Biotechnology Vol 8b Bio Transformation II, 2nd Ed

8/6/2019 Biotechnology Vol 8b Bio Transformation II, 2nd Ed

http://slidepdf.com/reader/full/biotechnology-vol-8b-bio-transformation-ii-2nd-ed 1/537



BiotechnologySecond Edition

Volume 8bBiotransformations I1



BiotechnologySecond Edition

Fundamentals

Volume 1Biological Funda mentals

Volume 2

Genetic Fundamentals andGene tic Engineering

Volume 3Bioprocessing

Volume 4

Measuring, Modelling and C ontrol

Products

Volume 5aReco mbinan t Proteins, MonoclonalAntibodies and Therapeutic Genes

Volume 5bGenom ics and Bioinformatics

Volume 6Products of Primary Metabolism

Volume 7Products of Secondary Metabolism

Volumes 8a and bBiotransformations I and I1

Special Topics

Volume 9Enzymes, Biomass, Food and Feed

Volume 10

Special Processes

Volumes 11a -cEnvironmental Processes 1-111

Volume 12Legal, Economic and

Ethical Dimensions

All volumes are also displayed on ourBiotech Website:httpJ/m.wiley-vch.delbooks/biotech



A Multi-Volume Comprehensive Treatise

BiotechnologySecond, Completely Revised Edition

Edited byH.-J.Rehm and G. Reed

in cooperation with

A.Puhler and P.Stadler

Volume 8b

Biotransformations I1

Volume Editor:D. R. Kelly

Industrial Consultant:

J. Peters

@WILEY=VCH

Weinheim .New York .Chichester .Brisbane .Singapore .Toronto



Series Editors:

Prof. Dr. H.-J. Rehm

Institut fur Mikrobiologie

Universitat Miinster

CorrensstraBe 3D-48149 Miinster

FRG

Prof. Dr. A. Piihler

Biologie VI (Genet ik)

Universitat Bielefeld

P.O. Box 100131

D-33501 Bielefeld

FRG

Dr. G. Reed

1029N. Jackson St. #501-A

Milwaukee, WI 53202-3226

USA

Prof. Dr. P.J.W. Stadler

Artemis Pharmaceuticals

Geschaftsfiihrung

Pharmazentrum Koln

Neurather Ring 1D-51063 Koln

FRG

Volume Editor:

Dr. D. R. Kelly

Department of Chemistry

Cardiff University

PO.Box 912

Cardiff CF 13TB

UK

Industrial Consultant:

Dr.J. Peters

Bayer AG

Geschilftsbereich Pharma

TO Biotechnologie

D-42096 Wuppertal

FRG

This book was carefully produced. Nevertheless, authors, editors and publisher do not warrant the information

contained therein to be free of errors. Readers are advised to keep in mind that statements, data, illustrations,

procedural detailsor other items may inadvertently be inaccurate.

Libraryof Congress Card No.: applied for

British Library Cataloguing-in-PublicationData:A catalogue record for this book is available from the British Library

Die Deutsche Bibliothek-CIP-Einheitsaufnahme

A catalogue record for this bookis available from Der Deutschen Bibliothek

ISBN 3-527-28324-2

0 WILEY-VCH Verlag GmbH, D-69469 Weinheim (Federal Republic of Germany), 2000

Printed on acid-free and chlorine-free paper.

All rights reserved (including those of translation into other languages).No part of this book may be reproduced in any

form-by photoprinting,microfilm, or any other means-nor transmittedor translated into a machine language without

written permission from the publishers. Registered names, trademarks, etc. used in this book, even when not specifi-

cally marked as such, are not to be considered unprotected by law.

Composition and Printing: Zechner Datenservice und Druck, D-67346 Speyer.

Bookbinding:J. Schaffer, D-67269 Griinstadt.

Printed in the Federal Republic of Germany



Preface

In recognition of the enormous advances inbiotechnology in recent years, we are pleasedto present this Second Edition of “Biotech-

nology” relatively soon after the introductionof the First Edition of this multi-volume com-prehensive treatise. Since this series was ex-tremely well accepted by the scientific com-munity, we have maintained the overall goalof creating a number of volumes, each de-voted to a certain topic, which provide scien-tists in academia, industry, and public institu-tions with a well-balanced and comprehensiveoverview of this growing field. We have fullyrevised the Second Edition and expanded it

from ten to twelve volumes in order to takeall recent developments into account.These twelve volumes are organized into

three sections. The first four volumes consid-er the fundamentals of biotechnology frombiological, biochemical, molecular biological,and chemical engineering perspectives. Thenext four volumes are devoted to products ofindustrial relevance. Special attention is givenhere to products derived from genetically en-gineered microorganisms and mammaliancells. The last four volumes are dedicated to

the description of special topics.The new “Biotechnology” is a reference

work, a comprehensive description of thestate-of-the-art, and a guide to the originalliterature. It is specifically directed to micro-biologists, biochemists, molecular biologists,bioengineers, chemical engineers, and foodand pharmaceutical chemists working in indus-try, at universities or at public institutions.

A carefully selected and distinguishedScientific Advisory Board stands behind the

series. Its members come from key institu-tions representing scientific input from abouttwenty countries.

The volume editors and the authors of theindividual chapters have been chosen fortheir recognized expertise and their contribu-tions to the various fields of biotechnology.Their willingness to impart this knowledge totheir colleagues forms the basis of “Biotech-nology” and is gratefully acknowledged.Moreover, this work could not have beenbrought to fruition without the foresight andthe constant and diligent support of the pub-lisher. We are grateful to VCH for publishing

“Biotechnology” with their customary excel-lence. Special thanks are due to Dr. Hans-Joachim Kraus and Karin Dembowsky, with-out whose constant efforts the series couldnot be published. Finally, the editors wish tothank the members of the Scientific AdvisoryBoard for their encouragement, their helpfulsuggestions,and their constructive criticism.

H.-J. RehmG.ReedA. Piihler

P. Stadler



Scientific Advisory Board

Prof Dr. M. . BekerAugust Kirchenstein Institute of MicrobiologyLatvian Academ y of SciencesRiga, Latvia New Delhi, India

Prof Dr. T K.GhoseBiochemical Engineering Research C entreIndian Institute of Technology

Prof Dr. J . D. Bu’LockWeizm ann Microbial Chemistry LaboratoryDepartment of ChemistryUniversity of Ma nchester Jerusalem, IsraelManchester, UK

Prof Dr. I.GoldbergDe partm ent of Applied MicrobiologyThe Hebrew University

Prof Dr. C. .Cooney

Department of Chemical E ngineeringMassachusetts Institute of Technology Alim enta ireCambridge, MA, US A

Prof Dr. G. Goma

DCpartement d e G dnie Biochimique e t

Institut National des Sciences AppliquCesToulouse, France

Prof Dr. H.WDoelleDepartment of MicrobiologyUniversity of QueenslandSt. Lucia, Australia

Prof Dr. . DrewsF. Hoffmann-La Roche AGBasel, S witzerland

Sir D.A.HopwoodDepartment of GeneticsJohn Inn es InstituteNorwich, UK

Prof Dr. E.H. HouwinkOrganon International bvScientific Development Gr oupOss.The N etherlands

Prof Dr.A. iechterInstitut fur BiotechnologieEidgenossische Technische Hochschule BiotechnologyZurich, Switzerland Lehigh University

Prof Dr.A. .HumphreyCe nter for M olecular Bioscience and

Bethlehem, PA, USA



VIII Scientific Advi sory Board

Prof Dr. I. KarubeResearch C enter fo r Advanced Scienceand TechnologyUniversity of Tokyo

Tokyo, Japan

Pro$ Dr. M .A. LachanceDepartment of Plant SciencesUniversity of Western OntarioLondon, O ntario, Canada

Prof Dr. Y : Liu

China National Ce nter for BiotechnologyDevelopmentBeijing, China

Prof Dr. J. E MartinDepartment of MicrobiologyUniversity of Le6nL e h , Spain

Prof Dr. B. Mattiasson

Department of BiotechnologyChemical CenterUniversity of LundLund. Sweden

Prof Dr. M . RoehrInstitut fur Biochemische Technologieund MikrobiologieTechnische Universitat WienWien, A ustria

Prof Dr. K . SchiigerlInstitut fiir Technische Che mieUniversitat HannoverHannover, Germany

Pro$ Dr. €? SensiChair of Fermentation Chemistryand Industrial MicrobiologyLepetit Research CenterGeren zano, Italy

Prof Dr.Y H. TanInstitute of Molecular and Ce ll BiologyNational University of SingaporeSingapore

Prof Dr. D . ThomasLabora toire d e Technologie Enzym atiqueUniversitC de CompiegneCom pibgne, France

Prof Dr.WVerstraete

Laboratory of Microbial EcologyRijksuniversiteit Ge ntGe nt, Belgium

Pro$ Dr. E. L.WinnackerInstitut fur BiochemieUniversitat MiinchenMunchen, Germany

Prof Dr. H. SahmInstitut fur BiotechnologieForschungszentrum JulichJulich, German y



Dr. Paul BentleyColumbia UniversityChemistry DepartmentHavemeyer HallMail Box 31353000 Broadway (119th Street)New York, NY 10027USAChapter 12

Contributors

Prof. Dr. G. Michael BlackburnDepartment of ChemistryThe University of SheffieldSheffield,S3 7HFUKChapter 11

Prof. Dr. Uwe T. BornscheuerInstitut fur Technische Chemieund BiotechnologieErnst-Moritz-Amdt-UniversitatSoldtmannstrasse 16D-17487 GreifswaldGermanyChapter 6

Dr. Antony D. M. CurtisDepartment of ChemistryKeele UniversityKeele, Staffordshire,ST5 5BGUKChapter 1

Dr. Sabine FlitschDepartment of ChemistryUniversity of EdinburghWest Mains RoadEdinburgh EH9 355ScotlandChapter 5

Dr. Ian FotheringhamNSC Technologies601 East Kensington RoadMt. Prospect, IL 60056USAChapter 8

Dr. Arnaud GargonDepartment of ChemistryThe University of SheffieldSheffield, S3 7HFUKChapter 11

Dr. SimonA. JackmanBiological LaboratoryUniversity of KentCanterbury, Kent CT27NJUKChapter 3



XI1 Contributors

Dr. Simon JonesDepartment of ChemistryBedso n BuildingUniversity of New castle upon Tyne

Newcastle upon Tyne, N E1 7R UUKandCancer Research InstituteArizona State UniversityBox 871604Tempe, A Z 85287-2404USAChapter 4

Dr. David R. Kelly

Department of ChemistryCardiff UniversityP.O. Box 912Cardiff CF 13T BUKChapter 2

Dr. Jane McGregor-Jones6 C artmel AvenueMaghull, Merseyside L31 9BJUK

andCancer Rese arch InstituteArizona State UniversityBox 871604Tempe,AZ 85287-2404USA

Chapter 10

Dr. Jassem MahdiDepartment of ChemistryUniversity of Wales

PO. Box 912Cardiff, CF 13 TBUKChapter 2

Dr. Jiirgen RabenhorstHaarmann & Reimer Gmb HPostfach 1253D-37601 Holzminden

GermanyChapter 9

Dr. Gary K. Robinson

Biological LaboratoryUniversity of KentCanterbury, Kent CT2 7NJUKChapter 3

Dr. Michael SchedelBayer AGPH-TO BiotechnologieBiotechnikum Mikrobiologie

Geba ude 226Friedrich-Ebert-Strasse 217D-42096 WuppertalGermanyChapter 7

Dr. Jane StratfordBiological LaboratoryUniversity of Ken t

Canterbury, Kent CT2 7N JUKChapter 3

Dr. Gregory M. WattDepartment of ChemistryUniversity of EdinburghWest Mains Roa dEdinburgh E H 9 355ScotlandChapter 5



Introduction

DAVID.KELLYCardiff,UK

The two years since the publication of Vol-

ume 8a have been a period of consolidation for

the business community.As large chemical and

pharmaceutical companies have merged intoever larger companies with increasing focus on

their primary mission (the “best of breed” con-cept), small companies specializing in bio-

transformations have prospered, by providing

expertise on demand.The lack of familiarity of

most synthetic chemists with biotransforma-

tions, has in fact worked to the advantage of

these small companies. Increasing demands for

enantiomerically pure materials will providefurther fuel for this trend. Many medium sized,fine and speciality chemical manufacturers are

adding biotransformations to their repertoireof skills by acquisitions or internal develop-

ments. Much of the impetus in the speciality

area, is the demand for so called natural or

“nature identical” products.

On the other hand the spectre of genetic

modification (GM) has undoubtedly acted as a

brake on the development of many new prod-

ucts and processes. It remains to be seen what

the public and legislators will tolerate. At-tempts to stop the growing of GM crops and

banish foodstuffs from GM organisms fromthe human food chain, lie outside the ambit of

biotransformations. But regulation of these ac-

tivities may well have a “knock on effect”. For

example enzymes derived from GM organ-

isms, are used in the beverage industry for de-hazing and clarification.Will these be regulat-

ed? Hydrolytic enzymes from GM organisms

are used in laundry. Will non-foodstuff uses es-

cape regulation?

On the academic front, enantioselective re-

actions which were once the sole preserve of

enzymes are now being encroached upon by

abiotic catalysts.%o good examples are the

Baeyer-Villiger reaction and ester forma-

tiodhydrolysis. Conversely the first enzymesand catalytic antibodies capable of catalyzing

pericyclic processes such as the Diels-Alderreaction are now emerging.

Volume 8b commences with the cornerstone

of organic synthesis: carbon-carbon bond for-

mation. Although there is no enzymatic equiv-

alent of the Grignard or organolithium re-

agent, aldol and benzoin condensationscan be

performed with exquisite stereochemical con-

trol. These reactions are constrained by a limit-

ed range of anionic components, but are fairly

promiscuous with electrophiles.Lyases (Chap-ter 2 ) are involved in a galaxy of reactions,

which includes just about every prosthetic

groupkofactor known or unknown.In connec-

tion with the latter, it has been discovered in

Biotechnology Second, Com pletely Revised E ditionH.-J. Rehm and G.Reed

copyright OWILEY-VCH Verlag GmbH ,2000



2 Introduction

the last year, that histidine ammonia-lyasewhich was long thought to contain a catalytic-ally active dehydroalanine residue, actuallycontains an unprece dented 4-methylene-imid-

azol-5-one residue. This result throws doubt onthe presence of the de hydroalanine residue inother enzymes, hought to contain it. Ch apter 2covers carboxy-lyases (many of which are in-volved in carbon-carbon bond fo rmation), hy-dro-lyases (including the intricate biosynthesisof deoxysugars) and th e am monia-lyases manyof which are involved in the biosynthesis ofamino acids. Halocompounds (Chapter 3)have a reputation as un-natural compoundswhich are persistent and damaging to the envi-ronment or at best as exotic marine natural

products. This reputation is unfounded. Themajority of halocompounds in the environ-ment originate from eological and naturalsources.Th e fascinating topic of the biosynthe-sis of th e ten know n natural organofluorocom-pounds provides an insight into an elementlargely spumed by nature. The cleavage andformation of phosphate bonds (Chapter 4) isneglected in most biotransformations reviews.Yet for most phosphorylated and pyrophos-phorylated cofactors the m ajority of the bind-

ing energy to the enzyme originates from thislinkage. Whereas, the traditional organicchem ist utilizes halosubstituents as nucleofug-es, nature utilizes phospha tes and their deriva-tives, because they can be selectively activatedby carefully placed protona tion sites.

Short polypeptide, RNA, and DNA se-quences (<18residues) are now routinely syn-thesized using automated machinery. In con-trast, the synthesis of even a trisaccharide re-quires a dedicated team of specialists and al-though spectacular advances have been made,

a general approach to the con trolled synthesisof polysaccharides is still lacking. Simultane-ous con trol of stereo- an d regioselectivity haveconfounded conventional chemical approach-es. Enzymatic glycosidation (Ch apter 5) pro-vides a powerful approach to escape this im-passe.

The “Applications” superchap ter covers se-lected industrial scale processes (commis-sioned by Dr. JORGPETERS)nd the use of bi-otransforma tion products as intermediates fororganic synthesis. Chap ter 6 reviews general

aspects of industrial biotransformations and

banishes some myths about their limitations.The carbohydrate theme of chapter 5 is contin-ued in chapter 7, which describes the develop-ment of the microbial oxidation of amino-

sorbitol, in the synthesis of l-deoxynojirimy-cin, a key prec ursor of the anti-diabetes drugM iglitoP. This is an important example of asynthetic route, which is fast an d efficient, be-cause the biotransformation step removes theneed to employ protecting groups to enforceregioselectivity.

The majority of amino acids are manufac-tured by fermen tation, although there a re alsoexamples of amino acids produce d em ployingisolated enzymes (e.g., L-fert. eucine). Devel-opments in molecular biology and genome

manipulation have enab led biosynthetic path-ways to be optimized to previously unimagin-able levels (Ch apter 8).This area has been fur-ther promoted by the continuing demand forthe dipeptide artificial sweetener, Aspartam e.

Many of the im portant natural flavorings orfragrances are extracted in small amountsfrom exotic plants an d th e supply is affected byseasonal variations due to the weather, croppests, and even political factors. Althoughmany flavorings and fragrances can be synthe-

sized via chemical routes, the resulting prod -ucts cannot be m arketed using the term “ natu-ral”, since EU and US legislation has definedthe term “natural”: The product must be pro-duced from n atural starting materials applyingphysical, enzym atic, or m icrobial processes fortheir final production. These ar e ideal circum-stances for the developm ent of biotransforma-tion routes, described in C hapter 9. For exam-ple, potential precursors of vanillin are foundin numerous plants and even waste products.Biotransformations can convert these materi-

als with low o r even negative value (i.e., dispo-sal costs) into valuab le products.

Biotransform ations have yielded a wealth ofnew routes to novel and known compounds.These have provided m arvellous oppo rtunitiesfor synthetic organic chemists. Chapter 10

gives brief descriptions of an enormous rangeof biotransformations and describes applica-tions of the products in th e synthesis of naturalproducts.

The final superchapter in Volume 8b pro-vides a glimpse into two are as which are likely

to play a m ajor role in future developments in



Introduction 3

biotransformations. Catalytic antibodies(Chapter 11) are literally artificial enzymes.They can be engineered to catalyze reactionswhich have no counterparts in the natural

world. Sophisticated strategies for raising anti-bodies and screening techniques for identify-ing active catalysts, have already yielded cata-lysts of exquisite selectivity. Intrinsic activityand turnover continue to be limited, but in thebest cases they are only a few orders of magni-tude lower than conventional enzymes.

The final chapter deals with wholly synthet-ic enzymes.These may be constructed from thefamiliar amino acid building blocks of conven-tional enzymes and may even borrow designfeatures, but the structures are, nevertheless,

totally alien. For example, poly-leucine cata-lyzes the enantioselective epoxidation ofenones by hydrogen peroxide and cyclic di-peptides catalyze the enantioselective additionof cyanide to aldehydes. This area will surelybenefit from the ever continuing improve-ments in computer processing power and newsoftware for molecular modeling.

All predictions of the future have one thingin common. They are wrong! But this does notabsolve us from trying to discern avenues that

are ripe for development. A Third Edition ofthis volume is likely to contain a wider rangeof reactions (possibly including pericyclic andphotochemical processes), more engineered

pathways and more site directed mutagenesis.Gene transfer and heterologous expression isnow common place. Are there better recipientorganisms than the ubiquitous E. coli? En-zymes from thermophiles were touted as theuniversal panacea for industrial processes atelevated temperatures, but their early promisehas not been fulfilled.Why? Enzyme catalyzedreactions in organic solvents are highly desir-able for hydrophobic substrates, but are cur-rently thwarted by low activity. Is it possible todesign and synthesize an enzyme which is sol-

uble in organic solvents (e.g., a lipase mutant)?These proposals represent reasonable exten-sions of current technology. However, it is cer-tain that the most spectacular advances will bemade in areas, that cannot be predicted atpresent.

I am pleased to acknowledge Dr. JORGPE-TERS’ elpful comments on this Introduction.

Cardiff, March 2000 D. R. Kelly



Carbon-Carbon Bond Formation,

Addition, Elimination and

Substitution Reactions



1 Carbon-Carbon Bond FormationUsing Enzymes

ANTHONY. M. CURTISKeele,U K

1 Introduction 62 Aldolases 6

2.1 DHAP Dependent Aldolases 72.2 Pyruvate, Phosphoenolpyruvate (PEP) Dependent Aldolases 15

2.3 2-Deoxyribose-5-PhosphateAldolase (DERA) 17

2.4 Glycine Dependent Aldolases 19

3 Transaldolase and Transketolase 204 Pyruvate Decarboxylase 235 Enzymes from Terpenoid and Steroid Biosynthesis 28

6 Other Enzyme CatalyzedC-C Bond Forming Reactions 30

7 Conclusion 328 References 33


copyright OWILEY-VCH Verlag GmbH, 2000



6 1 Carbon-Carbon Bond Formation Using Enzymes

1 Introduction

If the formation of carbon to carbon (C-C)

bonds is undeniably fundamental to organicchemistry then the ability to induce asymme-try in the constructive process must be the un-deniable goal of the modern organic chemist.Research into chemical methods for enantio-selective synthesis has yielded many elegantmethods for asymmetric C-C bond formationbut the concepts of efficiency and atom econo-my seem to have been ignored in many cases.Enantioselective methods are, of course, ofparamount importance to the industrialchemist, not least due to ever-tightening reg-

ulations governing single enantiomer pharma-ceuticals.

Historically, reviews concerning the utilityof enzymes in organic synthesis paid only cur-sory respect to the potential of enzymes to cat-alyze the formation of C-C bonds. However,trends in chemistry are not uncommon andthere is certainly a trend for such reviews todevote an increasing amount of space to thesubject. Indeed, as contemporary microbiolog-ical methods enable the study of enzymes from

different species this exciting field of researchhas witnessed an exponential period of devel-opment, as reflected by the increasing numberof articles devoted to the synthesis of C-Cbonds (RACKER,961a; OGURA nd KOYAMA,1981; FINDEISnd WHITESIDES,984;BUTT ndROBERTS, 986; JONES,1986; DAVIESet al.,1989; TURNER, 989,1994; SERVI, 1990; CSUKand GLANZER, 991; DRUECKHAMMERt al.,1991; WONG t al., 1991,1995a; SANTANIELLOet al., 1992; LOOK t al., 1993;FANGnd WONG,1994; WONG and WHITESIDES,994; WALD-

HOFFMANN,996;FESSNERnd WALTER, 997;TAKAYAMAt al., 1997a, b). The aim of thischapter is to present an overview of this rapid-ly growing field of research using appropriateexamples from the literature which exemplifythe versatility of the particular enzymes dis-cussed. As the majority of discoveries in thisfield have taken place within the past twodecades, this survey will focus predominantlyon work which has appeared in the primary lit-erature during that period. Readers are direct-

ed to the review articles cited above for a his-

MA”, 1995; WARD, 1995; GIJSEN t al., 1996;

torical perspective of developments within thisfield.

Many C-C bond forming reactions fall with-in Enzyme Commission subclass 4.1, car-

bon-carbon lyases. Lyases are enzymes whichcleave or form bonds other than by hydrolysisor oxidation. Systematic names for lyases areconstructed from the template “substrategroup-lyase”. Groups are defined at the sub-subclass level, e.g., decarboxylases (4.1.1),aldehydes (4.1.2), and oxoacids (4.1.3). Thus,threonine aldolase (EC 4.1.2.5), which catalyz-es the aldol condensation of glycine with acet-aldehyde, is systematically named “threonineacetaldehyde-lyase”. Reactions which haveonly been demonstrated in the carbon<arbon

bond forming direction or for which this direc-tion is more important are termed synthases.However, such designations have not foundfavor and trivial names are common. Othercarbon-carbon bond forming reactions whichappear in the following discussion may befound in classes 2 (transferases), 5 (isomeras-es), and 6 (ligases).

The laboratory use of enzymes for C-Cbond formation has developed in quite distinctareas, often determined by the class of enzyme

used. In recent years, a large amount of infor-mation concerning the effective use of aldolas-es in asymmetric synthesis has appeared, per-haps surpassing all of the other enzymesystems capable of forming C-C bonds. In anattempt to provide a balanced chapter, this re-view will present developments in the use ofmany enzymes, particularly on their use in thesynthesis of both naturally occurring and un-natural bioactive products. Reference will bemade to specialist reports in each area wherethey are available.

2 Aldolases

There are numerous chemical methods forachieving enantioselective aldol additions cit-ed in the literature, some simple and somequite sophisticated. Catalytic processes are, ofcourse, preferable over all others, provided thecatalyst is cheap and readily available. Howev-

er, stereoselective aldol reactions are a normal



2 Aldolases 7

metabolic process for the vast majority of or-ganisms and, given the skillsof the m icrobiolo-gist, naturally occurring catalysts have now be -com e widely available for use in the laborato-

ry. Several excellent accounts of the use of al-dolases (the general name applied to thisgrou p of lyases and synthases) in organic syn-thesis detail the progress made in this area,notably the prolific contributions by thegroups of WHITESIDESnd WONG WONG ndWHITESIDES,994; GIJSE N t al., 1996; FESSNERand WALTER,997;TAKAYAMAt al., 1997a,b).

At the time of writing, over 30 aldolaseshave been identified and isolated, most ofwhich catalyze the reversible stereospecific ad-dition of a keton e to an aldehyde. These en-

zymes are further classified according to themechanism by which they operate. Type I aldol-ases, typically associated with animals andhigher plants, form an imine intermediate w iththe donor ketone in the enzyme active site.Type I1 aldolases, on th e other ha nd, require aZn 2+ cofactor to act as a Lewis acid in theenzyme active site; such aldolases predomin-ate in bacteria and fungi. Type I1 aldolasestend to be mo re stable than their type I coun-terparts.

Aldolases are also characterized by theirspecificity for pa rticular do nor substrates andthe prod ucts they deliver:

(1) D ihydroxyac etone phosphate(D HA P) de pendent lyases yieldketose 1-phosph ates upon reactionwith an aldehyde acceptor.

(2) Pyruvate dependant lyases andphosphoenolpyruvate (PE P) depen-den t synthases yield 3-deoxy-2-keto-acids.

(3) 2-Deoxyribose 5-phosphate aldolase(D ER A ) is an acetaldehyde depen-den t lyase w hich yields 2-deoxyalde-hydes.

substituted a-am ino acids.(4) Glycine depe nde nt aldolases yield

Several aldolases are commercially avail-able and have been widely used. Other aldo-lases have not yet been subjected to such rigor-ous synthetic investigations, though their in

vivo reactions m ay give some indication as to

their utility.

2.1 DHAP Dependent Aldolases

The aldol addition of a ketone don or and a naldehyde acceptor can potentially yield four

stereoisomeric products which would requirecostly chromatographic separations. Fortu-nately for the synthetic chemist, aldolases existwhich can deliver all of the fo ur possible ster-eoisom ers. This is illustrated by the asym met-ric aldol addition of dihydroxyacetone phos-phate (D HA P) to D-glyceraldehyde 3-phos-phate or L-lactaldehyde (Fig. 1). In vivo , D-

fructose-1,6-diphosphateFDP) aldolase (EC4.1.2.13), L-rhamnulose-1-phosphate (Rha 1-P) aldolase (EC 4.1.2.19), L-fuculose-l-phos-phate (Fuc 1-P) aldolase (EC 4.1.2.17), and

tagatose 1,6-diphosphate (TDP) aldolase (EC4.1.2.-) form produ cts which differ in the rela-tive conformation of the hydroxy substituentson C-3 and C-4; FDP aldolase gives the(3S,4R) stereochem istry of L-fructose, Rh a 1-Paldolase gives (3R,4S) as in L-rham nulose, Fuc1-P aldolase gives (3R,4R) as in L-fuculose, an dTDP aldolase gives the (3S74S)configurationof L-tagatose.

Each of the above aldolases has been isolat-ed from a variety of mam malian and m icrobial

sources. FD P type I aldolase from rabbit m us-cle (rabbit muscle aldolase, R A M A ) has foundextensive use in organic synthesis and is com-mercially available, as are FDP type I1 aldo-lase, Rh a 1-P type I1 aldolase, and F uc 1-P typeI1 aldolase. All four complementary aldolaseshave been cloned and overexpressed (GAR-CIA-JUNCEDAt al., 1995).

The four DHAP-dependent aldolases haveeach been the subject of synthetic investiga-tions. Each aldolase will accom mod ate a w iderange of aldehydic substrates; the substrate

specificity of each these enzymes has beenelaborated in detail elsewhere (WO NG andWHITESIDES,994). Aliphatic aldehydes, in-cluding those with a-substituents, and mono-saccharides are generally suitable acceptorsubstrates, while aromatic, sterically hind ered ,and a$-unsaturated aldehydes are not gener-ally substrates. A recent review notes that over100 aldehydes have been used as substrates forD H A P depend ent aldolases in the synthesis ofmonosaccharides and the se quential use of a naldolase a nd a n isomerase facilitates the syn-

thesis of hexoses from the ketose aldol prod-




0

FDP aldolase TDP aldolase

D-Giyceraldehyde

3-phosphate

OP

D-Fructose 1,6-diphosphate D-Tagatose 1,gdiphosphate

Fuc 1-P aldolase $ Rha 1-P ldolase

0

L-Lac taldehyde

L-Fuculose 1-phosphate L-Rhamnulose 1-phosphate

Dihydroxyacetone 1-phosphate (DHAP)

Fig. 1. Four stereo-complementaryDHAP dependent aldolases.

ucts (WONG nd WHITESIDES,994;TAKAYAMA

et al., 1997b). In contrast, few analogs ofDHAP, for example, (1) and (2) (Fig. 2), aresubstrates for FDP aldolase, all being very

H 4 O P 2

H& SP 3

Fig. 2. Dihydroxyacetone phosphate (3) (DHAP)

analogs.

much less active than the natural ketone sub-strate (BISCHOFBERGERt al., 1988;BEDNAR-

SKI et al., 1989). Recently, however, the l-thio-analog of DHAP (3) was found to be a sub-strate and has found use in carbohydrate syn-thesis (FESSNERnd SINERIUS,994; DUNCAN

and DRUECKHAMMER,995).The aldolases discussed thus far require

DHAP as the ketone donor but the prohibi-tive commercial cost of DHAP may make thesynthetic use of such DHAP dependent al-dolases rather unattractive. However, manychemical and enzymatic methods have beenreported for preparing DHAP in quantitiessufficient for synthetic use (WONGnd WHITE-SIDES, 1994;FESSNERnd WALTER, 1997).A re-cent chemical method is able to give multi-gram quantities of DHAP from dimer (4) in61% overall yield (Fig. 3) (JUNG et al., 1994)and an enzymatic process using glycerol phos-

phate oxidase gives almost quantitative yields



2 Aldolases 9

1 (PhO)2FCC1,py, 96%yield

2 H l , PtO2 , 2411

3 HzO, 65"C, Sh

66% yie ld for t w o s t e p s

DHAP

Fig. 3. Synthesis of dihydroxyacetone phosphate

(DHAP).

facilitates the synthesis of carbohydrates assingle enantiomers (vide supra). Both enan-tiomers of D- and L-fructose were obtainedfrom the dihydroxylation of an a,@-unsaturat-

ed aldehyde (not itself a substrate for aldol-ases), followed by reaction with DHAP catal-yzed by the appropriate aldolase (HENDERSONet al., 1994a).

A new route to deoxysugars has becomeavailable through the use of the 1-thio-analog

of highly pure DHAP (FESSNERnd SINERIUS,1994).

Of all the aldolases discussed thus far, FDPtype I aldolase (rabbit muscle aldolase,RAMA) has proved to be a most useful cata-lyst. However, the stereochemical outcome ofreactions catalyzed by RAMA strongly de-pends upon the reaction conditions employed.

For example, under kinetically controlledconditions D-glyceraldehyde 3-phosphate waspreferred by the enzyme over L-glyceralde-hyde 3-phosphate with a selectivity of 20:1(BEDNARSKIt al., 1989). A model for this se-lectivity has been proposed (LEES nd WHITE-SIDES, 993). Dynamic resolution is possible inthermodynamically controlled reactions whenthe product is able to form a six-memberedcyclic hemiketal. Compounds with the leastnumber of diaxial interactions predominate(Fig. 4) (DURRWACHTERnd WONG, 1988;

BEDNARSKIt al., 1989).FDP aldolase, more particularly RAMA,

may be immobilized, used in a dialysis bag, oras the free enzyme in solution. Many examplesof its use have appeared, notably in the synthe-sis of carbohydrates and analogs thereof(WONGand WHITESIDES,994; GIJSEN t al.,1996).The aldehyde substrate may be used as aracemic mixture, although the use of singleenantiomers avoids the painful separation ofdiastereomeric product mixtures. Sharplessasymmetric dihydroxylation when used in tan-

dem with an aldolase catalyzed condensation aldolase.

&OPO >97%

OH OH

S

6HR

i R R

OH

4

It

OP <3%

HO

Fig. 4. Resolution of racemic aldehydes using FD P



10 I Carbon-Carbon Bond Formation Using Enzymes

S t e p s

* V O H

H 8 c OH

0 FDP aldolase

I C O H R

Glycol-

6 1-Deoxy-D -x ylulos e

FDP a ldola se___tt eps H

= . R7 PS 6 H 8 OM e

9 a - M e thy l Do l i vopyr ano s i de

Fig. 5. Applications of the thio-analog of dihydroxyacetone phosphate (3) in synthesis.

of DHAP (3) (Fig. 5 ) (DUNCANnd DRUECK-HAMMER,996). FDP aldolase catalyzed thecondensation of (3) with glycolaldehyde togive 1-deoxy-1-thio-D-xylulose -phosphate(S), furthe r elaboration of which yielded l-de -oxy-D-xylulose (6). Acetal (7) also reactedwith (3) in the presence of F D P aldolase to de-liver the diol (8), an intermediate in the syn-thesis of D-olivose methyl g lycoside (9).

RA M A catalyzed the addition of D H A P to

the semialdehyde (10) to give the diol (ll),which is an interm ediate in the synthesis of 3-deoxy-D-arabino-heptulosonic acid (12) (Fig.6) (TURN ERnd W HITESIDES,989).

Analogs of so-called “higher sugars” sialicacid and 3-deoxy-~-manno-2-octu~osoniccid(D- KD O), e.g., com pounds (13) and (14), canbe prepared by employing RA M A to catalyzethe reaction of D H A P and hexoses or pentos-es, respectively (BED NAR SKIt al., 1986). High-er sugars have bee n synthesized from simplerstarting materials bu t w ith unexpected results

(KIMand LIM,1996). R A M A catalyzed reac-tion of the aldehyde (W)with D H A P followedby treatment with phosphatase, surprisinglygave a 62% yield of ketose (16) (Fig. 7). Thishas (3S,4S)-configurations at the newlyformed chiral centers whereas R A M A cataly-sis normally delivers products with (3S,4R)-configurations, although other workers havenoted this change in stereoselectivity (GIJS ENand WON G, 995b; GIJSEN t al., 1996).

RA M A accepts aldehyde substrates bearinga variety of polar functionalities. 3-Azido-2-hydroxypropanal, either as the racem ate o r in

optically pure form, was used as the substratein a R A M A catalyzed synthesis of deoxynoji-rimycin (17) from L-3-azido-2-hydroxypropa-

nal and deoxymannojirimycin (18) from D-3-

10 DHAP

RAMA 37% yield1H 0

M a 2 CAcyJ+opOH 11

1NaHB( 0Ac) j

2 6 N HCl3 T r a n s a m i n a t i o nIH HO

OH

1 2 3-Deoxy-D-arabinohep t u l os o n i c ac id

Fig. 6. The synthesis of ketoses from amino acidsynthons.



2 Aldolases 11

Po

QOPOlll. 13

H O

I 1 RAMA, DHAP

' OH 16

Fig. 7. RAMA catalyzed synthesis of carbohy-drates.

azido-2-hydroxypropanal (Fig. 8); the samestrategy was adopted by both EFFENBERGERand WONG BDERSONt al., 1988,1990;ZIEG-

and has recently found use in the preparationof iminocyclitols (MoR~s-VARASt al., 1996).

Analogously, 3-azido-2-hydroxybutanalwasused as starting material for the synthesis of

several iminoheptitols (19-21) (STRAUBt al.,1990).

The synthesis of aza-analogs of 1-deoxy-N-acetylglucosamine (22) and I-deoxy-N-acetyl-mannosamine (23) employed optically pure 3-

azido-2-acetamidopropanal, (S)- or (R)- (U) ,respectively, as the aldehyde substrate in aRAMA catalyzed reaction with DHAP (Fig.9) (PEDERSONt al., 1990; KAJIMOTO t al.,1991a). Reduction of aldol Droducts orior to

LER et a]., 1988;VON DER @TEN et al., 1989)

17 Deoxynojir imycin

O H 18 Deoxymannoj i r imycin

vH

6 H 21

Fig. 8. Retrosynthetic analysis of the use of azidoaldehydes in the R A M A catalyzed synthesis of aza-sugars.

formation of 6-deoxy azasugars (KAJIMOTOtal., 1991b).

