Practical Methods for Biocatalysis and Biotransformations
-
Upload
peter-sutton -
Category
Documents
-
view
276 -
download
16
Transcript of 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
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
Avanzada de Cataluna, Consejo Superior de Investigaciones Cientificas, Jordi Girona
18-26, 08034 Barcelona, Spain
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
Ridge, Singapore 119260, Republic of Singapore
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
Ridge, Singapore 119260, Republic of Singapore
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
Avanzada de Cataluna, Consejo Superior de Investigaciones Cientificas, Jordi Girona
18-26, 08034 Barcelona, Spain
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
(POSTECH), San-31, Hyojadong, Pohang 790-784, Korea
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
Bioreactor Engineering, East China University of Science and Technology, Shanghai
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,
Gifu 501-1193, Japan
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,
13084-971, Campinas-SP, Brazil
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,
Gifu 501-1193, Japan
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
Science and Technology, Okayama University, Tsushima, Okayama 700-8530, Japan
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
(POSTECH), San-31, Hyojadong, Pohang 790-784, Korea
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,
45470 Mulheim/Ruhr, Germany
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
Bianco 9, 20131 Milano, Italy
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,
13084-971, Campinas-SP, Brazil
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,
45470 Mulheim/Ruhr, Germany
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
Bianco 9, 20131 Milano, Italy
Takashi Sakai, Division of Chemistry and Biochemistry, Graduate School of Natural
Science and Technology, Okayama University, Tsushima, Okayama 700-8530, Japan
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,
45470 Mulheim/Ruhr, Germany
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
Bioreactor Engineering, East China University of Science and Technology, Shanghai
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,
Montreal, Quebec H4P 2R2, 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
Bioreactor Engineering, East China University of Science and Technology, Shanghai
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,
Montreal, Quebec H4P 2R2, 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
Bioreactor Engineering, East China University of Science and Technology, Shanghai
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
Ridge, Singapore 119260, Republic of Singapore
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.)
4 Biotransformations in Small-molecule Pharmaceutical Development
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
6 Biotransformations in Small-molecule Pharmaceutical Development
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
8 Biotransformations in Small-molecule Pharmaceutical Development
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
1.3 Application of Biocatalysis in the Pharmaceutical Industry 9
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
10 Biotransformations in Small-molecule Pharmaceutical Development
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
1.3 Application of Biocatalysis in the Pharmaceutical Industry 11
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
12 Biotransformations in Small-molecule Pharmaceutical Development
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
1.3 Application of Biocatalysis in the Pharmaceutical Industry 13
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.)
14 Biotransformations in Small-molecule Pharmaceutical Development
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
1.3 Application of Biocatalysis in the Pharmaceutical Industry 15
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
16 Biotransformations in Small-molecule Pharmaceutical Development
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)
1.3 Application of Biocatalysis in the Pharmaceutical Industry 17
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.
18 Biotransformations in Small-molecule Pharmaceutical Development
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
1.3 Application of Biocatalysis in the Pharmaceutical Industry 19
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
20 Biotransformations in Small-molecule Pharmaceutical Development
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
1.3 Application of Biocatalysis in the Pharmaceutical Industry 21
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
22 Biotransformations in Small-molecule Pharmaceutical Development
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
1.3 Application of Biocatalysis in the Pharmaceutical Industry 23
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
24 Biotransformations in Small-molecule Pharmaceutical Development
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)
1.3 Application of Biocatalysis in the Pharmaceutical Industry 25
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
26 Biotransformations in Small-molecule Pharmaceutical Development
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
1.3 Application of Biocatalysis in the Pharmaceutical Industry 27
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
28 Biotransformations in Small-molecule Pharmaceutical Development
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 Application of Biocatalysis in the Pharmaceutical Industry 29
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
30 Biotransformations in Small-molecule Pharmaceutical Development
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
1.3 Application of Biocatalysis in the Pharmaceutical Industry 31
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
32 Biotransformations in Small-molecule Pharmaceutical Development
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
1.3 Application of Biocatalysis in the Pharmaceutical Industry 33
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.
34 Biotransformations in Small-molecule Pharmaceutical Development
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))
1.3 Application of Biocatalysis in the Pharmaceutical Industry 35
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
36 Biotransformations in Small-molecule Pharmaceutical Development
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
1.3 Application of Biocatalysis in the Pharmaceutical Industry 37
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
38 Biotransformations in Small-molecule Pharmaceutical Development
(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
1.3 Application of Biocatalysis in the Pharmaceutical Industry 39
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
40 Biotransformations in Small-molecule Pharmaceutical Development
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 Application of Biocatalysis in the Pharmaceutical Industry 41
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
42 Biotransformations in Small-molecule Pharmaceutical Development
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
1.3 Application of Biocatalysis in the Pharmaceutical Industry 43
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
44 Biotransformations in Small-molecule Pharmaceutical Development
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
1.3 Application of Biocatalysis in the Pharmaceutical Industry 45
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)
46 Biotransformations in Small-molecule Pharmaceutical Development
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
1.3 Application of Biocatalysis in the Pharmaceutical Industry 47
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
48 Biotransformations in Small-molecule Pharmaceutical Development
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
1.3 Application of Biocatalysis in the Pharmaceutical Industry 49
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
50 Biotransformations in Small-molecule Pharmaceutical Development
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
1.3 Application of Biocatalysis in the Pharmaceutical Industry 51
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
52 Biotransformations in Small-molecule Pharmaceutical Development
(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
1.3 Application of Biocatalysis in the Pharmaceutical Industry 53
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
54 Biotransformations in Small-molecule Pharmaceutical Development
• 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
56 Biotransformations in Small-molecule Pharmaceutical Development
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
1.4 Enzymes in Organic Solvent 57
(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
58 Biotransformations in Small-molecule Pharmaceutical Development
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
1.4 Enzymes in Organic Solvent 59
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
60 Biotransformations in Small-molecule Pharmaceutical Development
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
62 Biotransformations in Small-molecule Pharmaceutical Development
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
64 Biotransformations in Small-molecule Pharmaceutical Development
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
66 Biotransformations in Small-molecule Pharmaceutical Development
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
68 Biotransformations in Small-molecule Pharmaceutical Development
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.
70 Biotransformations in Small-molecule Pharmaceutical Development
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.
72 Biotransformations in Small-molecule Pharmaceutical Development
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.
74 Biotransformations in Small-molecule Pharmaceutical Development
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.
76 Biotransformations in Small-molecule Pharmaceutical Development
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.
78 Biotransformations in Small-molecule Pharmaceutical Development
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.
80 Biotransformations in Small-molecule Pharmaceutical Development
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.
82 Biotransformations in Small-molecule Pharmaceutical Development
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,
Practical Methods for Biocatalysis and Biotransformations Edited by John Whittall and Peter Sutton
� 2009 John Wiley & Sons, Ltd
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
86 Biocatalyst Identification and Scale-up: Molecular Biology for Chemists
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
2.2 Identifying Potential Biocatalysts for Chemical Synthesis 87
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
88 Biocatalyst Identification and Scale-up: Molecular Biology for Chemists
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
2.2 Identifying Potential Biocatalysts for Chemical Synthesis 89
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
90 Biocatalyst Identification and Scale-up: Molecular Biology for Chemists
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
2.2 Identifying Potential Biocatalysts for Chemical Synthesis 91
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
92 Biocatalyst Identification and Scale-up: Molecular Biology for Chemists
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.
