9

Nucleotide phosphorylases (NP) 30–2

Nucleotides 96

Odanacatib 42–3, 43

Olefins 355, 357

Oligosaccaride synthesis 227–9

Organic solvents 54–5

catalyst formulation 56–7

monophasic systems 55–6

solvent engineering 57–9

Origin of replication (ORI) 99

400 Index

Oxazolidines 173–4

Oxidation reactions 8, 11, 15–16,

299–304, 310–21, 323–6, 327–31,

333–4, 344–9, 351–4, 376–8, 385–9

P450, see Cytochrome P450

Palladium 148–52

Pasteur, Louis 83–4

PCR 103–5

Penicillin acylases 19

Penicillin G 19, 83–4, 84

pH memory effect 57

Phase I metabolic reactions 7, 8

Phase II metabolic reactions 8, 13–18

Phenylacetone monooxygenase

(PAMO-P3) 299–303

Phenylalanine dehydrogenase

(PheDH) 314–17

Photochemistry 299–304

Pig liver esterase (PLE) 93

�-piperidine-2-carboxylate reductase

(Pip2C) 310–12

Plantomycetes 117

Plasmids 98, 99

Polyesters 174–80, 179

Polymermatrices (as catalyst

support) 63–4

Polymerase chain reaction (PCR)

103–5

error-prone 106–7

Posaconazole 45

Pregabalin 36

Prodrugs 23–4

Product lifetimes 1

ProSAR (protein sequence activity

relationship) 6, 28–9

Proteases 121–4, 165

see also Bacillus subtilis protease

Prunus dulcis 236–9

Prunus mume 269–72

Pseudomonas 21

Pseudomonas fluorescens lipase

41, 125

Pseudomonas mendocina 379–80

Pseudomonas putida 13, 310–11

PubMed 86

Racemization, see Dynamic kinetic

resolution

RAMA 206–8

Rational enzyme design

109–10

rDNA 84–5, 98

see also Cloning; DNA

Reduction reactions 8carboxylic acids 295–7

ketones 48–52, 284–6,

288–90

photochemical 303–4

Regioselectivity 18–29

Retro-claisenase 341–2

Reverse transcription 95

Rhamnulose-1-phosphate aldolase

(rhAD) 203–5

Rhodococcus erythropolis NCIM

11540, 93, 186–8

Rhodococcus ruber 118–20

Ribavarin 24, 31

Riboflavin 84

RNA 96

Rosuvastatin 29

Roxifiban 43

Ruthenium 137–9

Saccharomyces cerevisiae, H402 x

pTKL1 218–20

Sepabeads EC-EP 63

Sertraline 49

Shotgun libraries 100

Shuttle plasmids 98

Simvastatin 25–6

Solvents 39

biphasic systems 59–61

monophasic systems 55–7

organic 54–5

stereoselectivity and 59

see also Ionic liquids; Organic

solvents; Supercritical fluids

Spirulina platensis 391–5

Start codon 96

Statins 25, 28–9

Stavudine 27

Stenotrophomonas maltophilia 355

Index 401

Stop codon 96

Streptomyces griseoplanus 367–8

Streptomyces griseus 351–4

Streptomyces lividans 93

Streptomyces species 9, 21

Subtilisin Carlsberg 55–6, 165

Suicide vectors 99

Sulfatases 117

Sulfation 8

Supercritical fluids 56

T4MO 379–83

Taq polymerase 106–7

Terpenes 327–9

Thermoanaerobacter ethanolicus 284

Thermobifida fusca 299

Thermomyces lanuginosus 36

Thiol conjugation 8

Tissue preparations 11

see also Liver cell microsomal

fractions

TMPase 27

Trametes versicolor laccase

243–4

Transfection 98

Transformation 102

Tyrosinase 382–3

Urethane polyesters 174–6

Uridine diphosphate glucuronide

transferase 14, 251–3

Uridine phosphorylase (URDP) 31

Valaciclovir 24

Vanillin 295–7

Vector promoters 99

Vectors 98, 99

Viruses 98

Vitamin C 84

W110A secondary alcohol

dehydrogenase 284–6

World Federation for Culture

Collections 87

Yeasts 112

Zanamavir 33

Zeolite beta 133–36

Zidovudine 27

402 Index

Practical Methods for Biocatalysis and Biotransformations

Documents

Transcript of Practical Methods for Biocatalysis and Biotransformations