NHAC S NHAc

-

NHAC s N H A C

( 3 - 2 4 1 RAMA,DHAP 2 2

2 Pase

3 H,, Pd/C

NHAc

Hg&J&-.,HAc -

( R ) - 2 4 23

Fig. 9. Azido aldehydes in the RAMA catalyzed

removal of the phosphate gioup lea& to the synthesisof azasugars.




1,4-Dideoxy-l,4-imino-~-arabitino125), apotent a-glucosidase inhibitor found in Arach-niodes sfandishii and Angylocalyx boutiquea-nus, and other polyhydroxylated pyrrolidines,

such as heterocycles (%a) and (26b) can beprepared using the RAMA catalyzed conden-sation of DHAP and 2-azidoaldehydes (Fig.10) (HUNG t al., 1991; LIU et al., 1991;KAJI-MOTO et al., 1991a;TAKAOKAt al., 1993).Ho-moaza sugars, e.g.,(27), may be prepared from3-azido-2,4-dihydroxyaldehydesy the samemethodology (HENDERSONt al., 1994b; HOLTet al., 1994).

Aldolase catalyzed reaction of thioalde-hydes with DHAP enables the preparation ofa range of thiosugars. 2-Thioglycolaldehyde(28) yielded enantiomerk 5-thio-~-fhreo-pen-

tulofuranoses (29) nd (30)when reacted withDHAP in the presence of FDP aldolase orRha 1-P aldolase, respectively (Fig. ll ) , whileRAMA catalyzed reaction of aldehyde (31)yielded thioketose (32),elaboration of whichgave l-deoxy-5-thio-~-glucopyranoseerace-tate (33) Fig. 12). Further thiopyranose com-pounds are obviously accessible by using FDPaldolase, Fuc 1-P aldolase, or Rha 1-P aldolase

OH

H OH

H

2 a 26b 27

Fig. 10. Azasugars.

OH 1 FDPaldolase 1 Rha 1-Pa ldo lase HoDHAP DHAP

29 2 2-Thioglycola ldehyde 30

Fig. 11. Aldolase catalyzed synthesis of thiofuranoketoses.

31 32StepsJ

33

Fig. 12. RA M A catalyzed synthesis of a 1-deoxythiopyranose. OAc



2 Aldolases 13

A O H 2

and enantiomers of (31) as required (EFFEN-

Cyclitols are interesting targets for enzym e

condensation of D H A P and chloroacetalde-hyde forms the basis of a g ene ral synthesis of arange of cyclitols and C-glycosides, for exam -ple, (34) and (35) (Fig. 13) (SCHMIDan dWHITESIDES,990; NICOTRAet al., 1993). Ni-

BERGER et al., 199 2a; CHOU e t a l., 1994 ).

catalyzed synthesis. The RAM A catalyzed0

DHAP 1RAMA, 20h, rt

trocyclitol(36)roxy-3-nitropropanalas preparednd D H A Py couplingatalyzed-hy- [~ ]by R A M A , followed by cyclization by an intra-

(CHOUet al., 1995). Similarly cyclopentene

- i(OEt)

molecular H enry re action in 51% overall yield i3H 0

I1 RAMA

2 Paw

DHAP I NC'

3 7Sweetpotato ac idaR=Pz lR = H phosphatase. pH 4 .8

Fig. 14. RAMA catalyzed aldol condensationin situ olefination.

with

(37a) was prepared using a one-pot strategythat required a R A M A catalyzed condensa-tion as the first step, followed by cy clization byHorner-Wadsworth-Emmons olefination atpH 6.1-6.8 (Fig. 14) (GIJSEN and WONG,1995a).

Unlikely as it may app ear, the use of aldolas-

es is not entirely restricted to carbohydratesynthesis. RAM A was used to install the

34 35 chirality requ ired for the synthesis of ( + ) - e m -brevicomin (38) Fig. 15), seem ingly a fav oredtarget for constructive biotransformations,from 5-oxohexenal (SCHULTZt al., 1990).TheC-9-C-16 fragment (39) of pentamycin hasbeen prepared from L-malic acid using R A M Ato catalyze the final elaboration; an alternativeroute starts from D-glucose, also employingRAMA (MATSUMOTOt al., 1993;SHIMAGAKIet al., 1993).The C-3-C-9 fragment of aspicilin

was similarly prepared in a stereoselective

--

HOlW

HO

aOHH

*q+0N

HO

_________________ - - - - - - - - - - - - - - - - -

36

% Ac

Fig. 13. The synthesis of carbocycles and C -glyco-

sides.




38 (+)-exeBrevicornin

39 40

Fig. 15. Synthetic targets prepared using RAMA

catalyzed aldol condensation.

manner using RAMA (CHENEVERTt al.,1997)and the synthesisof homo-C-nucleoside(40) also has a RAMA catalyzed condensation

as a key step (LIUand WONG, 992).FDP aldolase has been used to effect a bi-directional synthesis strategy in which a seriesof disaccharide mimics have been preparedfrom a,w-dialdehydes (EYRISCHnd FESSNER,

Hv1

.OH

H

Fig. 16. FD P aldolase catalyzed aldol condensationof a dialdehyde.

1995).For example, ozonolysis of cyclohexene(41) and subsequent enzymatic elaborationgave compound (42) (Fig. 16).The same “tan-dem aldolization” methodology was used to

convert cyclohexene (43) into disaccharide(44) Fig. 17) (FESSNERnd WALTER, 997).The strategy noted above has come under re-cent scrutiny: in the FDP aldolase catalyzeddesymmetrization of a series of dialdehydes inwhich the two carbonyl groups are remotefrom each other, only one aldehyde group re-acted with the enzyme and the stereochemis-try of the product did not appear to have beeninfluenced by any chirality present in the start-ing materials (KIMet al., 1997).

Rha 1-Paldolase, Fuc 1-Paldolase, and TDPaldolase have been used less widely than FDPaldolase, though this may well change now thateach enzyme has been overexpressed (GAR-C~A-JUNCEDAt al., 1995). Rha 1-P aldolaseand Fuc 1-P aldolase both accept a variety ofaldehydic substrates. The products from bothenzymes have the R-configuration at C-3 butthe stereochemistry at C-4can be less well de-fined (FESSNERt al., 1991).Both enzymes willeffect dynamic resolution of racemic 2-hydrox-yaldehydes (FESSNERt al., 1992).

Rha 1-P aldolase and Fuc 1-P aldolase havebeen used in the synthesis of a number of aza-sugars following analogous strategies to those

Fig. 17. FD P aldolase catalyzed aldol condensationof an azido dialdehyde, as a rou te t o aminosugars.



2 Aldolases 15

outlined above for FDP aldolase (LIU et al.,1991;KAJIMOTOt al., 1991b;LEES nd WHITE-SIDES,992). It should be noted that all of thesealdolases may be used in combination with

other enzymes to extend their synthetic utility.Aldolase catalyzed condensation and subse-quent reaction with a corresponding isomer-ase gives access to aldoses from ketose inter-mediates (FESSNER nd EYRISCH,992).

Recent investigations show that TDP aldo-lase also shows a low specificity for the alde-hyde substrate but in all cases examined a dia-stereomeric mixture of products resulted. Ofinterest here is the anomalous stereochemicaloutcome of the reactions; the major product ineach case had the (3S,4R) stereochemistry of

o-fructose and not the expected (3S,4S) ste-reochemistry of D-tagatose (compare this withanalogous observations made with FDP aldo-lase as noted above). Only D-glyceraldehyde,the natural substrate, delivered the expectedconfiguration (FESSNER nd EYRISCH, 992;EYRISCHt al., 1993). This lack of stereospeci-ficity has severely hindered the use of TDP al-dolase in organic synthesis.

2.2 Pyruvate, Phosphoenolpyruvate(PEP) Dependent Aldolases

These particular enzymes, while performingopposite biological functions and requiringdifferent substrates, can be examined togetheras they can both be used to synthesize keto-acids. However, only two aldolases from thisgroup will be discussed at length, the potentialof the other members not yet having beenrealized.

N-Acetylneuraminic lyase (NeuAc aldolase,pyruvate dependent, EC 4.1.3.3), also knownas sialic acid aldolase, and the correspondingsynthase (PEP dependent, EC 4.1.3.19) wouldboth appear to deliver the same aldol productsbut the latter enzyme has not yet been fullycharacterized (GIJSEN t al., 1996).A type I al-dolase, NeuAc aldolase catalyzes in vivo thereversible aldol condensation of N-acetyl-a-D-mannosamine (ManNAc, (45)) and pyruvateto yield the a-anomer of N-acetyl-5-amino-3,5-dideoxy-~-g~ycero-D-ga~acto-2-nonu~oson-

ic acid (NeuAc, (46))Fig. 18). The enzyme is

relatively stable and can be used in solution, indialysis tubing, or immobilized. It is specific forpyruvate and an excess of pyruvate is used tofavor the formation of aldol products (UCHIDA

et al., 1984,1985;BEDNARSKIt al., 1987; AWEet al., 1984,1985;AUGB nd GAUTHERON,987;KIMet al., 1988). The enzymes from Clostridiaand Escherichia coli are commercially avail-able.

A high conversion (>go%) of ManNAc(45) to NeuAc (46)s possible using the isolat-ed enzyme though the work-up can be tricky(Fig. 18) (KRAGL t al., 1991; LIN et al., 1992;FITZet al., 1995). A wide variety of aldehydesare tolerated by the enzyme; a recent reviewstates that over 60 aldoses are substrates pro-

viding certain structural motifs are displayed,though substrates containing less than fourcarbons are not tolerated (WONGand WHI-

1996). The stereochemical outcome of reac-tions catalyzed by NeuAc aldolase strongly de-pends upon substrate structure and seems to

be under thermodynamic control. Most sub-strates, like the natural substrate D-ManNAc(45), yield products with the S-configuration atthe new chiral center, arising from attack at the

si-face of the aldehyde substrate. Some sub-strates, however, such as L-mannose, L-xylose,and D-altrose, give the thermodynamicallymore stable products with the R-conforma-

TESIDES, 1994;FITZ t al., 1995; GIJSEN t al.,

Ho OH 46 NeuAc

Fig. 18. Ne uA c aldolase catalyzed aldol condensa-

tion of N-acetylmannosam ine with pyruvate.




tion, resulting from attack a t the re-face of thesubstrate (LINet al., 1992;FITZet al., 1995).

Many biologically relevant target com-pounds have been realized from NeuA c aldol-

ase catalyzed reactions, including a wide va-riety of NeuAc derivatives. For example, L-

NeuAc (47) (from L-ManNAc), D-KDO (48)(from D -arabinose), L-KD O (49) from L-arabi-

H&co2H ;&C02H

cN

O H OH47 L-NeuAc 48 DKDO

+ O2H H&

HO CO2HHOoH

OH49 L -KDO 50 L-KDN

Fig. 19. Higher sugars prepared using Ne uA c aldo-lase catalyzed aldol condensation.

nose and L-KDN (50) (from L-m annose) wereall synthesized using NeuA c aldolase (Fig. 19)(AuGE et al., 1990; LINet al., 1997).6-O-(N-Carbobenzyloxyglycinoyl)-N-acetyl

mannosamine (51) nderwent a NeuAc aldol-ase catalyzed transforma tion to yield the aldolproduct (52), from which the fluorescent sialicacid (53) was prep ared (Fig. 20) (FITZ andWONG, 994).Polyacry lamides with a -sialosideand poly-C-sialosides appendages have beensynthesized, which inhibit agg lutination by th einfluenza virus (SPALTENSTEINnd WHITE-SIDES,1991; NAGY nd BEDNA RSKI,991; SPE-VAK et al., 1993). Pyrrolidine (SS), derivedfrom the N euA c catalyzed reaction of protect-ed D-mannosamine (54) followed by catalytic

hydrogenation, was converted to 3-(hydroxy-methyl)-6-epicastanospermine 56) (Fig. 21)(ZH OU t al., 1993).

Again, despite performing distinctly differ-ent functions in vivo, 3-deoxy-~-manno-2-octulosonate lyase (2-keto-3-deoxyoctanoate

aldolase, K D O aldolase,EC 4.1.2.23) and 3-d e-oxy-~-manno-2-octulosonate -phosphate syn-thase (phospho-2-keto-3-deoxyoctanoateyn-thase, K D O 8-P aldolase) may be u sed to cata-lyze the form ation of the D-K DO skeleton; D-

K D O 8-P has been prepared on a preparativescale using K D O 8-P aldolase (BEDNARSK Ital., 1988). K D O aldolase isolated from E. coli

BocN

- HOlll.& O2HAcNH

NeuAc aldolase

0

51 oH A C 0 2 H Ho 52

1 HCI

2 Dansyl chloride,Me2N,@ lJ _.i a 2 C 0 3

HOlItl*& 0 2 H

</

AcN H

53 H O

Fig. 20. Ne uA c aldolase catalyzed aldol condensation in the synthesis of a fluorescent sugar.



2 Aldolases 17

OH

OH

55

OH

Steps

1Hs6

HO%- H

Fig. 21. NeuAc aldolase catalyzed aldol condensa-

tion of N-carbobenzyloxymannosamine (54) withpyruvate in the synthesis of 3-(hydroxymethyl)d-epicastanospermine (56).

will accept a number of different aldose sub-strates though the rate of reaction is muchslower than that for the natural substrateD-arabinose (GHALAMBORnd HEATH, 966).KDO aldolase from Aureobacterium barkeri,strain KDO-37-2, shows a much wider sub-

strate specificity. Substrates require the R-con-figuration at C-3 but the stereochemical re-quirements for C-2 are more relaxed, the S-configuration being preferred for kineticallycontrolled reactions and the R-configurationbeing favored under thermodynamic condi-tions. A number of carbohydrate derivativeswere prepared during the course of these stud-ies (SUGAI t al., 1993).

The 2-keto-4-hydroxyglutarate KHG) al-dolases from bovine liver and E. coli havebeen isolated, both enzymes having been

shown to accept a range of analogs of pyruvate

as the ketone donor (Rosso and ADAMS, 967;NISHIHARAnd DEKKER, 972; SCHOLTZndSCHUSTER,984; FLOYD t al., 1992).2-Keto-3-deoxy-6-phosphogluconateKD-

PG) aldolase from Pseudomonas fluorescenswill not accept simple aliphatic aldehydes assubstrates: rather, the only requirement ap-pears to be that there must be a polar substitu-ent a t C-2 or C-3. KDPG aldolase generates anew chiral center at C-4 with the S-configura-tion (ALLEN t al., 1992).

A number of other aldolases which are de-pendent on pyruvate or PEP have been isolat-ed but their substrate specificities have notbeen investigated in any depth (GIJSEN t al.,1996).

2.3 2-Deoxyribose-5-Phosphate

Aldolase (DERA)

2-Deoxyribose 5-phosphate aldolase (DE-RA, EC 4.1.2.4) is unique in that the donorspecies is an aldehyde. A type I aldolase, DE-RA catalyzes in vivo the reversible addition ofacetaldehyde and D-glyceraldehyde 3-phos-

phate to give 2-deoxyribose 5-phosphate. Theenzyme has been obtained from several differ-ent organisms but DERA from E. coli is avail-able in large quantities as a consequence ofcloning and overexpression (BARBAS t al.,1990; CHENet al., 1992; WONG t al., 1995b).DERA will also accept propanal, acetone, andfluoroacetone as donor moieties though theirrates of reaction are considerably slower thanthat of the natural donor. The enzyme will ac-cept a wide range of aldehyde acceptor sub-strates and is capable of effecting dynamic res-

olution of a racemic mixture of aldehydes,D-isomers reacting preferentially to L-isomers,and the new stereocenter created during reac-tion generally has the S-configuration. DERAhas been used to prepare a number of carbo-hydrate analogs from suitably substitutedaldehydes, examples being furanose (57),pyra-nose (58), thiosugar (59) , azasugar (a),ndsimple aldehyde (61) (Fig. 22).

Conveniently, when acetaldehyde is used asthe donor substrate the products from DERAcatalyzed reactions are themselves aldehydes

and are thus capable of acting as acceptor sub-




Fig. 22. Retrosynthetic analysisof the use of 2-de-

oxyribose 5-phosphate aldolase (DERA).

strates for further reaction (GIJSEN nd WONG,1994).As observed with a number of a-substi-tuted aldehydes, if the product from the firstaddition of acetaldehyde is unable to form a cy-clic hemiacetal then reaction with a secondmolecule of acetaldehyde proceeds to yield 2,4-

dideoxyhexoses, e.g., (62),cyclization of whichyields cyclic hemiacetal (63) thus preventingfurther addition (Fig. 23).The best substrate forsuch sequential additions appears to be succin-ic semialdehyde, the carboxylic acid groupmimicking the phosphate group of D-glyceral-dehyde 3-phosphate (WONG t al., 1995b).

In the same vein, DERA has also been usedin combination with FDP-aldolase (RAMA)to give a range of 5-deoxyketoses with threeaxial substituents (64) together with othercompounds which arise from thermodynamicequilibration (Fig. 24), and in combination

Fig. 23. 2-Deoxyribose 5-phosphate aldolase (DE-

RA) catalyzed acetaldehyde condensations.

Me-0 yH

1 D E W , RAMA

2 PaseIMe%o~ 64

OH OH

Fig. 24. One-pot multiple enzyme catalyzed syn-

thesis of sugars.



2 Aldolases 19

B 3 equiv.

1 DERA

I2 NeuAc aldolaseI y””’””

H&C02H OH 65

Fig. 25. Two step multiple enzyme catalyzed syn-thesis of sugars.

with NeuAc-aldolase to yield derivatives ofsialic acid (65) (55% yield) (Fig. 25) (GIJSENand WONG, 995b, c). In the latter reaction theconfiguration of C-4 is the opposite to thatnormally observed in NeuAc-aldolase cata-lyzed reactions and results from equilibratedcoupling.

2.4 Glycine Dependent Aldolases

Serine hydroxymethyltransferase (SHMT,serine aldolase) and threonine aldolase (bothEC 2.1.2.1) require pyridoxal5-phosphate andtetrahydrofolate to catalyze the reversibleaddition of glycine and an aldehyde to deliverP-hydroxy-a-amino acids. Despite their appar-ent synthetic potential, neither enzyme has re-ceived intensive investigative attention. Theterm “threonine aldolase” is also used to refer

to pyridoxal phosphate requiring enzymes(EC 4.1.2.5) which do not require tetrahydro-folate.In vivo SHMT catalyzes the reversible aldol

reaction of glycine with “formaldehyde” (inthe form of 5,lO-methylenetetrahydrofolate

and water) to give L-serine.SHMT can be iso-lated from numerous organisms and cloningexperiments have been successfully carriedout (FESSNERnd WALTER,997). SHMT willreact with other aldehyde substrates in the ab-sence of tetrahydrofolate, as demonstrated by

synthetic studies carried out using SHMT from

rabbit liver and pig liver (LOTZ t al., 1990;SAEED nd YOUNG,992).An excess of glycinewas required to favor formation of the newamino acids. L-Amino acids predominated but

long reaction times resulted in equilibration ofproduct stereochemistry, delivering the ther-modynamically more stable threo isomers.

Threonine aldolase from mammalian tissuescatalyzes in vivo the liberation of glycine andacetaldehyde from L-threonine, although L-

do-threonine appears to be a more active sub-strate for the enzyme (FESSNERnd WALTER,1997). The enzyme from Candida hurnicolashowed tolerance to a variety of aliphatic andaromatic aldehyde acceptors, although a,p-un-saturated aldehydes were not accepted (VASSI-LEV et al., 1995a).A large excess of glycine wasrequired to drive the preparative reactions andthe (2S,3S):(2S,3R) diastereoselectivity varieddepending upon the aldehyde used. L-Threo-nine aldolase was used in a synthesis of a-ami-no-/3-hydroxy-y-butyrolactone (66) (Fig. 26)(VASSILEVt al., 1995b). Kinetic and thermo-dynamic control has been applied to the reac-

L-Threon ne

aldolaseIlycine

1 H 2, Pd/C

2 HClIRR’H2.HCl 6

H

Fig. 26. The synthesis of a-amino-/3-hydroxy- -bu-

tyrolactone (66) using L-threonine aldolase.




L-Threonine aldolase15 minutes

Fig. 27. Kinetically controlled synthesis of amino

acids using L-threonine aldolase.

tions of L-threonine aldolase. Enzymic con-densation of ybenzyloxybutanal(67) (Fig. 27)and glycine for 15 min gave the (2S,3S)-aminoacid (68) s the major product (80% de, 18%yield), whereas over 15 h the thermodynami-cally more stable (2S,3R)-diastereoisomer pre-dominated (20% de, 70% yield) (SHIBATAt

al., 1996). Recent studies have focused on theuse of recombinant D- and L-threonine aldo-lases (KIMURAt al., 1997).

A threonine aldolase from Streptomycesamakusaensis has been purified and prelimi-nary synthetic investigations conducted, in-cluding the resolution of several racemicthreo-phenylserines to deliver enantiomerical-ly pure (2R,3S)-~-amino cids (HERBERT tal., 1993,1994).

3 Transaldolase and

Transketolase

Transaldolase (EC 2.2.1.2) and transketo-lase (EC 2.2.1.1) are enzymes found in boththe oxidative and reductive pentose phosphatepathway (RACKER, 961b, c; MOCALI t al.,1985) and an example of each enzyme ob-tained from bakers' yeast is commercially

available.

Transaldolase catalyzes in vivo the transferof a dihydroxyacetone unit between phos-phorylated metabolites; for example, theC-1-C-3 aldol unit is transferred from D-sedo-

heptulose 7-phosphate (69) to D-glyceralde-hyde 3-phosphate, delivering D-erythrose 4-phosphate (70) and D-fructose 6-phosphate(71) Fig. 28). However, despite its availabilityand potentially low acceptor substrate speci-ficity, transaldolase has not been fully exploit-ed in organic synthesis. This can probably beattributed to the fact that reactions catalyzedby transaldolase have the same stereochemicaloutcome as is obtained by using fructose 1,6-diphosphate (FDP) aldolase (RAMA).

Transaldolase has been used in a multien-zyme system for the conversion of starch to D-

fructose. The transaldolase catalyzed transferof dihydroxyacetone from D-fructose 6-phos-phate to D-glyceraldehyde enabled the finalstep of the synthesis, a step which could not besatisfactorily completed using a phosphatase(MORIDIANnd BENNER,992).

Transketolase, on the other hand, has seenrather more widespread use as a reagent in or-ganic synthesis. Transketolase catalyzes in vivo

the reversible transfer of a nucleophilic hy-

droxyacetyl unit from a ketose phosphate to

OH OH OH

69 DSedoheptulose 7-P D-Glyceralde-Iiyde 3-P

Transaldolase

- HJs+OP

OH OH OH

70 DErythrose 4-P 71 DFructose 6-P

Fig. 28. Transaldolase catalyzed transfer of a dihy-droxyacetone unit.



3 Transaldolase and Transketolase 21

an aldose phosphate, specifically the C-1-C-2ketol unit of D-xylulose 5-phosphate (72) to D-

ribose 5-phosphate (73),yielding D-sedohep-tulose 7-phosphate (69) and D-glyceraldehyde

3-phosphate (Fig. 29). D-Erythrose 4-phos-phate (70) also acts as a ketol donor, yieldingD-fructose 6-phosphate. The enzyme requiresboth thiamine pyrophosphate and Mg2' as co-factors.

P-Hydroxypyruvic acid may also be used asthe ketol donor for transketolase, though it istransferred to an aldose acceptor with an ac-tivity 4% of that of D-xylulose 5-phosphate(SRERE t al., 1958; BOLTE t al., 1987;HOBBSet al., 1993). The key advantage to using hy-droxypyruvic acid, however, is that the transfer

process is now irreversible, due to the elimina-tion of carbon dioxide subsequent to transferof the ketol unit.

Transketolases have been isolated from sev-eral different species; the enzymes used mostcommonly are obtained from bakers' yeast orfrom spinach leaves (BOLTE et al., 1987;SCHNEIDERt al., 1989; DEMUYNCKt al.,1990a, b; LINDQVISTt al., 1992), although re-cent reports have discussed the use of trans-ketolase from E. coli (HOBBS t al., 1993;HUM-PHREY

et al., 1995; MORRIS t al., 1996). The

7 3 DRibose 5-P 72 DXylulose 5-P

Transketolase

I t69 DSedoheptulose 7-P D-Glyceralde-

hyde 3- P

Fig. 29. Transketolase catalyzed transfer of a hy-

droxyacetyl unit.

substrate specificity of transketolase has beensurveyed and its synthetic potential comparedto the established chemistry of fructose 1,6-diphosphate (FDP) aldolase (RAMA) (DE-

MUYNCK et al., 1991; KOBORI t al., 1992; DAL-MAS and DEMUYNCK,993).A wide range ofaldehydes will act as ketol acceptors, the rateof reaction being increased if there is an oxy-gen atom a or P to the carbonyl group, and thenew chiral center formed always has the S-configuration. Transketolase can also be usedto effect a kinetic resolution of racemic 2-hy-droxyaldehydes as only the R-enantiomerreacts (EFFENBERGERt al., 1992b).

Recent investigations into the synthetic util-ity of transketolase from E. coli have yielded

promising results. The use of a geneticallymodified strain of E. coli to provide transketo-lase of sufficient purity for biotransformations,combined with an efficient synthesis of potas-sium hydroxypyruvate, enables the gram-scalesynthesis of trio1 (75) from aldehyde (74) in76% yield (Fig. 30) (MORRIS t al., 1996).

Several natural product syntheses employ-ing transketolase have been reported. Theglycosidase inhibitor 1,4-dideoxy-1,4-imino-~-arabinitol (25) was prepared by using the

transketolase catalyzed condensation of ra-cemic aldehyde (76) with lithium hydroxypyr-uvate and kinetic resolution as the key step.Condensation gave the azido compound (77)

in 71% effective yield, with further chemical

E coli transketolase

co2

i R

HE

75 5-OBenzyl-D-sylulose

Fig. 30. Transketolase catalyzed condensation of

P-hydroxypyruvate.




‘C -hH 76

Transke tolase

c o 2

H Y G H 25

Fig. 31. The synthesis of 1,4-dideoxy-1,4-imino-~-

arabitinol(25) using transketolase.

OH

TransketolassHaOzH

c o 2

StepsI38 (+)-exeBrevicomin

Fig. 32. Th e synthesis of a bark beetle pheromoneusing transketolase.

elaboration delivering the pyrrolidine deriva-

tive (25) n good overall yield (Fig. 31) (ZIEG-

(+)-exo-Brevicomin(38) epresents an ideal

target as it possesses the stereochemical motifobtained from a transketolase catalyzed trans-

formation. Taking advantage of the enzymatic

kinetic resolution, intermediate (79)was ob-

tained in 45% yield by reacting 2-hydroxybu-

tyraldehyde (78)with hydroxypyruvic acid in

the presence of transketolase and cofactors.

Ketone (79)may also be obtained from the

RAMA catalyzed condensation of propanal

and DHAP, followed by enzymic dephosphor-

ylation. Further chemical operations delivered

(+)-exo-brevicomin (38) (Fig. 32) (MYLES t

al., 1991).Yeast transketolase was used to prepare

trio1 (81) from 3-cyano-2-hydroxyaldehyde(80) s the key step in a synthesis of fagomine

(82) Fig. 33) (EFFENBERGERnd NULL,1992).

The same catalyst delivered 5-thio-D-threo-2-

pentulofuranose (29) rom the thiol-substitut-

ed aldehyde (31) (Fig. 34) (EFFENBERGERt

al., 1992a) which is also accessible from the

FDP aldolase condensation of DHAP and 2-thioglycolaldehyde (28) (Fig. 11) (EFFEN-

LER et al., 1988).

BERGER et al., 1992a;CHOU et al., 1994).

b C N 0

O H

Transketolase

c o 2

H2, Pd/CIH5h2 Fagomine

Fig. 33. The synthesis of fagomine (82) using trans-

ketolase.



4 Pyruvate Decarboxylase 23

O H

S

84 6-Deoxy-D-fructose

85 6-Deoxy-L-sorbose

Fig. 34. The synthesis of S-thio-~-thre0-2-pentulo- Fig. 35. Deoxysugars prepared using transketolase.furanose (29) using transketolase.

Transketolase catalyzed condensation of aracemic mixture of erythro- and threo-2d-di-hydroxybutanals with hydroxypyruvate wasspecific for the (2R)-stereoisomers, and yield-ed exclusively 6-deoxy-~-fructose 84) and 6-deoxy-L-sorbose (85) (Fig. 35) (HECQUETtal., 1994a, 1996). Similarly the triols (86) and(87) were prepared from the correspondingdithiane precursors. The corresponding alde-hydes are intermediates for the synthesis offagomine (82) and 1,4-dideoxy-l,44mino-~-arabitinol (25), respectively (Fig. 36) (HEC-QUET et al., 1994b).

4 Pyruvate Decarboxylase

The capability of bakers' yeast to catalyzethe acyloin condensation was first observedover 70 ago and has since been examined inseveral review articles (FUGANTI nd GRAS-

SELLI, 1985; DAVIES t al., 1989; SERVI, 990;CSUK nd GLANZER,991;WARD,1995;HOW-

In this classic biotransformation, a C-Cbond is formed by reaction of an aldehyde,e.g., benzaldehyde, and an acetaldehyde equiv-alent to yield an a-hydroxyketone (88);withbakers' yeast the acetaldehyde equivalentadds to the si-face of the aldehydic carbonylgroup (Fig. 37) (FUGANTIt al., 1984).The ket-01 product (88) is usually accompanied byvarying amounts of the corresponding diol

(89), derived from subsequent enzymatic re-

MA", 1996;FESSNERnd WALTER, 1997).

duction on the re-face of the carbonyl, and themajor by-product is benzyl alcohol. The reduc-tion step has been studied in some detail(CSUK nd GLANZER,991).

8 2 Fagomine

I,t

H

Fig. 36. The synthesis of fagomine (82) and 1,4-di-deoxy-1,4-imino-~-arabitinol25) from intermedi-

ates prepared using transketolase.




Sf-face

Bakers’ yeast(py ruv ate decarboxylase)

0 88 OH 89

Fig. 37. Pyruv ate decarboxylase m ediated acyloincondensation.

The acyloin conden sation is catalyzed by py-ruvate decarboxylase in the presence of thia-mine pyroph osphate an d M g*+. Decarboxyla-tion of pyruvic acid supplies the acetaldehydeequivalent and the condensation is highlyenantioselective. The caveat is that the bio-transform ation u sually gives low yields of thedesired conde nsation products. However, both

reactants and biocatalyst ar e cheap and the en -zymic condensation can deliver chiral com-pounds which may not be readily obtainedfrom the chiral pool.

The main reason why this particular bio-transformation continues to be of interest isthat the venerable condensation of benzalde-hyde with pyruvic acid to give ~-3-hydroxy-3-phenylpropan-2-one (phenylacetyl carbinol,PAC) (88) is used in the industrial synthesisof L-( - -ephedrine and D-pseudoephedrine(ROSE , 1961). In fact, the vast majority of

recent publications concerning the enzymaticacyloin condensa tion discuss the production ofPAC. Despite this, all these reports have indi-cations for the laboratory-scale p roduction ofa-hydroxyketones.

It has been repo rted that the judicious addi-tion of acetone to the reaction mixture sup-presses diol formation, by undergoing reduc-tion and inhibiting further reduction of PAC(SHIMAZU,950). M oreover, the use of carbo-hydrate-free yeast partially surpresses the for-mation of several by-products (GLA NZE R t

al., 1987).

During their investigations using isolatedyeast pyruvate decarboxylase, CROUTet al.(1991) observed that aldeh ydes actually inhib-ited the enzyme. Indeed, at a concentration of

0.05 mol L-’ benzaldehyde completely inhib-its the decarboxylation reaction, thus explain-ing why the large-scale production of PAC isdisfavored if the concentration of benzalde-hyde in the reactor is too high. Various poly-meric supports have been investigated for theimmobilization of yeast cells during PAC pro-duction (NIKOLOVAnd W ARD, 994). In a fed-batch process, immobilized Cundidu utilis ledto a PAC yield of 15.2 g L - l (SHINand Ro-GERS, 1995 ),while the use of isolated pyruvatedecarboxylase from C. utilis delivered an opti-

mum yield of 28.6 g L-l (SHIN nd ROGERS,1996). Most recently, the large-scale produc-tion of PAC has been optimized further t o givea “cleaner” process and higher yields of PAC(OLIVER t al., 1997).

An alternative enzyme, benzylformate de-carboxylase isolated from Pseudomonus puti-du AT CC 12633, has potential for use in theproduction of PAC and other a-hydroxyke-tones by condensing benzylformate with ace-taldehyde (WILCOCKSt al., 1992; WILCOC KS

and WA RD, 992). Benzylformate decarboxyl-ase from Acinerobucter calcoacericus will alsodeliver PAC from benzylformate an d acetalde-hyde in an optimized yield of 8.4 g L-’ (PRO-SEN and WARD, 1994).

The yeast catalyzed acyloin condensation isnot restricted to the use of benzaldehyde andpyruvic acid, although pyruvate decarboxylasedoes appear to show some substrate specific-ity. Variously substituted cinnamaldehydeshave been used extensively as substrates forpyruvate decarboxylase to provide chiral com-

pounds since the sem inal observation that diol(91a) may be isolated from a ferm enting mix-ture containing cinnamaldehyde (Wa) (Fu-GANTI and GRASSELLI,977). For example,aldehydes (9Oa-d) gave the correspondingdiols (91a-d) in 30%, 50%, 30%, and 10%yield, respectively (Fig. 38). Larger substitu-ents a o the carbonyl group inhibited the con-densation but a ddition of ace taldehyde to thereaction mixture was reported to dramaticallyincrease the yield of condensation products(FUG AN TI t al., 1979). Diols (93a, b) were

obtained in 15% and 10’30,respectively, from




&Q0

cR1=H,R2=Me

d R’ = Me, R2 = H

(pyruvatedecarboxylase)

Fig. 38. Bakers’ yeast mediated acyloin condensa-tion of cinnamaldehyde to give erythro-diols (91).

2- and 3-methyl-5-phenylpenta-2,4-dien-l-a1

(92a, b) (Fig. 39) (FUGANTInd GRASSELLI,1978).Recently, the condensation of cinnamal-dehyde to give (2S,3R)-5-phenylpent-4-en-2,3-diol in the presence of bakers’ yeast was op-timized using statistical methods to modify the

experimental design (EBERT t al., 1994).a-Ketoacids other than pyruvic acid canparticipate in the condensation with benzalde-hyde to yield the corresponding diols in>95% ee, though the range of such alternative

a R1 H, R2 = M e

b R1 = Me, R2= H

Bakers’ yeast(pyruvatedecarboxy lase)

’ R’ OH

Fig. 39. Bakers’ yeast mediated acyloin condensa-tion of 2- and 3-methyl-5-phenylpenta-2,4-dien-l-al

(92a, b).

donors appears extremely limited (FUGANTItal., 1988).

CROUTet al. provided firm experimentalevidence that pyruvate decarboxylase is in-

deed responsible for a-hydroxyketone forma-tion by carrying out a series of biotransforma-tions with the commercially available enzyme(CROUT t al., 1991; KRENet al., 1993). Pyru-vate decarboxylase was subsequently the sub-ject of a molecular modeling study, based uponthe published crystal structure of the enzymefrom Saccharomyces uvarum (DYDA et al.,1993;LOBELLnd CROW,1996).

Other species have been shown to have py-ruvate decarboxylase activity. The enzymefrom Zymomonm mobi l i s has been isolated

and purified; while by-products resulting fromreduction were suppressed, the purified en-zyme does not give higher yields than purifieddecarboxylases from other species (BRINGER-MEYERand SAHM,1988). Interestingly, thestereoselectivity of pyruvate decarboxylasefrom brewers’ yeast differed from that ob-

tained from Z . mobilis or from wheat germ inreactions with acetaldehyde (Fig. 40) (CHEN

ee 2 8-5 4%

a;-HYeast

ee 19-22%0vH

t

CO2H

Z. mobilisJ

ee 24-29% ee 524 1%

Fig. 40. Enantiocomplemen tary acyloin condensa-

tions.




OH and JORDAN,1984; CROUTt al., 1986; ABRA-

HAM an d STUMPF, 987; BORNEMANNt al.,1993).