94 Biocatalyst Identification and Scale-up: Molecular Biology for Chemists
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
2.3 Molecular Biology for Improved Biocatalysts 95
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).
96 Biocatalyst Identification and Scale-up: Molecular Biology for Chemists
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
2.3 Molecular Biology for Improved Biocatalysts 97
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
98 Biocatalyst Identification and Scale-up: Molecular Biology for Chemists
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
2.3 Molecular Biology for Improved Biocatalysts 99
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
100 Biocatalyst Identification and Scale-up: Molecular Biology for Chemists
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
2.3 Molecular Biology for Improved Biocatalysts 101
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
102 Biocatalyst Identification and Scale-up: Molecular Biology for Chemists
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)
2.3 Molecular Biology for Improved Biocatalysts 103
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
104 Biocatalyst Identification and Scale-up: Molecular Biology for Chemists
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
2.3 Molecular Biology for Improved Biocatalysts 105
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
106 Biocatalyst Identification and Scale-up: Molecular Biology for Chemists
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
2.3 Molecular Biology for Improved Biocatalysts 107
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
108 Biocatalyst Identification and Scale-up: Molecular Biology for Chemists
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
2.3 Molecular Biology for Improved Biocatalysts 109
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
110 Biocatalyst Identification and Scale-up: Molecular Biology for Chemists
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
112 Biocatalyst Identification and Scale-up: Molecular Biology for Chemists
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.
114 Biocatalyst Identification and Scale-up: Molecular Biology for Chemists
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
Practical Methods for Biocatalysis and Biotransformations Edited by John Whittall and Peter Sutton
� 2009 John Wiley & Sons, Ltd
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
3.1.2.1 Materials and Equipment
• 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.
120 Kinetic Resolutions Using Biotransformations
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
3.2.1.1 Materials and Equipment
• 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).
122 Kinetic Resolutions Using Biotransformations
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.
124 Kinetic Resolutions Using Biotransformations
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
3.3.1.1 Materials and Equipment
• 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.
3.3 Desymmetrization of Prochiral Ketones Using Enzymes 125
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
3.3.2.1 Materials and Equipment
• 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
• rotary evaporator
• 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.
126 Kinetic Resolutions Using Biotransformations
3.3.3 Procedure 3: Resolution of (–)-4-Cyano-4-(30,40-dichlorophenyl)cyclohex-1-
enyl Acetate (8)
3.3.3.1 Materials and Equipment
• (–)-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
• magnetic stirrer plate
• sintered vacuum filtration funnel
• rotary evaporator
• equipment for column chromatography.
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
3.3 Desymmetrization of Prochiral Ketones Using Enzymes 127
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.
128 Kinetic Resolutions Using Biotransformations
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
3.4.1.1 Materials and Equipment
• 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
• rotary evaporator
• 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
3.4.2.1 Materials and Equipment
• 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)
• anhydrous magnesium sulfate
• 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
130 Kinetic Resolutions Using Biotransformations
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.
Practical Methods for Biocatalysis and Biotransformations Edited by John Whittall and Peter Sutton
� 2009 John Wiley & Sons, Ltd
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
+ +
4.1.1.1 Materials and Equipment
• (–)-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
• rotary evaporator.
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
4.1.2.1 Materials and Equipment
• (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
• hotplate with magnetic stirrer
• oil bath
• analytical balance
• separation funnel, 50 mL
• rotary evaporator.
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.
136 DKR for the Synthesis of Esters, Amides and Acids Using Lipases
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
138 DKR for the Synthesis of Esters, Amides and Acids Using Lipases
• 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
• magnetic stirrer plate
• filter paper
• rotary evaporator
• 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.
140 DKR for the Synthesis of Esters, Amides and Acids Using Lipases
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
4.3.1.1 Materials and Equipment
• Pentamethylcyclopentadienyliridium(III) chloride dimer (4.57 g)
• sodium iodide (8.55 g)
• argon cylinder
• acetone (<100 ppm water) (525 mL)
• dichloromethane (500 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
• rotary evaporator
• 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
142 DKR for the Synthesis of Esters, Amides and Acids Using Lipases
4.3.2.1 Materials and Equipment
• 3-Methoxyphenol (6.47 g)
• tetrabutylammonium iodide (131.9 mg)
• dichloromethane (40 mL)
• 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
• rotary evaporator
• 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.
4.3 DKR of 6,7-Dimethoxy-1-methyl-1,2,3,4-tetrahydroisoquinoline 143
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
4.3.3.1 Materials and Equipment
• Toluene (30 mL)
• distilled water (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
• dichloromethane (100 mL)
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.
144 DKR for the Synthesis of Esters, Amides and Acids Using Lipases
• 1.0 M hydrochloric acid (100 mL)
• 1.0 M sodium hydroxide (100 mL)
• anhydrous sodium sulfate
• 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
• measuring cylinder, 50 mL
• measuring cylinder, 10 mL
• Buchner flask, 250 mL
• Buchner funnel
• filter paper
• rotary evaporator
• 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)
4.3 DKR of 6,7-Dimethoxy-1-methyl-1,2,3,4-tetrahydroisoquinoline 145
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
146 DKR for the Synthesis of Esters, Amides and Acids Using Lipases
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.
4.3 DKR of 6,7-Dimethoxy-1-methyl-1,2,3,4-tetrahydroisoquinoline 147
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
4.4.1.1 Materials and Equipment
• 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
• 50 mL round-bottom flask
• magnetic stirrer plate
• 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.
148 DKR for the Synthesis of Esters, Amides and Acids Using Lipases
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
4.4.2.1 Materials and Equipment
• Dried and degassed toluene (8 mL)
• 1-phenylethylamine (97 mg, 0.80 mmol)
• Pd/AlO(OH) (99 mg)
• Novozym 435 (96 mg)
• ethyl acetate (234 mL)
• 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
• rotary evaporator
• 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
4.4.3.1 Materials and Equipment
• Dried and degassed toluene (3 mL)
• 1-methyl-3-phenylpropylamine (90 mg, 0.60 mmol)
• Pd/AlO(OH) (891 mg)
• Novozym 435 (72 mg)
• ethyl acetate (176 mL)
• 50 mL round-bottom flask
• glass filter (pore size:20�30 mm)
• a balloon filled with H2 gas
• silica gel (Merck: silicagel 60 A 200�400 mesh), 10 g
• rotary evaporator
• 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.
150 DKR for the Synthesis of Esters, Amides and Acids Using Lipases
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)
152 DKR for the Synthesis of Esters, Amides and Acids Using Lipases
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
4.5.1.1 Materials and Equipment
• 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
• magnetic stirrer plate
• 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)
4.5.2.1 Materials and Equipment
• 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)
• one-necked reaction flask equipped with a magnetic stirrer bar, 25 mL
• magnetic stirrer plate
• cooling equipment
• rotary evaporator
• equipment for column chromatography.
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).
154 DKR for the Synthesis of Esters, Amides and Acids Using Lipases
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.
156 DKR for the Synthesis of Esters, Amides and Acids Using Lipases
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
4.6.1.1 Materials and Equipment
• (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
• rotary evaporator
• 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
4.6 Chemoenzymatic Dynamic Kinetic Resolution of (S)-Ibuprofen 157
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
4.6.2.1 Materials and Equipment
• 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.