4I, :A 5

OH Several acyclic unsaturated aldehydes under-

went a condensation reaction in the presenceOH of fermenting Mucor circinelloides CBS

39468 ; (2S,3R)-diols ( 94%) (Fig. 41) we re ob-

tained from t he corresponding aldehydes, pre-R & sumably the reduction products of initially

formed a-hydroxyketones (STUMPFnd KIES-

Saccharom yces fermentati an d S. delbrueckiipromoted the acyloin condensation betweenbenzaldehyd e and pyruvic acid, giving productyields which were higher than with other mi-

Fig. 41. Mucor circinelloides CBS 39468 mediated CroorganismS, bu t did no t offer any advantageacyloin condensation and reduction products. ov er Saccharomyces cerevisiae when chal-

OH

LICH, 1991).

UY96R &OH

97 D(-)-a l loMuscar ine 98 L-Amicetose 99 L-Mycarose 100 L-Olivomycose

H&OH @OH woHH2N HONH2 HO NH2

101 L-Acosamine 102 L-Daunosamine 103 L-Ristosamine 104 L-Vancosamine

-

& 107 (-)-Frontalin 38 (+)-exo-Brevicomin105 4-HexanolideOH

106 (3S,4S)-4-Methyl-3-heptanol

Fig. 42. Natural products synthesized using the biotransformationproducts of cinnamaldehydes.




& akers'yeastyruvate ~&H ~ W O H

111 112ecarboxylase1 1 0

I1 1 3 a-Tocopherol, Vitamin E

Fig. 43. Biocatalytic synthesis of a-tocopherol(113).

lenged with other aldehydes or oxoacids(CARDILLOt al., 1991). The yeast catalyzedacyloin condensation has been used as a keystep in many syntheses of natural products, theFUGANTIroup having eminently dem onstrat-ed that biotransformation of cinnamaldehydederivatives can provide useful chiral startingmaterials (FUGA NTI nd GRASSE LLI, 977,1985). The following are just som e of the tar-

gets that have b een prepared:D-( -

-d o -mu s -carine (97) (FRONZAt al., 1978), L-amicetose(98), L-mycarose (W), nd L-olivomycose(100) (FUGANTInd GRA SSELLI,978;FUGAN-TI e t al., 1979),N-acyl derivatives of aminohex-oses including L-acosamine (101), L-daunos-amine (102),L-ristosamine (103) an d L-vancos-amine (104) (FRO NZ A t al., 1980,1981,1982,1985, 1987; FUGANTIt al., 1983a), (+)- and(-)-4-hexanolide (105)(BERNARDIt al., 1980;FUGANTIt al., 1983b), (3S,4S)-4-methyl-3-heptanol (106), a pheromone from Scolytus

multisfriutus(FUGANTIt al., 1982), (-)-fron-talin (107) (FUGA NTI t al., 1983c), (+)-em-

brevicomin (38) BERNARDIt al., 1981;Fu-GANTI et al., 1983b), LTB, (108) (FUGANTItal., 1983d), and (+)- and (-)-rhodinose (109)(SER VI, 985) (Fig. 42).

Following an earlier synthesis, both the diol(111) and by-product alcohol (1l2), obtainedin 20% and 10% yield, respectively, from reac -tion of a-methyl-~-(2-furyl)acrolein110) infermenting yeast, were used in a synthesis of a-tocopherol (vitamin E, (113)) (Fig. 43) (Fu-

GANTI and GRASSELLI,979,1982).

The synthesis of solerone (5-0x0-4-hexano-lide, (116)) was achieved using pyruvate d ecar-boxylase from bakers' yeast in a key step.Ethyl 4-oxobutanoate (114),a novel substratefor the enzymic acyloin condensation, yieldedethyl 4-hydroxy-5-oxohexanoate 115),whichwas readily converted into solerone (116) byacid (Fig. 44) (HA RIN G t al., 1997).

0 1 1 4

Bakers' yeastpyr uva te

decarboxylaseIH

t

Fig. 44. Biocatalytic synthesis of solerone (116).




9%yield OH

= (R)-117

1 Pig liver farnesyl pyrophosphate synthetase2 Alkaline phosphatae

1 2 % yield OH

(S)-117

Steps

5 Enzymes from Terpenoid

and Steroid Biosynthesis

AL+ 119

Examples of the laboratory use of enzymesfrom the biosynthetic pathways which lead to

spite abundant microscale biosynthetic stud-ies. Pig liver farnesyl pyrophosphate synthet-

ase has been used to prepare both enantiom-ers of 4-methyldihomofarnesol (117). ( S ) -

(117)was used in the synthesis of juvenile hor-mone (118) (Fig. 45) (KOYAMAet al., 1985,1987).

Several enzymes from the biosynthesis ofsteroids are of use in organic synthesis. Pigliver 2,3-oxidosqualene cyclase has long beenknown to catalyze the cyclization of 2,3-oxi-dosqualene (119) and derivatives substitutedat the Al8-19 double bond, to yield the expect-ed steroidal derivatives, in this case lanosterol(120) Fig. 46) (DJERASSI,981;GOAD,981).

terpenes and steroids are relatively sparse, de- 0

2,3-oxidosqualenelanosterol cyclase

1 2 0 Lanosterol

Fig. 46. Polycyclization of 2,3-epoxy-squalene.



5 Enzym es from Terpenoid and Steroid Biosynthesis 29

Pig liver

2,3-osidosqualene

lanosterol cyclaset

121

a R1 = Me, R2 =

@

H

b R1 = Et, R2 = M e

c R1 Me , R2 = Et

124 R 3 = w 4= H

Fig. 47. Polycyclizationof 2,3-epoxy-squalene nalogs (121).

More recently, the enzyme has been used to

prepare steroids (l22b, c) or tricyclic com-

pounds (123) and (l24)rom the correspond-

ing oxidosqualene analogs (l21a-c) which

possess the unnatural Z-configuration at the

A18-19 double bond (Fig. 47) (HERINet al.,

1979; KRIEF t al., 1987).

Bakers’ yeast has been used as a source of

sterol cyclase, allowing the efficient cyclization

of squalene oxide derivatives to the expected

steroidal compounds. The cyclization is accom-

panied by a kinetic resolution and is curiouslypromoted by ultrasonication; whether ultra-

sound increases cyclase activity or simply dis-

tributes the substrate more efficiently is debat-

able (BUJONSt al., 1988; MEDINA nd KYLER,

1988; MEDINAt al., 1989;X A O and PRESTWICH,

1991).Interestingly, n the transformation of ra-

cemic compound (125) to optically pure steroid

(U),he vinyl group rearranges from C-8 to C-

14 in the same way as a hydrogen does in the

natural substrate (l21a) (Fig.48).

A comparison has been made between pig

liver 2,3-oxidosqualene cyclase and the ultra-

Bakers‘ yeast

2,3-oxidosqualene

lanosterol cyclase

Fig. 48. Epoxy-polyene cyclization.




‘ 12 2,3-Dihydrosqualene

Cyclase fromT. pyriformis

Fig. 49. Polyene cyclization.

sonicated bak ers’ yeast preparation in cataly-sis of th e cyclizations men tioned above (KR IEFet al., 1991).A protozoal cyclase from Tetrahy-menu pyriformis has been used to synthesizethe unnatural steroid euph-7-ene (128) rom2,3-dihydrosqualene (127) (Fig. 49) (ABE andROHMER,991).

Alder-ase”, could not be identified (LASCH AT,1996).In fact, it may readily be concluded thata D iels-Alder-ase does not exist, based uponthe difficulty encountered in isolating such an

enzyme. Several reports of biocatalytic Diels-Alder reactions have appeared in the litera-ture. For example, in the presence of bakers’yeast the reaction of maleic acid and cyclo-pentadiene in water gave exclusively the exo-

cycloadduct (129),whereas in the absence ofyeast the endo-cycloadduct (130)predominat-ed (Fig. 50) (RA M ARAOet al., 1990).Whetherthis effect may be attributed to a specific e n -zyme is, howev er, uncertain.

The long-awaited breakthrough came whenOIKAWA,CHIHARA,nd coworkers detected

enzym ic activity for a Diels-Alder reaction ina cell-free extract from Alternuria soluni, apathogenic fungus which causes early blightdisease in potatoes and tomatoes (OIKAWAtal., 1995; ICHIHARAnd OIKAWA,997). Fol-lowing on from previous work in which theyproposed that solanapyrones (131-134), phy-totoxins produced by A . solani, were biosyn-thesized via a [4+21 cycloaddition prosola-napyrone 111 (135)was treated with the cell-free extract a t 30°C for 10 min t o yield cyclized

products (131) nd (132)with an endo:ex0 ra-

6 Other Enzyme Catalyzed I QC-C Bond FormingReactions

t

129

It is well established that enzyme catalyzedpericyclic reactions fo rm p art of many biosyn-thetic pathways. The D iels-Alder reaction, un-doubtedly one of the most powerful synthetictransformations available to the organic chem -ist, has long been speculated as being a key re-action in the biosynthesis of many naturalproducts but an enzyme responsible for f a d -

itating such a reaction, a so-called “Diels-

4 0 2 H 30

COzH

fig. 50.tion.

Bakers’ yeast mediated Diels-Alder reac-



6 OtherEnzj!m eCatalyzed C-C Bond Forming Reactions 31

tio of 47:53 (Fig. 51). When compound (135)

was reacted in th e absence of the cell-free ex-tract, the uncatalyzed reaction delivered prod -ucts (131) and (132) with an endo:exo ratio of

97:3; a similar endo:exo ratio was obtainedwhen a den ature d cell-free extract was used asthe reaction m edium. Overall, the enzym e cat-alyzed Diels-Alder reaction proceeds with aconversion of 15% and an endo:exo ratio of13:87 (this translates to an ee of >92% in fa-vor of solanapyrone (132)). Such enzymic cy-cloadditions ar e particularly interesting from amechanistic point of view: usually, the endo cy-

cloadduct predom inates when Diels-Alder re-actions are carried out in the absence of ca ta-lyst or if chemical methods of catalysis are

used. It must be noted that, at this stage, nosingle enzyme was isolated t o which these re-sults could b e attributed.

Since the above report, the Japanese grouphas offered a second example of a bio-catalyzed Diels-Alder reaction (OIKAWA tal., 1997). Pyrone (136) was converted to mac-rophomic acid (137) by reaction with phos-phoenolpyruvate in a cell-free extract fromMacrophoma cornmelinae; the proposedmechanism involves an initial [4+21 cycload-dition, followed by loss of carbon dioxide to

131 R = C H O 132 R = CHO

133 R= CH iO H 134 R = CHzOH

?Me

O 0

135 Prosolanapyrone I11

Fig. 51. Intramolecular Diels-Alder reaction.

137 Macrophomic acid

Fig. 52. Diels-Alder reaction of phosphoenolpyru-

vate?

deliver the product (137) (Fig. 52). H owever,the implied involvement of a Diels-Alder-asein this reaction is not certain a s an alternativebiosynthetic route m ay be envisaged.

C-Cbond formation has been achievedusing certain enzymes from the biosynthesis

of a min o acids. A variety of P -substituteda-amino acid derivatives were prepared fromp-chloroalanine by using E. coli tryptophansynthase and tyrosine ph enol lyase as reactioncatalysts (NAGASAWAnd YAMADA,986).

Reactions using yeast have delivered somerathe r unexpected products. The cyclic hemi-acetal (139) was isolated from a co ndensationof cinnam aldehyde in the presence of ferm ent-ing yeast with added acetaldehyd e (BERTOLLI

et al., 1981). It is probably fo rmed by M ichaeladdition of the enolate of ketol (138) to cin-namaldehyde, followed by hemiacetal forma-tion (Fig. 53).

In ano ther ap parent M ichael addition, a$-unsaturated ketones and esters react with2,2,2-trifluoroethanol in the presence of fer-men ting bakers' yeast to give products in highenantiom eric excess, but unknown absolutestereochemistry. For example, ethyl acrylate(140) reacts to give lactone (142) in 47% yieldand 79% ee via the addition product (141)

(Fig. 54) (KITAZUMEnd ISHIKAWA,984).




Bakers’ yeastIFig. 53. Bakers’ yeast mediated Michael additionto cinnamaldehyde.

0 140

CF3CH20H Bakers‘ yeastIt

F3CG O 42

Fig.54. Bakers’ yeast mediated Michael addition.

A strange enzymatic alkylation reaction wasobserved during the reduction of 3-oxobuty-ronitrile (143)using fermenting bak ers’ yeastin ethanol (ITOH t al., 1989). The biotrans-formation did not deliver the expected 3-hy-droxybutyronitrile but gave, instead, the dia-stereomeric (3S)-2-ethyl-3-hydroxybutyroni-

triles (144) nd (145) n a syn:anti ratio of 2: 1

(Fig. 55) each of >99% ee. The alkylated ke to

Bakers’ yeast,rt, 48h

t

144 145

aCCNt 146

Fig. 55. An unusual bakers’ yeast mediated alkyla-tion.

nitrile intermediate (146) ould be ob tained ifthe reaction if was terminate d before it went tocompletion. A mechanism to account for theseobservations was not proposed but alkylationappears to precede the reductive process. O nepossibility is C-acetylation by acetyl-CoA, fol-lowed by reduction, elimination, and furtherreduction of the alkene bond.

7 Conclusion

C-C bond formation is fast becoming th e

sole subject of many review articles in the p ri-mary chemistry literature, a fact which bearswitness to th e steady increase in th e number ofreports concerning th e use of enzymes to per-form such a biotransformation. The majorityof organic chemists will thus recognize the useof, for example, bakers’ yeast as a laboratoryreagent but, unfortunately, do not seek to useit (likewise other enzymic systems), in th e be-lief that it is inconvenient and somewhatmessy. However, as noted above, even lowlybakers’ yeast can accomplish asymmetric C-C

bond formation reactions. While this present



8 References 33

work provides an account of the state of th eart, a healthy com bination of brea kthroughs inmodern microbiology and a growing aware-ness among chemists of the potential benefits

of using enzymes will give future reviewersmuch to write a bout.

8 References

ABE, ., ROHMER,M. (1991), Enzymatic cyclization

of 2,3-dihydrosqualene into euph-7-ene by a cell-

free system from the protozoan Tetrahymenapyriformis, J. Chem. SOC., Chem. Commun.,

ABRAHAM,.-R., STUMPF,. (1987), Enzymatic acy-

loin condensation of acyclic aldehydes, Z. Natur-forsch. 42C, 559-566.

ALLEN,S. T., HEINTZELMAN,. R., TOONE,E. J.(1992),Pyruvate aidolases as reagents for stereo-

specific aldol condensation, J. Org. Chem. 57,

426-427.

AuGE, C., GAUTHERON,. (1987), The use of an im-

mobilized aldolase in the 1st synthesis of a natu-

ral deaminated neuraminic acid, J . Chem. SOC.,Chem. Commun., 859-860.

AuGE, C., DAVID, ., GAUTHERON,. (19&4),Synthe-

sis with immobilized enzyme of the most impor-

tant sialic acid, Tetrahedron Lett. 25,4663-4664.

AuGE,C., DAVID,., GAUTHERON,., VEYRIBRES,.

(1985), Synthesis with an immobilized enzyme of

N-acetyl-9-0-acetylneuraminic acid, a sugar re-

ported as a component of embryonic and tumor

antigens, Tetrahedron Lett. 26,2439-2440.

AuGE, C., GAUTHERON,., DAVID,. (1990), Sialyl

aldolase in organic synthesis: from the trout egg

acid, 3-deoxy-D-g~ycero-D-ga~acto-2-nonu~osonic

acid (KDN), to branched-chain higher ketoses as

possible new chirons, Tetrahedron46,201-214.

BARBAS, . F. I., WANG,Y.-F., WONG,C.-H. (1990),

Deoxyribose- 5-phosphate aldolase as a synthetic

catalyst, . Am . Chem.SOC.1U,2013-2014.

BEDNARSKI,. D., WALDMANN,. J., WHITESIDES,.

M. (1986), Aldolase-catalyzed synthesis of com-

plex C-8 and C-9 monosaccharides, TetrahedronLett. 27,5807-5810.

BEDNARSKI,. D., CHENAULT,. K., SIMON,E. S.,

WHITESIDES,. M. (1987), Membrane-enclosed

enzymatic catalysis (MEEC) - a useful, practical

new method for the manipulation of enzymes in

organic synthesis, J . Am. Chem. SOC. 109,

1283-1 285.

BEDNARSKI,. D., CRANS, . C., DICOSIMO, ., SI-

MON, . S., STEIN,. D., WHITESIDES,. M. (1988),

902-903.

Synthesis of 3-deoxy-~-manno-2-octulosonate-8-

phosphate (KDO-8-P) from D-arabinose- gener-

ation of ~-arabinose-5-phosphate sing hexokin-

ase, Tetrahedron Lett. 29,427-430.

BEDNARSKI,. D., SIMON,E. .,BISCHOFBERGER,N.,

FESSNER, .-D., KIM,M.-J. et al. (1989), Rabbit

muscle aldolase as a catalyst in organic synthesis,

J . Am. Chem.SOC.111,627-635.

BERNARDI,., FUGANTI,., GRASSELLI,.,MARINO-

NI , G. (1980), Synthesisof the enantiomeric forms

of 4-hexanolide (y-caprolactone) from the opti-

cally active5-phenyl-4-pentene-2,3-diolrepared

from cinnamaldehyde and bakers’ yeast, Syn-thesis, 50-52.

BERNARDI,., FUGANTI,C., GRASSELLI,. 1981), On

the steric course of addition of Gngnard reagents

on to a,&dialkoxy erythro and threo chiral alde-

hydes. Synthesis of (+)- and ( - ) -ex0 and endo-brevicomin, Tetrahedron Lett. 22,40214024.

BERTOLLI,., FRONZA, ., FUGANTI,., GRASSELLI,

P., MOJORI, ., SPREAFICO,. (1981), On the cr-ket-

01s obtained from @unsaturated aromatic alde-

hydes and bakers’ yeast, Tetrahedron Lett. 22,

BISCHOFBERGER,., WALDMANN,., SAITO,T., Sr-MON, . s.,LEES,W. et a]. (1988), Synthesisof ana-

logs of 1,3-dihydroxyacetonephosphate and gly-

ceraldehyde-3-phosphate for use in studies of

fructose-1,6-diphosphate ldolase, J . Org. Chern.

BOLTE, .,DEMuYNcK,C.,AMAKI,H.1987), Utiliza-

tion of enzymes in organic chemistry: transketo-

lase catalyzed synthesis of ketoses, TetrahedronLett. 28,5525-5528.

BORNEMANN,. , CROUT,D. H. G., DALTON,H.,HUTCHINSON,. W., DEAN,G. et al. (1993), Ster-

eochemistry of the formation of lactaldehyde and

acetoin produced by the pyruvate decarboxylases

of yeast (Saccharomyces sp.) and Zymomonusmobilis - different Boltzmann distributions be-

tween bound forms of the electrophile, acetalde-

hyde in the enzymatic reactions, J . Chem. Soc.,Perkin Trans. , 309-311.

BRINGER-MEYER,., SAHM,H. (1988), Metabolic

shifts in Zymornonas mobilis in response to

growth conditions, FEMS Microbiol. Rev. 54,

BUJONS,.,GuAJARDo,R.,KYLER,K.S.1988),Enan-

tioselective enzymatic sterol synthesis by ultra-

sonically stimulated bakers’ yeast, J . Am . Chern.SOC.110,604-606.

B u n , S., ROBERTS,. M. (1986), Recent advances in

the use of enzyme-catalyzed reactions in organic

research - he synthesis of biologically-active nat-

ural products and analogs, Nut. Prod. Rep. 3,

CARDILLO,., SERVI, ., TINTI,C. (1991), Biotrans-

formation of unsaturated aldehydes by micro-

965-968.

53,3457-3465.

131-142.

489-503.



34 I Carbon-Carbon Bond Formation Using Enzym es

organisms with pyruvate decarboxylase activity,

Ap pl. Microbiol. Biotechnol. 36,300-303.

CHEN,G. C., JORDAN,. (1984), Brewers' yeast pyru-

vate decarboxylase produces acetoin from acetal-

dehyde - a novel tool to study the mechanism of

steps subsequent to carbon dioxide loss, Bio-chemistry 23,3576-3582.

CHEN, ., DUMAS, . P., WONG, .-H. (1992), Deoxy-

ribose-5-phosphate aldolase as a catalyst in asym-

metric aldol condensation,J . Am . Chem SOC.114,

CHENEVERT,., LAVOIE, ., DASSER, . (1997), Use

of aldolases in the synthesis of non-carbohydrate

natural products. Stereoselective synthesis of as-

picilin C-3-C-9 fragment, Can.J . Chem.75,68-73.

CHOU,W.-C., CHEN,L., FANG, .-M., WONG,C.-H.

(1994), A new route to deoxythiosugars based on

aldolases, . A m .Chem. SOC.116,6191-6194.

CHOU,W.-C., FOTSCH, ., WONG, .-H. (1995), Syn-

thesis of nitrocyclitols based on enzymatic aldol

reaction and intramolecular nitroaldol reaction,

J. Org. Chem. 60,2916-2917.

CROUT,D. H. G., LITTLECHILD,., MORREY,. M.

(1986), Acetoin metabolism - stereochemistry ofthe acetoin produced by the pyruvate decarboxyl-

ase of wheat-germ and by the a-acetolactate de-

carboxylase of Klebsiella aerogenes, J . Chem.SOC ., erkin Trans. I , 105-108.

CROUT,D. H. G., DALTON, ., HUTCHISON, . W.,

MIYAGOSHI,. (1991), Studies on pyruvate de-

carboxylase: acyloin formation from aliphatic, ar-

omatic and heterocyclic aldehydes,J. Chem. SOC.,

Perkin Trans. I , 1329-1334.

CSUK, ., GLANZER,. I. (1991), Bakers' yeast-medi-

ated transformations in organic chemistry,Chem.Rev. 91,49-97.

DALMAS, ., DEMUYNCK,. (1993), Study of the

specificity of a spinach transketolase on achiral

substrates, Tetrahedron: Asymmetry 4, 2383-

2388.

DAVIES, . G., GREEN,R. H., KELLY, . R., ROBERTS,

S. M. (1989), Biotransformations in PreparativeOrganic Chemistry, pp. 221-231. London: Aca-

demic Press.DEMUYNCK,., FISSON,., BENNANI-BAITI,., SAMA-

KI , H., MANI,.-C. (1990a), Immunoaffinity purifi-

cation of transketolases from yeast and spinach

leaves,Agric. Biol. C hem. 54,3073-3078.

DEMUYNCK,., BOLTE,., HECQUET,., SAMAKI, .

(1990b), Enzymes as reagents in organic chemis-

try - ransketolase-catalyzed synthesis of D-[1,2-

'3C,]xylulose, Carbohydr. Res. 206,7945.

DEMUYNCK,., BOLTE, ., HECQUET,., DALMAS,V.

(1991), Enzyme-catalyzed synthesis of carbohy-

drates: synthetic potential of transketolase, Tetra-hedron Lett. 32,5085-5088.

DJERASSI, . (1981), Recent studies in the marine

sterol field,Pure Ap pl. Chem. 53,873-890.

741-748.

DRUECKHAMMER,. G., HENNEN,W. J., PEDERSON,

R. L., BARBAS11,C. F., GAUTHERON,. M. et al.

(1991), Enzyme catalysis in synthetic carbohy-

drate chemistry,Synthesis,499-525.

DUNCAN,., DRUECKHAMMER,. G. (I995),A pseu-

doisomerization route to aldose sugars using al-

dolase catalysis,J. Org. Chem. 60,7394-7395.

DUNCAN,., DRUECKHAMMER,. G. (1996), Prepar-

ation of deoxy sugars via aldolase-catalyzed syn-

thesis of 1-deoxy-thioketoses, . Org. Chem. 61,

438-439.

DURRWACHTER,. R., WONG, .-H. (1988), Fructose

1,6-diphosphate aldolase catalyzed stereoselec-

tive synthesis of C-alkyl and N-containing sugars

- thermodynamically controlled C-C bond for-

mations,J. Org. Chem. 53,41754181.

M A , ., FUREY,W., SWAMINAWN,.,SAX,M.,FAR-RENKOPF, B., JORDAN,. (1993), Catalytic centers

in the thiamine diphosphate dependent enzyme

pyruvate decarboxylase at 2.4 8, esolution, Bio-chemistry 32,6165-6170.

EBERT, ., GARDOSSI,., GIANFERRARA,., LINDA,

P., MORANDINI,. (1994),A multivariate reexam-

ination of experimental condition effects on acy-

loin-type condensation mediated by Saccharo-myces cerevisiae, Biocatalysis 10,15-23.

EFFENBERGER,., NULL,V. (1992), Enzyme-cata-

lyzed reactions 13. A new, efficient synthesis of

fagomine,Liebigs Ann. Chem .,1211-1212.

EFFENBERGER,., STRAUB,., NULL,V. (1992), En-

zyme-catalyzed reactions. 14. Stereoselective syn-thesis of thiosugars from achiral starting com-

pounds by enzymes, Liebigs Ann. Chem.,

EFFENBERGER,., NuLL,V.,ZIEGLER,T.1992), Prep-

aration of optically pure L-2-hydroxyaldehydes

with yeast transketolase, Tetrahedron Lett. 33,

EYRISCH, ., FESSNER,W.-D. (1995), Disaccharide

mimetics by enzymatic tandem aldol reaction,

Angew. Chem. (Int. Edn. Engl.) 34,1639-1641.

EYRISCH,.,SINERIUS,., FESSNER, .-D. (1993),Fa-

cile enzymatic de novo synthesis and NMR spec-

troscopic characterization of D-tagatose 1,6-bis-

phosphate, Carbohydr. Res. 238,287-306.

FANG,.-M., WONG,C.-H. (1994), Enzymes in organ-

ic synthesis: alteration of reversible reactions to

irreversible processes, Synlett, 393402.

FESSNER, .-D., EYRISCH,. 1992), Enzymes in or-

ganic synthesis. 2. One-pot synthesis of tagatose

1,6-bisphosphate by diastereoselective aldol

addition, Angew. Chem. (Int. Edn. Engl.) 31,

56-58.

FESSNER, .-D., SINERIUS,. (1994), Enzymes in or-

ganic synthesis. 7. Synthesis of dihydroxyacetone

phosphate (and isosteric analogs) by enzymatic

oxidation - sugars from glycerol, Angew. Chem.(Int. Edn. Engl.) 33,209-212.

1297-1301.

5157-5160.



8 References 35

FESSNER, .-D., WALTER, . (1997), Enzymatic C-C

bond formation in asymmetric synthesis, Top.

Curr. Chem. 184,97-194.

FESSNER,W.-D., SINERIUS, ., SCHNEID ER,., DREY-ER , M., SCHULZ, . E. et al. (1991), Enzymes in or-

ganic synthesis. 1. Diastereoselective enzymatic

aldol additions -L-rhamnulose and L-fuculose1-phosphate aldolases from Escherichia coli, An-

gew. Chem. (Int. Edn. Engl.) 30,555558.

FESSNER, .-D., BADIA, ., EYRISCH, ., SCHNEIDER,A., SINERIUS, . (1992), Enzymes in organic syn-

thesis. 5. Enzymatic synthesis of rare ketose 1-phosphates, Tetrahedron Lett. 33,5231-5234.

FINDEIS, . A., WHITESIDES,. M . (1984), Enzymic

methods in organic synthesis, Ann. Rep. Med.

Chem. 19,263-272.

FITZ,W., WONG,C.-H. (1994), Combined use of sub-

tilisin and N-acetylneuraminic acid aldolase for

the synthesis of a fluorescent sialic acid, J . Org.

Chem. 59,8279-8280.

FITZ,W., SCHWARK,.-R., WONG,C.-H. (1999, Aldo-

tetroses and C(3)-modified aldohexoses as sub-

strates for N-acetylneuraminic acid aldolase - a

model for the explanation of the normal and the

inversed stereoselectivity, J . Org. Chem. 60,

FLOYD, . C . ,LIEBSTER, . H.,TURNER, . J. (1992),

A simple strategy for obtaining both enantiomers

from an aldolase reaction - preparation of L-4-

hydroxy-2-ketoglutarate and ~-4-hydroxy-2-

ketoglutarate, J . Chem. SOC., Perkin Trans. I ,

FRONZA,G., FUGANTI,C., GRASSELLI,. (1978),

Noncarbohydrate-based enantioselective synthe-

sis of D -( )-allO-muscarine, Tetrahedron Lett. 19,

FRONZA, ., FUGANTI,.,GRASSELLI,. (1980), Syn-

thesis of N-trifluoroacetyl-L-acosamine and L-

daunosamine, J . Chem. Soc., Chem. Commun.,

4 4 2 4 .

FRONZA, .,FUGANTI, .,GRASSELLI,., PEDRO CCHI-FANTONI, . (1981), Synthesis of the N-benzoyl

derivatives of L-arabino, L-xylo, and L-lyxo

(L-vancosamine) isomers of 2,3,6-trideoxy-3-C-

methyl-3-aminohexose from a non-carbohydrate

precursor, Tetrahedron Lett. 22,5073-5076.

FRON ZA, ., FUGANTI,., GRASSELLI,., MAJO RI, .,

PEDROCCHI-FANTONI,., SPREAFICO,. (1982),

Synthesis of enantiomerically pure forms of N-

acyl derivatives of C-methyl analogs of the ami-

nodeoxy sugar L-acosamine from noncarbohy-

drate precursors,J . Org. Chem. 47,3289-3296.

FRONZA , ., FUGANTI,.,GRASSELLI,.,PEDROCCHI-FANTONI, . (1985), Carbohydrate-like chiral syn-

thons. Synthesis of the N-trifluoroacetyl deriva-

tives of 4-amino-2,4,6-trideoxy-~-lyxo,-amino-

2,4,6-trideoxy-~-arabino,and 4-amino-2,4,6-tri-

deoxy-L-ribo-hexose from the (2S,3R) 2,3-diol

3663-3670.

1085-1 086.

3941-3942.

formed from cinnamaldehyde in fermenting

bakers’ yeast, Carbohydr. Rex 136,115-124.

FRONZA,G., FUGANTI,C., PEDROCCHI-FANTONI,.,SERVI,S. (1987), Stereochemistry and synthetic

applications of the products of yeast reduction

of 3-hydroxy-3-methyl-5-phenylpent-4-en-2-one,

J . Org. Chem. 52,1141-1144.

FUGANTI, ., GRASSELLI,. (1977), Transformations

of non-conventional substrates by fermenting

bakers’ yeast: production of optically active

methyl-diols from aldehydes, Chem. Ind. (Lon-

don), 983.

FUGANTI,C., GRASSELLI,. (1978), Stereospecific

synthesis from non-carbohydrate precursors of

the deoxy- and methyl-branched deoxy-sugars

L-amicetose, L-mycarose and L-olivomycose, J.

Chem. Soc., Chem. Commun., 299-300.

FUGANTI, ., GRASSELLI,. (1979), Efficient stereo-

selective synthesis of natural a-tocopherol (vi-

tamin E), J . Chem. SOC.,Chem. Commun.,

FUGANTI, ., GRASSELLI,. (1982), Synthesis of the

C , , chromanyl moiety of natural a-tocopherol

(vitamin E), J . Chem. SOC., Chem. Commun.,

205-206.

FUGANTI, ., GRASSELLI,. (1985), Stereochemistry

and synthetic applications of products of fer-

mentation of a,punsaturated aromatic alde-

hydes by bakers’ yeast, in: Enzymes in Organic

Synthesis, Ciba Foundation Symposium III (POR-TER, R., CLARK, ., Eds.), pp. 112-127. London:

Pitman.

FUGANTI,C., GRASSELLI,., MARINON I, . (1979),

Further studies on the transformations of unsatu-

rated aldehydes by fermenting bakers’ yeast: a fa-

cile synthesis of L-olivomycose,Tetrahedron Lett.

FUGANTI, . , GRASSELLI,., SERVI,S. , ZIROTTI, .

(1982), Synthesis of the enantiomeric forms of cis

and trans l-benzyloxy-2,3-epoxybutane nd of

(3S,4S) 4-methyl-3-heptano1, Tetrahedron Lett.

23,4269-4272.

FUGANTI, ., GRASSELLI,., PEDRO CCHI-FAN TONI,.(1983a), Stereospecific synthesis of N-benzoyl-L-

daunosamine and L-ristosamine,J. Org. Chem. 48,

FUGANTI,C., GRASSELLI,., PEDROCCHI-FANTONI,G., SERVI, ., ZIROTI-I,. (1983b), Carbohydrate-

like chiral synthons.Preparation of (R) y-hexano-

lide, (5R,6S,7S) 6,7-isopropylidendioxy-&octa-

nolide and (+)-exo-brevicomin from (2S,3S,4R)

2,3-isopropylidendioxy-4-benzyloxyhept-6-ene),

Tetrahedron Lett. 24,3753-3756.

FUGANTI, ., GRASSELLI,., SERVI, . (1983c), Syn-

thesis of (-)-frontalin from the (2S,3R)-diol pre-

pared from a-methylcinnamaldehyde and fer-

menting bakers’ yeast, J . Chem. SOC., Perkin

Trans. I,241-244.

995-996.

20,1161-1164.

909-910.



36 I Carbon-Carbon Bond F ormation Using Enzym es

FUGANTI,., SERVI,S.,ZIRO TTI, . (1983d), Non-car-bohydrate based synthesis of natural LIB 4, Tetra-hedron Lett. 24,5285-5288.

FUGANTI,., GRA SSELLI,., SERVI, ., SPREAFICO,.,ZIROT TI, . (1984), On the steric course of bakers’yeast reduction of a-hydroxyketones, J. Org.Chem. 49,40874089.

FUGANTI,., GRA SSELLI,., POLI,G., SER VI, ., ZOR -ZELLA, . (1988), Decarboxylative incorporationof a-oxybutyrate and a-oxovalerate into (R)-a-hydroxyethyl ketones and norm al-propyl ketoneson reaction with aromatic and a#-unsaturatedaldehydes in bakers’ yeast, J. Chem. Soc., Chem .Commun.,1619-1621.

GARCIA-JUNCEDA,., SHEN,G.-J., SUGA I, ., WONG,C.-H. (1995),A new strategy for the cloning, over-expression and one-step purification of 3 DHAP-dependant aldolases - rhamnulose 1-phosphatealdolase, fuculose 1-phosphate aldolase and taga-tose 1,6-diphosphate aldolase, Bioorg. Med.Chem. 3,945-953.

GHA LAMB OR, . A., HEATH, E. C. (1966), Thebiosyn thesis of cell wall lipopolysaccharide inEscherichia coli.V. urification and properties of3-deoxy-~-manno-octu~osonateldolase, J . Biol.Chem. 241,3222-3227.

GIJSEN, . J. M., WO NG , .-H. (1994), Unprecedent-ed asymmetric aldol reactions with three alde-hyde substrates catalyzed by 2-deoxyribose-5-phosphate aldolase, J. Am. Chem. SOC. 116,8422-8423.

GIJSEN, . J. M., WO NG , .-H. (1995a), Synthesis of acyclitol via a tandem enzymatic aldol - intra-molecular Homer-Wadsw orth-Emmons reaction,Tetrahedron Lett. 36,7057-7060.

GIJSEN,H. J. M., WONG ,C.-H. (1995b), Sequentialone-pot aldol reactions catalyzed by 2-deoxyri-bose-5-phosphate aldolase and fructose-1,6-diphosphate aldolase, J. Am. Chem. SOC. 117,

GIJSEN,H . J. M., WONG,C.-H. (1995c), Sequentialthree- and four-substrate aldol reactions cata-lyzed by aldolases, J. Am. Chem. SOC. 117,

GIJSEN,H. J. M., QIAO,L., FITZ, W., WO NG ,C.-H.(1996), Recent advances in the chem oenzymaticsynthesis of carbohydrates and carbohydrate mi-metics, Chem.Rev.%,443-473.

GLANZER, . I., FABER,K., GRIENGL, . (1987),Enantioselective hydrolyses by bakers’ yeast. 3.Micro bial resolution of alkynyl esters using lyo-philized yeast, Tetrahedron43,5791-5796.

GOAD,L. J. (1981), Sterol biosynthesis and metab-olism in marine invertebrates, Pure Appl. Chem.53,837-852.

HAR ING, ., SCHREIER,., HER DER ICH, . (1997),Rationalizing the origin of solerone (5-0x0-4-hexanolide): biomimetic synthesis and identifica-

2947-2948.

7585-7591.

tion of key metabolites in sherry wine, J . Agric.Food Chem. 45,369-372.

HECQ UET, ., BOLTE,J., DEMUYNCK,. (1994a),Chemoenzymatic synthesis of 6-deo xy-~ -fructos e

and 6-d eoxy-~ -sorbose sing transketolase,Tetra-

hedron 50,867743684.HECQUET,., LEMA IRE, .,BoLTE, ., DEM UY NC K,.

(1994b), Chemo-enzym atic synthesis of precur-sors of fagomine and 1,4-dideoxy-1,4-imino-~-

arabitinol, Tetrahedron Lett. 35,8791-8794.HECQUET,.,BoLTE, .,DEMUYNCK,. (1996), Enzy-

matic synthesis of “natural-labeled” 6-deoxy-~-sorbose precursor of an important food flavor,Tetrahedron52,8223-8232.

HENDERSON,., SHARPLESS,. B., WONG,C.-H.(1994a), Synthesis of carbohydrates via tandemuse of the osmium-catalyzed asymmetric dihy-droxylation and enzyme-catalyzed aldol additionreactions,J. Am. Chem. SOC.116,558-561.

HENDERSON,., LASLO,K., WONG,C.-H. (1994b),Chemoenzymatic synthesis of homoazasugars,Tetrahedron Lett. 35,359-362.

HERB ERT, . B., WILKINSON,., ELLAMES,. J., Ku-NEC, E. K. (1993), Stereospecific ysis of a range ofP-hyd roxy-a-amino acids catalyzed by a novel al-dolase from Streptomyces amakusaensis,J. Chem.SOC.,Chem. Commun., 205-206.

HERBERT,R. B., WILKINSON,., ELL AM ES, . J.

(1994), Preparation of (2R, 3 s) P-hydroxy-a-ami-no acids by use of a novel Streptomyces aldolaseas a resolving agent for racemic material, Can. J.Chem. 72,114-117.

HERIN,M., SANDRA, KRIEF,A. (1979), Stereospe-cific enzym e cyclization of a synthetic 2,3-oxido-squalene analog bearing an 182 carbon<arbondouble bond, Tetrahedron Lett. 33,3103-3106.

HO BBS , . R., LILLY,M. D., TURNER,. J., WAR D,.M., WILLETS,A. J., WOODLEY,. M. (1993), En-zyme-catalyzed carbon-carbon bond formation:use of transketolase from Escherichia coli, J.Chem. SOC.,Perkin Trans. I , 165-166.

HOWMANN,. 1996), Bakers’ yeast - a live reagentfor synthesis in organic chemistry,Chemie in Un-serer Zeit

30,201-203.HOLT,K. E., LEE PER ,E., HANDA,. (1994), Synthe-sis of P-1-homono rjirimycin and P-l-hom oma n-norjirimycin using the enzyme aldolase, J. Chem.SOC.,Perkin Trans.I , 231-234.

HUMPHREY,. J.,TURN ER, . J., MCCAGU E, .,TAY-LOR, S.J. C. (1995), Synthesis of en antiom ericallypure a-hydroxyaldehydes from the correspond-ing a-hydroxycarboxylic acids: novel substratesfor Escherichia coli transketolase, J. Chem. Soc.,Chem. Commun., 2475-2476.

HUNG,R. R., STRAUB,. A., WHITESIDES,. M.(1991), a-Aminoaldehyde equivalents as sub-strates for rabbit muscle aldolase - synthesis of1,4-dideoxy-~-arabitinol and 2(R),S(R)-bis(hy-



8 References 37

droxymethyl)-3(R),4(R)-dihydroxypyrrolidine, J .

Org. Chem. 56,3849-3855.ICHIHARA,., OIKAWA,. (1997), Biosynthes is of

phytotoxins from Alternaria solani, Biosci. Bio-tech. Biochem. 61,12-18.

ITOH,T., TAKAGI,., FUJISAWA,. (1989), A novelcarbon-carbon bond formation in the course ofbakers’ yeast reduction of cyanoacetone, Tetrahe-dron Lett. 30,3811-3812.

JONES,. B. (1986), Enzymes in organic synthesis,Tetrahedron42,3351-3403.

JUNG, .-H., JEONG,.-H., MILLER,., WONG,C.-H.

(1994), An efficient m ultigram-scale preparationof dihydroxyacetone phosphate, J. Org. Chem. 59,

KAJIMOTO,., CHEN, ., LIU,K. K.-C., WONG,C.-H.

(1991a), Palladium-mediated stereocontrolled re-ductive amination of azido sugars prepared fromenzymatic aldol condensation - a general ap-proach to the synthesis of deoxy azasugars,J.A m .Chem. SOC.113,6678-6680.

KAJIMOTO,T.,IU,K. K.-C., PEDERSON,. L., ZHONG,

Z., ICHIKAWA,. et al. (1991b), Enzyme-catalyzedaldol condensation for asymmetric synthesis ofazasugars -synthesis, evaluation, and m odeling ofglycosidase inhibitors, J . Am. Chem. SOC. 113,618741196.

KIM,M.-J., LIM, .T. (1996), Syn thesisof unsaturatedC8-C9 sugars by enzymatic chain elongation,Synlett, 138-140.

KIM,M.-J., HENNEN,. J., SWEERS,. M., WONG, .-

H. (1988), Enzym es in carbohyd rate synthesis -N-acetylneuraminic acid aldolase catalyzed reac-tions and preparation of N-acetyl-2-deoxy-~-neu-raminic acid-derivatives,J. Am . Chem. SOC. 110,6481-6486.

KIM,M.-J., LIM, .T., KIM,H.-J., WONG, .-H. (1997),Enzymatic single aldol reactions of remote dial-dehydes, Tetrahedron:Asymmetry 8,1507-1509.

KIMURA,., VASSILEV,. €?,HEN, .-J., WONG, .-H.

(1997), Enzyma tic synthesis of P-hydroxy-a-amino acids based on recombinant D- and L-

threonine aldolases, J. Am. Chem. SOC. 119,11734-11742.

KITAZUME,T.,SHIKAWA,. (1984), Induc tion of cen -ter of ch irality into fluoro compounds by micro-bial transformation of 2,2,2-trifluoroethanol,Chem. Lett., 1815-1818.

KOBORI,., MYLES, . C., WHITESIDES,. M. (1992),Substrate specificity and carbohydrate synthesisusing transketolase, J. Org. Chem. 57,5899-5907.

KOYAMA,T.,ATSUBARA,., OGURA,. (1985),Iso-

prenoid enzyme systems of silkworm. 11.Forma-tion of the juvenile hormone skeletons by farne-syl pyrophosphate synthetase 11, J . Biochem. Jpn.

98,457-463.KOYAMA,T.,GURA, ., BAKER,. C., JAMIESON,.

C., SCHOOLEY,. A. (1987), Synthesis and abso-

7182-7184.

lute co nfiguration of 4-methyl uvenile horm one I

(4-MEJH I) by a biogenetic approach - a combi-nation of enzyma tic synthes is and biotransforma-tion, J.Am . Chem.SOC.109,2853-2854.

KRAGL,U., GYGAX, ., GHISALBA,.,WANDREY,.

(1991), Enzymatic 2-step synthesis of N-acetyl-neuraminic acid in the enzyme membrane reac-tor,Angew. Chem. (Int. Edn. Engl.) 30,827-828.

KREN,~.,ROUT, . H. G.,DALTON,., HUTCHINSON,

D. W., KONIG,W. et al. (1993), Pyruvate decarbox-ylase: a new enzyme for the production of acy-loins by biotransformation, J. Chem. SOC.,Chem.Commun., 341-343.

KRIEF, ., SCHAUDER,.-R., G U I ~ T ,., HERVE U

PENHOAT,., LALLEMAND,.-Y. (1987),About themechanism of sterol biosynthesis,J. Am . Chem.SOC.109,791&7911.

KRIEF,A., ASAU,., QuBRI~,. (1991), Comparisonof the behavior of oxidosqualene cyclases frompig liver and yeast to tow ard epoxysqualene ana-logs possessing a A18-19 Z or E C, C doublebond, Bioorg. Med. Chem. Lett. 1,365-368.

LASCHAT,. (1996), Pericyclic reactions in biologicalsystems - does nature know about the Diels-Al-de r reaction? Angew. Chem. (Int. Edn. Engl.) 35,

LEES,W. J., WHITESIDES,. M. (1992),The enzymaticsynthesis of 1,5-dideoxy-l,5-diimino-D-talitolnd1-deoxygalactostatin using fuculose 1-phosphatealdolase,Bioorg. Chem. 20,173-179.

LEES,W. J., WHITESIDES,. M. (1993), Diastereose-lectivity (enantioselectivity) of aldol condensa-tions catalyzed by rabbit muscle aldolase at C-2of RCHOHCHO if R has an appropriatelyplaced negatively charged group, J. Org. Chem.

LIN,C.-H., SUGAI,., HALCOMB,. L., ICHIKAWA,.,WONG, .-H. (1992), Unu sual stereoselectivity insialic acid aldolas e-catalyzed aldol condensations- synthesis of both enantiomers of high-carbonmonosaccharides, . Am . Chem. SOC. 114,lO138-10 145.

LIN,C.-C., LIN,C.-H.,WONG, .-H. (1997), Sialic acidaldolase-catalyzed condensation of pyruvate andN-substituted mannosam ine: a useful method forthe synthes is of N-substituted sialic acids, Tetruhe-dron Lett. 38,2649-2652.

LINDQVIST,., SCHNEIDER,., ERMLER, ., SUND-STROM, M. (1992), 3-Dimensional structure oftransketolase, a thiamine diphosphate d epend entenzyme, at 2.5 A resolution, E M B O J. 11,

LIU,K. K.-C., WONG, .-H. (1992),A new strategy forthe synthesis of nucleoside analogs based o n en-zyme-catalyzed aldol reactions, J. Org. Chem. 57,47894791.

LIU, K. K.-C., KAJIMOTO,T., CHEN,L., ZHONG, .,

ICHIKAWA,., WONG, .-H. (1991), Use of dihy-

289-291.

58,1887-1894.

2373-2379.




droxyacetone phosphate dependent aldolases in

the synthesis of deoxyazasugars,J.Org. Chem. 56,

628M289.

LOBELL, ., CROUT, . H. G. (1996), Pyruvate decar-

boxylase: a molecular modeling study of pyruvate

decarboxylation and acyloin formation, J. A m .Chem.SOC.118,1867-1873.

LOOK,G. C., FOTSCH, . H., WONG,C.-H. (1993),

Enzyme-catalyzed organic synthesis - practical

routes to azasugars and their analogs for use as

glycoprocessing inhibitors, Acc. Chem. Res. 26,

LOTZ,B. T., GASPARSKI,. M., PETERSON,., MIL-

LER,M. J. (1990), Substrate specificity studies of

aldolase enzymes for use in organic synthesis,

J . Chem.SOC.,Chem . Cornrnun.,1107-1109.

MATSUMOTO,., SHIMAGAKI,., NAKATA,T., ISHI,

T. (1993), Synthesis of acyclic polyol derivatives

via enzyme-mediated aldol reaction, TetrahedronLett. 34,4935-4938.

MEDINA,. C., KYLER,K. S. (1988), Enzymatic cy-

clization of hydroxylated surrogate squalenoids

with bakers' yeast, J. Am. Chem. SOC. 110,

48184821.

MEDINA,. C., GUAJARDO,., KYLER,K. S. (1989),

Vinyl group rearrangement in the enzymatic cy-

clization of squalenoids - synthesis of 30-oxyster-

ols,J. Am . Chem.SOC.111,231~2311.

MOCALI,A., ALDINUCCI, ., PAOLETTI,. (1985),

Preparative enzymic synthesis and isolation of

~-threo-2-pentulose5-phosphate (D-XylUlOSe 5-phosphate), Carbo hydr. Res. 143,288-293.

MORADIAN,., BENNER,. A. (1992), A biomimetic

biotechnological process for converting starch to

fructose - hermodynamic and evolutionary con-

siderations in applied enzymology,J. Am . Chem.SOC.114,6980-6987.

Enzymatickhemical synthesis and biological

evaluation of 7-membered iminocyclitols,J. Am.Chem.SOC.118,7647-7652.

MORRIS, . G., SMITH,M. E. B., TURNER,. J., LILLY,

M. D., MITRA,R. K., WOODLEY,. M. (1996),

Transketolase from Escherichia coli: a practicalprocedure for using the biocatalyst for asymmet-

ric carbon-carbon bond synthesis, Tetrahedron:Asymmetry 7,2185-2188.

MYLES,D. C., ANDRULIS11, P. J., WHITESIDES,.

M. (1991), A transketolase-based synthesis of

(+)-exo-brevicomin, Tetrahedron Lett. 32,

4835-4838.

NAGASAWA,., YAMADA,. (1986), Enzymatic trans-

formations of 3-chloroalanine into useful amino-

acids,Ap pl. Biochem. Biotechnol. 13,147-165.

NAGY,. O.,BEDNARSKI,. D. (1991),The chemical-

enzymatic synthesis of a carbon glycoside of

N-acetylneuraminic acid, Tetrahedron Lett. 32,

182-190.

MORfS-VARAS, F., QIAN,X. N., WONG, C.-H. (1996),

3953-3956.

NICOTRA, ., PANZA,L., Russo, G., VERANI,A.

(1993), Simple and stereoselective chemoenzy-

matic synthesis of an a-C-mannoside, Tetra-hedron:Asymmetry 4,1203-1204.

NIKOLOVA,., WARD,0.P.

(1994), Effect of support

matrix on ratio of product to by-product forma-

tion in L-phenylacetyl carbinol synthesis,Biotech-nol. Lett. 16,7-10.

NISHIHARA,., DEKKER, . E. (1972), Purification,

substrate specificity and binding, /3-decarboxyl-

ase activity, and other properties of Escherichiacoli 2-keto-4-hydroxyglutarate ldolase, J . Biol.Chem. 247,5079-5087.

OGURA, ., KOYAMA,. (1981), Applications of en-

zymes to organic synthesison the enzymatic re-

actions of C-C bond formation, J. Synth. Org.Chem.Jpn. 39,459-466.

OIKAWA, ., KATAYAMA,., SUZU KI,~ .,CHIHARA,A.(1999, Enzymatic activity catalyzing exo-selective

Diels-Alder reaction in solanapyrone biosynthe-

sis,J. Chem.SOC.,Chem. Commun., 1321-1322.

OIKAWA, ., YAGI,K., WATANABE,., HONMA, .,

ICHIHARA, . (1997), Biosynthesis of macro-

phomic: plausible involvement of intermolecular

Diels-Alder reaction,J. Chem.SOC.,Chem. Com -mun., 97-98.

OLIVER,A. L., RODDICK,. A., ANDERSON,. N.

(1997), Cleaner production of phenylacetyl carbi-

no1 by yeast through productivity improvements

and waste minimization, Pure Appl. Chem. 69,

PEDERSON,. L., KIM,M.-J., WONG,C.-H. (1988), A

combined chemical and enzymatic procedure for

the synthesis of 1-deoxynojirimycinand l-deoxy-

mannojirimycin, Tetrahedron Lett. 29,4645-4648.

PEDERSON,. L., LIU,K. K.-C., RUTAN,. F., CHEN, .,

WONG, .-H. (1990), Enzymes in organic synthe-

sis - synthesis of highly enantiomerically pure

1,2-epoxy aldehydes, epoxy alcohols, thiirane,

aziridine, and glyceraldehyde-3-phosphate, J.Org. Chem. 55,4897-4901.

PROSEN,., WARD,0.P. (1994), Optimization of re-

action conditions for production of S-( )-2-hy-

droxypropiophenone by Acinetobacter calcoacet-icus,J. Ind. M icrobiol. 13,287-291.

RACKER,. (1961a), Cleavage and synthesis of car-

bon to carbon bonds, in: The Enzymes (BOYER,.

D., LARDY, ., MYRBACK,., Eds.), pp. 305-317.

New York: Academic Press.

RACKER,. (196lb),Transketolase, in: The Enzymes(BOER, P. D., LARDY,H., MYRBACK,., Eds.),

pp. 407-412. New York: Academic Press.

RACKER, . (1961c),Transaldolase, n: The Enzymes(BOYER, . D., LARDY, ., MYRBACK,., Eds.),

pp. 397406. New York: Academic Press.

RAMA AO,A., SRINIVASAN,. N., BHANUMATHI,.

(1990), A stereoselective biocatalytic Diels-Al-der reaction, Tetrahedron Lett. 31,5959-5960.

2371-2385.



8 References 39

ROSE,A. H. (1961), Industrial Microbiology. Lon-

don: Butterworths.

Rosso, R. G., ADAMS, . (1967), 4-Hydroxy-2-keto-

glutarate aldolase of rat liver- purification, bind-

ing of substrates and kinetic properties, J. Biol.Chem. 242,5524-5534.

SAEED,A.,YOUNG,D.. (1992), Synthesisof L-p-hy-

droxyaminoacids using serine hydroxymethyl-

transferase, Tetrahedron48,2507-2514.

SANTANIELLO,., FERRABOSCHI,., GRISENTI, .,

MANZOCCHI,. (1992),~heiocatalytic approach

to the preparation of enantiomerically pure chiral

building blocks,Chem. Rev. 92,1071-1140.

SCHMID, ., WHITESIDES,. M. (1990), A new ap-

proach to cyclitols based on rabbit muscle aldol-

ase (RAMA),J. Am . Chem. SOC.112,9670-9671.

SCHNEIDER,., SUNDSTROM,., LIDQVIST,. (1989),

Preliminary crystallographic data for transketo-

lase from yeast,J.Biol. Chem.264,21619-21 620.

SCHOLTZ,. M., SCHUSTER,.M. (1984), Substrates of

hydroxyketoglutarate aldolase,Bioorg. Chem. 12,

SCHULTZ, ., WALDMANN,., VOGT,W., KUNZ,H.

(1990), Stereospecific C-C bond formation with

rabbit muscle aldolase - a chemoenzymatic syn-

thesis of (+)-exo-brevicomin, Tetrahedron Lett.

SERVI,S. (1985), 2,2,5-Trimethyl-1,3-dioxolane-4-

carbaldehyde as a chiral synthon: synthesis of the

two enantiomers of methyl 2,3,6-trideoxy-a-~-

threo-hex-2-enopyranoside,ey intermediate in

the synthesis of daunosamine. and of (+)- and

(-)-rhodinose, J. Org. Chem. 50,5865-5867.

SERVI,. (1990), Bakers’ yeast as a reagent in organ-

ic synthesis,Synthesis, 1-25.

SHIBATA, ., SHINGU, ., VASSILEV,. P., NISHIDE,

K., FUJITA, . et al. (1996), Kinetic and thermo-

dynamic control of L-threonine aldolase cata-

lyzed reaction and its application to the synthesis

of mycestericin D, Tetrahedron Lett. 37,

SHIMAGAKI,., MUNESHIMA,H.,UBOTA, ., OISHI,

T. (1993), Chemoenzymatic carbon+arbon bond

formation leading to noncarbohydrate derivative

- stereoselective synthesis of pentamycin C- ll-

C-16 fragment, Chem . Pharm. Bull. 41,282-286.

SHIMAZU,. (1950), Formation and decomposition

of D-( -)-phenylacetylcarbinol by yeast,J. Chem.SOC.Jpn. 71,503-505 (Chem.Abstr. 45,9004).

SHIN,H. S., ROGERS,. L. (1995), Biotransformation

of benzaldehyde to L-phenylacetyl carbinol, an

intermediate in L-ephedrine production, by im-

mobilized Candida utilis, Ap pl. Microbiol. Bio-technol. 44,7-14.

SHIN,H. S., ROGERS,. L. (1996), Production of L-

phenylacetylcarbinol (L-PAC) from benzalde-

hyde using partially purified pyruvate decarbox-

ylase (PDC), Biotechnol. Bioeng. 49,5242.

229-234.

31,867-868.

2791-2794.

SPALTENSTEIN,., WHITESIDES,. M. (1991), Poly-

acrylamides bearing pendant a-sialoside groups

strongly inhibit agglutination of erythrocytes by

influenza virus,J . Am. Chem. SOC.113,68&687.

SPEVAK, ., NAGY,. O., CHARYCH,. H., SCHAEFER,

M. E., GILBERT,. H., BEDNARSKI,. D. (1993),

Polymerized liposomes containing C-glycosides

of sialic acid -potent inhibitors of influenza virus

in vitro infectivity, J. Am. Chem. SOC. 115,

1146-1147.

SRERE, ., COOPER,. R., TABACHNICK,., RACKER,

E. (1958), Oxidative pentose phosphate cycle.

I. Preparation of substrates and enzymes, Arch.Biochem. Biophys. 74,295-305.

STRAUB, ., EFFENBERGER,., FISCHER,. (1990),

Enzyme-catalyzed reactions. 4. Aldolase-cata-

lyzed C-C bond formation for stereoselective

synthesis of nitrogen-containing carbohydrates,

J. Org. Chem. 55,3926-3932.

STUMPF, ., KIESLICH,. (1991), Acyloin condensa-

tion of acyclic unsaturated aldehydes by Mucorspecies,Ap pl. Microbiol. Biotechnol. 34,598-603.

SUGAI, ., SHEN,G.-J., ICHIKAWA,., WONG,C.-H.

(1993), Synthesis of 3-deoxy-D-manno-2-octulo-

sonic acid (KDO) and its analogs based on KDO

aldolase-catalyzed reactions, J. Am. Chem. SOC.115,413421.

TAKAOKA,., KAJIMOTO, ., WONG,C.-H. (1993),

Inhibition of N-acetylglucosaminyltransferen-

zymes - chemical-enzymatic synthesis of new 5-membered acetamido azasugars,J. Org. Chem. 58,48094312.

TAKAYAMA,., MCGARVEY, . J., WONG,C.-H.

(1997a), Microbial aldolases and transketolases:

new biocatalytic approaches to simple and com-

plex sugars,Ann. Rev. Microbiol. 51,285-310.

TAKAYAMA,. , MCGARVEY, . J., WONG,C.-H.

(1997b), Enzymes in organic synthesis: recent de-

velopments in aldol reactions and glycosylations,

Chem. SOC .Rev. 26,407-415.

TURNER,. J. (1989), Recent advances in the use of

enzyme-catalyzed reactions in organic synthesis,

Nat. Prod. Rep. 6,625-644.

TURNER,. J. (1994), Recent advances in the use of

enzyme-catalyzed reactions in organic synthesis,

Nut. Prod. Rep. 11,l-15.

TURNER,. J., WHITESIDES,. M. (1989), A com-

bined chemical-enzymatic synthesis of 2-deoxy-

D-arabino-heptulosonic acid 7-phosphate, J. Am.Chem.SOC.111,624-627.

UCHIDA, ., TSUKADA,., SUGIMORI,. (1984), Pur-

ification and properties of N-acetylneuraminate

lyase from Escherichia coli, J. Biochem. 96,

UCHIDA,Y.,TSUKADA,Y.,UGIMORI,T.1985), Distri-

bution of N-acetylneuraminic lyase in bacteria

and its production by Escherichia coli, Agric. Biol.Chem. 49,181-187.

507-522.




VO N DER OSTEN, . H.,SINSKEY,A.., BARBAS, . F.,PEDERSON, . L., WANG,Y.-F., WONG, C.-H.

(1989), Use of a recombinant bacterial fructose-

1,6-diphosphate aldolase in aldol reactions -

preparative syntheses of 1-deoxynojirimycin,1-deoxymannojirimycin, 1,4-dideoxy-l,4-imino-~-arabinitol, and fagomine, J . Am . Chem. SOC.111,

3924-3927.

VASSILEV,. P., UCHIYAMA,., KAJIMOTO,.,WONG,

C.-H. (1995a), L-Threonine aldolase in organic

synthesis: preparation of novel P-hydroxy-a-ami-

no acids, Tetrahedron Lett. 36,40814084.

VASSILEV,. P., UCHIYAMA,., KAJIMOTO,., WONG,

C.-H. (1995b),An efficient chemo-enzymatic syn-

thesis of a-amino-P-hydroxy-ybutyrolactones,

Tetrahedron Lett. 36,5063-5064.WALDMANN,. (1999, Enzymatic carbon-carbon

bond formation, in:Organic Synthesis HighlightsI1 (WALDMANN,., Ed.), pp. 147-155. Weinheim:

VCH.

WARD,0.P. (1995), Application of bakers’ yeast in

bioorganic synthesis,Can.J . Bot. 73, S1043-Sl048.

WILCOCKS,., WARD,0.P. (1992), Factors affecting

2-hydroxypropiophenone formation by benzyl-

formate decarboxylase from Pseudomonas puti-da , Biotechnol. Bioeng. 39,1058-1063.

WILCOCKS,., WARD, .P., COLLINS,.,DEWDNEY,N.

J., HONG, . P., PROSEN,. (1992), Acyloin forma-

tion by benzylformate decarboxylase from Pseu-domonas putida, Appl. Environ. Microbiol. 58,

1699-1704.

WONG,C.-H., WHITESIDES,. M. (1994), Enzymes inOrganic Chemistry, Tetrahedron Organic Series12, pp. 195-251. Oxford: Pergamon Press.

WONG, .-H., SHENG,.J., PEDERSON,. L.,WANG,.F.,HE”EN, W J. (1991), Enzymatic catalysis n or-

ganic synthesis,Methods Enzym ol. 202,591620.

WONG,C.-H., HALCOMB,. L., ICHIKAWA,., KAJI-MOTO T. (1995a), Enzymes in organic synthesis:

application to the problems of carbohydrate rec-

ognition (part l), Angew. Chem. (Int. Edn. Engl.)

34,412432.

co, O., GIJSEN, . J. M., STEENSMA,. H. (1995b),

Recombinant 2-deoxyribose-5-phosphate aldo-

lase in organic synthesis: use of sequential two-

substrate and three-substrate aldol reactions,

J . Am . Chem. SOC.117,3333-3339.

XIAO, X.-Y.,PRESTWICH,. D. (1991), Enzymatic

cyclizations of 26-hydroxy-2,3-oxidosqualenes

and 29-hydroxy-2,3-oxidosqualenesive 19-hy-

droxylanosterols and 21-hydroxylanosterols, Te-trahedron Lett. 32,68434846.

ZHOU,P., SALLEH, . M., HONEK,. F. (1993), Facile

chemoenzymatic synthesisof 3-(hydroxymethy1)-

6-epicastanospermine,J . Org. Chem. 58,264-266.

ZIEGLER,., STRAUB, A., EFFENBERGER,. (1988),

Enzyme-catalyzed reactions. 3. Enzyme-cata-

lyzed synthesis of 1-deoxymannonorjirimycin,1-deoxynorjirimycin and 1,4-dideoxy-1,4-imino-

D-arabinitol,Angew. Chem. (Int. Edn. Engl.) 27,

716-717.

WONG, C.-H., GARCIA-JUNCEDA, E., CHEN,L.,BLAN-



2 Lyases

JASSEMG. MAHDI

DAVID. ELLYCardiff, Wales, UK

1 Introduction 43

2 Conventions 433 Enzyme Commission Classification 434 Examples of Lyases 47

4.1 Carbon-Carbon Lyases 474.1.1 Carboxy-Lyases 47

4.1.1.1 Pyruvate Decarboxylase 49

4.1.1.2 Oxalate Decarboxylase 554.1.1.3 Oxaloacetate Decarboxylase 554.1.1.4 Acetoacetate Decarboxylase 584.1.1.5 Acetolactate Decarboxylase 624.1.1.6 Aconitate Decarboxylase 644.1.1.7 Benzoylformate Decarboxylase 654.1.1.8 Oxalyl-CoA Decarboxylase 67

4.1.1.9 Malonyl-CoA Decarboxylase and Related Malonate DecarboxylatingEnzymes 68

4.1.1.10 Aspartate a-Decarboxylase 72

4.1.1.11 Aspartate 4-Decarboxylase 744.1.1.12 Valine Decarboxylase 764.1.1.13 Glutamate Decarboxylase 764.1.1.14 Ornithine Decarboxylase 78

4.1.1.15 Lysine Decarboxylase 794.1.1.16 Arginine Decarboxylase 80

4.1.1.17 Diaminopimelate Decarboxylase 804.1.1.18 Histidine Decarboxylase 81

4.1.1.19 Orotidine-5 ‘-Phosphate Decarboxylase 82

4.1.1.20 Tyrosine Decarboxylase 824.1.1.21 Aromatic L-Amino Acid Decarboxylase 83

4.1.1.22 Phosphoenolpyruvate Carboxylases and Carboxykinases 844.1.1.23 Phenylpyruvate Decarboxylase 85





42 2 Lyases

4.1.1.24 Tar trona te Sem i-Aldehyd e Synthase 854.1.1.25 D ialkylglycine Deca rboxylase 864.1.1.26 Indo lepyruva te Deca rboxylase 87

4.1.2.1 Threon ine Aldolase 894.1.2.2 Mandelonitrile Lyase and Hyd roxym andelonitrile Lyase 90

4.1.3.1 Isocitra te Lyase 904.1.3.2 Ac etolac tate Synthase 91

4.1.4 O the r Carbon-Carbon Lyases 914.1.4.1 Tryptophanase 914.1.4.2 L-Tyrosine Phenol-L yase 94

4.1.2 Aldehy de-Lyases 89

4.1.3 Oxo-A cid-Lyases 90

4.2 Carbon-Oxygen Lyases 984.2.1 Hyd ro-Lyases 98

4.2.1.1 Carbonic Anh ydrase 994.2.1.2 Fumarase 100

4.2.1.3 Aconitase 1094.2.1.4 a,P-Dihy droxy-A cid De hyd ratase 1114.2.1.5 3-Deh ydroquinate Dehy dratase 1114.2.1.6 Enolase 1124.2.1.7 L-Serine Dehy dratase, D-Serine Dehy dratase an d L-Threonine

Dehydratase 1134.2.1.8 Imidazoleglycerol Pho spha te Dehyd ratase 1144.2.1.9 M aleate Hy drata se 114

4.3.1 Carboh ydra te Hyd ro-Lyases 1154.3 Carbo hydrate Lyases 115

4.3.1.1 Monosaccharide Carbo xylic Acid De hyd ratases 115

4.3.1.2 3,6-Dideoxysugars 1174.3.2 Lyases which Act on Po lysaccharides4.3.2.1 Pectin and Pectate Lyases 1224.3.2.2 Pec tate Lyases 1244.3.2.3 Pectin Lyases 125

122

4.4 Carbo n-Nitrogen Lyases 1254.4.1 Am mo nia Lyases 125

4.4.1.1 Aspartase 1264.4.1.2 /3-Methylaspartase 1294.4.1.3 Histidase 1324.4.1.4 Phen ylalanine Am mo nia-Lyase 137

4.5 O the r Carbon-Nitrogen Lyases 142

4.6 Carbo n-Halide Lyases 1424.7 O the r Lyases 143

4.7.1 Alkyl Mercury-Ly ase 1435 References 143



3 Enz yme Commission Classification 43

1 Introduction

Lyases are defined as enzy me s cleaving C-C,

C - 0 , C-N and other bonds by other meansthan by hydrolysis or oxidation (Anonymous,1984,1992). In m ost cases the produc ts are un-saturated molecule (frequently an alkene)plus an X-H species.The lyases are a disparategroup of enzymes, catalysing a w ide range ofprocesses with few features in common, be-yond their assignment to the same enzymecomm ission group. Lyases utilize an extraordi-nary range of prosthetic groups which includethiamine pyrophosp hate (JORDA N, 1999;LINDQVISTnd SCHNEIDER,993), iron sulfur

clusters (FLINTand ALLEN,1996), NAD(P)(H), pyridoxal5 ’-phosphate (JANSON IUS,998;JOH N, 1995; HA YAS HI, 995; ALEXA NDERtal., 1994), biotin (TOH e t al., 1993) and back-bone dehydroalanine (GU LZA R t al., 1997),or 4-methylidene-imidazole-5-one esidues(SCHWEDEt al., 1999).

They have been rarely (if ever) reviewed to-gether and indeed one major biotransforma-tions text doe s not eve n mention them as a sin-gle group in the index (DRAUZ nd WALD-MA“, 1995).The fu ndam ental biochemistry oflyases (including protein crystallography) is asubject of frenetic activity at present. Severallyases are used in industrial processes, such asthe production of am ino acids, however, labor-atory applications are very rare. It is hopedtha t this review will bring this subject to a wid-er au dience and prom ote exploitation of thesefascinating enzymes.

2 Conventions

Eac h lyase is discussed in a sepa rate section,according to Enzym e C omm ission class num-bers. Each section is presented in the se-quence: definition of the biotransformation,background, enzymology and enzyme struc-ture studies, stereochemistry, synthetic appli-cations, technological asp ects. Th e last of thesesubsections covers large scale reactions andmethodology leading to manufacturing pro-

cesses. In shorter sections these headings are

not shown explicitly. The conven tions for re-ferring to figures, references and compoundsused through out this, and th e previous volume,are adhered to in this chapter. However, if

these a re m arked “cf.” the attribution refers toa similar, related or parallel process, ratherthan that d irectly under discussion at th e time.Tab. 1summ arizes the princ ipal classes of lyas-es.

3 Enzyme Commission

Classification

Lyases are enzym es that cleave C-C, C-0 ,C-N and oth er bonds by elimination to formmultiple bo nds or rings. In m any cases this in-volves cleavage of C-X and C-H bon ds on ad-jacent carbon atoms by an E l or E2 mecha-nism. R eactions which involve hydrolysis (butnot h ydration) o r oxidation (e.g., desaturases)are excluded. Lyases are assigned to EnzymeCom mission class 4. The subclass (1-6 and 99)is assigned according to the bond broken andthe subsubclass according to the group elimi-nated. As usual for enzyme comm ission codes,the subsubsub class is a u nique design ator foreach enzym e. For example phenylalanine am-monia-lyase is assigned to E C 4.3.1.5. This in-dicates a carbon-nitrogen lyase (E C 4.3),which eliminates amm onia from the substrate(E C 4.3.1).

The systematic names for lyases are con-structed according to the pattern substrategroup-lyase, e.g., phenylalanine ammonia-lyase. The hyphen is an im portant co mpo nent

of th e systematic nam e because it distinguishesthis class of enzym es from o the rs with similarnames, e.g., hydro-lyases and hydrolases. How-ever, this rule is relaxed for recommendednames, e.g., aldolase, decarbo xylase and dehy-dratase.

In som e cases elimination is followed imme-diately by addition and hence overall the reac -tion is appare ntly a substitution. If there is ev-idence of an unsaturated interm ediate the en-zym e is nevertheless classified as a lyase.

The intramolecular lyases are a sm all group

of enzymes which ar e classified as isomerases



44 2 Lyases

Tab. 1.Representative Examples of Lyase Catalyzed Reactions~~~

E C Code ReactionSystematic Names

4.1 Carbonxarbon lyases4.1.1 Carboxy-lyases 0

Pyruvate decarboxylase

0 01 Pyruvate COZ 2 Acetaldehyde

"

26 Oxalate COZ 27 Formate

0 Oxaloacetate 0decarboxylase

EC4.1.1.3

0

COZ 1 Pyruvate

0 0

28 Oxaloacetate

41 Acetoacetate?@ decarboxylase

40 Acetoacetate 39 Acetone

0 0 Acetolactate 0

HOC 4.1.1.5

5 (R)-AcetoinCOZ

.65 3-Acetolactate

78 Benzylformate 4 Benzaldehyde

Histidinedecarboxylase

NH, EC 4.1.1.22N N.H

137 L-Histidine

H138 Histamine



3 Enzyme Commission Classification 45

Tab. 1 (continued)

EC Code React ion

Systematic Name s

4.1.2 Al de hy de- lya ses Orotidine 5'-phosphatedecarboxylase

13 9 R =coy 140EC 4.1.1.23

COZ

$ %

HO OH

141 142a R = H; L-Tyrosineb R = OH; L-DOPA

a R = H; Tyramineb R = OH; Dopamine

acid decarboxylase

NH H

143 144a R = H; L-Tryptophanb R = OH; 5-Hyhxy-L-tryptophan

a R = H; Tryptamineb R = OH ; Serotonin

4.1.3 0x0-ac id- lyases 0

AH H$ACOY

2 Acetaldehyde 158 Glycineo2

166 L-Thon ine

4.1.4 Other c ar bo nsar bon 0lyases

H A C O P

lsocitrate lyase 151 Glyoxylate

EC4.1.3.1+

198 Succinate68 Isocitrate



46 2 Lyases

Tab. 1 (continued)

EC Code Reaction

Systematic Nam es

4.2 Carbon-oxygen lyases4.2.1 Hydrolyases Carbonic anhydrase

EC 4.2.1.1

t HCOPCOZ

196 Fumarate

H20

197 L-(S)-(-)-Malate

3-Dehydroquinatedehydratase

EC4.2.1.100 0

-OH OH

257 3-Dehydrcquinate 258 3-Dehydroshikimate

I

4.2.2 Acting on -olysaccharides

I

-OC0, R Pectin and pectate lyases Ro

OHRO

RO RO

323 ";"21 -



4 Examples of Lyases 47

Tab. 1 (continued)

EC Code ReactionSystematic Names

4.3 Carbon-nitrogen lvases4.3.1 Amm onia lyases

Aspartase

EC 4.3.1.1

101 L-(3-(+)-Aspartate96 Fumarate NH,O

* Ooz& s co:P-Methylaspartase

EC 4.3.1.2

NH,@

335 Mesaconate 336 L-threo-(21i,33)-3-Methylaspartate

137 L-Histidine 368 Urocanate

Phenylalanine 0a m m o n i a 7 ~ r c 0 2

EC 4.3.1.5

NH,O388 L-Phenylalanine 389 tram-Cinnamate

(E C 5). They catalyze reactions in which ag r o w is eliminated from one Dart of a mole-

4 Examples of Lyasescule 'to leave a mu ltiple bond \;bile remainingcovalently attached to th e m olecule.

A rema rkable nu mb er of lyases are involvedin central metabolism in th e citric acid (K rebscycle), glycolysis and neoglucog enesis. These Many of the enzym es und er this heading are

are sum marized in Fig. l. covered in Chapter 1, this volume: Car-bon-Carbon Bond Formation Using Enzymes.