158 DKR for the Synthesis of Esters, Amides and Acids Using Lipases
4.6.3 Procedure 3: Enzymatic Kinetic Resolution of (R,S)-2-Ethoxyethyl Ibuprofen
Ester with Immobilized Lipase
where KS�KR.
4.6.3.1 Materials and Equipment
• (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 Chemoenzymatic Dynamic Kinetic Resolution of (S)-Ibuprofen 159
4.6.4 Procedure 4: DKR of (R,S)-2-Ethoxyethyl Ibuprofen Ester
4.6.4.1 Materials and Equipment
• 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)
160 DKR for the Synthesis of Esters, Amides and Acids Using Lipases
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
References and Notes
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.
4.6 Chemoenzymatic Dynamic Kinetic Resolution of (S)-Ibuprofen 161
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
4.7.1.1 Materials and Equipment
• 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)
162 DKR for the Synthesis of Esters, Amides and Acids Using Lipases
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
4.7.2.1 Materials and Equipment
• 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.
164 DKR for the Synthesis of Esters, Amides and Acids Using Lipases
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.
Practical Methods for Biocatalysis and Biotransformations Edited by John Whittall and Peter Sutton
� 2009 John Wiley & Sons, Ltd
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
5.1.1.1 Materials and Equipment
• 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
5.1.2.1 Materials and Equipment
• Alcalase (0.3 mL) (Novo Industrial, Denmark; 2.5 AU mL�1)
• Z-Asp-OMe (28.1 mg, 0.1 mmol)
• Ser-NH2 (72.8 mg, 0.7 mmol)
• triethylamine (56 mL, 0.4 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
process was monitored at 220 nm. The fractions collected were lyophilized to afford the
desired product (29.3 mg). The HPLC purity and the yield of the product were 90.6 %
and 75.5 % respectively.
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
5.1.3.1 Materials and Equipment
• Alcalase (0.3 mL) (Novo Industrial, Denmark; 2.5 AU mL�1)
• Bz-Arg-OEt�HCl (34.3 mg, 0.1 mmol)
• Gly-Asp-(NH2)2 (94 mg, 0.5 mmol)
• triethylamine (70 mL, 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
process was monitored at 220 nm. The fractions collected were lyophilized to afford the
168 Enzymatic Selectivity in Synthetic Methods
desired product (34.2 mg). The HPLC purity and the yield of the product were 96.6 %
and 73.6 % respectively.
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
5.2.1.1 Materials and Equipment
• Vinyl chloroformate (50 mmol)
• acetone oxime (35 mmol)
• anhydrous pyridine (10 mL)
• dichloromethane (3 � 20 mL)
• brine
• anhydrous sodium sulfate
• N2 gas
• filter paper
• one 25 mL Schlenk
• one addition funnel
• one separatory funnel
• rotary evaporator
• 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
170 Enzymatic Selectivity in Synthetic Methods
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
5.2.2.1 Materials and Equipment
• 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
• rotary evaporator
• 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.
172 Enzymatic Selectivity in Synthetic Methods
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
5.3.2.1 Materials and Equipment
• 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
• hotplate with magnetic stirrer
• 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
174 Enzymatic Selectivity in Synthetic Methods
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).
5.3.4.1 Materials and Equipment
• 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
• water-cooled condenser
• vacuum pump with pressure control system.
5.3 Use of Lipase Enzymes for the Synthesis of Polymers 175
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
5.3.6.1 Materials and Equipment
• Dihydroxyethyl hexamethylene bis-carbamate (15.23 g, 0.05 mol)
• 1,6-hexanediol (50.37 g, 0.43 mol)
• adipic acid (69.92 g, 0.48 mol)
• Novozym 435 (1.3 g)
• diphenyl ether (136 mL)
• nitrogen
• four-necked flange flask with Heidolf mechanical stirrer
• 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
176 Enzymatic Selectivity in Synthetic Methods
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
5.3.8.1 Materials and Equipment
• Dihydroxyethyl hexamethylene bis-carbamate (7.31 g, 0.025 mol)
• 1,6-hexanediol (29.39 g, 0.249 mol)
• adipic acid (40.17 g, 0.274 mol)
• Novozym 435 (1.2 g)
• n-heptane (40 mL)
• nitrogen
• four-necked flange flask with Heidolf mechanical stirrer
• oil bath
• hotplate with magnetic stirrer
• water-cooled condenser with Dean–Stark trap
• vacuum pump with pressure control system.
5.3 Use of Lipase Enzymes for the Synthesis of Polymers 177
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
5.3.10.1 Materials and Equipment
• 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)
178 Enzymatic Selectivity in Synthetic Methods
• Novozym 435 (1.04 g)
• four-necked flange flask with stirrer bar
• oil bath
• hotplate with magnetic stirrer
• water-cooled condenser
• vacuum pump with pressure control system.
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
5.3.12.1 Materials and Equipment
• 1,6-Hexanediol (42.84 g, 0.363 mol)
• adipic acid (52.97 g, 0.363 mol)
• toluene (20 g)
5.3 Use of Lipase Enzymes for the Synthesis of Polymers 179
• Novozym 435 (1.44 g)
• nitrogen
• four-necked flange flask with Heidolf mechanical stirrer
• oil bath
• hotplate
• water-cooled condenser
• vacuum pump with pressure control system.
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.
180 Enzymatic Selectivity in Synthetic Methods
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.
5.3 Use of Lipase Enzymes for the Synthesis of Polymers 181
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
5.4.1.1 Materials and Equipment
• 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)
• distilled water (1300 mL)
• 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
182 Enzymatic Selectivity in Synthetic Methods
• 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
5.4.2.1 Materials and Equipment
• 3-Cyanopyridine (166.4 g)
• 0.1 M potassium phosphate buffer pH 8.0 (1000 mL)
• hydrochloric acid (conc.) (100 mL)
• distilled water (1000 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.
184 Enzymatic Selectivity in Synthetic Methods
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
5.5.1.1 Materials and Equipment
• 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%)
• ethyl acetate (100 mL)
• heptane (200 mL)
• MgSO4 (500 mg)
• incubator
• flask (100 mL)
• glass fritt
• separation funnel
• rotary evaporator
• equipment for column chromatography
• thin-layer chromatography (TLC) plates (silica-gel-coated glass plates Merck 60 F254)
186 Enzymatic Selectivity in Synthetic Methods
• 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
5.5.2.1 Materials and Equipment
• 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
• ethyl acetate (600 mL)
• MgSO4
• diazomethane solution in diethyl ether
• acetic acid
• ethanol
• petroleum ether
5.5 Enzyme-promoted Desymmetrization of Prochiral Dinitriles 187
• incubator
• flask (500 mL)
• glass filter
• rotary evaporator
• equipment for column chromatography
• 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
5.5.3.1 Materials and Equipment
• 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
188 Enzymatic Selectivity in Synthetic Methods
• flask (100 mL)
• rotary evaporator
• equipment for column chromatography
• 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.
5.5 Enzyme-promoted Desymmetrization of Prochiral Dinitriles 189
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.
5.6.1.1 Materials and Equipment
• 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.
190 Enzymatic Selectivity in Synthetic Methods
• dimethyl formamide (DMF; anhydrous, 65 mL)
• NH4Cl solution (saturated aqueous, 35 mL)
• ethyl acetate (80 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)
• magnetic stirrer plate
• one 300 mL separatory funnel
• rotary evaporator
• equipment for column chromatography
• cooling equipment.