4.1 Carbon-Carbon Lyases(EC 4.1)

4.1.1 Carboxy-Lyases (E C 4.1.1.X)

This subsubclass contains enzymes whichcatalyze decarboxylation (IDING t al., 1998;CROUT t al., 1994). Most carboxy-lyases uti-lize thiamine pyrophosphate (JORDAN,999;

LINDQUISTnd SCHNEIDER,993), pyridoxalphosphate or biotin (TOH t al., 1993)as cofac-

tors.



48 2 Lyases

decarboxylase

1Pyruvate 0 2 Acetaldehyde

C ~ A S H , AD@

74 Itaconate

COZOxaloacetate

decarboxylase decarboxylase

56 Acetyl-CoA Citrate

Citrate SpthaSe Aconitase

28 Oxaloacetate (33)EgDH'valate

NADH7 ehydrogenase

& 197 L-(S)- Malate (199,

200,205, u)6,210,214,- 215,217,222,231,243)H

0

0 2

Fumarase

H2O

0Ooz+COz 196 Fumarate

(213,218,221)

Y 20

73 cis-Aconitate

Aconitase \

0 15 1 Glyoxylate

Isocitrate

168 Isocitrate (249)

Isocitratedehydrogenase

NADH

a-Ketoglutarate "O~C\/K c$

dehydrogenaseSuccinyl CoA

synthetase NADH, Ho

0 Succinyl CoA198 Succinate (169)

GTP GDP

Fig. 1. Lyases in the citric acid (Krebs) cycle and relate d biosynthetic sequences. Bold nu m bers in bracketsrefer t o labeled analogs.




Bakers' yeast

decarboxylase)(pymvate

4.1.1.1 Pyruvate Decarboxylase(EC 4.1.1.1)

IntroductionPyruvate decarboxylase catalyzes the decar-

boxylation of pyruvate (1)to acetaldehyde (2)and acyloin type condensation reactions (Fig.2) (OBAand URITANI,982). Synthetic applica-tions of pyruvate decarboxylase (PDC) are de-scribed in Chapter 1 of this volume and havebeen reviewed recently elsewhere (SCHORKENand SPRENGER,998; FESSNER,998;FESSNERand WALTER, 997).This section will describemechanistic aspects and biotechnological ap-plications. PDC is widely distributed in plants

and fungi, but is rare in prokaryotes and ab-sent in animals (CANDYnd DUGGLEBY,998;VA N WAARDE,991).The major metabolic roleof PDC is as a component of the anaerobicpathway from pyruvate to ethanol. Pyruvate(1,Fig. 2) undergoes decarboxylation to giveacetaldehyde (2) which is reduced by alcoholdehydrogenase to ethanol. The NAD+ pro-duced by the reductive step is then utilized incellular metabolism (PRONK t al., 1996). Thispathway is of course the basis of the brewingindustry and much effort has been made to en-hance its productivity for both beverage pro-duction and for ethanol for fuel (INGRAMtal., 1999; DENG nd COLEMAN,999; SCOPES,1997). The operation of this pathway is essen-tial for the survival of crop plants which aresubjected to flooding, such as rice (TADEGE tal., 1999;Su and LIN,1996; SACHS t al., 1996;PESCHKEnd SACHS, 993).

Mechanism and Enzyme StructurePyruvate decarboxylase is of considerable

synthetic importance because the acetalde-hyde a-anion equivalent (3) can be intercept-ed by exogenous aldehydes to give acyloins(e.g., 5 , 6 ) .Surprisingly, he reaction with acet-aldehyde (2), reproducibly gives 3-hydroxy-butanon-2-one (acetoin) with a low enantio-meric excess. Whole yeast and the purifiedPDC from yeast both give (R)-3-hydroxybuta-non-2-one (5) of about 50% enantiomeric ex-cess, whereas Zymomonas mobilis PDC givesthe opposite enantiomer, but with essentiallythe same enantiomeric excess (BORNEMANNt

P

2

0 1Pyruvate

Bakers' yeast(pyruvate

co2 decarboxylase)

O f ]

3

w0

5 circa 50%ee 6 99%ee

Steps

' 7 L-Ephedrine

Fig. 2. Pyruvate decarboxylase mediated d ecarboxy-lation and acyloin condensation.

to racemic 3-hydroxybutanon-2-one, whereasin the presence of exogenous acetaldehyde,the (S)-enantiomer is produced with a low

al., 1993).Wheat germ PDC converts pyruvate enantiomeric excess (CROUTet al., 1986).



50 2 Lyases

None of these reactions are of appreciablepractical value, however, interception by ben-zaldehyde (4) is highly stereoselective and the(R)-phenylacetyl carbinol ( 5 ) so formed has

been used in the manufacture of L-ephedrine(7) since 1932 (Knoll process; SHUKLA ndKULKARNI999; NIKOLOVAnd WARD, 991).Similarly high enantiomeric excesses havebeen achieved with other aromatic and hetero-aromatic aldehydes using both yeast PDC(KREN t al., 1993, CROUT t al., 1991b; CAR-DILLO et al., 1991) and Zymom onas m obil isPDC (BORNEMANNt al., 1996). Moreover, ali-phatic aldehydes (Cz-Clz) are also good sub-strates (MONTGOMERYt al., 1992).

The active site of PDC contains thiamine

pyrophosphate (TPP, 8, Fig. 3), which is alsoknown as thiamin diphosphate (ThDP). Re-cent advances in the study of TPP linked en-zymes are reported in a collection of articles involume 1385 of Biochimica et Biophysica Acta(Various, 1998). TPP (8) mediates both decar-boxylation and the acyloin condensation(SPRENGERnd POHL, 1999; SCHORKENndSPRENGER,998). Magnesium ions or calciumions are required for activity, as part of the py-rophosphate binding site (VACCARO t al.,

1995). Fundamental chemical aspects of themechanism are well understood and have beenengineered into abiotic chemical models (IM-PERIALI et al., 1999;KNIGHTnd LEEPER, 998;cf. KLUGER t al., 1995) and catalytic antibod-ies (TANAKAt al., 1999). PDC catalyzes therate of decarboxylation of pyruvate by a factorof about 10’’ above the spontaneous rate,which is equivalent to transition state stabiliza-tion of some 70 kJ mol-’ (HUHTA t al., 1992).

TPP (8) is bound tightly, but non-covalentlyto PDC (Fig. 4) and is converted to the ylide

(9) by intramolecular deprotonation (HUBNERet a1 1998) of C -2 of the thiazolium ring, by theamino group of the aminopyrimidine substitu-ent (KERN t al., 1997;FRIEDEMANNnd NEEF,1998),which is involved in a proton relay, with

a glutamate side chain (KILLENBERG-JABStal., 1997). The thiazolium ring, the aminopy-rimidine ring and the intervening methylenegroup adopt an unusual V conformation which

brings C-2 of the thiazolium ring and the ami-no substituent into close proximity. This ar-rangement increases the rate of deprotonationby at least a factor of lo4, relative to that ob-served in solution. The ylide (9) (SCHELLEN-BERGER, 1998) undergoes nucleophilic addi-tion to the ketone group of pyruvate (1) togive the zwitterion ( lo ) ,which evolves carbondioxide. The product (11) acts as an ena-minehinyl sulfide rather than an enol andhence undergoes either protonation (12) r al-kylation (13) adjacent to the oxygen to give an

imminium salt. Cleavage of the C-2 thiazoliumbond and restitution of the carbonyl groupyields either acetaldehyde (2) or the acyloinproduct (6).The stereochemistry of the inter-mediates has been inferred from molecularmodeling studies (LOBELLnd CROUT, 996b).A complete catalytic decarboxylation cycle re-quires the consumption of one proton, which isappended to C-1 of acetaldehyde. Single turn-over experiments with [2-3H]-TPP, demon-strate that 43% of the tritium label (*H, Fig.5)

is incorporated into [l-3H]-acetaldehyde and54% into [3H]-water (HARRISand WASHA-BAUGH, 995a, b). It is beyond doubt that theylide (9) exists at the active site at some stage,however, it has not been detected by 13C-NMR(SCHELLENBERGER,998; LI and JORDAN,1999). It is, therefore, reasonable to speculate,that protonation of the ylide is part of a relaywhich brings the “consumed proton” into theactive site and that ylide formation is linkedwith binding of pyruvate as substrate or allos-terically. Further aspects of Fig. 5 are discussed

below.S. cerevisiae, PDC consists of the products of

the genes P D C l , P DCS and PDC6. The pre-dominant isozyme monomer is the gene prod-uct of P D C l , which is 80% identical with the

Fig. 3. The structure of thiamine pyro-phosphate,R’ nd R2 re abbreviations

used in the following tw o figures.Thiamine pyrophosphate (TPP)




H 8 TPP bound to yeast

pyruvate decarboxylaseI

HO&

\ lPyruvate

R l H q R 2 ]

HO

11-1 11-I I 11-111

Fig. 4. “Minimal mechanism”for the pyruvate decarboxylase (PDC) catalyzed decarboxylation of pyruvate.

Stereochemical assignm ents are based on the molecular modeling studies of LOBELLnd CROUT 1996b)(cf. Fig. 3 , for R’ nd R’).

PDCS gene product. Normally these mon-omers associate into mixed tetramers (a4,

a&, a& etc.; FARRENKOPFnd JORDAN,

1992) which can be separated by ion-exchangechromatography, however, by using a non-PDC.5 expressing mutant, a homotetramericpdcl (MW 240,000 Da) has been isolated di-rectly (KILLENBERG-JABSt al., 1996). The ex-

pression of PDC.5 is repressed by thiamine but

P D C l is unaffected, hence under thiaminelimiting conditions both are expressed (MULL-ER et al., 1999).PDC6 is only expressed at verylow levels and appears to be metabolically un-important under normal circumstances (HOH-MANN, 1991a, b). PDC2 is a regulatory genewhich controls basal expression of P D C l(HOHMANN, 1993). pdcl and pdc5 both have

four cysteine residues per monomeric unit,



52 2 Lyases

0 0

H Ph 13

Fig. 5. Com plete catalytic sequence for the pyru vate decarboxylase (PDC) catalyzed decarboxylation of pyr-

uvate, incorporating stereochemical assignments (L OBE LLnd CROUT, 996b), [2-3H]-TPP abeling studies,(HARRIS nd WASHABAUGH,995a,b) and opening and closing of the active site (ALVAREZt al., 1995) (cf.Fig. 3 , for R' and R').

whereas pdc6 only has one (Cys221;BABURINA

et al., 1996). This residue which is located on

the surface of the protein at an interface seems

to be involved in allosteric activation by pyru-

vate and other carbonyl containing com-

pounds, which triggers conformational chang-

es (LI and JORDAN, 1999;JORDAN et al., 1998;

BABURINAt al. 1998a,b).

PDC undergoes slow allosteric activation by

pyruvate (1) or substrate surrogates such as

pyruvamide or ketomalonate. This may take

several seconds to minutes and causes an ap-

proximately ten fold increase in activity. Each

allosteric binding of a pyruvate molecule suf-fices to activate several thousand catalytic cy-

cles (ZENG t al., 1993;BABURINAt a1.,1994).It has been proposed that addition of the cys-

teine-221 thiol group (14) to pyruvate ( l ) ,ive



4 Examples of Lyasrs 53

structure of Zyrnornonas mobilis is also a tet-ramer (DOBRITZSCHt a l., 1998), but tightpacking of the sub units and the absen ce of ap-propriate cysteine residues precludes alloster-

ic activation by pyruvate and the enzyme islocked in an activated conformation (SUNetal., 1995; FU REY t al., 1998; KON IG 998).

BiotransformationsThe Knoll process for the manufacture of

(R)-phenylacetyl carbinol (5) from benzalde-hyde (4) (Fig. 7) using yeast continues to be themost efficient route to this material, neverthe-less, there ar e still significant problem s to besolved (OLIVE R t al., 1997).A high concentra-tion of benzaldehy de is essential for high pro-

ductivity in batch processes, but it is toxic toyeast and reduced to benzyl alcohol (NIK OLO -VA and WA RD, 991; WA RD, 995) which com-plicates the work-up. Increasing the supply ofbenzaldehyde when PDC activity is maximaland reducing the supply when alcohol dehy-drogenase activity is maximal m inimizes theseeffects (OL IVER t al., 1997). Imm obilized andsemi-continuous production gives similar ben-efits (TR IPATH I t a l., 1997). Reduc tion of L-

phenylacetyl carbinol (6 ) to (2R,3S)-phenyl-propan-2,3-diol (16) s also a significant sidereaction (MOCHIZUKIt al., 1995), althoughthe form er can be isolated by bisulfite adduc tformation ( SHUKLAnd KULKARNI,999).

In principle, use of purified P D C has the po-tential to avoid all these problems, but thus farthis has not been achieved on a practical scale,moreover benzaldehyde also inhibits PDC(CHOWet al., 1995). Partially purified PD Cfrom Candida utilis, was used to produce L-

phenylacetylcarbinol ( 6 ) at a final concentra-tion of 28.6 g L-'(190.6 mM ) (SHIN nd RO G-

ERS,1996a). When this enzyme w as imm obi-lized in spherical polyacrylamide beads, higherconcentrations of benzaldehyde were tolerat-ed (300 mM) and the half life of the enzymewas increased to 32 days (SHINand ROGERS,1996b). Similarly the thermal stability ofbrewers' yeast PD C can be improved by cova-lent conjugation with am ylose via a glycylgly-cine spacer (OHB A t al., 1995).

Z. rnobilis is an anaerobic gram-negativebacterium which is widely used in tropical re-gions for the ferm entation of alcoholic bever-

ages (D OELL E t al., 1993). Unlike most other

a he mithioketal(15) triggers the initial slow al-losteric activation of PDC (Fig. 6). However,recent resu lts suggest that pyruvate binding atthe regulatory site, without thiol ad dition caus-

es slow initial activation, and tha t hem ithioke-tal forma tion occurs during th e catalytic cycle,rathe r tha n prior to it. In this model, thiol addi-tion opens t he active site for admission of pyr-uvate and elimination closes it for the decar-boxylation, protonation and acetaldehyde for-mation steps. Finally, addition again enablesrelease of acetaldehyde and carbon dioxide(ALVAREZt al., 1995; Fig. 5).A closed confor-mation for the decarboxylatiodprotonation

stages of the cycle is consistant with the needto sequester the highly basic ylide (9) (pK,

17.5 in solution), away from solvent and the [2-3H]-TPP labeling study, which indicates mini-mal hydrogen exchange with the solvent (cf.th e waterproof active site, LOBEL Lnd CROUT,1996a).

The X-ray crystal structures of wild typeSaccharomyces cerevisiae PDC (from P D C l ,a4), DC from P D C l expressed in E. coli andS. uvariurn are PDC, have been determined(Lu et al., 1997; ARJU NANt al., 1996;DYDAtal., 1990, 1993). Both enzymes have almostidentical sequences and are tetrameric. Eachmonomeric unit consisting of a single peptidechain, is tightly associated into a dimer withthe TPP s situated at the interface. Two dimericpairs associate to give the tetramer (KONIG,1998;KON IG t al., 1998).Each m onomeric unitconsists of three domains: the N-terminal, a-

domain which binds the pyrimidine group ofTPP, the cen tral p-regulatory dom ain and theC-terminal y-domain which binds the pyro-phosphate group of TPP. The X-ray crystal

14 Closed 15 Open

Fig. 6. It has been prop osed that addition of cyste-ine-221 to pyruvate opens a lid which enables ano-ther molecule of pyruvate to enter the active site

(ALVAREZt al., 1995;LOBELLnd CROUT,996b).



54 2 Lyases

Bakers' yeast(pyruvatedecarboxylase)

Ph& 0

6 99%ee

1

1Steps

P h RL/N-

7 L-Ephedrine

6 H

16

Ph/' OH

17

Fig. 7. Byproducts formed in the the Knoll processfor the preparation of L-phenyl acetyl carbinol ( 6 ) ,which is used in the manufacture of L-ephedrine(7).

plants and fungi it utilizes the Entner-Dou-doroff pathway for the conversion of glucoseto py ruvate. This is much less efficient than gly-

colysis and hence large amo unts of ethanol ac-cumulate and PDC is one of the most abun-dant proteins produced by the organism (GU -NASEKARAN and RkT, 1999). Unlike yeastPDC, Z .mobilisPD C maintains the tetramericform above p H 8, when catalytic action is lostdue to dissociation of TPP and magnesiumions. It also has norm al M ichaelis-Menten ki-netics ( K , pyruvate 0.3-4.4 m M), whereas allother PDCs to date have sigmoidal kinetics(CANDY nd DUGG LEBY,998; POHL et al.,1995). Overall these factors render 2. mobilis

PDC a more effective pyruvate decarboxyla-

tion catalyst (1.7-lox) than yeast PDC , partic-ularly at low pyruvate concentrations (SUN tal., 1995).

The E. coli strain DH1 has been trans-

formed with the plasmid pLOI295, harboringthe gene encoding PD C from 2.mobilisATCC31821.27% of the soluble cell protein consist-ed of active recombinant Z . mobilis PDC,which was purified to homogenerity. The en-zyme was somewhat less active than yeastPDC , with benzaldehyde (Tab. 2, entry l) , andother arom atic aldehydes (Tab. 2, entries 2-9),however, the relative reactivities were similar.Heteroaromatic aldehydes (Tab. 2, entries10-12) and propanal were poor substrates(BORNEMANNt al., 1996). Cinnamaldehydes

are good substrates for yeast mediated acyloinformation. In virtually all cases the initallyform ed acyloins are stereospecifically reducedto the corresponding diols. These substratesare covered in detail in the chapter on C-Cbond formation.

Fluoropyruvate (20) ndergoes decarboxyl-ation by yeast PD C to give acetate (21), atherthan fluoroacetaldehyde (22) Fig. 8). De carb-oxylation procee ds normally (cf. Figs. 4 ,5 ) togive the enol (23) Fig. 9), which undergoes

elimination of fluoride to give the imminiumalkene (U),hich presumably undergoes hy-dration (25) and cleavage of the C-2 thiazoli-um bond to give acetate (21) nd the ylide ( 9 )(GISH et al., 1988). Fluoropy ruvate (20) alsoacts irreversibly at the allosteric site (SUNetal., 1997) and hence it would b e of great bene-fit to investigate the kinetics of decarboxyla-tion of fluoro pyru vate by Z . mobilis PDC. The

0

20 Fluoropyruvate

0

22 Fluoroacetaldehyde

Fig. 8.The decarboxylation of fluoropyru vate(20)toacetate (21) (GISH t al., 1988;SU Net al., 1997) cata-

lyzed by pyruvate decarboxylase.




Tab. 2. Acyloin Condensation Catalyzed by Whole Yeast, Yeast Pyruvate Decarboxlase (KRENet al., 1993)

and Z . mobih Pyruvate Decarboxylase (BORNEMANNt al., 1996)

0AH SociiiiIpyteAr

pyruvate 0

19ecarboxylase

18

Entry Ar Yeast Yeast PDC” Z . mobilis PDC

% ee Yield % ee Yield % ee

12

3

4

5

6

7

8

9

10

11

12

Phenyl2-fluorophenyl

3-fluorophenyl

4-fluorophenyl

2-chlorophenyl

3-chlorophenyl

4-chlorophenyl

2,3-difluorophenyl

2,6-difluorophenyl

2-fury1

3-f~ryl

3-thienyl

9787

95

97

81

86

77

97

87-

-

-

-60

25

31

32

25

40

-

-

-

-

-

9999

99

99

98

99

99

99

92-

-

-

1560

30

35

5

2035-

-

low

low

0

989897

98

98

98

98-

-

80-90

80-90-

practical impact of this result is that pyruvatesbearing potential nucleofuges are unsuitablesubstrates for alkylation.

4.1.1.2 Oxalate Decarboxylase

(Oxalate Carboxy-Lyase,

EC 4.1.1.2)

Oxalate decarboxylase catalyzes the decar-boxylation of oxalate (26) to formate (27) (Fig.10). Oxalate occurs as a byproduct of photo-respiration in plants (spinach is a rich source)and is produced by many microorganisms andfungi. Several species of Penicillium and As-

pergillus convert sugars into oxalic acid in highyields. Hyperoxaluria is a major risk factor inhuman urinary stone disease (SHARMAt al.,1993). The enzyme is readily available and isused in kits for determining oxalate in urine(SANTA-MARIAt al., 1993).The biotechnolog-

ical potential of this enzyme remains to be ex-

ploited, however, differential binding of car-boxylate substituted biomimetic dyes indicatesthat substrates other than oxalate may be ac-cepted (LABROU nd CLONIS,995). Furtheraspects of oxalate decarboxylation are coveredin Sect. 4.1.1.8,Oxalyl-CoA Decarboxylase.

4.1.1.3 Oxaloacetate

Decarboxylase (Oxaloacetate

Carboxy-Lyase, EC 4.1.1.3)

Oxaloacetate plays a central role in primarymetabolism as an intermediate in the citric ac-id (Krebs) cycle (Fig. 1) and as the “startingmaterial” for gluconeogenesis.All amino acidsexcept for leucine and lysine can be convertedto oxaloacetate. The majority are converted tooxaloacetate via citric acid cycle intermedi-ates. Aspartate and asparagine are convertedvia other pathways and the remainder are con-

verted via pyruvate and carboxylation to oxa-



56 2 Lyases

9 Ytide0

decarboxylase

20Fluoropyruvate

I0HO

H20- l

0

24

2s

9 n i d e

Oxalate

CO,26 Oxalate 27 Formate

Fig. 10. The decarboxylation of oxalate (26) to for-mate (27) catalyzed by oxalate decarboxylase.

loacetate. There are two classes of enzymeswhich are capable of the interconversion ofcarbon dioxide and pyruvate (1)with oxalo-acetate (28) (Fig. 11). Pyruvate carboxylases(EC 6.4.1.1) are ligases and hence carboxyla-tion (which is thermodynamically disfavored),

is coupled to the hydrolysis of ATP to ADP.Whereas oxaloacetate decarboxylases (OAD,oxaloacetate carboxy-lyase, EC 4.1.1.3), cata-lyzes the rev erse an d thermodynamically morefavorable reaction. The latter enzym es are notable to catalyze the ne t carboxylation of pyru-vate.

The re are four classes of enzym es that cata-lyze the decarboxylation of oxaloacetate. Themost important are the OADs that requiremagnesium(I1) or m anganese(I1) ions and a reconsequently sensitive to ED TA (e.g., Micro-

coccus lysodeikticus O A D et al., 1999;Coryne-bacterium glutamicum O A D et al., 1995) andthe b iotinylated protein (TOH t al., 1993) lo-cated on the cytoplasmic membrane, which

0 0 28 Oxaloacetate

n 1PvruvateFig. 9. The mechanism for the decarboxylation offluoropyruvate (20) to acetate (21) y pyruvate de-carboxylase (GISH t al., 1988;SUN t al., 1997); cf.Figs.3,4,5 or R' and R2 he sequence 110 20 .

Fig. 11.The decarboxylation of oxaloacetate (28) topyruvate (1) atalyzed by oxaloacetate decarboxyl-

ase.




acts as a sodium ion pump (DIMROTH,994;DIMROTHnd THOMER, 1993). In addition L-

malate dehydrogenase (L-malic enzyme, EC1.1.1.38; EDENS t al., 1997; SPAMINATOnd

ANDREO,995) and pyruvate kinase in m uscle,both have ox aloacetate decarboxylating activ-ity, but oxaloacetate is not a substrate in vivo(KRAUTWURSTt al., 1998).

The mechanism of the OAD catalyzed de-carboxylation of oxaloacetate (28) involvesthe formation of a magnesium(I1) or manga-nese(I1) a-ketoacid complex (29) (Fig. 12),which undergoes retro-aldol cleavage to givethe m agnesium enolate (30) lus carbon diox-ide.This mechanism is quite different t o that ofoth er P-k etoacid deca rboxylases (e.g., acetoac-

eta te decarboxylase) which involve imine for-mation by the ketone to initiate decarboxyla-tion. Imine based OADs, have been preparedby rationale pep tide design, but the rates of re-action are much slower than those of com par-able enzymes (ALLERTt al., 1998; JOHNSONet al., 1993).

Oxaloacetate (28) enolizes rapidly in solu-tion and hence special precautions are re-quired to e lucidate the stereochemistry of de-carboxylation. This was achieved in a rema rk-able sequence catalyzed by five enzymes. Incu-bation of (2S)-[2-3H,3-3H2]-asparticcid (31)with L-aspartase in deuterium oxide gave ster-eospecifically labeled (2S,3S)-[2-3H,3-2H,3H]-aspartic acid (32) ia fum arate (Fig. 13).Trans-amination yielded (3S)-[3-'H,3H]oxaloacetate(33),which was immediately decarboxylatedin sifu with the OAD from Pseudomonas puti-da to give (3S)-[3-lH,2H,3H]-pyruvate(34). oprev ent$ nolization an d loss of the radiolabels,this was reduced with L-lactate dehydro genaseto (2S,3S)-[3-lH,2H,3H]-lactate35), with

which was isolated by anion exchange chrom a-tography.Oxidative decarboxylation with L-lactate

oxidase gave (2S)-[2-'H:H,3H]-acetate (36)

(Fig. 14). The chirality of w hich was de ter-mined by CORNFORTH'Sethod (CORNFORTH

et al., 1969;LUTHYt al., 1969), which utilizes afurther four enzym es. Decarboxylation byOAD proceeds with inversion of stereochem-istry, whereas the corresponding reaction withthe biotin dependent pyruvate carboxylasesand malic enzyme all proceed with retention

of configuration (PICCIRILLIt al., 1987).

0 0 28Oxaloacetate

Oxaloacetatedecarb0 y1asIq2+

$.0

0

0 1 Pyruvate

Fig. 12.Mechan ism for the decarboxy lation of oxa-loacetate (28) to pyruvate (1)catalyzed by oxaloac e-tate decarboxylase. Magnesium(I1) bound to OAD

is shown w ithout ligands for the pu rposes of simplic-ity only. It is m ost likely that water or carboxylate li-gands are displaced when oxaloacetate is bound.



58 2 Lyases

AspartateI

32 (2S,3S)-[2-3H,

3-ZH,3H]-Aspartate

Glutamate-oxdoacetatetransaminase

a-Ketoglutarate

k -Glutamate

Oxdoacetate

Oxaloacetatedecarboxylase

co2

Pyruvate

NADH

L-Lactate dehydrogenase

NAD@

ZH,3H]-Lactate

Fig. 13. Stereoselectivity of the decarboxylation oflabeled oxaloacetate (33) catalyzed by Pseudomo-nus putid a OAD.

2H,3H]-Lactate

0 2

CO,, H2O

L-Lactate oxidase

36 (2s)-[2-'H;H,3H]-AcetateFig. 14. Oxidative decarboxylation of lacta te (35)yields acetate (36) f known stereochemistry.

4.1.1.4Acetoacetate Decarboxylase

(Acetoacetate Carboxy-Lyase,

EC 4.1.1.4)

Acetoacetate decarboxylase (AAD) cata-lyzes the decarboxylation of acetoacetate toacetone, a process which parallels the abioticdecarboxylation of acetoacetic acid (37s) (Fig.15). The salts and esters of p-ketoacids aregenerally stable to decarboxylation, whereasthe corresponding acids decarboxylate sponta-neously. Consequently, this reaction is most of-ten employed during the acid catalyzed hy-drolysis of p-ketoesters (37c).The mechanismof th e abiotic reaction is fairly well understood

(HUAN G t al., 1997;BACH nd CANEPA,996).Intramolecular pro ton transfer yields the enol-ic form of the ketone (%a), plus carbon diox-ide, in a process which is in effect a retro-a ldolreaction. The obligatory nature of the protontransfer can b e inferred from the correspond-ing reaction of trimethylsilyl p-ketoesters(37b), which gives isolable trimethysilyl enolethers (38b). It might be anticipated that theenzyme catalyzed process would simply in-volve protonation of the p-ketoacid saltswhich a re the predominant form at physiolog-

ical pH. However, early experiments, demon-




strated that exchange of the ketonic oxygenwas an obligatory step in the decarboxylationof acetoacetate by AAD (HAMILTONndWESTHEIMER,959), which is not consistant

with a simple protonation mechanism. It isnow well established, tha t reaction of acetoac -etate (40) (Fig. 16) with a lysine residue ofAAD 41), gives a proton ated imine (Schiff’sbase) derivative (42), which undergoes decar-boxylation to give the ena mine (43), which inturn is hydrolyzed to acetone (39). The pH op-timum for Clostr idium acetobutylicum AAD s5.95, wherea s the pKa for the e-am ino group oflysine in solution is 10.5, hence the effectivepK, at the active site must be reduced by a fac-tor of at least 20,000 to enable the &-amino

group of lysine to act as an effective nucle-ophile (WESTHEIMER,995). Clostridium ace-

tobutylicum AAD ears adjacent lysine resi-dues at the active site. Lysine-115 acts as thenucleophile, whereas the &-am monium groupof lysine-116 serves to reduce the pKa of theadjacent residue by electrostatic effects(HIGHBARGERt al., 1996). This red uction inpKa has also been exploited in the design of1,3-diarninonorbornane decarboxylation cata-lysts (OGINO t al., 1996). Catalytic an tibodie s

which are c apable of catalysing aldol reactionsand decarboxylation, also bear deeply buriedlysine residues w hich are essential for catalyticactivity (BARBAS11 e t al., 1997;BJORNESTEDTet al., 1996)

The decarboxylation of (2R) and (2S)-[2-3H]-ace toacetate in deuterium oxide catalyzedby AAD ives racemic [2-’H,2H,3H]-acetate.A eries of control experiments dem onstratedunambigously that this was a consequence ofthe decarboxylation step and was not due tohydrogen exchange reactions w ith the solvent

(ROZZELLr and BENNER, 984). In contrastthe homolog, racemic 2-methyl-3-oxobutyrate(44) (45) (Fig. 17) is decarboxylated enantiose-lectively. Treatm ent of th e crude reac tion m ix-ture with sodium borohydride gives a mixtureof diastereomeric alcohols (46) 48) whichboth have th e (2S)-configuration and w hen thereaction is run in deuterium oxide, (3R)-[3-’H1-2-butanone (47) is produced, indicatingthat decarboxylation occurs with retention ofstereochemistry (BENNERt al., 1981).Similar-ly, racemic 2-oxocyclohexanecarboxylate(49)

(50) (Fig. 18) undergoes enantioselective de-

a R = H; Acetoacetic acidb R = Me3Sic R = alkyl

+

0 38a R = H

b R = Me3SiIR = H

1 9Acetone

Fig. 15.Abiotic mechanism for the decarboxylationof acetoacetic acid (37a).

carboxylation as demo nstrated by recovery ofthe (1s)-diastereomeric alcohols (51) (53) af -ter sodium cyanoborohydride reduction. De-carboxylation of the deuterio-derivative (54)in water gives (2S)-[2-*H]-cyclohexanone55)(Fig. 19), again with rete ntion of stereochemis-try (BENNER nd MORTON,981). The bio-

transformations potential of AAD as thus faronly been investigated with a few “unnaturalsubstrates”. The ability to d ecarboxylate sub-strates such as 2-oxocyclohexanecarboxylate

(49) (SO), which are structurally far removedfrom the natural substrate (40) suggests it maybe app licable to a diverse range of P-ketoacids.

In th e first half of the twentieth century, ace-tone was manufactured by fermentation, butthis technology was supp lanted in the 1950s bymore cost effective abiotic processes based o npetrochemicals. With increasing concerns

about enviromental impacts and the need for



60 2 Lyases

40 Acetoacetate

R- NH 2

4 1 Acetoacetatedecarbox ylase

R H

h42Irnine

Ico2R H

N'I

A 43Enamine

k H 2 0-NH,

41 Acetoacetatedecarboxylase

0A 39Acetone

I (-)-44- (+)-45

Acetoacetatedecarboxylase

NaBH4 (2H20)

'I

I 48

Fig. 17.Enantioselective decarboxylation of racemic2-methyl-3-oxobutyrate (44) 45).

(+)-49 (-)-SO

Acetoacetate

decarboxylas

NaCNBH3

1

82co:

ig. 16. Mechanism for th e decarboxylation of ace- OHtoacetate (40) y acetoacetate decarboxylase.

the development of renewable resources, the ( + ) A

fermentation process is now being reexamined(Reviews: DUERRE, 1998; GIRBALnd Sou-CAILLE, 1998;SANTANGELO and DUERRE, 1996;WOODS, 995;DUERREt al., 1992). The origi-

based fermentation was Clostridium acetobu-tylicum, however, later strains were grown onsucrose (molasses) and this has led to some ge-netic drift such that 39 cultures from various

collections can now be divided into four

nal organism used for the starch (corn mash) (-)-53

Fig. 18.Enantioselective decarboxylation of racemic2-oxocyclohexanecarboxylate 49) (50).



62 2 Lyases

Acetoacetyl-CoA:acetateibutyrate P-hydroxybutyvl-

CoA dehydrogenase HO 0CoA transferase e e

58OASH

56 57 A cetoacetyl-CoA

Crotonase

CoASH

1

00 40AcetoacetatedSCoA

59

Butyvl-CoAdehydrogenase

co2

El9 Acetone

Axo*

I I 6o

CoA-acy latingaldehyde

dehydrogenaseButanol

63 Butanol 62 Butyraldehyde

Butyraldehydedehy drogenase

61 Butyrate

Fig. 20. The production of acetone (39) and bu tanol(63) by C.acetobutylicum and related species. Isopropa-no1(64)s most frequently formed by C. beijerincki (CoASH = coenzyme A).

4.1.1.5 Acetolactate Decarboxylase

(2-Hydroxy-2-Methyl-3-

Oxobutyrate Carboxyl-Lyase,

EC 4.1.1.5)

Acetolactate decarboxylases from Klebsiel-la aerogenes (formerly Aer obacfe r aerogenes)

catalyzes the decarboxylation of (S)-a-aceto-

lactate (65) to give (W-acetoin (5) (Fig. 21).(2S)-a-Acetolactate (65) is decarboxylatedwith inversion of configuration (CROUT t al.,1984; ARMSTRONGt al., 1983) and the carbox-ylate group is substituted by a proton from thesolvent (DRAKE t al., 1987). Surprisingly,(2R)-a-acetolactate (66) is also decarboxylat-ed to (R)-acetoin (5), but at a slower rate(CROUT t al., 1986).




Acetolactatedecarboxylase

co2

Fig. 21. Decarboxylation of a-acetolactate [(2-hy-droxy-2-methyl-3-oxobutanoate]65) (66) to ( R ) -acetoin (3-hydroxy-2-butanone) (5), catalyzed by ac-etolactate decarboxylase.

A clue to the mechanism was provided bythe decarboxylation of racemic a-acetohy-droxybutyrate (67) (68) with acetolactate de-carboxylase from a Bacillus sp. (Fig. 22). The(S)-enantiomer (67) was decarboxylated rap-idly to give the k eto l(6 9), which was followed

by slow decarboxylation of the @ )-enantiom-e r (68) o give the isomeric ketol (71). Bothketols were form ed with high enantiomeric ex-cesses. This apparently anomo lous result canbe rationalized by a stereospecific enzyme cat-alyzed tertiary ketol rearrangement of (68)with transfer of the carboxylate group to there-face of the ketone to give the a-propiolac-tate (70). This compound which is a homologof (S)-a-acetolactate (65) then undergoes de-carboxylation by th e “normal” pathway. W hen(R)-a-acetolactate (66) s subjected to the ter-

tiary ketol rearrangement, the enantiomeric(S)-a-acetolactate (65) is formed and hencethis also explains the ability of this enzyme totransform both enantiomers of a-acetolactate(CROUTand RATHBONE,988). An abioticbase catalyzed tertiary ketol rearrangement(CROUTnd HEDGECOCK,979) was excludedby control experiments.

a-Acetolactate (65) is produced biosynthet-ically by the condensation of two m olecules ofpyruvate (1 ) with concomittant decarboxyla-tion, catalyzed by acetolactate synthase [ace-

tolactate pyruvate-lyase (carboxylating); EC


co2

’V E tO

p 70

Et

HO H 71 HO ’+

Fig. 22. Decarboxylation of (2S)-a-acetohydroxy-butyrate [(2S)-2-ethyl-2-hydroxy-3-oxobutanoate](67),nd tertiary ketol rearrangementof (2R)-a-ac-etohydroxybutyrate(a),o (2S)-2-hydroxy-2-meth-yl-3-oxopentanoate (70) and decarboxylation cata-

lyzed by acetolactate decarboxylase.