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
5.6.2.1 Materials and Equipment
• 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)
• ethyl acetate (90 mL)
• hexane (1100 mL)
• Na2SO4 (anhydrous, 10 g)
• silica gel 60 (spherical; 100–210 mm, 37558-79, KANTO Chemical Inc., 425 g)
• two-necked reaction flask (100 mL) equipped with a magnetic stirrer bar
• magnetic stirrer plate
• one 100 mL separatory funnel
• rotary evaporator
• equipment for column chromatography.
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.
3. The crude residue (5.30 g) was purified by silica gel column chromatography. Elution
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
5.6.3.1 Materials and Equipment
• (–)-1-Benzyloxymethyl-1-methyloxirane (4.5 g, 25.2 mmol)
• B. subtilis Tamy-2 strain cell suspension (19.5 mL)
192 Enzymatic Selectivity in Synthetic Methods
• 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)
• ethyl acetate (550 mL)
• hexane (600 mL)
• Na2SO4 (anhydrous, 20 g)
• silica gel 60 (spherical; 100–210 mm, 37558-79, KANTO Chemical Inc., 160 g)
• 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)
• round-bottomed reaction flask (200 mL) equipped with a magnetic stirrer bar
• magnetic stirrer plate
• one 100 mL separatory funnel
• one 200 mL separatory funnel
• rotary evaporator
• equipment for column chromatography
• 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.
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
5.6 Catalyzed Synthesis of (R)-3-Benzyloxy-2-methylpropane-1,2-diol 193
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
5.6.4.1 Materials and Equipment
• (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)
194 Enzymatic Selectivity in Synthetic Methods
• 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)
• ethyl acetate (600 mL)
• hexane (500 mL)
• Na2SO4 (anhydrous, 10 g)
• silica gel 60 (spherical; 100–210 mm, 37558-79, KANTO Chemical Inc., 100 g)
• two-necked reaction flask (100 mL) equipped with a magnetic stirrer bar, an argon
balloon and a thermometer (�100 �C to þ50 �C)
• magnetic stirrer plate
• one 100 mL separatory funnel
• rotary evaporator
• equipment for column chromatography
• cooling equipment.
• (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)
• ethyl acetate (450 mL)
• hexane (600 mL)
• Na2SO4 (anhydrous, 15 g)
• silica gel 60 (spherical; 100–210 mm, 37558-79, KANTO Chemical Inc., 38 g)
• round-bottomed reaction flask (200 mL) equipped with a magnetic stirrer bar and a
thermometer (�100 �C to þ50 �C)
• magnetic stirrer plate
• one 200 mL separatory funnel
• rotary evaporator
• equipment for column chromatography
• cooling equipment.
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.
5.6 Catalyzed Synthesis of (R)-3-Benzyloxy-2-methylpropane-1,2-diol 195
3. The crude residue (2.96 g) was purified by silica gel column chromatography. Elution
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.
6. The crude residue (0.77 g) was purified by silica gel column chromatography. Elution
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
5.6.5.1 Materials and Equipment
• (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)
• ethyl acetate (500 mL)
• hexane (500 mL)
• Na2SO4 (anhydrous, 5 g)
• silica gel 60 (spherical; 100–210 mm, 37558-79, KANTO Chemical Inc., 21 g)
• round-bottomed reaction flask (30 mL) equipped with a magnetic stirrer bar
• magnetic stirrer plate
• one 50 mL separatory funnel
• rotary evaporator
• equipment for column chromatography.
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.
196 Enzymatic Selectivity in Synthetic Methods
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.
5.6 Catalyzed Synthesis of (R)-3-Benzyloxy-2-methylpropane-1,2-diol 197
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.
198 Enzymatic Selectivity in Synthetic Methods
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
+
Halohydrin dehalogenase
NaCN, buffer
Halohydrin dehalogenase
40% yield, 96.8% ee
41% yield, 95.2% ee
5.7.1.1 Materials and Equipment
• 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
5.7.2.1 Materials and Equipment
• Methyl 4-chloro-3-hydroxybutanoate (0.50 g, 3.3 mmol)
• tris-SO4 buffer (62 ml, 0.5 M, pH 7.5)
200 Enzymatic Selectivity in Synthetic Methods
• NaCN (322 mg, 6.6 mmol)
• HheC-W249F (15 mg purified mutant halohydrin dehalogenase in 3 ml buffer)
• NaCl
• ethyl acetate
• anhydrous magnesium sulfate
• 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
• magnetic stirrer
• 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
• rotary evaporator
• 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.
202 Enzymatic Selectivity in Synthetic Methods
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.
Practical Methods for Biocatalysis and Biotransformations Edited by John Whittall and Peter Sutton
� 2009 John Wiley & Sons, Ltd
6.1.1 Procedure 1: Expression of RhaD
6.1.1.1 Materials and Equipment
• 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
6.1.2.1 Materials and Equipment
• Crude RhaD preparation (2.4 g wet weight)
• distilled water (160 mL)
• 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
• rotary evaporator.
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
6.2.1.1 Materials and Equipment
• 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.
206 Aldolase Enzymes for Complex Synthesis
• 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
• ethyl acetate (60 mL)
• dichloromethane (600 mL)
• methanol (80 mL)
• silica gel 60 (40–63 mm, Merck) (25 g)
• thin-layer chromatography (TLC) plates (silica gel 60 F254, Merck)
• one-necked reaction flask equipped with a magnetic stirrer bar, 25 mL
• one-necked reaction flask equipped with a magnetic stirrer bar, 100 mL
• magnetic stirrer plate
• sintered glass funnel
• filtered flask
• water bath and thermostat
• microtube Eppendorf, 1.5 mL
• microcentrifuge
• pH meter
• rotary evaporator
• 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
6.2.2.1 Materials and Equipment
• 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)
• ethyl acetate (100 mL)
• 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)
• one-necked reaction flask equipped with a magnetic stirrer bar, 50 mL
• magnetic stirrer plate
• sintered glass funnel
• filtered flask
• syringe (10 ml)
• Daicel Chiracel OD column (25 cm � 4.6 mm ID)
• rotary evaporator
• 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.
208 Aldolase Enzymes for Complex Synthesis
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 %
6.2.3.1 Materials and Equipment
• (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)
• methanol (20 mL)
• TLC plates (silica gel 60 F254, Merck)
• Parr apparatus
• ultrafiltration membrane
• glass funnel filtration system with sintered glass membrane support
• filtered flask
• rotary evaporator
• equipment for column chromatography.
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).
6.2 Straightforward RAMA-mediated Synthesis of Aminocyclitols 209
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
210 Aldolase Enzymes for Complex Synthesis
References and Notes
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.
6.2 Straightforward RAMA-mediated Synthesis of Aminocyclitols 211
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
6.3.1.1 Materials and Equipment
• 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
212 Aldolase Enzymes for Complex Synthesis
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
6.3.2.1 Materials and Equipment
• 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
• anhydrous sodium sulfate
• filter paper
• one 500 mL round-bottom flask
• one Buchner funnel, diameter 10 cm
• one Buchner flask, 500 mL
• one 500 mL separation funnel
• rotary evaporator.
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
214 Aldolase Enzymes for Complex Synthesis
(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
6.3.3.1 Materials and Equipment
• N-Benzyloxycarbonyl-3-amino-1-propanal 1 (4.7 g)
• DHA (2.1 g)
• dimethylformamide (DMF, 40 mL)
• methanol (200 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
• one Buchner flask, 500 mL
• rotary evaporator
• 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
6.3 Catalysed Synthesis of D-Fagomine 215
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
216 Aldolase Enzymes for Complex Synthesis
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.
References and Notes
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.
6.3 Catalysed Synthesis of D-Fagomine 217
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
6.4.1.1 Materials and Equipment
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
218 Aldolase Enzymes for Complex Synthesis
• 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.
6.4 Chemoenzymatic Synthesis of 5-Thio-D-xylopyranose 219
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
6.4.2.1 Materials and Equipment
• 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
220 Aldolase Enzymes for Complex Synthesis
• 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
• anhydrous magnesium sulfate
• centrifuge
• heating block
• shaker
• rotary evaporator.
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 Chemoenzymatic Synthesis of 5-Thio-D-xylopyranose 221
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
• Round-bottom flask (100 mL)
• 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)
• methylene chloride (analytical grade)
• Merck 60 F254 silica gel TLC plates
• Merck 60/230–400 and 60/40–63 mesh silica
• heating block
• shaker
• anhydrous magnesium sulfate
• rotary evaporator.
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
222 Aldolase Enzymes for Complex Synthesis
(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
• Round-bottom flask (50 mL)
• sodium hydrosulfide hydrate (NaSH�xH2O) from Aldrich (0.532 g)
• distilled water (70 mL)
• methanol (analytical grade)
• methylene chloride (analytical grade)
• Merck 60 F254 silica gel TLC plates
• Merck 60/230–400 and 60/40–63 mesh silica
• centrifuge
• heating block
• shaker
• anhydrous magnesium sulfate
• rotary evaporator.
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
6.4 Chemoenzymatic Synthesis of 5-Thio-D-xylopyranose 223
6.4.4.1 Materials and Equipment
• 5 mL capped vials
• distilled water (70 mL)
• 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)
• Merck 60 F254 silica gel TLC plates
• Merck 60/230–400 and 60/40–63 mesh silica
• Millipore� ultrafiltration system
• heating block
• shaker
• rotary evaporator
• 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.
224 Aldolase Enzymes for Complex Synthesis
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).
6.4 Chemoenzymatic Synthesis of 5-Thio-D-xylopyranose 225
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
Practical Methods for Biocatalysis and Biotransformations Edited by John Whittall and Peter Sutton
� 2009 John Wiley & Sons, Ltd
7.1.1.1 Materials and Equipment
• 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)
• methanol (15 mL)
• dichloromethane (25 mL)
• saturated NaHCO3 solution (25 mL)
• anhydrous MgSO4 (3 g)
• ethyl acetate (150 mL)
• hexanes (150 mL)
• silica gel 60 (particle size 40–63 mm, Merck)
• thin-layer chromatography (TLC) plates (silica gel 60 F254, Merck)
• 50 mL reaction flask
• 50 mL separatory funnel
• rotary evaporator
• equipment for flash column chromatography.
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
7.1 Glycosynthase-assisted Oligosaccharide Synthesis 229
7.1.2.1 Materials and Equipment
• �-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)
• methanol (10 mL)
• dichloromethane (25 mL)
• saturated NaHCO3 solution (25 mL)
• anhydrous MgSO4 (3 g)
• ethyl acetate (150 mL)
• hexanes (150 mL)
• silica gel 60 (particle size 40–63 mm, Merck)
• TLC plates (silica gel 60 F254, Merck)
• 50 mL reaction flask
• 50 mL separatory funnel
• rotary evaporator
• equipment for flash column chromatography.
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
230 Enzymatic Synthesis of Glycosides and Glucuronides
(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.
7.1 Glycosynthase-assisted Oligosaccharide Synthesis 231
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.
7.2.1.2 Materials and Equipment
• 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
232 Enzymatic Synthesis of Glycosides and Glucuronides
• 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.
234 Enzymatic Synthesis of Glycosides and Glucuronides
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
236 Enzymatic Synthesis of Glycosides and Glucuronides
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
7.3.1.1 Materials and Equipment
• Almond (200 g)
• apricot kernel (200 g)
• peach kernel (200 g)
• apple seed (100 g)
• distilled water (1000 mL)
• ethyl acetate (1000 mL)
• 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
7.3.2.1 Materials and Equipment
• Glucose
• almond meal (200 g) or apricot kernel (200 g) or peach kernel (200 g) or
• apple seed (100 g)
• distilled water (1000 mL)
• dioxane
• tert-butyl alcohol
• CH3CN
• magnetic stirrer
• ethyl acetate (1000 mL)
• CH3OH (100mL)
• thin-layer chromatography plates (silica gel 60 F254)
• silica gel (300-400 mesh)
• acetone (500mL)
• rotary evaporator
238 Enzymatic Synthesis of Glycosides and Glucuronides
• one 50 mL Buchi funnel
• equipment for column chromatography.
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).
References and Notes
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.
240 Enzymatic Synthesis of Glycosides and Glucuronides
7.4.1.1 Materials and Equipment
• 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)
• ethyl acetate (800 mL)
• 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
• UV–vis spectrophotometer
• Merck 60 F254 thin-layer chromatography (TLC) plates
• rotary evaporator
• 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
7.4 Laccase-mediated Oxidation of Natural Glycosides 241
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
7.4.2.1 Materials and Equipment
• Amygdalin (200 mg, 0.437 mmol)
• TEMPO (13.6 mg, 0.087 mmol)
• 50 mM acetate buffer, pH 4.5 (13 mL)
• Laccase from T. versicolor purchased from Fluka (30 mg, 440 U)
• ABTS (5.8 mg, 0.01 mmol)
• ethyl acetate (800 mL)
• MeOH (400 mL)
• (NH4)6Mo7O24 x 4H2O (42 g)
• Ce(SO4)2 (2 g)
• H2SO4 concentrated (62 mL)
• vial (50 mL)
• shaker
• UV–vis spectrophotometer
• Merck 60 F254 TLC plates
• rotary evaporator
• silica gel (Merck 60, 230–400 mesh)
• column chromatography equipment.
242 Enzymatic Synthesis of Glycosides and Glucuronides
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
7.4.3.1 Materials and Equipment
• 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)
• 50 mM acetate buffer, pH 4.0 (6 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.
7.4 Laccase-mediated Oxidation of Natural Glycosides 243
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
7.4.4.1 Materials and Equipment
• Thiocolchicoside (200 mg, 0.354 mmol)
• TEMPO (11 mg, 0.07 mmol)
• 50 mM acetate buffer, pH 4.5 (13 mL)
• 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)
• H2SO4 concentrated (62 mL)
• 50 mL vial
• shaker
• Merck 60 F254 TLC plates
• UV–vis spectrophotometer
• 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.
244 Enzymatic Synthesis of Glycosides and Glucuronides
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
7.5.1.1 Materials and Equipment
• 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
7.5.2.1 Materials and Equipment
• Freshly prepared substrate solution (100 mM in 20 % dimethylsulfoxide (DMSO))
• tris-HCl buffer (1 M, pH 8.0)
246 Enzymatic Synthesis of Glycosides and Glucuronides
• 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
• rotary evaporator
• 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
7.5 Biocatalysed Synthesis of Monoglucuronides 247
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
7.5.3.1 Materials and Equipment
• 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
• reciprocal shaker
• rotary evaporator
• 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).
248 Enzymatic Synthesis of Glycosides and Glucuronides
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).