4.1.3.181. The enzyme is TPP dependent andhence th e mechanism is very similar to that ofpyruvate decarboxylase described in Sect.4.1.1.1 (CARROLLt al., 1999; CHIPMANt al.,1998; HILL t al., 1997) (Fig. 23). a-Acetolac-tate (65) is the biological precursor of diacetyl(72), which is a desirable aroma in some fer-mented diary products, but an off-flavor inbeer. Acetoin has no flavoriimpact on beer(HUGENHOLTZ,993).The conversion of a-ac-

etolactate (65) to diacetyl (72) is an oxidativedecarboxylation. Plausibly the anionic inter-mediate of decarboxylation of a-acetolactate(65) is intercepted by oxygen (BOUMERDASSIet al., 1996) o r peroxide acting as electrophiles.This process apparently d oes not occu r duringdecarboxylation by acetolactate decarboxyl-ase. a-Acetolactate (65) undoubtedly under-goes spontaneous, extracellular oxidative de-carboxylation (SEREBRENNIKOVt al., 1999),but th e evidence is accumulating for a parallelintracellular enzyme (RONDAGSt al., 1997,

1998; JORDAN et al., 1996) or quinone cata-



64 2 Lyases

0 lpyruvate

Acetolactatesynthase

co2

[a]

1 Pyruvate t

O,?

co2 coz


5 (R)-Acetoin 72 Diacetyl

Fig. 23. The biosynthesis of (2S)-c~-acetolactate(S)-

2-hydroxy-2-methyl-3-oxobutanoate]65) from pyr-uvate ( l ) ,atalyzed by acetolactate synthase. No n-oxidative decarboxylation to give @)-acetoin (5), iscatalyzed by acetolactate decarboxy lase. Oxidativedecarboxylation to diacetyl (72) occurs spontane-ously, but may also be catalyzed by an unidentified

lyzed process (PARK t al., 1995). Conversionof acetoin to diacetyl or vice versa does notseem to be a significant pathway in most mi-croorganisms.

The conversion of a-acetolactate (65) to ace-toin (5) reduces the “pool” of a-acetolactate,available for conversion to diacetyl, hence byregulating the levels of acetolactate decarbox-

ylase, it should be possible to control the levelsof diacetyl (RENAULTt al., 1998;BENITEZtal., 1996; KOK,1996). However, selection forLactobacillus rhamnosus mutants with lowlevels of acetolactate decarboxylase activityhad no affect on diacetyl levels (MONNETndCORRIEU,998;PHALIPt al., 1994).Acetolac-tate accumulates, but is not converted to di-acetyl unless the redox potential is high (MON-NE T et al., 1994).In contrast incorporation ofthe acetolactate decarboxylase gene fromAcetobacter aceti into the genome of brewers’yeast successfully reduced the levels of diacet-yl (TAKAHASHIt al., 1995; YAMANO t al.,1994, 1995). Moreover the free enzyme fromBacillussubtilis containing the structural genefor ALDC production originating from a Ba-cillus brevis strain is now used directly in beerbrewing to reduce diacetyl levels (D EBOER tal., 1993;GERMAN,992).

There is considerable unrealized potentialfor the use of both acetolactate synthase andacetolactate decarboxylase in synthesis

(CROUT t al., 1994)

4.1.1.6 Aconitate Decarboxylase

(cis-Aconitate Carboxy-Lyase,

E C 4.1.1.6)

Aconitate decarboxylase from Aspergillusterreus catalyzes the decarboxylation of cis-aconitate (73) to itaconate (74) (Fig. 24).Therehas been considerable interest in the produc-

tion of itaconate (74) by various strains of A .terreus for use as a monomer for plastics man-ufacture. However, the enzymology has largelybeen neglected, because of the instability ofthe enzyme (BENTLEYnd THIESSEN,957).The generally accepted mechanism for the de-carboxylation involves an allylic shift of the al-kene bond with protonation on the (2-re,3-se)face of cis-aconitate (73) (RANZI t al., 1981).However, recent work in which the enzymewas immobilized in a silicate matrix, has indi-cated that the initial decarboxylation yields ci-traconate (75), which is subsequently isomer-




co:

O o 2 4 z 73 cis-Aconitate

I \'H,A 0 *H

Aconitatedecarboxylase

-2H 74 Itaconate

75 Citraconate

Fig. 24. Decarboxylationof cis-aconitate (73) to itac-onate (74) catalyzed by aconitate decarboxylase.

ized to itaconte (74) by a previously unknownenzyme (BRESSLER nd BRA UN, 996). Thismechanistically unlikely result rem ains to beconfirmed.

cis-Aconitate is a citric acid (Krebs) cycleintermediate (Fig. 1) which is produced pre-dominantly in the mitochondria, whereasaconitate decarboxylase is located exclusivelyin the cytosol (JAKLITSCHt al., 1991). Mu chwork has b een devoted to increasing the pro-ductivity an d efficiency of produ ction of itaco-nate (74) from carbohy drate foodstocks (BON-NARME et al., 1995). Selection o n itaconic acidconcentration gradient plates yielded A. terre-

us TN-484, which gave 82 g L-' itaconic acidfrom 160 g L-' glucose (51% efficiency) in ashaking flask (YAHIRO t al., 1995). Althoughthis is a goo d level of produ ctivity and efficen-cy, the challenge is to achieve the same resultswith cheaper foodstocks. Corn starch gave thebest results for a range of starchy materialswith A . terreus NR R L 1960, but the productiv-ity was only 18g L-' and the efficiency 34%(PETRUCCIOLIt al., 1999).Pretreatment of thethe corn starch with nitric acid enabled morethan 60 g L-' itaconic acid to be produced

from 140 g L-' corn starch (50% efficiency)

using A. terreus TN-484; a result which is al-most equivalent to th at achieved with glucose(YAH IRO et al., 1997). Protoplast fusionbetween A. terreus and A. usamii (a gluco-

amylase producer), produced fusant mutantswhich were subcultured to confirm stability.The F-112 mutant maximally produced 35.9g L-' of itaconic acid from 12 0 g L -' solublestarch (KIRIMUR At al., 1997).The other ma-jor factor in itaconate production is aeration.Interuption of a eration for five minutes causedthe cessation of itaconate prod uction an d thiswas only restored after a furthe r 24 hours a er-ation. Moreover when protein synthesis wasalso inhibited after interuption of aeration n ofurther itaconate production occurred (GYA-

Given these results it is not surprising that airlift reactors have bee n favored for large scaleconversions (PARK t al., 1994; OKABEt al.,1993) O n a small scale a 10 mm air lift reactorhas been mo nitored directly by 31P-NM R at161 MH z (LYNSTRADnd GRASDALEN993).

MERAH et al., 1995; BONNARMEt al., 1995).

4.1.1.7 Benzoylformate

Decarboxylase (BenzoylformateCarboxy-Lyase, EC 4.1.1.7)

Benzoylformate decarboxylase (BFD) is arare enzyme which catalyzes the decarboxyla-tion of benzoylformate (phenyl glyoxalate;78)to benzaldehyde (4) (Fig. 25). The most in-tensely investigated example is the B FD fromthe mandelate pathway of Pseudomonas puti-da .

Phenylacetate (76) is a common intermedi-

ate in the microbial metabolism of various ar-omatic compounds, including phenylalanine.In Pseudomonas putida, phenylacetate is de-graded by benzylic hydroxylation to mande-late (77), oxidation to benzoylformate (78)

(LEHOUnd M ITRA, 999),decarboxylation t obenzaldehyde (4) catalyzed by BFD and oxi-dation to give benzoate (79),which is oxidizedfurthe r by ring fission (Ts0 u et al., 1990). Som eother organisms utilize nominally similar path-ways, but with a different complement of en-zymes. For exam ple, the denitrifying bacterium

Thauera arom atica utilizes a m emb rane bound



66 2 Lyases

0. O

76 Phenylacetate

Co Aderivatives

I OH

( 2 9 - 7 7

Mandelate

NAD(P)H

CoAderivatives

0 78 Benzylformate

Benzo ylformatelyase

CO,

n

Benzaldehydedehydrogenase

79 Benzoate

NAD(P)

NAD(P)H

- o@

Fig. 25. Generalized pathway for the catabolism ofphenylacetate (76).

molybdenium-iron-sulfur prote in, which con-verts phenylacetyl-CoA directly to benzylfor-mate (78) in the absence of dioxygen (RHEE

and FUCHS, 999).An other denitrifying bacte-rium A zocarus evensii, catalyzes the oxidativedecarboxylation of benzylformate (78) to ben-zoyl-Co A with a six subu nit iron-sulfur en-

zyme (HIRS CH t al., 1998).BFD from Pseudomonas putida is a tetra-

meric TP P (thiamine pyrophosphate) depen-dent decarboxylase composed of 57 kD amonomers (HASSONt al., 1995,1998).As w ithother TPP dependent decarboxylases theintermediate enamine can be intercepted byexogen ous aldehydes to give acyloins (cf. pyru-vate decarboxylase, Sect. 4.1.1.1). The acyloinconden sation with acetaldehyde (2) ives (S) -

(-)-2-hydroxypropiophenone (80) with up to92% enantiomeric excess (Fig. 26). Thus far,

only p-substituted benzylformates have beenshown to act as substrates an d pyruvate; a-keto-glutarate and a-ketobutryrate are not sub-strates (REYNOLDSt al., 1988; WEISSet al.,1998).The fa irly limited synthetic applicationsof BFD have been reviewed (IDING t al., 1998;SCHORKENnd SPRENGER,998; SPRENGERand POHL, 1999). [p-(Bromomethy1)ben-zoyllformate (81s) eacts with BFD at about1% of the ra te of benzy lformate (78) and caus-es inhibition which can be reversed by the ad-

dition of TPP. The “enamine” (82) which isforme d by the “n orm al” mechanistic sequence,undergoes elimination of bromide to give atransient quinone-dimethide (83), which tau-tomerizes to the acyl derivative and und ergoesslow hydrolysis via the h ydra te (84) to give p-methylbenzoate (85) (Fig. 27). This mechanismof this reaction parallels the conversion of 2-fluoropyruvate (20) o acetate (21) Fig. 9) bypyruvate decarboxylase, however, [p-(fluo-romethyl)benzoyl]formate (81b) undergoesdecarboxylation without elimination of fluo-

ride, which suggests that the intervening phe-nyl group imposes an energy barrier w hich im-pedes elimination (DIRMAIERt al., 1986;RE-

The production of (S)-(-)-2-hydroxypropi-opheno ne from benzylformate (78) and acetal-dehyde (2) by A cinetobacter calcoaceticus, hasbeen optimized at p H 6.0 and 30 O C. Maximalproductivity was 8.4 g L-’ after two hours. Inone hour, 6.95 g L-’ of (S)-(-)-2-hydroxyprop-iophenone (80) was formed which corre-sponds to a productivity of 267 mg per g of dry

cells per hour (PROSEN nd WAR D,1994).

YNOLDS et al., 1998).




n

&O@ 78 Benzylformate

Benzoy lformatedecarboxylase

0 2 Acetaldehydeo2

80 (9-(-)-2-Hydroxypropiophenone

Fig. 26. Acyloin condensation of benzylformate (78)

and acetaldehyde (2) catalyzed by benzylformatedecarboxylase.

Whole cell and cell extracts of Pseudomonasputida ATCC 12633 also catalyze this transfor-mation, but benzaldehyde was a major by-product together with small amounts of benzyl

alcohol (WILLCOCKSt al., 1992, WILLCOCKSand WARD 992).

4.1.1.8 Oxalyl-CoA Decarboxylase

(Oxalyl-CoA Carboxy-Lyase,

EC 4.1.1.8)

Oxalyl-CoA decarboxylase (OCD) catalyz-es the decarboxylation of oxalyl-CoA (86) toformyl-CoA (87Fig. 28) and is TPP depen-

dent. This offers the prospect that the “ena-mine” intermedaite (cf. 11, Fig. 4), which isfunctionally equivalent to a formate anion (88)could be trapped by electrophiles, such as alde-hydes (cf. Sect. 4.1.1.1, pyruvate decarboxyl-ase). The most heavily investigated OCD isthat from Oxalobacter formigenes, which is agram-negative anaerobe found in the gut,which is able to grow using oxalate as the soleenergy and carbon source, if small amounts ofacetate are also available (CORNICKnd ALLI-SON, 996; CORNICKt al., 1996).In humans the

absence of 0.formigenes, in the gut is a major

MR2R t N&S 9 Ylide

0

Benzylformate fr Goyodecarboxylase

81

a X = B rb X = F

Br@/I 82

83

84

’ ’ 85 p-methylbenzoateR1/ ‘LS

6 9 Ylide

Fig. 27. The mechanism for the decarboxylation of[p-(bromomethyl)benzoyl]formate@la) to p-meth-ylbenzoate (85) by benzylformate decarboxylase(B FD ) cf. Figs. 3,4,5 or R’ nd R2 nd the sequence9 -+ 10 --t 11+ 12 -+ 9 and Figs. 8,9 for the corre-

sponding reaction of fluoropyruvate (20).



68 2 Lyases

Oxalyl-CoAdecarboxylase

COZ

0

CoASA H 87Formyl-CoA

Fig. 28.Oxalyl-CoA decarboxylase.

risk factor for hyperoxaluria and calcium oxa-late kidney stone disease (SIDHU t al., 1998,1999).

O CD in 0. ormigene is located in the solu-ble cytoplasmic fraction (BAET Z nd ALLISON ,

1992) and is linked to a transport system(OxlT; RU AN t al., 1992),which exchanges ex-tracellular oxalate for intracellular formate.Decarboxylation of oxalyl-CoA (86 ) to for-myl-CoA (87), followed by hydrolysis con-sumes a proton and hence transport of for-ma te out of th e cell develops a n internally neg-ative mem brane p otential. In effect, formate isacting as a proton carrier and hence transp ortplus decarboxylation performs the function ofan indirect proton pump (Fu and MALONEY,1998). However, there is no obligatory rela-

tionship between decarboxylation and thetransport system, because functional OC D wasproduced when a PCR fragment containingthe gene (oxc) was overexpressed in E. coli(BA ETZ nd PEC K, 992; cf. SIDH U t al., 1997;LU NG t al., 1994).

It is interesting to sp eculate if th e formateanion equivalent (88) is involved in the biosy n-thesis of com mo n cellular con stituents such asCitric acid (K rebs) cycle interm ediate s (Fig. 1)and amino acids. 14C labeling studies indicatethat some 54 % of the total cell carbon is de-

rived from oxalate and at least 7% from ace-

tate. Carbo nate is assimilated but not formate.Apparently, there are multiple pathways forthe assimilation of oxalate (COR NICKnd AL-LISON, 996; CORNIC Kt al., 1996).The strictly

anerobic, obligatory oxalotrophic bacterium,Bacillus oxalophilus, utilizes tartronate semi-aldeh yde synthase (E C 4.1.1.47; Sect. 4.1.1.24)or the serine pathway to assimilate oxalate andOCD and formate dehydrogenase for oxida-tion (ZAITSEVt al., 1993).

4.1.1.9 Malonyl-CoADecarboxylase (M alonyl-CoACarboxy-Lyase,EC 4.1.1.9)andRelated Malonate D ecarboxylatingEnzymes

Malonyl-CoA decarboxylase (MCD) cata-lyzes the decarboxylation of malonyl-CoA(90) to acetyl-CoA (56;Fig. 29). Some forms ofthis enzyme are also able to act as malonate-C oA transferase (E C 2.8.3.3), which catalyzesthe transfer of coenzyme A from acetyl-CoA(56) to malonate (89). M CD is not able to cat-

alyze the net carboxylation of acetyl-CoA.This requires the biotin dependent enzyme,acetyl-CoA carboxylase, in which the thermo -dynam ically disfavored carb oxylation is linkedto the hydrolysis of ATP to A D P (EC 6.4.1.2).

The mechanism for th e abiotic decarboxyla-tion of malonic acid (92) proceeds via a six-membered transition state (Fig. 30; HUANGtal., 1997) in which a pro ton is transferred fromthe carboxylate group undergoing decarboxy-lation to the incipient carboxylate group of ac-etate (21) n a concerted sequen ce of bond mi-

grations (GOPOLAN,999). This m echanism isessentially identical to that of the decarboxyla-tion of acetoacetate (37a;Fig. 15). Despite therelative ease of the a biotic reaction an d the po-tential to catalyze it by protonation and aug-mented proton transfer, there d o not app ear tobe any decarboxylases in which malonateundergoes direct decarboxylation. The enzy-matic decarboxylation of malonate apparentlyrequires obligatory formation of the corre-sponding thioester to facilitate the decarboxyl-ation. Th e ability of th e sulfur atom of the thi-

oester to acts as an electron sink and hence fa-




0 0

CoAS 0" 90Malonyl-CoA

Malonyl-CoA

decarboxylaseCOZ

0 0

CoAS 91 Acetyl-CoA enolate

CoASA 6 Acetyl-CoA

Fig. 29. Malonyl-CoA decarboxylase catalyzes thedecarboxylation of malonyl-CoA (90) to acetyl-CoA enolate (91) which undergoes protonation togive acetyl-CoA (56).Some forms of the enzyme arealso able to catalyze metathesis of m alonate (89) andacetyl-CoA (56)with malonyl-CoA (90) and acetate

(21).

cilitate the reaction is well established (BACHand CANEPA,996), nevertheless, the absenceof enzymes able to catalyze the direct decar-boxylation of malonate is surprising.

MCDs have been found in mammalian,avian and plant tissues, plus various microor-ganisms (KOLA-I-I-UKUDYt al., 1981). In hu-mans, MCD deficiency results in mild mentalretardation, seizures, metabloic acidosis andexcretion of malonate in the urine (GAOet al.,

1999b;FITZPATRICKt al., 1999;SACKSTEDERt

xHI

HO0 93

Fig. 30.Abiotic m echanism for the decarbox ylationof m alonic acid (37a).

al., 1999). In mammals,MCD and acetyl-CoAcarboxylase regulate the levels of malonyl-CoA in muscular tissues as part of a complexsystem regulating glucose and fatty acid me-tabolism (GOODWINnd TAEGTMEYER,999;ALAMnd SAGGERSON,998).

MCD activity has been described for Myco-

bacterium tuberculosis, Pseudomonas ovalis(TAKAMURAnd KITAYAMA,98l), Pseudo-monas fluorescens (KIMet al., 1979),Pseudo-monas putida (CHOHNANt al., 1999),Sporo-musa malonica, Klebsiella oxytoca and Rho-

dodbacter capsulatus (DEHNINGnd SCHINK,1994), Citrobacter divs. (JANSSENnd H AR-FOOT, 1992),Acinetobacter calcoaceticus (KIMand BYUN,1994),Malonomonas rubra (DEH-NING and SCHINK,989) and Klebsiella p neu -moniae.

There is some controversy over the exact

nature of the MCD activity in microorganisms,



70 2 Lyases

which has arisen from work with two speciesthat utilize malonate as sole energy source(DIMROTH nd HILBI, 1997). Malonomomasrubra is a microaerotolerant fermenting bacte-

rium which can grow by decarboxylation ofmalonate to acetate under anaerobic condi-tions. It is phylogenetically related t o the sulfurreducing bacteria (KO LB t al., 1998).Klebsiel-la pneumoniae is able t o grow aerobically onmalonate, or anaerobically if supplementedwith yeast extract. Both of these species areable to decarboxylate malonate without theintermediacy of malonyl-CoA. In each case,ma lonate is decarboxylated as the thioester ofan acyl-carrier protein (ACP), which bearsan acetylated 2‘-(5 ”-phosphoribosyl)-3 ‘-de -

phosphocoenzyme A prosthetic group (94b)(Fig. 31) (BE RG et al., 1996; SCHM ID t al.,1996). This gro up w as previously identified asa component of K . pneumoniae citrate (pro-3s)-lyase (ROBINSO Nt al., 1976).The K .pneu-moniae malonate decarboxylase consists offour subunits (a-8) nvolved in the catalyticcycle (Fig. 32), plus a furthe r enzym e ( M d c H )which transfers malonyl-CoA (90) to the thiolform of the ACP (94a) (HOENKEnd DIMROTH1999). Decarboxylation then yields the acetyl

derivative (94b),which undergoes trans-acyla-tion with m alonate (89) to comm ence the cata-lytic cycle (HO EN KE t al., 1997). In this se-quence, there is no m eans by which the energyreleased by decarboxylation (DG” ’= - 7.4

kJ mo l-l) can be “harvested” and it is lost asheat. This is in any case, unnecessary becausemore than sufficient energy is created by theaerobic oxidation of acetate to carb on dioxide

and w ater in th e citric acid cycle and respirato-ry chain.

The decarboxylation system of the ane robeM . rubra consists of water soluble cytoplasmiccom ponents (some of which resemble those ofK . pneurnoniae) plus membrane bound com-ponents which are involved in energy harvest-ing. Acetate and ATP are required for thepriming step in which ACP-SH (94a) is con-verted to the acetyl form (94b) (Fig. 33). Thisundergoes metathesis with malonate to givemalonyl-ACP (94c) and decarboxylation is

linked to the transfer of carbon dioxide toMadF, the biotin carrier protein (BCP) (96)

with regeneration of acetyl-ACP (94b). Thecarboxybiotin protein (95) is believed to dif-fuse to the mem brane, where it is decarboxyl-ated by MadB in an exo gonic reaction which islinked to sodium ion pumping across themembrane. The Na+ gradient so formed isthen exploited for ATP synthesis by the FlFOATP synthases. This process has been termeddecarboxylative phosphorylation and is re-

sponsible for all ATP synthesis in several an-aerobic bacteria (DIM ROT Hnd SCHINK,998).It is claimed that the app aren t MC D activity

of Acinetobacter calcoaceticus, Citrobacterdivs., K . oxytoca, and some Pseudomonas spp.

b R = A ccR = malonyl

Fig. 31.2‘45 ” -phosphoribosyl)-3 -dephosphocoenzymeA and derivatives (94a-c) sterified with a serine

residue of the acyl camer protein. This is the prosthetic group of the acyl carrier protein of the malonate

decarboxylases of Malonomomas rubra (BERGet al., 1996) and Klebsiella pneum oniae (SCHMIDt al., 1996).

It was first identified as the prosthetic group of Klebsiella pneumoniae citrate (pro-3S)-lyase

(ROBINSONt al., 1976).



4 Examples ofLyases 71

89 Malonate

21 Acetate

0

MdcA (a ) MdcD. E (P, r)

94c

HSCoA

MdcH 1 0

' .SCa4 90Malonyl-CoA

HS-ACP 94a

Fig. 32. Propose d reaction mechanism for malonate decarboxylation by Klebsiella pneurnoniae. The genecluster cosists of eight consecutive genes which hav e been termed rndcABCDEFGH and th e divergently or-ientated rndcR gene.The anticipated functions of the gene products a re as follows (functional subunits of theenzyme ar e shown in brackets): M d c A , acetyl-S-acyl carri er protein: malo nate acyl carrier pro tein transferase(a); dcB, involved in the synthesis and attachment of the prosthetic group; MdcC, acyl carrier protein(ACP)(8); MdcD, E , malonyl-S-acyl carri er protein decarbo xylase ( p ,7);MdcF, mem brane pro tein possiblyinvolved in malonate transport; M d c C , involved in the synthesis and attachment of the prosthetic group;M d c H , malonyl-CoA:acyl carr ier prot ein-S H transacylase; M d c R is a prote in of the LysR reg ulato r family.Itis probably a transcriptional regulator of the rndc genes (HOENKEt al., 1997).

HS-ACP 94a 0

0

89 Malonate

@O8,21 Acetate

0

A A C P

Ma&

MadB

H@

BCP co296

94c

Fig. 33. Proposed reaction mechanism for malonate decarboxylation by Malonornonas rubra .The gene clus-ter contains 14 onsecutive genes which have be en term ed rnadYZGBAECDHKFLMN.7he functions of th egene products are as follows:M a d A , acetyl-S-acyl carrier protein: m alonate acyl cam er protein-SH transfe-rase; MadB, carboxybiotin decarboxylase; MadC, D carboxy-transferase subunits; M a d E acyl carrier protein(ACP); MadF, biotin carrier prote in (BCP); MadC may be involved in th e biosynthesis of MadHlM a d H , de-acetyl acyl carrier protein: acet ate ligase; Mad,!. and MadM are m emb rane pro teins which could function as

a malonate carrier.The function of rnadl: Z , K and N is currently unknown (BERGet al., 1997).



72 2 Lyases

is due to the use of assays in which malonate ispresent in addition to malonyl- and acetyl-CoA . This remains t o be established definitive-ly, however, there are striking parallels

betwee n som e of the subunits identified in theMC Ds of these species and those from M . ru -bra and K. pneumoniae (DIMROTHnd HILBI,1997).

The stereochemistry of decarboxylation of(R)-[l-13C, 2-3H]-malonate by the biotin de-pendent “M CD ” of M . rubra (MICKLEFIELDtal., 1995) and th e biotin independe nt “M CD s”of l? fluorescens and Acinetobacter calcoaceti-cus (HA ND A t al., 1999) in d euterium oxidehave been determined. In every case the car-boxylate group was replaced by hydrogen with

retention of configuration. Methods for thesynthesis of ch iral malon ates used in this studyare described in the section on fumarases(Sect. 4.2.1.2).

The decarboxylation of malonate to acetatehas few biotechnological merits othe r than forthe removal of malonate from situations inwhich it may be viewed as a contaminant. Nev-ertheless the organisms discussed above pro-duce large amounts of the decarboxylasesystem and there is great potential for their use

with “un natur al substrates”.There has been considerable interest in thedecarboxylation of 2,2-disubstituted malo-nates, which can po tentially give chiral carbox-ylic acids (cf. MUSS ON St al., 1999; for relatedabiotic catalysts). OHTA’S roup screened mi-croorganisms from soil samples on phenylma-lonic acid. This stud y yielded Alcaligenes bron-chisepticus KU 1201,which is capable of asym-metric decarboxylation of arylmalonates withgood t o excellent enantiome ric excesses (Tab.3) (MIYAMOTOnd OHTA, 990,1991).

The gene for the arylmalonate decarboxyl-ase was identified and expressed in E. coli(MIYAMOTOnd OHT A, 1992). Remarkably,the am ino acid sequenc e had no significant ho-mologies with known proteins and the thiolnucleophile was furnished by a cysteine resi-due rather than coenzyme A (KAWA SAKIt al.,1995). Hyd rophobic interactions with the aro-matic ring were shown to b e especially impor-tant for substrate recognition (KAWASAKItal., 1996a, 1997;OHTA,997).Whole cell medi-ated decarboxylation gave good results with a-fluoro-a-phenylmalonates, particularly when

electron withdrawing substituents werepresent on the the arom atic ring (MIYAM OTOet al, 1992). The generally po orer results withall substrates bearing arom atic rings with elec-

tron donating substituents (e.g., OM e) is prob-ably only partially a function of slow er decar-boxylation, but is also due to the presence ofalternative degradation pathways (MIYAMOTOand OHTA,1990, 1991. Th e purified recom bi-nan t enzym e quantitatively converts a-fluoro-a-phenylm alonic acid (99) to enantiomericallypure (R)-a-fluorophenylacetic acid (100) (Fu-KUYAMA et al., 1999).

4.1.1.10 Aspartate a-Decarboxylase

(L-Aspartate 1-Carboxy-Lyase,

EC 4.1.1.11)

Aspartate a-decarboxylase ( A aD ) catalyzesthe decarboxylation of L-aspartate (101) to p-alanine (102) (Fig. 35) (WILL IAM SON ,985).p-Alanine (102) is a biosynthetic precursor ofpantothenate in bacteria (DUSCH t al., 1999)and is a deg rada tion produ ct of uracil in plantsand fungi (GAN I nd YO UNG, 983).

A a D is a tetramer in which each monom erconsists of a six stranded double $ p-barrelcapped by small a-helices (ALBERTet al.,1998). This motif is foun d in several pro teinsuperfamilies and includes enzymes such asDM SO reductase, forma te dehydrogenase an dthe aspartic proteinases (CASTILLO t al.,1999). In the X-ray crystal structure catalyticpyruvoyl groups were fou nd in thre e subunitsand an ester in the fourth. The enzyme is in-itially produced as an inactive proenzyme (T-

protein) (103;Fig. 36), which undergoes N O -

acyl migration between Gly-24 and Ser-25 togive the amino ester (104). p-elimination ofthe p-protein (105) and hydrolysis yields thecatalytically active a-subunit with a terminalpyruvoyl residue (107). Purified recombinantenzyme consisted predominantly of T-protein(103).Incubation of this at 50O C for 49 hoursgave a fully processed tetrame ric enzyme con-taining three subunits bearing pyruvoylgroups. Unchanged serine was found at theterminus of some of the a-subunits (106), inthe tetramer, but the exact proportion could

not b e determ ined (RAMJEE t al., 1997).




Tab. 3. Decarboxylationof a-Aryl-a-Methylmalonates MIYAMOTOnd OHTA,1990)

Alcaligenes bronchisepticus KU 1201

ArkyiL ? Ar

97 COZ 98

2 H2NzR = H

b R = C H 3

No. Substrate Substrate Yield ee ConfigurationConcentration% % %

1

23

45

67

8

9

1011

0.3 870.4 93

0.5 90

0.1 480.2 35

0.3 950.5 96

0.3 95

0.5 85

0.3 980.5 97

9196

98

9997

>95>95

9897

9591

R

R

R

RR

RR

R

R

SS

It has been known for over a hundred yearsthat a-ketoacids catalyze the decarboxylationof a -am ino acids, nevertheless some aspects ofthe reaction ar e still controversial. In principle,two zwitterionic imine rotomers (108) (109)

can be formed. The “syn” rotomer (108) bene-fits by an intramolecular hydrogen bond(BACH and CANEPA,1997), whereas the“anti”rotome r is able to undergo cyclization tothe uncharged oxazolidin-5-one(110), hich isan intermediate in the abiotic reaction. Bothrotome rs and the oxazolidin-5-one are cap able

of eliminating carbon dioxide to give the

azom ethine ylide (111 to 125) which has beentrapp ed in abiotic mo del studies (GRIGGt al.,1989).AaD has not yet found applications inbiotransformations, however, developmentssuch as the the X-ray crystal structure d etermi-nation and the isolation of large amounts ofenzyme, should encourage its exploitation.Histidine decarboxylase, S-adenosylmethio-nine decarboxylase and phosphatidylserinedecarboxylase all bear catalytic pyruvoylgroups and hence AaD provides an exemplarof this class of enzym es.



74 2 Lyases

N.0 -acy l migration

(y::9mlmalonate

decarboxylaseexpressed

co2 in E . coli

Fig. 34. Alcaligenes bronchisepticus KU 1201 aryl-malonate decarboxylase expressed in E. coli (MIYA-MOTO and OHTA, 992) catalyzes the decarboxyla-tion of a-fluoro-a-phenylmalonic acid (99) to enan-tiomerically pure (R)-a-fluorophenylacetic acid(100) in quantitative yield! (FUKU YAM At al., 1999).

cop

0CO Y 101 L-Aspartate

H 3N Icoz/1, Aspartate a-decarboxy lase

H3N 102 0-Alanine

Fig. 35. L-Aspartate a-decarboxylase catalyzes the

decarboxylation of L-aspartate (101) to p-alanine(102).

P-Elimination

0 10 4

Ester hydrolysis

10 5 P-Protein(2.8 KDa)

@-Elimination

t

- &L0 107 a-Protein

(11.8 kDa)

Fig. 36. Processing of a-aspartate decarboxylase (T-protein) (103) to catalytically active a-protein.

4.1.1.11 Aspartate 4-Decarboxylase(L-aspart ate P-Carboxylase,L-aspartate 4-Carboxy-Lyase,Desulfinase,EC 4.1.1.12)

Aspartate 4-decarboxylase catalyzes the de-carboxvlation ofL-amartate (101) o L-alanine(111) Fig.38) ( C H A ~Gt al.,' 1982).Th's reac- - , - -

tion has been used in various forms for themanufacture of L-alanine, which is difficult toproduce by fermentation. The process hasbeen optimized with resting cells of Pseudo-monas dacunhae grown at pH 7.0-7.5 on gluta-mate (CALIK t al., 1997) and immobilized on

K-carrageenan (CALIK t al., 1999). By using




0

H

RZ

101L-Aspartate

H2O@ O - - - H 0 108 H 0 109

5-endo-trig

cyclisation

R2

H 110

0

107 a-Protein

0o,c*o

NH3

102 p-AlanineH2O

H 0 125 Azomethineylide

Fig. 37. Mechanism of decarboxylation by a-aspartate decarboxylase (107).

D,L-aspartate as substrate both L-alanine and ported on acylonitrile activated silica gelD-aspartate were produced using whole cells (ABELYAN,999; ABELYANt al., 1991; SENU-immobilized in carrageenan or enzyme sup- MA et al., 1989). Coimmobilized aspartase and

cells with aspartate decarboxylase activity,converted fumarate to aspartate which was de-carboxylation in sifu (ABELYANnd AFRIK-YAN, 997).Remarkably the combined catalyst

was stable for 450-460 days. Aspartate 4-de-carboxylase contains a pyridoxal5 ‘-phosphateprosthetic group, which is discussed in detail inSect. 4.1.1.13, L-Glutamate Decarboxylase.

All a-amino acid decarboxylases and mostother amino acid decarboxylases, rely on imineformation between the amino group of theamino acid and the carbonyl group of a pyru-voyl (108) (109), pyridoxal 5’-phosphate orsome other group to promote decarboxyla-tion. Imine formation is also the first step intransamination catalyzed by aminotransferas-

es.?hus by impeding the hydrogen shift step in

101L-Aspartate

Aspartate 4-d ecarboxylase

Fig. 38. Aspartate-4-decarboxylase.



76 2 Lyases

transamination and promoting the protona-tion requ ired for decarboxylation, it should bepossible to switch an enzym e from transamina-tion to decarboxylation. This has been

achieved with an E. coli aspartate am inotrans-ferase triple mutant (Y225R/R292K/R386A),which catalyzes the 4-decarboxylation ofaspa rtate eight times as fast as transam ination.This is a greater than 25 million fold increasein decarboxylation relative to the wild type en-zyme (GRAB ERt al., 1999)

4.1.1.12 Valine Decarboxylase(L-Valine Decarboxylase,

EC 4.1.1.14)

Valine decarboxylase catalyzes the decar-box ylation of L-valine(112) o 2-m ethylpropan-amine (113) Fig. 39) and also acts on L-leu-cine, norvaline and isoleucine. It has a pyridox-a1 5 ’-phosphate (PLP ) (116) rosthetic group,but seems to have been largely overlookedsince its discovery (SUTTON nd KIN G, 962).

4.1.1.13 Glutamate Decarboxylase(L-Glutamate l-Carboxy-Lyase,EC 4.1.1.15)

Glutamate decarboxylase (G A D ) catalyzesthe decarboxylation of L-glutamate (114) o 4-amino butanoate, which is more widely knownas y-aminobutyric acid (GABA) (115) (Fig.

coz1 aline decarboxylase

H 1132-Methylpropanamine

Fig. 39. Valine decarboxylase.

40). GABA plays a role in pH regulation, ni-trogen storage and defence in plants (SHE LP tal., 1999) and as a neurotransmitter in mam-mals (WAAGEPETERSENt al., 1999; SCH WA RTZ

et al., 1999). GAD also decarboxylates L-as-partate (101) o p-alanine (102).

The prosthetic group of GAD is nominallypyridoxal 5’-phosphate (PLP) (116a), ut incommon with many other PLP dependent en-zymes, the usual form of the holoenzyme con-tains the cofactor bonded t o the peptide chainvia an aldimine (116b) ormed with an ca m i-no group of a lysine residue (reviews, JAN SON -IUS, 1998;JOHN, 1995; HAY ASH I,995; ALEX-

The mechanism and stereochemistry of re-actions catalyzed by E. coli GA D have beendetermined by detailed isotope studies (Fig.41). The “norm al” decarboxylation pathway isinitiated by imine (117) orma tion between thePLP-GA D aldimine derivative (116b) andglutamate (114). Decarboxylation yields aquinoid type interm ediate (118)which can re-proto nate at two sites. Protonation of the for-mer C-a-center occurs on the sam e face as theC4’-Si side of the quinoid intermedaite andhence th e carboxylate gro up is replaced by the

incoming proto n w ith retention of configura-tion.The imine (119) so formed, undergoesmetathesis to give the PLP-GAD aldimine de-rivative (116b) nd GABA (115).Decarboxyl-ation may also occur concommitantly with

ANDER et al., 1994).

colutamate decarboxylase

Fig. 40.L-Glutamate decarboxylase.