7.5 Biocatalysed Synthesis of Monoglucuronides 249
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.
250 Enzymatic Synthesis of Glycosides and Glucuronides
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
7.6.1.1 Materials and Equipment
• 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)
• refrigerated centrifuge
• 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.
7.6 Synthesis of the Acyl Glucuronide of Mycophenolic Acid 251
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
7.6.2.1 Materials and Equipment
• 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
• rotary evaporator
• 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
252 Enzymatic Synthesis of Glycosides and Glucuronides
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.
7.6 Synthesis of the Acyl Glucuronide of Mycophenolic Acid 253
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.
254 Enzymatic Synthesis of Glycosides and Glucuronides
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
Practical Methods for Biocatalysis and Biotransformations Edited by John Whittall and Peter Sutton
� 2009 John Wiley & Sons, Ltd
8.1.1 Procedure 1: (S)-3-Hydroxytetrahydrothiophene-3-carbonitrile
S
OH
CN
8.1.1.1 Materials and Equipment
• 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
• rotary evaporator.
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
8.1.2.1 Materials and Equipment
• (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
• magnetic stirrer plate
• 50 mL separatory funnel
• frit
• rotary evaporator.
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
8.1.3.1 Materials and Equipment
• (S)-3-Hydroxytetrahydrothiophene-3-carboxylic acid (300 mg, 2 mmol)
• Raney nickel (5.5 g moist, 50 % Ni)
• distilled water (28 mL)
• 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
• magnetic stirrer plate
• 50 mL separatory funnel
• frit
• rotary evaporator.
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.
258 Synthesis of Cyanohydrins Using Hydroxynitrile Lyases
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
8.2.1.1 Materials and Equipment
• 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)
• ethyl acetate (25 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.
260 Synthesis of Cyanohydrins Using Hydroxynitrile Lyases
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
8.3.1.1 Materials and Equipment
• 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.
262 Synthesis of Cyanohydrins Using Hydroxynitrile Lyases
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.
264 Synthesis of Cyanohydrins Using Hydroxynitrile Lyases
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
8.4.1.1 Materials and Equipment
• 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.
266 Synthesis of Cyanohydrins Using Hydroxynitrile Lyases
• nitrogen gas
• fume hood
• cyanide detector
• 25 mL flask
• magnetic stirrer
• rotary evaporator.
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.
268 Synthesis of Cyanohydrins Using Hydroxynitrile Lyases
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
8.5.1.1 Materials and Equipment
• 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
• thin-layer chromatography (TLC) plates (silica gel 60 F254, Merck)
• 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
8.5.2.1 Materials and Equipment
• 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
• anhydrous Na2SO4
270 Synthesis of Cyanohydrins Using Hydroxynitrile Lyases
• magnetic stirrer plate
• filter paper
• separatory funnel
• rotary evaporator
• 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.
272 Synthesis of Cyanohydrins Using Hydroxynitrile Lyases
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
9.1.1.1 Materials and Equipment
• 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)
Practical Methods for Biocatalysis and Biotransformations Edited by John Whittall and Peter Sutton
� 2009 John Wiley & Sons, Ltd
• 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
9.2.1.1 Materials and Equipment
• 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
276 Synthesis of Chiral sec-Alcohols by Ketone Reduction
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
9.3.1.1 Materials and Equipment
• 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)
• ethyl acetate (200 mL)
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
278 Synthesis of Chiral sec-Alcohols by Ketone Reduction
• saturated NaCl solution (70 mL)
• anhydrous MgSO4 (3 g)
• pH meter
• one-necked reaction flask equipped with magnetic stirring bar, 250 mL
• magnetic stirring plate
• one 250 mL separatory funnel
• rotary evaporator.
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
9.3.2.1 Materials and Equipment
• 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-A1B (50 mg) (Codexis Inc.)
• NaOH solution (2 M)
• ethyl acetate (200 mL)
9.3 Synthesis of a-Alkyl-�-hydroxy Ketones and Esters by Isolated KREDS 279
• saturated NaCl solution (70 mL)
• anhydrous MgSO4 (3 g)
• pH meter
• one-necked reaction flask equipped with magnetic stirring bar, 250 mL
• magnetic stirring plate
• one 250 mL separatory funnel
• rotary evaporator.
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 %).
1H NMR (CDCl3; 500 MHz) � 5.74–5.82 (m, 1H), 5.02–5.12 (m, 2H), 4.02–4.07
(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
9.3.3.1 Materials and Equipment
• 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-108 (70 mg) (Codexis Inc.)
280 Synthesis of Chiral sec-Alcohols by Ketone Reduction
• NaOH solution (2 M)
• ethyl acetate (200 mL)
• saturated NaCl solution (70 mL)
• anhydrous MgSO4 (3 g)
• pH meter
• one-necked reaction flask equipped with magnetic stirring bar, 250 mL
• magnetic stirring plate
• filter paper
• one 250 mL separatory funnel
• rotary evaporator.
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 %).
1H NMR (CDCl3; 500 MHz) � 5.67–5.76 (m, 1H), 5.03–5.11 (m, 2H), 3.92–3.97
(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 Synthesis of a-Alkyl-�-hydroxy Ketones and Esters by Isolated KREDS 281
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]
282 Synthesis of Chiral sec-Alcohols by Ketone Reduction
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.
9.3 Synthesis of a-Alkyl-�-hydroxy Ketones and Esters by Isolated KREDS 283
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
9.4.1.1 Materials and Equipment
• 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
284 Synthesis of Chiral sec-Alcohols by Ketone Reduction
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
9.4.2.1 Materials and Equipment
• Ketone substrate (0.34 mmol)
• 2-propanol (600 mL)
• hexane (2 mL)
• ethyl acetate (4 mL)
• anhydrous Na2SO4
• 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
• rotary evaporator
• equipment for column chromatography.
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.
286 Synthesis of Chiral sec-Alcohols by Ketone Reduction
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)
• nicotinamide adenine dinucleotide phosphate (NADPþ, 40 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)
• separatory funnel
• rotary evaporator.
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
288 Synthesis of Chiral sec-Alcohols by Ketone Reduction
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
9.5.2.1 Materials and equipment
• 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.
290 Synthesis of Chiral sec-Alcohols by Ketone Reduction
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
9.6.1.1 Materials and Equipment
• 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
• UV–vis spectrophotometer
• 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
9.6.2.1 Materials and Equipment
• 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)
• nicotinamide adenine dinucleotide phosphate (NADPþ, 10 mg)
• 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
292 Synthesis of Chiral sec-Alcohols by Ketone Reduction
• ethyl acetate
• 100 mL test tube (2.7 cm diameter)
• water bath with a thermostat
• magnetic stirrer
• vortex mixer
• pH indicator paper
• centrifuge
• rotary evaporator
• 200 mL Erlenmeyer flask
• 200 mL round-bottom flask
• 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)).
294 Synthesis of Chiral sec-Alcohols by Ketone Reduction
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
Practical Methods for Biocatalysis and Biotransformations Edited by John Whittall and Peter Sutton
� 2009 John Wiley & Sons, Ltd
10.1.1.1 Materials and equipment
• 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.
298 Reduction of Functional Groups
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 Light-driven Stereoselective Biocatalytic Oxidations and Reductions 299
10.2.1 Procedure 1: Expression and Purification of PAMO-P3
10.2.1.1 Materials and equipment
• 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
300 Reduction of Functional Groups
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
10.2.2.1 Materials and Equipment
• 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.
10.2 Light-driven Stereoselective Biocatalytic Oxidations and Reductions 301
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
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 320 �C with 8 �C min�1, then
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
Achiral method. Instrument: Agilent Technologies 6890N; carrier gas: 0.6 bar H2; column:
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).
302 Reduction of Functional Groups
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
10.2.3.1 Materials and Equipment
• 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
10.2.4.1 Materials and equipment
• YqjM
• ketoisphorone
10.2 Light-driven Stereoselective Biocatalytic Oxidations and Reductions 303
• 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
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 110 �C with 5 �C min�1, then
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.
304 Reduction of Functional Groups
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.
10.2 Light-driven Stereoselective Biocatalytic Oxidations and Reductions 305
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
10.3.1.1 Materials and Equipment
• 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)
306 Reduction of Functional Groups
• 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
• magnetic stirrer
• 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.
References and Notes
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
308 Reduction of Functional Groups
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
10.4.1.1 Materials and Equipment
• 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
310 Reduction of Functional Groups
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
10.4.2.1 Materials and Equipment
• 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)
• reciprocal shaker
• 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.
312 Reduction of Functional Groups
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
10.5.1.1 Materials and Equipment
• Hydantoin (5.2 g, 52 mmol)
• 4-fluorobenzaldehyde (5 mL, 47 mmol)
• piperidine (9.9 mL, 100 mmol)
• cyclohexane
• ethyl acetate
• distilled water (200 mL)
• nitrogen gas
• HCl 12 M (�20 mL)
• NaOH 5 M aqueous (21 mL)
• thin-layer chromatography (TLC) plates (silica gel 60 F254, Merck)
• 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
• one 250 mL separatory funnel
314 Reduction of Functional Groups
• flask
• rotary evaporator.
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
10.5.2.1 Materials and Equipment
• 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
• rotary evaporator.
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 %).
316 Reduction of Functional Groups
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.
Practical Methods for Biocatalysis and Biotransformations Edited by John Whittall and Peter Sutton
� 2009 John Wiley & Sons, Ltd
11.1.1 Procedure 1: Preparation of the Biocatalyst
11.1.1.1 Materials and Equipment
• 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
11.1.2.1 Materials and Equipment
• 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
• rotary evaporator.
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.
322 Enzymatic Oxidation Chemistry
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
+
11.2.1.1 Materials and Equipment
• 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
• rotary evaporator.
*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
11.2 Glucose Oxidase-catalysed Synthesis of Aldonic Acids 323
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
C1 (in accordance with literature data7).
[�]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
C1 (in accordance with literature data9).
[�]D ¼ �8.9 (c ¼ 10, H2O) (lit.8 [�]D ¼ �8.8).
324 Enzymatic Oxidation Chemistry
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
C1 (in accordance with literature data12).
[�]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.
11.2 Glucose Oxidase-catalysed Synthesis of Aldonic Acids 325
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.
326 Enzymatic Oxidation Chemistry
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
11.3.1.1 Materials and Equipment
• (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
• magnetic stirrer.
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
11.3.2.1 Materials and Equipment
• 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
• magnetic stirrer.
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
328 Enzymatic Oxidation Chemistry
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
11.4.1.1 Materials and Equipment
• 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
• anhydrous magnesium sulfate
• diethyl ether
• filter paper
• one 10 mL test tube with a screw cap and equipped with a magnetic stirrer bar
• magnetic stirrer plate
• one 25 mL separatory funnel
• rotary evaporator.
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.
330 Enzymatic Oxidation Chemistry
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
332 Enzymatic Oxidation Chemistry
11.5.1 Procedure 1: Preparation and Purification of CHMO
11.5.1.1 Materials and Equipment
• 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
11.5.2.1 Materials and Equipment
• 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
• magnetic stirrer plate
• separatory funnel
• rotary evaporator
• high-performance liquid chromatography (HPLC) equipment
• Chiracel OD column (Daicel, Illkirch, France)
• column chromatography equipment.
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.
334 Enzymatic Oxidation Chemistry
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.
336 Enzymatic Oxidation Chemistry
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
11.6.1.1 Materials and Equipment
• 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
11.6.2.1 Materials and Equipment
• 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
• anhydrous sodium sulfate
• one 1 L shake flask with a cotton plug
• shaker
• photometer
• centrifuge
• one separatory funnel
• rotary evaporator.
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.
338 Enzymatic Oxidation Chemistry
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
11.6.3.1 Materials and Equipment
• 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
• anhydrous sodium sulfate
• silica gel
• thin-layer chromatography plates (silica gel 60 F254)
• reaction flask, 250 mL
• water bath
• magnetic stirrer
• separatory funnel
• rotary evaporator
• equipment for column chromatography.
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.
340 Enzymatic Oxidation Chemistry
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
11.7.1.1 Materials and Equipment
• 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
11.7.2.1 Materials and Equipment
• 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)
• ethyl acetate (150 mL)
• anhydrous MgSO4 for drying
• thin-layer chromatography (TLC) plates (silica gel 60 F254, Merck)
• 250 mL round-bottomed flask with a magnetic stirrer bar
• magnetic stirrer plate
• filter paper
• 250 mL separatory funnel
• rotary evaporator.
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).
342 Enzymatic Oxidation Chemistry
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.
References and Notes
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]
11.8.1.1 Materials and Equipment
• 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
• 50 mL Erlenmeyer flask
• 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
344 Enzymatic Oxidation Chemistry
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
11.8.2.1 Materials and Equipment
• 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)
• anhydrous sodium sulfate
• hexane
• ethyl acetate
• 500 mL baffled Erlenmeyer flask
• 500 mL separatory funnel
• rotary evaporator
• 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.
11.8.3.1 Materials and Equipment
• 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.
346 Enzymatic Oxidation Chemistry
• 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
• magnetic stirrer plate
• chiral-phase GC, �-Dex 225 column (Supelco Inc.)
• 50 mL and 500 mL separatory funnels
• rotary evaporator
• equipment for column chromatography.
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
11.8 Synthesis of Optically Pure Chiral Lactones by Cyclopentadecanone 347
11.8.4.1 Materials and Equipment
• 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
• anhydrous sodium sulfate
• 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)
• refrigerated centrifuge
• separation funnel, 3000 mL
• rotary evaporator.
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.
348 Enzymatic Oxidation Chemistry
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.
11.8 Synthesis of Optically Pure Chiral Lactones by Cyclopentadecanone 349
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.
350 Enzymatic Oxidation Chemistry
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
12.1.1.1 Materials and Equipment
• 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.
Practical Methods for Biocatalysis and Biotransformations Edited by John Whittall and Peter Sutton
� 2009 John Wiley & Sons, Ltd
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
12.1.2.1 Materials and Equipment
• 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
• anhydrous sodium sulfate
• 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.
• rotary evaporator
• 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).
References and Notes
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.
354 Whole-cell Oxidations and Dehalogenations
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
12.2.1.1 Materials and Equipment
• Luria–Bertani (LB) broth
• tryptone (35 g)
• yeast extract (17.5 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
12.2.2.1 Materials and Equipment
• 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
• rotary evaporator
• 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.
356 Whole-cell Oxidations and Dehalogenations
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
12.2.3.1 Materials and Equipment
• Distilled water (350 mL)
• hydrochloric acid (5 mL)
• 5-hydroxyl IMI (0.3 g)
• ethyl acetate (100 mL)
• anhydrous sodium sulfate (5 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.