77Examples ofLyases

!E0

Hyx H3N@ 11 4 L-Glutamate H

Glutamate

ip* 116 decarboxylase

-7-/NI

a X = 0, PLP

b X = N-&-lysine u n,.

H,N@ 115 GABA

HzO r H3N-E-lysine

o

0

H\Ht/JIOcozI

C-a rotonation

Non-oxidative11 8

11 9 decarboxylation HI

H

f?.oo

121 Succinate

semialdehyde

2 0

0 3 p

H 122PMP H 120

Fig. 41.The m echanism of decarboxylation by L-glutamate decarboxylase. Abbreviations: G AB A, y-amino-butyric acid (115); LP, pyridoxal 5'-phospha te (116); MP, pyridoxamine 5'-phosphate (122). In this andothe r diagrams showing PMP derivatives, the ph enolic hydroxyl group-imine interaction [(117)-(120)] isshown as a hydrogen bo nd (dotted line) or the purposes of simplicity. O n the basis of solution phase pK,'s, aprot ona ted imine-phenoxide interaction would be a mo re accurate depiction.



78 2 Lyases

transamination (cf. KHRISTOFOROVt al.,1995). C-4 -%face proto nation of the quinoidintermedaite (119) gives the imine (120),which hydrolyzes to succinate semialdehyde

(121) and stereospecifically labeled PM P (122)in a single turnover process (TILL EY t al.,1992,1994). A similar pathway occurs with a-

methylglutamate (123) (Fig. 42), which is oxi-datively deaminated to levulinate (124) n amulti-turnover catalytic process which re-quires dioxygen (BERTOLD It a l., 1999).

PLP dependent aminotransferases have aSer-X-Ala-Lys sequence in the active site,whereas decarboxylases have a Ser-X-His-Lyssequence. The lysine residue is responsible foraldimine formation and C-4 %-face protona-

tion of the quinoid intermedate (118),whereasthe histidine binds the a-carboxylate groupsand protonates at C -a (GANI,1991, SMITH tal., 1991). Hum an G A D consists of two iso-forms (GAD 65 and G AD67; SCHWARTZt al.,1999), which closely resemble aspa rtate trans-aminase and ornithine decarboxylase (Qu etal., 1998).

G A D has been used extensively for the thestereospecific synthesis of GABA hydrogenisotopomers (115) (YO UN G, 991; BA'ITERSBYet al., 1982b) an d for t he synthesis of D-gluta-mate. L-glutamate produced by fermentationwas racemized with recombinant glutamate

CO'i> 123 a-Methylgiutamate

Glutamate decarboxylase

Fig. 42. Oxidative decarboxylation of a-methylgluta-

mate (123)by glutamate decarboxylase.

racemase from Lactobacillus brevis ATCC-8287. The L-glutamate was then converted toGA BA using GA D from E. coli ATCC-11246which enabled the unreactive D-glUtamate to

be isolated (YAGASHInd OZA KI, 998).

4.1.1.14 Ornithine Decarboxylase

(EC 4.1.1.17)

Ornithine decarboxylase (ODC) is a PLPdepende nt enzyme, which catalyzes the decar-boxylation of L-ornithine (127) to putrescine(128) 1,4-diaminobutane) (Fig. 43). It also de-carboxy lates L-lysine, but more slowly (SW AN -SON et al., 1998). O D C has been heavily inves-tigated because putrescine (128) and the nexthighest hom olog, cadaverine , (which is derivedfrom L-lysine) are precursors of polyamines(e.g., spermideine and spermine; MORGAN,1999), which ar e accumulated in cancer cellsand stimulate proto-oncogene expression (TA -BI B and BACHRACH,999). O D C is the ratelimiting enzyme in polyamine biosynthesis andits activity is down regulated by increasingpolyamine levels which accelerate d egradation(TOTHand COFFINO, 999; D I G AN GI t al.,1987). O D C is widely distributed a nd well es-tablished methods are available for its isola-tion from E. coli (MORRIS nd BOEKER ,983),Physarum polycephalum (MITCHELL, 983),yeast (TYAG I t al., 1983), germinated barleyseeds (KYRIAKIDISt al., 1983), mouse kidney(SEELY nd PEGG,1983), rat liver (H AYA SHIand KAM EJI, 983)

X-ray crystal structures have also been de-termined for mouse O D C (KE RN t al., 1999),Lactobacillus-30a ODC (L30a OrnDC; Mo-

MANY et al., 1995) and Trypanosoma bruceiO D C as the native enzyme and as a complex(GRISHINt al., 1999) with the suicide inhibitora-difluoromethylornithine (SEELY t al., 1983).Trypanosoma brucei is the causative agent ofAfrican sleeping sickness and its OD C is a rec-ognized drug target (MC CA NN t al., 1983).These a re large enzym es with disparate, com-plex, multimeric structures.The active site liesat the interface of a dimers and PL P is boundas an aldimine with the & -aminogroup of a ly-sine res idue (e.g., Lys-60 for Trypanosoma bru-

cei O D C O ST ER MA Nt al., 1999, 1997) as is



? N H 3


FNH3H3N/t-Cop 129 a-Methylomithine

Omithine decarboxylase

co2I0

INH3H1Nol 128Putrescine

Omithine decarboxylase

L

f- H 3

06 130Fig. 43. Ornithine decarboxylase (ODC) catalyzeddecarboxylation of L-ornithine (127) to putrescine H3@4128).

found for GA D and most other a-amino aciddecarboxylases.

L-Ornithine (127) undergoes decarboxyla-

figuration (ORRnd GOU LD, 982) This reac-

tion by ODC, such that the carboxylate groupis replaced by hydrogen with retention of con- C L 1312-Methyl- I-py rroline

tion has been for the stereospecific ‘yn-fig. 44.Ornithine decarboxylase catalyzed oxidative

thesis of putrescine hydrogen isotopomers(YOU NG, 991). a-Methylornithine (129) (Fig.44) undergoe s oxidative decarboxvlation w ith

decarboxylation of a-methylornithine (129) to 2-methyl-l-pymoline (131).

ODC in t i e presence of dioxygen to give theammonium ketone (130) which cyclizes to 2-methyl-l-pyrroline (131) (BERTOLDI t al.,1999).

Selenomonas ruminantium lysine decarboxyl-ase, catalyzes both reactions with similar ki-netics pa ram eters (TAKATSUKA t al., 1999a,

b). Most workers have utilized lvsine decar-4.1.1.15 Lysine Decarboxylase

(L-Lysine Carboxy-Lyase,

EC 4.1.1.18)

Lysine decarboxylase is a PLP dependentenzyme, which catalyzes the decarboxylationof L-lysine (132) to cadaverine (133) (Fig. 45)in a reaction wh ich is homologo us with that ofornithine decarboxylase (Fig. 43). Most formsof both enzymes are capable of catalysing both

reactions, to some degree, however, unusually

boxylase from Bacillus cadaveris-or the con-stitutive enzyme from E. coli (BOEKER andFISCHER, 983), but a second form has beenidentified recently (LEMONNIERnd LANE,1998; YAM AM OTOt al., 1997). Lysine decar-boxylases decarboxylates L-lysine (132) suchthat the carboxylate group is replaced by hy-drogen with retention of configuration (ORRet al., 1982; BATTERSBYt al., 1982a).This reac-tion has been used for the stereospecific syn-thesis of cadaverine hydrogen isotopomers

(YOUNG, 991).



80 2 Lyases

I

("'H

Lysine decarboxylase

COZ

I "

i-133Cadaverine

Arginine decarboxylase

COZ

("'H

Jig. 45. Lysine decarboxylase (ODC) catalyzed de-carboxylation of L-lysine (127) to cadaverine (128)

(1,5-diaminopentane). H,N 135 Agmatine0

Fig. 46. Arginine decarboxylase (ADC) catalyzeddecarboxylation of L-arginine (134) to agmatine

4.1.1.16 Arginine Decarboxylase (135) 4-aminobut~lguanidine).

(L-Arginine Carboxy-Lyase,EC 4.1.1.19)

Arginine decarboxylase catalyzes th e decar-boxylation of L-arginine (134) to agmatine(135) 4-aminob utylguanidine) (Fig. 46). It is aPLP dependent enzyme which decarboxylatesL-arginine such that the carboxylate group isreplaced by hydrogen with retention of config-uration (O RR et al., 1982). As with the otherbasic amino acid decarboxylases, it has been

used to prepare agmatine hydrogen isotopom-ers (YOUNG,1991; RICHARDS nd SPENSER,1983).

4.1.1.17 D iaminopimela e

Decarboxylase (meso-2,6-

Diaminoheptanedioate Carboxy-

Lyase, EC 4.1.1.20)

Diaminopimelate decarboxylase catalyzes

th e decarboxylation of meso-2,6-diaminopim-

elate (136) o L-lysine (132) Fig. 47), which isthe last step in the biosynthesis of L-lysinefrom L-aspartate and pyruvate (SCAPINandBLANCHARD,998; C H A ~ R J E E ,998). meso-2,6-Diaminopimelate (136) s a cross-linkingconstituent of the peptidoglycans of virtuallyall gram-negative an d som e gram-positive bac-teria. L-Lysine is an essential dietary amin o ac-id for mammals, because they lack the di-aminopimelate pathway, hence inhibitors of

the enzymes of this pathway could have anti-microbial or herbicidal properties with lowmamm alian toxicity (SONG t al., 1994).

Incubation of rneso-2,6-diaminopimelate(136)with diaminopimelate decarboxylase indeuterium oxide yields (2S,6R)-[6-*Hl]-lysineand hence the decarboxylation occurs suchthat the incoming hydrogen replaces the car-boxylate group with inversion of stereochem-istry (A SA DA et al., 1981; KELLARD t al.,1985). This result is unusual because PLP de-pendent enzymes, such as diaminopimelate d e-

carboxylase generally catalyze decarboxyla-




H 137L-Histidine

Diaminopimelate decarboxylase

C 0 2 1

Fig. 47. Diaminopimelate decarboxylase catalyzeddecarboxylation of meso-2,6-diaminopimelatem e-so-2,6-diaminoheptanedioate) 136) to L-lysine(132).

tion with retention of configuration. However,the center undergoing decarboxylation has the(R)-configuration rather the (S)-configurationwhich is typical of most natural amino acids,consequently with a similar mode of bindingfor the two enantiomers the hydrogen is deliv-ered from the same side. Stereospecific syn-thetic routes to meso-2,6-diaminopimelate(136) and other stereoisomers (WILLIAMSndYUAN, 1992) have been reported and decar-boxylatian has been used to prepare a range of

hydrogen isotopomers (ASADAet al., 1981;KELLARDt al., 1985; YOUNG, 991), but othersynthetic opportunities remain to be exploit-ed.

4.1.1.18 His tidine Decarboxylase

(Histidine Carboxy-Lyase,

EC 4.1.1.22)

Histidine decarboxylase (HDC) catalyzes

the decarboxylation of L-histidine (137) to his-

Histidine decarboxylase

COZ

H 138Histamine

Fig. 48. Th e histidine decarboxylase catalyzed de car-boxylation of L-histidine(l37) o histamine (138).

tamine (138) (Fig. 48). Most forms of the en-zyme have a pyruvoyl group at the active site,however, PLP dependent forms are present inKlebsiella planticola, Enterobacter aerogenes(KAMATH t al., 1991) and Morganella morga -nii (SNELL nd GUIRARD,986).

Pyruvoyl-dependent HDCs are produced byLactobacillus buchneri, Clostridium perfring-ens, Micrococcus sp.n., Leuconostoc oenos9204 (COTON t al., 1998) and Lactobacillus30a (SNELL, 986) All of these enzymes showhigh sequence homology and have probablyevolved from a common ancestral protein.Lactobacillus 30a HDC has been studied ex-tensively and the X-ray crystal structure deter-mined (PARKS t al., 1985; GALLAGHERt al.,1993).Lactobacillus 30a HDC is formed fromhexameric prohistidine decarboxylase (n6;

SNELL nd HUYNH, 986), by autoproteolysisbetween Ser-81 and Ser-82 and elimination togive a “N”-terminal pyruvoyl group derivedfrom Ser-82 (the a-chain; cf. Fig. 36.), plus theremainder of the peptide chain with a normalterminal carboxylate group (the P-chain). Thea- and P-chains are tightly bound together andthree of these subunits form a trimer, whichbinds to a second trimeric unit to give a dumb-bell shaped hexamer; (ab)6 (GALLGGHER etal., 1993).Each pyruvoyl group lies in an activesite at the interface between two subunits in a

trimer (GELFMANt al., 1991; PISHKOnd RO-



82 2 Lyases

BERTUS, 1993). Decarboxylationof L-histidine(137) by HDC occurs with retention of config-uration (BATTERSBYt al., 1980; BATTERSBY,

1985; SAN TAN IELLOt al., 1981; YOU NG, 991).

4.1.1.19 Orotidine-5 -Phospha teDecarboxylase (ODC ase, Orotidine5 ' Monophosphate Decarboxylase[OM P], Orotidine-5 -PhosphateCarboxy-Lyase, EC 4.1.1.23)

Orotidine-5 -phosphate decarboxylase(O PD ) catalyzes the decarboxylation of oroti-dine 5'-ph osph ate (139) to uridine 5 '-phos-phate (140) (Fig. 49), which is the final step inde n o w pyrimidine biosynthesis (G AO et al.,1999a). This re action is especially notab le be-cause OPD catalysis increases the rate by afactor of 1017over the spontaneous rate, which

. .$

HO OH 139 Orotidine5'-phOsphate

Orotidine 5'-phosphatedecarboxy ase

COZ

J .+HO OH 140Uridine

5'-phosphate

Fig. 49. The orotidine 5 -phosphate decarboxylase(OD C) catalyzed decarboxylation of orotidine 5 ' -phosphate(l39) o uridine 5'-phosphate (140).

is higher than that of any other enzyme de-scribed thus far. To appreciate the magnitudeof catalysis;consider that oro tic acid isestimat-

ed to have a half life for decarboxylation ofsome 78 million years, in neutral aque ous solu-tion at room temperature (MILLERet al.,1998). The precise origin of the catalytic effi-ciency is currently an enigma (EH RLIC Ht al.,1999).

4.1.1.20 Tyrosine Decarboxylase(L-Tyrosine Carboxy-Lyase,

EC 4.1.1.25)Tyrosine decarboxylase (TDC)catalyzes the

decarboxylation of L-tyrosine (141a) to tyra-mine (142a) and L-DO PA (141b) to dopamine(142b).(Fig. 50). TDC is a common plant en-zyme involved in the synthesis of numeroussecondary metabolites. Tyramine (142a) is adesirable flavoring in small amoun ts a range of

food p roducts includ ing wine, cheese and sau-sages (MORENO-ARRIBASnd LONVAUD-Fu-NEL , 999), and (together w ith dopamine) a bio-synthetic precursor of numerous alkaloids in

141a R = H; L-Tyrosineb R =OH; -DOPA

C O 2 1 yrosine decarboxylase

14 2a R = H; Tyramineb R =OH; Dopamine

Fig. 50. The tyrosine decarboxylase catalyzed decar-boxylation of L-tyrosine (1411) to tyramine (142a)and L-DOPA (3,4-dihydroxyphenylalanine) 141b)to doparnine (142b).




plants (FACCHINI,998). T D C is a P LP depen-dent enzyme and catalyzes decarboxylationsuch that the departing carboxylate group isreplaced by hydrogen with retentio n of config-

uration. This reaction has been used to p repa retyramine hydrogen isotopomers (BELLEAUtal., 1960; BELL EAUnd BURB A, 960; YOUN G,1991).

TDC has been utilized in a multi-enzymeprocess for the production of dopam ine (142b)from catechol. Cond ensation of catechol, pyru-vate and ammonia catalyzed by tyrosine phe-nol-lyase gave L-DO PA (14lb),which w as de-carboxylated w ith rat liver L-DOP A decarbox-ylase or Streptococcus fueculis TDC. The L-

DOPA decarboxylase was inhibited by L-

DOPA concentrations greater than 1 mM,whereas TDC was unaffected up to 40 mM,consequently TD C was selected for further d e-velopment. Decarboxylation was carried outin an enzyme reactor a t 37 “ C or 12 h , with pHcontrol by addition of hydrochloric acid andsupplem ents of PL P to compensate for decar-boxylative transamination. At 100 mM L-

DOPA (141b),conversion to dopam ine (142b)was quantitative. When the decarboxylationwas run with in situ synthesis of L-DOP A, the

productivity was much lower than with twoconsecutive “single pot” reactions. (LE E et al.,1999a, b). The sam e process has also been runwith whole cells expressing the relevent en-zymes. Erwiniu herbicolu cells with phenol ty-rosine-lyase activity and Streptococcus faeculiscells with T D C activity (AN DER SON t al.,1987) were coimmobilized in glutaraldehydecrosslinked porcine gelatin beads and placedin a pack ed bed reactor together with catechol,pyruvate a nd am monia, Dopam ine was pro-duced but the conversion was low (AND ER SON

et al., 1992a, b). These processes are discussedin grea ter detail in th e section describing phe-nol tyro sine lyase (S ect. 4.1.4.2).

4.1.1.21 Aromatic L-Amino Acid

Decarboxylase (DOPA

D ecarboxylase, Tryptophan

Decarboxylase, HydroxytryptophanDecarboxylase,EC 4.1.1.28)

Aromatic L-amino acid decarboxylase(AAAD) catalyzes the decarboxylation of L-

tryptophan (143a) to tryptamine (144a); 5-hy-droxy-L-tryptophan (143b) to serotonin (144b)(Fig. 51) and L-DO PA (3,4-dihydroxyphenylal-anine) (141b) to dopamine (142b) (Fig. 50).Prior to 1984 the enzyme commission recog-nized D OP A decarboxylase (E C 4.1.1.26) and

tryptophan decarboxylase (EC 4.1.1.27). H ow -ever, these distinctions are not warranted bythe selectivity of the enzymes and they havenow been incorporated into the AA A D sub-subsubclass.A A A D s also generally, have somephenylalanine decarboxylase (EC 4.1.1.53) ac-tivity, an d conversely,phenylalanine decarbox-ylases generally have some activity with L-DOPA (141b), tryptophan (143a) and 5-hy-droxy-L-tryptophan (143b). AAADs are also

H 143a R = H; L-Tryptophanb R =OH; 5-Hydr0xy-L-tryptophan

IAromatic L-amino

acid decarboxylase

IH 144

a R = H; Tryptamineb R = OH; Serotonin

Fig. 51. The aromatic L-amino acid decarboxylasecatalyzed, decarboxylationof L-tryptophan (143a) totryptamine (144s) and 5-hydroxy-L-tryptophan

(143b) to serotonin (5-hydroxytryptamine)(144b).



84 2 Lyases

frequently described as L-DOPA decarboxyl-ases. Typically, the ratio of activity of anAAAD between L-DOPA (141b) and 5-hy-droxy-L-tryptophan (143b) is 5: 1 JEBAI t al.,

1997). The plurality of s ubstra tes decarbo xyl-ated by AAADs suggests that a mixtures ofenzymes or isozymes may be responsible forthe activity, but at least with the w ell character-ized enzymes from pig kidney and rat liver thevast majority of the da ta indicates that activityis due t o a single protein.

A A A D is present in fungi (NIEDEND t al.,1999), plants, insects, fish (N AG AI t al., 1996)and mamm als (ZHU and JUORIO, 1995;POUPON t al., 1999). In mammals, AA A Dclearly plays a role in the biosynthesis of the

neurotransmitters; serotonin (144b) and L-

DOPA (141b),but other roles have been sug-gested including the prevention of apoptosis(BERR Y t al., 1996;ZHU nd JUOR IO, 995).

There are reliable methods for the extrac-tion of pig kidney AAAD (VOLTATTORNItal., 1987; DOM INICIt al., 1993), bu t instabilityand low specific activity limit the availability.Expression of the gene in E. coli gives prote inwith twice the specific activity of the extractedenzyme and each dimer binds two molecules

of PLP rather than one as with the extractedenzyme. It appears that the latter contains PLPbound covalently in a n no n-catalytically activeform (MO ORE t al., 1996) and th e differencesbetween the two forms are ap pare nt in recent-ly reported preliminary crystal structure data(MALASHKEVICH,999). Extra ction of rat liveror kidney AAAD (SOURKES,987) has alsobeen supplanted by expression of the gene inE. coli, but in this case the recombinant pro-tein is kinetically identical to the extractedprotein (JEBAI t al., 1997).

As with all other PL P d ependen t, L-aminoacid a-decarboxylases, decarboxylation occurssuch that the carboxylate group is replaced byhydrogen with retention of configuration (e.g.,5-hydroxy-~-tryptophan, 143b)BATERSBY etal., 1990). Despite the ability of A A A D t o ac-comodate a wide ring off substrates it hasbarely been exploited for biotransformations,presumably because of its limited availabilityuntil recently. a-Methyl-L-DOPA (145) under-goes decarboxylation as expected to give a-

methyldopamine (146) (Fig. 52), this is oxida-

tively deam inated to give 3,4-dihydroxyphenyl-

cop

HO

145 a-Methyl-L-DOPA

Aromatic L-aminoacid decarbox ylase

co2

146 a-Methyldopamine

Aromatic L-aminoacid decarbox ylase

0 2

147 3.4-Dihydroxyphenylacetone

Fig. 52. The oxidative deamination of a-methyl-DO-PA (145) produces 3,4-dihydroxyphenylacetone(147), hich deactivates AAAD.

acetone (147), which irreversibly inactivatesthe enzyme by formation of a ternary adductwith PLP and an active site lysine residue(BERTOLDIt al., 1998).

4.1.1.22 PhosphoenolpyruvateCarboxylases and Carboxykinases

(EC 4.1.1.31,32,38,49)

There are four lyases which catalyze the car-boxylation of phosphoenolpyruvate (PEP)(148) (Fig. 53) to oxaloacetate (28) and the re-verse reaction (Tab. 4). PEP is an excellentphosphate don or which is frequently used forrecycling AD P to A n . A n efficient mole scalechemical synthesis has been reported

(HIRSCHBEINt a1,1982).




lize four nucleotides with the followingV,, /K, values (M-ls-') ATP ( 6 4 , GTP(1.30), CTP (0.87), ITP (0.66). Decarboxyla-tion is fully reversible and hence it may be

used to incorporate isotopic labels (e.g., "C)into oxaloacetate (28) (SALVARREYt al., 1995;TARI t al., 1996).nolpyruvate

I t

4.1.1.23 Phenylpyruvate

Decarboxylase (Phenylpyruvate

Carboxy-Lyase, EC 4.1.1.43)

Phenylpyruvate decarboxylase catalyzes thedecarboxylation of phenylpyruvate (149) to

0 0 28Oxaloacetate phenylacetaldehyde (150) (Fig. 54) and alsohas Some activity with indole pyruvate (seeSect. 4.1.1.26). This reaction is the second stepin the catabolism of phenylalanine, which com-mences with transamination to form phenyl-pyruvate. Moreover the decarboxylation is ho-

Fig. 53. The carboxylation of phosphoenolpyruvate(14) o give oxaloacetate (28) catalyzed by phos-phoenolpyruvate carboxylases (PEP-CL) and car-boxykinases (PEP-CK). Cofactors are not shown,see Fig. 54 .

Tab. 4. Phosphoeno lpyruvate Carboxylases and Carboxykinases

E C N o. Name Phosphate Donor Other ProductsPhosphoenolpyruvate-

4.1.1.31 -carboxylase Orthophosphate H204.1.1.32 -carboxykinase GTP GDP4.1.1.38 -carboxylase Pyrophosphate Orthophosphate4.1.1.49 -carboxykinase ATP A D P

PEP-carboxylases (CL) are present inplants, algae and bacteria, but the propertiesvary widely depending on the source (YANO tal., 1995). PEP-CL is especially important inC, plants, where it catalyzes the initial step of

photosynthetic carbon fixation (Hatch-Slackpathway; LEPINIECt al., 1994). PEP-CL willaccept phosphate donors other than PEP, suchas 2-phosphoenolbutyrate and phosphoenol-3-fluoropyruvate, but dephosphorylation with-out carboxylation is the predominant reaction(GONZALEZnd ANDREO, 988).

The carboxykinases (CK) are assigned onthe basis of their ability to utilize GTP or ATP,but both nucleotides or others may show someactivity with a given enzyme. An extreme ex-ample is provided by the PEP-CK from the

halophile Vibrio costicolu,which is able to uti-

mologous to benzoylformate decarboxylationwhich is a later step in the same pathway (Sect.4.1.1.7) (SCHNEIDERt al., 1997; BARROWMANand FEWSON,985). This reaction is virtuallyunexplored, but there is the prospect that trap-

ping with an aldehyde could be used to forman acyloin as observed with benzylformate de-carboxylase (Fig. 26).

4.1.1.24 Tartronate Semi-Aldehyde

Synthase [Glyoxylate Carboxy-

Lyase (dimerizing), EC 4.1.1.471

Tartronate semi-aldehyde synthase catalyz-es the decarboxylation of glyoxylate (El) nd

condensation with another molecule of glyox-



86 2 Lyases

enolpyruvate

0 0 280xaloacetate

Fig. 54. The phenylpyruvate decarboxylase catalyzeddecarboxylation of phenylpynwate (28) to phenyl-acetaldehyde (148).

ylate to give tartronate semi-aldehyde (152)

(Fig.55) , n a reaction analogous to the acyloincondensation of pyruvate (Fig.2).The enzymeis reported to be a flavoprotein, although aTPP prosthetic group would be expected. Thisreaction forms part of the tartronate pathway,

which is utilized by many organisms for themetabolism of glyoxylate. The strictly anerob-ic, obligatory oxalotrophic bacterium, Bacillusoxalophilus, utilizes tartronate semi-aldehyde

n

Tartronate Isemi-aldehyde

synthase

COZ 0 151 Glyoxalate

H

on

152 Tattronate semi-aldehyde

Fig. 55. The acyloin condensation of glyoxalate (151)to give tartron ate semi-aldehyde (152), atalyzed by

tartro nate semi-aldehyde synthase.

synthase or the serine pathway to assimilateoxalate and oxalyl-CoA decarboxylase andformate dehydrogenase for oxidation of oxa-late (ZAITSEVt al., 1993).Glycine fed to Pseu-

dom onas fluoresens, is converted to glyoxy-late, and proccessed to the antibiotic obafluor-in via the tartronate semi-aldehyde pathway(HERBERTnd KNAGGS,992).

4.1.1.25 Dialkylglycine

Decarboxylase [2,2-Dialkylglycine

Decarboxylase (Pyruvate),

EC 4.1.1.641

Dialkylglycine decarboxylase (DAGD)from Pseudomonas cepacia catalyzes the oxi-dative decarboxylation of dialkyl glycines(153) to ketones (39) (Fig. 56) and the PMP(122)form of the enzyme.Transamination witha-ketoacids such as pyruvate (1) regeneratesthe “internal imine” (116b) and an amino acidsuch as L-alanine (111).Thus this enzyme cata-lyzes two classical PLP-dependent reactions(cf. Fig. 41), with different substrates.

DAGD is a tetramer which resembles apar-tate aminotransferase. In solution, it consists oftwo conformers of differing activity (ZHOU

and TONEY, 999). The equilibrium betweenthe two conformers is perturbed by alkali met-al ions. Li+ and Na+, act as inhibitors, whereasK+ and R b+ are activators. Similarly the en-zyme adopts two different conformationalforms in the crystalline state, depending on thealkali metal ion present (TONEY t al., 1995;HOHENESTER,t al., 1994). It is generally pre-sumed that the conformers observed in the

crystal structures correspond to those in solu-tion. However, crystal structures of DAGDbound to other inhibitors (e.g., l-aminocyclo-propane carboxylate (162)),show only the “ac-tive” K+/Rb+ conformer (MALASHKEVICH,1999).

As with other PLP-dependent decarboxyl-ases, non-oxidative decarboxylation may alsooccur with DAGD. The ratio between non-oxi-dative and oxidative decarboxylation is deter-mined by the propensity of the enzyme to pro-tonate the quinoid intermediate (118 Fig. 41)

at C-a and C-4’, which in turn is a function of




(entries 4,5), which has low and comparablereactivity to L-phenylglycine(157) (entry 6). D-

Phenylglycine is not a substrate.Most extraordinarily, 1-aminocyclopropane

carboxylate (162), forms an imine with thePLP group, but does not undergo decarboxyla-tion. The X-ray crystal structure of this adductshows the aldimine bond is out of the plane ofthe pyridinium ring (cf.117,Fig. 41) and thus isunable to act as an electron sink for the pair ofelectrons produced by decarboxylation (MAL-

This enzyme has tremendous potential forthe resolution of a-methyl and other a-alkylamino acids, which are poor substrates withamidases and proteases.

ASHKEVICH, 1999).

153Dimethylglycine 39 Acetone

I t -

H 116 H 122PMP

a X =0, LP

b X = N-E-lysine

111 L-Alanine 1 Pyruvate

Fig. 56. Dialkylglycine decarboxylase (DA GD ) cata-lyzed oxidative decarboxylation of dimethylglycine

(153) o acetone (39). Pyruvate (1) s the preferredsubstrate for transaminating the PMP-form of theenzyme t o the “internal” imine (116b).

the fit of the substrate to the active site. Di-methylaminoglycine (153) is the best substratefor DAGD (Tab. 5, entry 1). Larger alkylgroups can be accommodated (entries 2 and3), but replacement of an a-alkyl group by hy-drogen (entries 4-7) or carboxylate (entries 9,10) results in appreciable non-oxidative decar-boxylation (SUN et al., 1998a, b). L-Alanine

(111) is more reactive than D-alanine (156)

4.1.1.26 Indolepyruvate

Decarboxylase (EC 4.1.1.72)

Indolepyruvate decarboxylase (IDPC) cata-lyzes the decarboxylation of indole-3-pyruvate(163) to indole-3-acetaldehyde (164) (Fig. 57).Phenylpyruvate decarboxylase (EC 4.1.1.43)also catalyzes this reaction. In bacteria, L-tryp-tophan (143a) is converted to indole-3-aceticacid (165) by two pathways; the indole-3-acet-amide pathway (EMANUELEnd FITZPATRICK,1995) and the indole-3-pyruvate pathway(MANULISt al., 1998). The latter commenceswith the transamination of L-tryptophan(143a)to indole-3-pyruvate (163) (KOGA t al.,1994; GAOet al., 1997), followed by decarbox-ylation to indole-3-acetaldehyde (164) and ox-idation to indole-3-acetic acid (165).In yeast, asimilar pathway is followed except that indole-3-acetaldehyde (164) is reduced to indole-3-

ethanol (tryptophol, IRAQUI t al., 1999;FURU-KAWA et al., 1996).The biosynthesis of indole-3-acetic acid (165) in plants has been exten-sively investigated because of its importanceas a plant growth regulator, nevertheless im-portant aspects of this pathway remain to beelucidated (KAWAGUCHIt al., 1996). Conse-quently, much of the discussion that followswill deal with bacterial species, particularlythose that are plant pathogens or are root as-sociated. Many of these organisms overpro-duce indole-3-acetic acid (165) in culture when

fed supplementary L-trytophan (143a) and



m m

T

5.D

e

b

a

o

a

T

a

min

o

C

ay

b

Diak

gy

n

De

b

a

E

y

S

ae

De

b

a

o

No

Oxd

v

T

a

min

o

De

b

a

o

h

k

k

&

M-S

(%)

k

6-7

k

&

(M-k

1

1

2

1

72

0

5

o

<00

3

1

23

02

4

3

5

1

02

00

28 5

6

157

02

00

2

5

7

158

12

0

1

10-~ 52

09

<15

9

160

79

(0

8

1

3

1102

5

9

1

161

2

6

.

01

0

0

9

6

22

02

1

<00

<00

2

1

09

00

4

1

33

0

3

o

47

07

98

30

37

02

58(08

8

1

n

n

n

E

mae

e

o

a

esh

wn

nb

a

s

SUN e

a

1

b

n

n

n

n

n

n n

<15

n

n

25

(0

3

1

4

72

05

.

n

n

n

n35

06

. 1

22

110

2

1

3

77

10

. o

HN"Xc0N0 u 0 HNO

HdCOFHAo-

H0 YCO?

153 D

me

h

gy

n

154 1 -A

in

o

a

155 1-Am

in

o

1LA

a

n

156 D

A

a

n

1

LP

gy

n

c

b

ae

c

b

ae

161 A

in

mao

e

162 1 -A

in

o

o

c

b

ae




1998), the plan t gall inducer, Erwinia herbicola(BRANDLnd LINDOW,996,1997), seven spe-cies of Enterobacteriaceae (ZIMMERt al.,1994) and Enterobacter cloacae (KOGA t al.,

1991) and ID PC activity has been detected inBradyrhizobium elkanii (MINAMISAWAt al.,1996).The I DP C gene from th e plant associat-ed bacterium Enterobacter cloacae, has beenexpressed in E. coli and the enzyme character-ized. It is a tetramer which is TP P and magne-sium depen dent. ID PC is inactive with indole-3-lactate and oxaloacetate (33), but does de-carboxylate pyruvate (1)(19% activity relativeto indolepyruvate) (KOGA,1995;KOGA t al.,1992). Although there have be en no syntheticapplications of this enzyme thus far, the activ-

ity with pyruvate suggests the enzyme shouldbe fairly substrate permissive and clearly thereare excellent prospects for acyloin formation.Whole organism multi-enzyme processes mayalso be possible. The am inotransferase fromCandida maltosa accepts a range of fluoro-and methyl-tryptophans (BODE and BIRN-BAUM, 1991). Indolepyruv ate ferredoxin oxi-doreductase catalyzes the oxidative decarbox-ylation of arylpyruv ates (SIDDIQUIt al., 1997,1998).

ONH,

H 143a L-Tryptophan

O0

H 163 ndole-3-pyruvate

Indolepyruvate lyase

co24

H 164 Indole-3-acetaldehyde

&

H 165Indole-3-acetic acid

Fig. 57. The indole-3-pyruvate pathway for the con-version of L-tryptophan (143a) to indole-3-acetic ac-

id (165).

4.1.2 Aldehyde-Lyases (EC 4.1.2)

Aldehyde lyases are enzymes which cata-lyze the cleavage of alcohols to carbanions a ndaldehydes. The vast majority are aldolases, in-volved in carbohydrate metabolism. Two ex-amples (involving non-carbohydrates) areprovided h ere for illustrative purposes.

4.1.2.1 Threonine Aldolase(L-Threonine Acetaldehyde-Lyase,

EC 4.1.2.5)

Threonine aldolase is a PLP-dependent en-zyme which catalyzes the retro-aldol cleavageof L-threonine (166) to acetaldehyde (2) andglycine (158) (Fig. 58). A similar reaction oc-curs with serine, but the enzyme is both PLPand tetrahydrofolate dependent and is classi-fied as serine (or glycine) hydroxymethyl-

transferase (E C 2.1.2.1). This enzyme also acts



90 2 Lyases

Threonine aldoiase

COZ

H 2 Acetaldehyde

Fig.58. The threonine a ldolase catalyzed retro-aldolcleavage of L-threonine (166) to acetaldehyde (2)and glycine (158).

on L-threonine (166) (OGAWA nd FUJIOKA,1999). Perhaps surprisingly, enzymes havebee n isolated that can transform all four ster-eoisomers of threonine. In most cases, recogni-

tion of the a-am ino acid center (i.e., D- or L-) isfairly specific, bu t reco gnition of the stereo -chemistry of the carbo n bearing the hydroxylgroup is more relaxed (WADA t al., 1998; KA-TAOKA et al., 1997). These enzymes a re dis-cussed in detail in Chapter 1, his volume: Car-bon-Carbon Bond Formation Using Enzymes.

4.1.2.2 Mandelonitrile Lyase

(EC 4.1.2.10) and

Hydroxymandelonitrile Lyase

(EC 4.1.2.11)

Mandelonitrile lyase and hydroxymande-lonitrile lyase catalyzes the add ition of cyanideto benzaldehyde (4a) or p-hydroxybenzalde-hyde (4b) o give mandelonitrile (167a) or (p-hydroxy)mandelonitrile (16%) respectively(Fig. 59). These enzymes are oftern termedoxynitrilases. They occur pred ominantly inplants (WAJANT and EFFENBER GER,996),

where catalyze the form ation and elimination

4 a R = H 167

b R = O H

Fig. 59. The mandelonitrile lyase or hydroxymande-lonitrile lyase catalyzed addition of cyanide to benz-aldehyde (4a) or p-hydroxybenzaldehyde (4b) togive mandelonitrile (167a) or (p-hydroxy)mande-lonitrile (167b).

of cyan ogenic glycosides which are utilized forprotection against predators (WA GNE R t al.,1996). The synthetic potentia l of this reactionhas been reviewed in Chapter 6 of Volume 8a

of th e Biotechnology series (BUN CH,998) andelsewhere (POCHLAUER,998; FESSN ER,998;FESSNERnd WALTER,1997; EFFENBERGER ,1999).