358 Whole-cell Oxidations and Dehalogenations
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
12.3.1.1 Materials and Equipment
• Malt extract (2.25 g)
• casitone (0.375 g)
• agar (1.13 g)
• distilled water (75 mL)
• 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
12.3.2.1 Materials and Equipment
• Distilled water (16.5 L)
• glucose (281 g)
• Lab-Lemco (Oxoid) (42 g)
• peptone from casein (52.5 g)
• yeast extract (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)
• methanol (10 mL)
• XAD-16 adsorber resin (16 g) (Rohm and Haas France S.A.S., Lauterbourg, France)
• isopropanol (2 L)
• ethyl acetate (4.4 L)
• saturated NaCl solution (800 mL)
• 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
360 Whole-cell Oxidations and Dehalogenations
• 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
• rotary evaporator
• 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.
12.3 Biocatalytic Synthesis of 6-Hydroxy Fluvastatin 361
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
12.3.3.1 Materials and Equipment
• Distilled water 23 L
• Tween 80 (70 mg)
• glycerol (14 g)
• cellulose powder (4 g)
• oat meal (2 g)
362 Whole-cell Oxidations and Dehalogenations
• 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.
12.3 Biocatalytic Synthesis of 6-Hydroxy Fluvastatin 363
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
364 Whole-cell Oxidations and Dehalogenations
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
12.3 Biocatalytic Synthesis of 6-Hydroxy Fluvastatin 365
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
References and Notes
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.
366 Whole-cell Oxidations and Dehalogenations
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
12.4.1.1 Materials and Equipment
• Glucose (0.4 g)
• malt extract (1 g)
• yeast extract (0.4 g)
• tap water 100 mL
• malt extract (1 g)
• yeast extract (1 g)
• 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
12.4.2.1 Materials and Equipment
• Adamantane (41 mg, 0.3 mmol)
• Tween 60 (900 mg)
• ethyl acetate
• anhydrous magnesium sulfate
• n-hexane and ethyl acetate
• filter paper
• silica gel (Wakogel C300 45–75 mm), 15 g
• one 300 mL separatory funnel
• rotary evaporator.
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.
368 Whole-cell Oxidations and Dehalogenations
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
12.5.1.1 Materials and Equipment
• 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)
• anhydrous sodium sulfate (5 g)
• 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
• magnetic stirring plate
• 250 mL separating funnel
• rotary evaporator
• equipment for continuous liquid–liquid extraction
• equipment for flash column chromatography using a glass column (2.5 cm� 25 cm)
• thin-layer chromatography (TLC) plates (silica gel 60 F254, Merck)
• 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
370 Whole-cell Oxidations and Dehalogenations
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
12.5 Enantioselective Benzylic Microbial Hydroxylation of Indan and Tetralin 371
12.5.2.1 Materials and Equipment
• �-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
• anhydrous sodium sulfate (5 g)
• silica gel (400–200 mesh, Aldrich) (23 cm height in glass column)
• orbital shaker
• centrifuge
• 500 mL Kitasato
• Buchner funnel
• filter paper
• magnetic stirring plate
• 250 mL Separating funnel
• rotary evaporator
• equipment for continuous liquid–liquid extraction
• equipment for flash column chromatography using a glass column (2.5 cm� 25 cm)
• TLC plates (silica gel 60 F254, Merck)
• 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.
372 Whole-cell Oxidations and Dehalogenations
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
12.5 Enantioselective Benzylic Microbial Hydroxylation of Indan and Tetralin 373
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.
374 Whole-cell Oxidations and Dehalogenations
References and Notes
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.
12.5 Enantioselective Benzylic Microbial Hydroxylation of Indan and Tetralin 375
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
12.6.1.1 Materials and Equipment
• 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
376 Whole-cell Oxidations and Dehalogenations
• yeast extract (3 g)
• 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
12.6.2.1 Materials and Equipment
• Malt extract (30 g)
• glucose (10 g)
• peptone (10 g)
• yeast extract (30 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.
378 Whole-cell Oxidations and Dehalogenations
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
12.7.1.1 Materials and Equipment
• 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)
12.7 The Biocatalytic Synthesis of 4-Fluorocatechol from Fluorobenzene 379
• 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
12.7.2.1 Materials and Equipment
• Stock solution 1 (9 mL)
• deionized water (439 mL)
• stock solution 2 (450 mL)
• stock solution 3 (450 mL)
• stock solution 4 (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)
• stock solution 6 (250 mL)
380 Whole-cell Oxidations and Dehalogenations
• 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
12.7.3.1 Materials and Equipment
• 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 The Biocatalytic Synthesis of 4-Fluorocatechol from Fluorobenzene 381
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
12.7.4.1 Materials and Equipment
• 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.
382 Whole-cell Oxidations and Dehalogenations
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.
12.7 The Biocatalytic Synthesis of 4-Fluorocatechol from Fluorobenzene 383
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.
384 Whole-cell Oxidations and Dehalogenations
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)
12.8.1.1 Materials and Equipment
• 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)
12.8.2.1 Materials and Equipment
• 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.
386 Whole-cell Oxidations and Dehalogenations
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
12.8 Synthesis of Enantiopure (S)-Styrene Oxide by Selective Oxidation of Styrene 387
(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)
12.8.4.1 Materials and Equipment
• 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
388 Whole-cell Oxidations and Dehalogenations
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.
12.8 Synthesis of Enantiopure (S)-Styrene Oxide by Selective Oxidation of Styrene 389
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.
390 Whole-cell Oxidations and Dehalogenations
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
392 Whole-cell Oxidations and Dehalogenations
12.9.2.1 Material and Equipment
• S. platensis in SOT medium (100 mL)
• 2-bromoacetophenone (100 mg, 0.50 mmol)
• ethyl acetate
• ether
• hexane
• anhydrous sodium sulfate
• filter paper
• one 200 mL Erlenmeyer flask
• silica gel (Kieselgel 60 40–63 mm), 15 g
• fluorescent lamp
• shaker
• rotary evaporator.
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
+
12.9.3.1 Material and Equipment
• S. platensis in SOT medium (100 mL)
• 2,6-dibromo cyclohexanone (100 mg, 0.39 mmol)
• ethyl acetate
• ether
• hexane
• anhydrous sodium sulfate
12.9 Biotransformation of a-Bromo and a,a0-Dibromo Alkanone 393
• filter paper
• one 200 mL Erlenmeyer flask
• silica gel (Kieselgel 60 40–63 mm), 15 g
• fluorescent lamp
• shaker
• rotary evaporator.
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).
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.
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
+ +
12.9.4.1 Material and Equipment
• S. platensis in SOT medium (100 mL)
• 2,6-dibromo menthone (100 mg, 0.32 mmol)
• ethyl acetate
• ether
• hexane
• anhydrous sodium sulfate
• filter paper
394 Whole-cell Oxidations and Dehalogenations
• one 200 mL Erlenmeyer flask
• silica gel (Kieselgel 60 40–63 mm), 15 g
• fluorescent lamp
• shaker
• rotary evaporator.
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).
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.
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.
12.9 Biotransformation of a-Bromo and a,a0-Dibromo Alkanone 395
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.
396 Whole-cell Oxidations and Dehalogenations
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
Practical Methods for Biocatalysis and Biotransformations Edited by John Whittall and Peter Sutton
� 2009 John Wiley & Sons, Ltd
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