4.1.3 0x0-Acid-Lyases (EC 4.1.3)

0x0-acid-lyases catalyze the cleavage of a-hydroxy carboxylic acids to a-carbonyl car-boxylic acids and carbanions in a retro-aldoltype reaction. This subsubclass includes seve r-al enzymes from the citric acid (Krebs) cycleand re lated pathways.

4.1.3.1 Isocitrate Lyase (Isocitrate

Glyoxylate-Lyase, Isocitrase,

EC 4.1.3.1)

Plants (but not animals) are able to convertacetyl-CoA (56) to oxaloacetate (33) via theglyoxylate pathway. The first step in this pa th-way is the c leavage of isocitrate (168) to glyox-ylate (151) and succinate (169) catalyzed byisocitrate lyase, which can b e rationalized as aretro-aldol reaction (Fig. 60) (R EHM AN ndMCFADDEN,996). Wh en the reaction is car-ried o ut in deu terium oxide, (2S)-[2-2H,]-suc-cinate (169) is prod uced, hence C-C bon dcleavage occurs with inversion of configura-

tion (DUCROCQt al., 1984).




168 Isocitrate

1 2H20

169(2S)-[2-’H, 1-Succinate 151 Glyoxylate

Fig. 60. The isocitrate lyase catalyzed retro-aldolcleavage of isocitrate to succinate and glyoxylate indeuterium oxide.

4.1.3.2 Acetolactate Synthase

Acetolactate Pyruvate-Lyase

(carboxylating), EC 4.1.3.181

a-Acetolactate (65) is produced biosynthet-ically by the con dens ation of two m olecules ofpyruvate (1)with concommittant decarboxyla-tion, catalyzed by acetolactate synthase (Fig.61). The enzyme is TP P dependent and hencethe mechanism is very similar to th at of pyru-vate decarboxylase described in Sect. 4.1.1.1(CAR ROL L t al., 1999; CHIPM ANt al., 1998;HILL t al., 1997) (Fig. 23).

4.1.4 Other Carbon-Carbon Lyases

4.1.4.1 Tryptophanase

[L-Tryptophan Indole-Lyase

(deaminating), EC 4.1.99.21

Tryptophanase is a PLP -depende nt enzym e,which catalyzes the cleavage of L-tryptophan(143a) to indole (170), yruvate (1) and am-monia (Fig. 62). It also catalyzes the net re-

verse reaction, as well as a,p-elimination,

0 1Pvruvate

- t

wooO ‘*

65 (2S)-cc-Acetolactate

Fig. 61. Acetolactate synthase catalyzes the decar-boxylative aldol condensation of two molecules ofpyruvate (l) , o give (2s)-a-acetolac tate (65).

H 143a L-Tryptophan

Tryptophanase

N H ~

H 0

170 Indole 1Pyruvate

Fig. 62.Tryptophanase catalyzed cleavage of L-tryp-tophan (143a) to indole (170),pyruvate (1) nd am-monia.

P-replacement an d a-hydro gen exchange withvario us L-amino acids.The p-subunit of trypto-phan synthase or desmolase (EC 4.2.1.20) cat-alyzes similar reactions and the coupling of in-dole w ith serine to give L-tryptophan. How ev-

er, this is a bifun ctional enzym e which utilizes



92 2 Lyases

the a-subunit to cleave 3-indolyl-~-glycerol3’-phosphate t o indole and glyceraldehyde 3‘ -phosphate (WOEHL and DU NN , 999). Tryp-tophanase is unable to accomplish this latter

reaction.Tryptophanase is a tetramer, consisting of

identical units (52,000 D a) each of which con-tains one molecule of PLP. It is activated bypotassium ions and inactivated by cooling,which causes release of the PLP and dissocia-tion into dimers (EREZt al., 1998).The X -raycrystal structure resembles that of aspartateaminotransferase and tyrosine phenol-lyase,with the active site at a dimer interface (Isu-PO V et al., 1998).

The mechanism of cleavage proceeds viaformation of an “external” aldimine (171)

HYxH

143a L-Tryptophan

(Fig. 63), between the substrate a nd the tryp-tophanase bound PLP group (116b). Protontransfer between the a-carbon atom and pro-tonation of the indole gives the quinoid inter-

mediate (172), which is poised for c leavage ofthe exocyclic indole C-C bond to give the am i-noacrylate (173).This undergoes hydrolysis tocom plete the cycle and yield the initial form ofthe enzyme (116b), pyruvate (1 ) and amm oni-um ions (IKUSH IROt a l., 1998;PHILLIPS,989).The transfer of hydrogen from the C-a centerto th e 3-position of th e indole ring involves abase with pKa = 7.6 and deprotonation/proto-nation of the indolic nitrogen involves a basewith pKa = 6.0. The mechanism of action oftryptophanase has a num ber of similarities to

that of tyrosine phenol-lyase and the section

H@A

0H,O or H,N-E-lysine

H 170 Indole

173 H 172

Fig.63.Minimal mechanism for the tryptopha nase catalyzed cleavage of L-tryptophan (143a) to indole (170),

pyruvate (1) nd ammo nia.Tryptophanase also catalyzes the net reverse reaction, but this is no t indicated for

the purposes of clarity.




salts and coenzymes were passed through acolumn packed with immobilized alanine racem-ase, D-amino acid oxidase and tryptopha-nase. The alanine racemase ensures that the

[3-"C]-~-alanine (175) s equilibrated with[3-11C]-~-alanine(174),which is oxidativelydecarboxylated to [3-"C]-pyruvate (176) ndamm onium ions. Both of these then react with5-hydroxyindole(177)catalyzed by tryptophan-ase to give [3-"C]-5-hydroxy-~-tryptophan(178) Fig. 64) in 60% yield (IKEMOTOt al.,1999; AXELSON et a l., 1992). Tryptophanasefrom E. coli has been assayed with a broadrange of "tryptophans" in which the benzenering has been replaced by pyridine or thio-phene. 4-Aza, 5-aza, 6-aza and 7-a za- ~-tr yp to-

phan (143a) Fig. 65) were all very slow sub-strates (kcat <1% of L-tryptophan), whereasP-indazoyl-L-alanine (179)was a better sub-strate and the thiophene analogs approachedthe reactivity of L-tryptophan (SLOANan dPHILLIPS,996; PHILLIPS,989).

on the latte r (Sect. 4.1.4.2) should be consultedfor furthe r details.

Curren tly, trypto phan is resolved industrial-ly by the enzyme catalyzed hydrolysis of ra-

cemic N-carbamoyl or hydantoinyl derivatives,but mo re direct rou tes would be desirable. TheEnterobacter aerogenes tryptophanase genewas expressed in the pyruvate producing E.coli strain W14851ip2. A fter 32 hours in cu lturethe 28.6 g L-' of pyruvate had been produc ed.Addition of ammonia and indole initiated L-

tryptophan production which reached 23.7g L-' after a further 36 hours (KAW ASAKIt al.,1996b).

The synthesis of com pounds containing l lCis particularly dem and ing because of the shor t-

half life (tl,z = 20.34 minutes) and the limitedavailability of co stly labeled precursors. The seand a number of other problems were solvedin a ne at syn thesis of [3-"C]-5-hydroxy-~-tryp-tophan (178).Racemic [3-"C]-alanine (174)(175), -hydroxyindole (177) lus ammonium

Alanine1;c racemase "C-f - - - B 0

H,N CO, H,N CO,

174 [3-"C]-D-Alanine 175 [3-"C]- L-Alanine

H 2 0 ~ c ~ $-A; acid oxidase

"CI

H H

177 5-Hydroxyindole 17 8 [3'-"C]- 5-Hydroxy-L-tryptophan

Fig. 64.Multi-enzyme system for the preparation of [3-1'C]-5-hydroxy-~-tryptophan178).



94 2 Lyases

H

143a L-Tryptophan

- N-IH

179 P-Indazoyl-L-alanine

Fig. 65. L-Tryptophan substitution and analogs

4.1.4.2 L-Tyrosine Phenol-Lyase(deaminating) (TPL, P-Tyrosinase,EC 4.1.99.2)

Tyrosine phenol-lyase (TPL , P-tyrosinase) isa PL P d epende nt enzyme, which catalyzes the

racemization, a,p-elimination and P-substitu-tion of L-tyrosine (141a)and other amino acidssuch as L- or D-serine (Ma), S-alkyl L-cys-teines (184c,d) and p-chloro-L-alanine (184e)(SUNDARARAJUt al., 1997). It also catalyzesthe net synthesis of L-tyrosine from pyruvate( l) ,ammonia and phenol (lma,Fig. 66) at highconcentrations of ammonium pyruvate. This

a R =H; henol

b"80=OH; atechol

0

NH,@GTyrosine phenol-lyase

a R =H; -Tyrosine

H3N 9 0 H 1 4 1R =OH; -DOPA

0

Fig. 66. Tyrosine phenol-lyase catalyzes the synthesis

of L-tyrosine (141a) from phenol (18Oa), ammonia

and pyruvate (1). By using catechol (18Ob) in placeof phenol L-DOPA(141b, 3-(3,4-dihydroxyphenyl)-L-alanine) can also be prepared.

latter reaction has attracted considerable at-tention because TPL accepts a range of aro-matic substrates in place of phenol (Tab. 6).

Tab. 6. Products from Tyrosine Phenol-Lyase(TPL) Catalyzed Reactions

Product Reference

L-Tyrosine (141a) cf. SUNDARARJUt al., 1997

~-[P-'~C]-tyrosine AXELSON et al., 1992

L-DOPA(141b) LEEet al., 1999a; SUZUKIt al., 1992

L-(/~-"C]-DOPA IKEMOTOt al., 1999;

Dopamine (142b) LEEet al., 1999a, ANDERSONt al., 1992a, b

(2S,3R)-/3-methyltyrosine KIMand COLE, 999

Fluorinated tyrosines PHILLIPSt al., 1997; VONTERSCHt al., 1996; FALEEV

et al., 1995,1996; URBANnd VON TERSCH,999

2-Chloro-~-tyrosine FALEEVt al., 1995:

2-Methyl- and 3-methyl-~-tyrosine FALEEVt al., 1995:

2-Azido-~-tyrosine HEBEL t al, 1992

l-Amino-2-(4-hydroxyphenyl)ethyl hosphinic acid KHOMUTOVt al., 1997




For example, substitution of catechol (18Ob)enables the synthesis of L-DOPA (141b),which is used in the treatment of Parkinson’sdisease (LEE et al., 1999a).

TPL is found primarily in enterobacteriaplus a few arthropods and plays a role in thebiosynthesis of shikimates (DEWICK1998).TPL‘s have been identified in Erwinia herbico-la (and expressed in E. coli, FOORt al., 1993),Citrobacter freundii (POLAK and BRZESKI,1990), Citrobacter intermedius (NAGASAWA,1981) and Symbiobacterium sp. (LEE et al.,1997). X-ray crystal structures have been de-termined for the TPLs from Erwina herbicola(PLETNEV, 997, 1996) and that from Citro-bacter freundii with bound 3-(4 ’-hydroxyphe-

ny1)propionic acid (SUNDARARAJUt al., 1997)and without (ANTSONt al., 1992,1993).

The mechanism of action of tyrosine phe-nol-lyase is similar to that of tryptophanase.Aldimine formation between the “internal”PLP-aldimine (116) (Fig. 67) and L-tyrosine(141a) yields an aldimine which is deprotonat-ed by “base-1’’ (pK, = 7.6) to give the qui-nomethide (181). Concurrent deprotonationof the phenolic hydroxyl group by “base-2’’(pK, = 8.0) and C-4 protonation of the phenol

by “base-1-H+”gives the quinone (182), whichundergoes protonation by “base-2-H+” andrate limiting bond cleavage to phenol (MOa),plus the pyruvate enamine derivative of PLP(173) (cf. Fig. 63). Hydrolysis restores the rest-ing state of the enzyme ( A X E U S O N et al.,1992). The elimination and substitution reac-tions of serine derivatives bearing a p-nucleo-fuge (184) may be rationalized by an abbrevi-ated form of this mechanism. In this case aldi-mine (185) (Fig. 68), formation and deprotona-tion results in p-elimination to give the the py-

ruvate enamine derivative of PLP (173) direct-ly (CHENand PHILLIPS, 993). The identity ofbase-1 remains uncertain despite the X-raycrystal structure data. In Citrobacter freund ii,TPL it is probably Lys-257 or a water moleculebound to Lys-256. Base-2 is almost certainlyArg-381 (SUNDARARAJUt al., 1997).

TPL has been identified in several species(mainly Enterobacteriaceae), grown on mediacontaining L-tyrosine (lBOa), but TPLs aregenerally inhibited by catechol (180b) whichlimits their application to the synthesis of L-

DOPA (141b). A thermostable TPL, which is

tolerant of catechol, was identified in a ther-mophile, Symbiobacterium sp. SC-1, however,this only grows in coculture with Bacillus sp.SK-1, which makes large scale cultivation ex-

tremely difficult. The gene for TPL was identi-fied and overexpressed in E. coli (15% of solu-ble proteins) and isolated in 48% yield (LEEetal., l996,1997).The enzyme is a tetramer (M W202 kDa), consisting of four identical subunits,each of which contains one molecule of pyri-doxal 5 ’-phosphate (PLP). Remarkably, theenzyme retains full activity at 70 ”C for at least30 minutes (LEE et al., 1999b).Symbiobacteri-um thermophilum is another obligate, symbiot-ic, thermophile that only grows in coculturewith a specific thermophilic,Bacillus sp .,strainS . It produces both tryptophanase and TPL(SUZUKI t al., 1992). Expression of the genefor TPL in E. coli yielded a 375 times increasein TPL production relative to that of S. thermo-philum (HIRAHARAt al., 1993).The amino ac-id sequences of the TF’Ls from the two Symbi-obacterium sp. differ at only three positions(LEEet al., 1997).

Symbiobacterium sp. SC-1TPL has been uti-lized in a multi-enzyme process for the pro-duction of dopamine (142b) from catechol.Condensation of catechol(180b), pyruvate (1)and ammonium salts catalyzed by TPL gave L-

DOPA (141b) (Fig. 69), which was decarboxy-lated with rat liver L-DOPA decarboxylase orStreptococcus faecalis tyrosine decarboxylase.The L-DOPA decarboxylase was inhibited byL-DOPA concentrations greater than 1 mM,whereas tyrosine decarboxylase was unaffect-ed up to 40 mM, consequently tyrosine decar-boxylase was selected for further develop-ment. Decarboxylation was carried out in anenzyme reactor at 37 “C or 12 hours, with pH

control by addition of hydrochloric acid andsupplements of PLP to compensate for decar-boxylative transamination. At 100 mM L-

DOPA (141b), conversion to dopamine (142b)was quantitative. When the decarboxylationwas run with in situ synthesis of L-DOPA, theproductivity was much lower than with twoconsecutive “single pot” reactions. (LEE et al.,1999a, b). Erwinia herbicola TPL and Strepto-coccus faecalis tyrosine decarboxylase havebeen investigated as components of a multi-enzyme system. Kinetic studies were run at pH

7.1 which is a compromise value dictated by



96 2 Lyases

NH,O-,

0

1 Pyruvate

2 0

____) 0 3 ' 9 H1 H.0

:TOHH' I 141a L-TyrosineOH

/ N \

I

116

aX =0;LP

b X = N-E-lysine t Tyrosine phenol-

Base-2

Base-1

18 1

2 0

I

0 0 3 P 182

H I

1832 0

0 3 p y

Fig.67. Mechanism of action of tyrosine phenol-lyase (TPL) with L-tyrosine.



4 Examples of L y a m 97

0

0

1 Pyruvate

116

a X = O ;PLP-H,O

b X = N-&-lysine

184

0 a R = O Hb R = O B zc R = S M ed R = S E tOe R = C I

Tyrosine phenol-

Base-2Base-1

185

2 0

0

Tyrosine phenol-\ yase

186

I 173

Fig. 68. Mechanism of action of tyrosine phenol-lyase (TPL) with p-substitutedL-alanines(184).

differing values for the pH optima. The TPLfollowed pseudo 1st orde r kinetics with cate-chol, pyruvate and amm onium salts, whereas,the tyrosine decarboxylase was inhibited bydopamine and combinations of pyruvate andcatechol (ANDERSO Nt al., 1992a). The sameprocess has also been run w ith whole cells ex-pressing the relevent enzymes. Erwinia herbi-cola cells with phenol tyrosine-lyase activityan d Streptococcus faecalis cells with tyrosinedecarboxylase activity (AND ERSO Nt al., 1987)were co-immobilized in glutaraldehyde cross-

linked porcine gelatin beads (mean diameter

2.8 mm). The beads were placed in a packedbed reactor together with catechol, pyruvateand ammonia. Dopamine was produced, butthe conversion was low (AND ERSO N t al.,1992b). Whole Erwinia herbicola cells weremicroencapsulated in alginate-poly-L-lysine-alginate mem braned m icrocapsules,and testedfo r TPL activity in a rotary shak er incubator.An agitation rate of at least 240 rpm was re-quired to ensure that the microencapsulatedcells achieved the activity of the free cells(LLOYD-GEORGEnd CHAN G, 993,1995).



98 2 Lyases

180bCatecholHO

0

OH

coy

HO

141b L-DOPA

Rat liver L-DOPAdecarboxylase or

Streptococcus aecalistyrosine decarboxylase

?H

+

b"

oluene dioxygenase

187Toluene cis-glycol

dehydrogenase

H3N0Po;PL

Pyruvate 1

0

.OH

141b L-DOPA NH4@ 180b Catechol

Fig. 70. Hybrid pathway for the synthesis of L-

DOPA (141b),expressed in E. coli and Pseudomo-nus aeruginosa (PARK t al., 1998).

expressing TPL reduced 100 mM phenol to8 mM within 24 hours at 37 O C (LEEet al.,

HO 1996).

142bDopamine

4.2 Carbon-Oxygen Lyases

(E C 4.2)Fig. 69. The multi-enzyme synthesis of dopamine(142b) (LEE t al., 1999).

In a neat approa ch, toluene dioxygenase, o-luene cis-glycol dehydrog enase and TPL fromCitrobacter freundii (BAI and SOMERVILLE,1998) have been coexpressed in E. coli. The

maximum concentration of L-DOPA (141b)(Fig. 70) achieved in 4 hours was only 3 mMdue t o the toxicity of benzene (187). However,with Pseudomonas aeruginosa 14 mM L-

DOPA was achieved in 9 hours (PARK t al.,1998).

Carbon-oxygen lyases catalyze the cleavageof a carbon-oxygen bo nd, with the form ationof unsatu rated products. Th e vast majority ofenzymes in this subclass are hydro-lyases (E C4.2.1) which catalyze the elimination of water.Elimination of sugars from ca rbohydrates ( EC

4.2.2) a nd a few othe r (mo stly complex) cases(E C 4.2.99) complete the subclass

4.2.1 Hydro-Lyases (Hydratases

and Dehydratases,EC 4.2.1)TPL have been suggested as a m eans for re-

moving phenol (166a) from waste water byconversion to L-tyrosine (141a) which is insol-uble. Optimal concentrations of substrateswere 60 mM ph enol, 0.1M pyruvate a nd 0.4 Mamm onia and th e reaction was effective up to

70O C, p H 6.5-9.0. Intac t or aceton e dried cells

Hydro-lyases are a group of enzymes thatcatalyze cleavage of carbon-oxygen bondswith the elimination of water and typically(but not exclusively) the formation of an al-kene. In most cases the elimination of water

occurs adjacent to a carboxylic acid, ketone or




ester (189) (Fig. 71). Deprotonation occurs ad-jacent to the carbonyl group to give an enolate(190) which undergoes @-elimination o givean a,@-unsaturatedcarbonyl compound (191).

This is the common ElcB mechanism (mono-molecular elimination from the conjugatebase) or in Guthrie-IUPAC nomenclatureAxhDH DN (GUTHRIE and JENCKS,1989).When there is a hydroxyl group adjacent to thecarbonyl group (as commonly occurs in sug-ars), tautomerization of the enol (191, X =

OH) occurs to give the 2-ketocarboxylic ester(192).

There are two stereochemical classes of hy-dratase-dehydratase enzymes.Those that cata-lyze the addition of water to a,@-unsaturated

thioesters have syn-addition-elimination ster-eochemistry, whereas those that catalyze theaddition of water to a,@-unsaturatedcarboxyl-ates have unti-stereochemistq. In a limited in-vestigation of the spontaneous reaction, addi-tion of deuterium oxide to a$-unsaturatedthioesters was moderately selective for unti-over syn-addition (4.3: ) , whereas a,@-unsatu-rated esters only showed a minute preference(1.3: 1) (MOHRIGt al., 1995).Thus the stereo-chemical preferences of the enzyme catalyzed

reaction do n o reflect the innate stereochemi-cal preferences of the solution phase reaction.

189 I 190

0

192

0

19 1

Fig.71.Mechanisms for the elimination of P-hydrox-yl-esters (189) o @unsaturated esters (191) nd a-

ketoesters (192).

4.2.1.1 Carbonic Anhydrase

(Carbonate Dehydratase,

EC 4.2.1.1)

Carbonic anhydrase (CA) is a zinc depen-dent metalloprotein which catalyzes the hy-dration of carbon dioxide to carbonate. It is aubiquitous enzyme present in animals, plants,algae and some bacteria and is frequentlypresent as several different forms in the sameorganism. There are at least 14 different formsin human tissues (THATCHERt al., 1998) andmuch effort has been devoted to efforts to de-velop inhibitors, principially for eye conditions(DOYONnd JAIN, 999;KEHAYOVAt al., 1999;

WOODRELLt al., 1999). Three evolutionarilydistinct classes(a, , 7 )have been defined andwere originally thought to be characteristic ofanimals, plants and bacteria, however, as morehave been discovered, it has emerged that theyare distributed indiscriminately (LINDSKOG,1997; HUANG t al., 1998; ALBER t al., 1999).Extraction of bovine CA has been recom-mended as an undergraduate experiment(BERING t al., 1998).

The CA catalyzed hydration of carbon diox-

ide is one of the fastest known enzyme reac-tions. Typical kinetics parameters are Khl 1.2 .lo- ' M, k,,, 1.0 . 106 S-' , VM,,/K,, 8.3 . 10'M-ls-'. Although KM is unremarkable, thevalue for k,,, is extremely high and only ex-ceeded by a few enzymes (e.g., catalase). Theactive site of CA consists a shallow depressioncontaining a zinc ion bound to three histidinesand a hypernucleophilic hydroxide ligand(193) (Fig. 72). This adds to carbon dioxide togenerate bicarbonate (194),which is displacedby water (195) and deprotonated in the rate

limiting step to regenerate the hydroxide li-gand (193) (QIANet al., 1999;EARNHARDTtal., 1999; MUGURUMA,999).CA also catalyzesthe hydration of aldehydes and ketones, andthe hydrolysisof esters, phosphates (SHERIDANand ALLEN, 981) and cyanamide (BRIGANTIet al., 1999). There is considerable unrealizedpotential for the use of CA as an esterase. Bo-vine CA I11 increases the rate of decarboxyla-tion of amino acids by L-arginine, L-lysine or L-ornithine decarboxylases used in a biosensor(BOTRE nd MAZZEI, 999).



100 2 Lyases

0

I

Imd- Zn- 9194

fmd H

HCOYHzolimd- Zn- 9

I

Imd H 19'

Fig. 72. Minimal mechanisms for the hydration of

carbon dioxide by carbonic anhydrase. Zinc ispresent as Zn(I1). Hydroxide, bicarbonate and waterligands are depicted without charge transfer to thezinc atom.

4.2.1.2 Fumarase (EC 4.2.1.2)

IntroductionFum arase catalyzes the addition of w ater to

the alkenic bond of fumarate (1%) to give L-

malate (197, ig. 73). It is one of a nu mber ofrelated enzymes which catalyze additions tothe alkene bond of fumarate, such as aspartase(E C 4.3.1.1, amm onium, S HI et al., 1997), andthe amidine lyases; arginosuccinase (EC4.3.2.1, L-arginine, CO HE N-K UPEICt a l., 1999;TURNERt al., 1997) and adenylsuccinase (E C4.3.2.2, A MP, LE E e t al., 1999). Many of theseenzymes have high sequence homology, eventhough they catalyze different reactions ornone at all (WOODS et a l., 198 6,1988a).For ex-ample 81-crystallin from duck lens is catalyti-

cally inactive (PIATIGORSKYnd WISTOW,

196 Fumarate

0& OF 197 L-(8-(-)-Malate

Fig. 73. Fumarase catalyzes the interconversion offumarate (1%)and L-malate (197).

02

1991; VALLEEet al., 1999), even thoug h itshares 90% sequence homology with argino-succinase (SIMPSO Nt al., 1994; cf. W u et al.,1998).

Fum arase is a remarkably efficient enzyme.It has been estimated that the rate enhance-men t relative to spon taneous hydration of fu-ma rate is som e 3.5 - 1015 fold w hich equ ates toa transitio n state stabilization of -30 kcalmol-'. On this basis it is th e seco nd most effi-cient enzyme known (BEAR NE nd W OLFEN-DEN , 995) after orotidine 5 ' monophosphatedecarboxylase (-3 2 kcal mol-', EHRLICHtal., 1999).

Fumarase is widely distributed in animals,plants (BEHAL nd OLIVE R, 997; KOBA YASHIet al., 1981), yeast (KER UCH ENK Ot al., 1992),fung i and bacteria (BATTATet al., 1991) and ar-chaebacteria (COLOMBOt al., 1994). In hu -mans, mutations of fumarase (CO UG HLINt al.,1998) ar e associated with diseases such as fu-maric aciduria and progressive encephalopa-thy (DEVIVO, 993; BOU RGE Nt a l., 1994).

In E. coli, fumarase activity arises fromthree distinct genes, fum A, fumB and fumC(BELL t al., 1989).The gene produc ts o f f u m AandfurnB are fumarase A (FUM A) and fuma-rase B (FU MB ), which are exam ples of class Ifumarases. These ar e iron depen dent (4Fe-4S),heat labile, dimeric enzymes (MW 120 kDa,UED A t al., 1991),whereas fumC yields fuma-rase C (FUMC) a class I1 fumarase. These areiron-independent, heat stable, tetrameric en-zymes (MW 200 kD a), which include yeast fu-marase, Bacillus subtilis fumarase (c i tC ) and

mammalian fumarases. Class I1 fumarases



4 Examples of Lyases 10 1

that fumarase C m ay act as a back-up to fuma-rase A under conditions of iron deprivation(HASSETT et al., 1997). The am ino acid se-quences of fumarase A and fumarase B have

90% sequence homology to each other, but al-most no homology to the class I1 fumarases.Each monomeric unit of fumarase A and Bcontains a [4Fe-4S] cluster as do Euglena gruc-ilis fumarase (SHIBA TAt al., 1985; RIK IN ndSCHWARTZBACH,989) and Rhodobacter cap-sulutus fumarase. Other hydrolyases whichcontain a non-redox active [4Fe-4S] cluster(review,FLINTnd A LLE N, 996) include acon-itase (G RU ER t al., 1997;BEINERTt al., 1996),dihydroxy -acid dehy dratas e (FLINT et al., 1996;LIMBERG nd THIEM,1996), isopropylmalate

isomerase, phosphogluconate dehydratase,tartrate dehydratase, maleate dehydratase, lac-tyl-CoA dehydratase, 2-hydroxyglutaryl-CoAdehydratase and Peptostreptococcus asacchar-olyticus serine dehydratase (HOFMEISTERtal., 1994).There is little or no sequenc e homol-ogy between aconitase and fumarase A, al-though both bear the dipeptide moietyCys-Pro which is often found in the sequenc esof iron-sulfur proteins. Bo th are oxidized byoxygen, but fum arase A is more susceptible to

further oxidation and “reactivation” by ironand a reductant takes longer (FLINTet al.,1992b). Fumarase A, fumarase B, aconitaseand E. coli di-hydroxy-acid dehydra tase ar e in-activated by superoxide (OF )with release ofiron and by hyperbaric oxygen (FLINTet al.,1993; UED A t al., 1991).

The addition of water to fuma rate catalyzedby fumarase A is stereospecific (Fig. 75). Thehydroxyl group and the hydrogen undergotrans-addition with the incoming hydrogenadded to the (3R)-position (199).Prolonged

incubation does not result in the form ation of[’HI-fumarate, which testifies to the stereo-chemical fidelity of the reaction (FLJNT t al.,1994).There is no hydrogen isotope effect andhence hydrogen abstraction is not the ra te de-termining step.

Fumarase A catalyzes the transfer of l80

from (2S)-[2-’80,]-malate (200) (Figs. 76, 77)and ’H from (2S,3R)-[3-2H,]-malate (206) toacetylene dicarboxylate (202) to give “0 and*H labeled oxaloacetate enol (203) (208)

(FLINT t al., 1992a; FELL t al., 1997), which

isomerizes to the keto-form of oxaloacetate

have high sequence homology and have beenextensively studied, m echanistically an d kinet-ically (W OO DS t al., 1988b).

Class I FumarasesIn E. coli, fumarase A and fumarase C are

expressed under aerobic cell growth condi-tions, whereas fumarase B is more abundantduring anaerob ic growth (TSENG , 1997;WO ODS nd GUEST, 987). Fumarase A (andC) act in the citric acid cycle during aerobicmetabolism, whereas fumarase B enables fu -marate to act as an anaerobic electron accep-tor in the reductive sequ ence leading from ox-aloacetate (28) to succinate (198) (Fig. 74)(WOOD S t al., 1988b).There is some evidence

0

0

28 Oxdoacetate2

NADH + H@

Malate dehydrogenase

NAD@

OH

197 L-Mdate

0

0 2196Fumarate

FADHZ

Succinate dehydrogenase

FA D

198 Succinate

Fig. 74. The role of fumarase in anaerobic metab-

olism.



102 2 Lyases

0

'02& 'O2 196 Fumarate

H 'H 19 9 (2S,3R)-[3-'H1]-Malate

Fig. 75. Fumarase A, catalyzed addition of the ele-ments of water occurs stereospecifically n the trans-

orientation.

(204) (209). Further oxygen or hydrogen ex-change was prevented by reduction with sodi-um borodeuteride to the malates (205) (210).It was calculated that a t an infinite concen tra-tion of acetylene dicarboxylate (202), some33% of th e l80 nd 100% of the 'H would betransfered. This indicates that these atom s arestrongly boun d at th e active site, presumably

with the oxygen acting as a ligand to the [4Fe-4S]-cluster and th e hydrogen bound to a basicresidue (211) (Fig. 78. FLINT nd MCKAY,1994). Moreover fumarase also catalyzes theisomerization of oxaloacetate enol(208) to theketone form (209), at the same active sitewhich is responsible for hydration-dehydra-tion. The rea ction occurs with retention of theenolic oxygen and is stereoselective with thehydrogen delivered to the 3-pru-(S)-position(FLINT, 993).

Fumarase A has high substrate specificity

which is typical of fumarases in general. Theonly substrates which have reactivity which iscomparable to fuma rate (relative rate = 100,10 mM substrate) are L-malate (41), acetylenedicarboxylate (97) and (2S,3S)-tartrate (11).2-

Fluorofumarate (13) is a reasonably reactivesubstrate. It undergoes addition of th e hydroxygroup a t the 2-position to give a 2,2-fluorohy-drin which spontaneo usly eliminates hydrogenfluoride to give oxaloacetate. This reaction isidentical to that catalyzed by th e class I1 fuma-rases and is described in detail in the subse-

quent section.

H"0

200 (2S)-[2-'801]-Malate

Fumarase A

H

.oz&co,.

H 196Fumarate201 Fumarase A-[ '801]

@o,c-cEc-co,o252 Acetylene dicarboxylate

Fumarase A

Fumarase A

N&H,I180 2H

205 rac-[2-*HI,2- '*0,]-Malate

02c

Fig 76. The hydration of acetylene dicarboxylate cat-alyzed by fumaraseA-["Ol] (201) from E. coli.

In summ ary, class I fumarases are rare, un -stable enzymes bearing a [4Fe-4S] clusterwhich has a non-redox role.The stereochemis-try of hydration of fumara te is identical to that

of the class I1 fumarases




OH

206 (2S,3R)[3-'H1]-Malate

H 'H

0 0,c- e op

202Acetylene dicarboxylate

F u m e A

'A 208 3-ZH,]-Oxaloacetate nol

Fumarase A

I

211

BI

.r.Ahr

0 2 , Fig. 78. Outline mechanism for the hydration of fu -2H 2w 3R)-[3-ZH1]-0xaloacetate

marate (1%) by a [4Fe-4S]-cluster.

N ~ BH.,1H 'H 210 2RS,3R)-[2JHl, 3-'Hl]-

Malate

Fig77.The hydration of acetylene dicarboxylatecat-alyzed by fumaraseA-['H,] (Un)rom E. coli.

Class I1 Fumarases

In S. cerevisiue both cytoplasmic and mito-

chondrial fumarases (class 11) are encoded by

a single nuclear gene FUM1.The mitochondri-

a1 enzyme is produced with a 23-residue trans-

location sequence, which is responsible for di-

recting it to the mitochondria and is removedduring importation into the mitochondria

(KNOXet al., 1998). Similar systems are em-

ployed in humans, pigs (O'HAREand Doo-NAN, 1985), rats (SUZUKIt al., 1989;TUBOItal., 1990) and mice, which apparently only uti-

lize class I1 fumarases. In S. cerevisiue, some

30% of the enzyme activity is associated with

the mitochondria1 fraction and the remaining

70% with the post-ribosomal fraction, al-

though the specific activity of the former is

some 3-4 fold higher. (W u and TZAGOLOFF,

1987).



10 4 2 Lyases

StereochemistryEarly work demonstrated that the addition

of deuterium oxide to fumarate catalyzed byfumarase was stereospecific (FISHERet al.,

1955). Initially the evidence suggested that theaddition occumed with syn-stereochemistry(FA RR AR t al., 1957; ALBERTYnd BENDER,1959). How ever, the synthesis of stereospecifi-cally labeled standards, clearly indicated anti-stereochemistry with the hydrogen a tom in the(3R)-position (GAWR ON nd FONDY ,1959;ANET,1960; GAW RONt al., 1961), which wasconfirmed by a neu tron diffraction study (BAUet al., 1983). This stereochemical ou tcome isidentical to that of the class I fumarases (Fig.75).

Enzymology and StructuresKinetic analysis of fumarase hydration is

complex.A t low substrate concentrations, clas-sical Michaelis-Menten kinetics are followed,at substrate concentrations greater than fivetimes K M substrate activation occurs andabove 0.1 M inhibition occurs. Although theconversion of substra te is fast, recycling of th eenzyme t o the active state is slow and is cata-lyzed by anions, including the substrates (R o-

SE,1997,1998;BEHAL nd OLIVER, 997).Thus,although there is no deuterium isotope effectin the hydration reaction, k,, is redu ced if d eu-terium oxide is used in place of water as sol-vent. X-ray crystal structures for E. coli(WEAVERnd BANASZAK,996;WEAVER t al.,1995) and yeast (WEAV ER t al., 1998; KE RU -CHENKO et al., 1992) fumarases have revealed abinding site for anions, structurally close to th eactive site, which acts allosterically and is ap-parently responsible for the unusual kineticsnoted above. M utation of a histidine residue t o

asaparagine in the active site (H188N), result-ed in a large decrease in specific activity, butthe same change at the second site (H129N)had essentially no effect, however, both m uta-tions reduced the affinity for carboxylic acidsat the respective sites (WEAVERt al., 1997). Itmay be that the asparagine residue a t the sec-ond site in the mutant acts as a surrogate forthe ligand’s carboxylate group and hence com-pensates for the lack of normal allosteric ac-tion. The pig liver enzyme also has a secondsite, as deduced from a kinetics study (BEE CK-MANS and VAN DRIESSCHE,998). A prelimi-

nary report of crystallization and diffractiondata for this enzyme was reported, however,thus far there has been no crystal structure(SACCHETTINIt al., 1986).

The extreme stereochemical fidelity of fu-marase catalysis has enabled it to be used inthe synthesis of stereospecifically labeled ma l-ates (review, YOU NG, 991).As has been notedpreviously, (2S,3R)-[3-2Hl]-malate (199) canbe prepared by fumarase catalyzed addition ofdeuterium oxide to fumarate (1%) (AXELSSONet al., 1994) (Fig. 75). Th e co nverse reaction inwhich deu terated fumarate (213) s hydratedby water gives (2S,3S)-[2,3-’H2]-malate(214)(Fig. 79) (EN GLA RDnd HANSON969; AXE LS-SONet al., 1994).Alternatively, this can b e p re-pared by malate dehydrogenase catalyzedstereospecific reduction of fully deu terated ox-aloacetate (216),with (4R)-[4-’H1]-NADH, togive fully deuterated malate (215) and ex-change of the (3-pro -R)-deuteron with hydro-gen ions catalyzed by fumarase (CLIFFORDtal., 1972; YO UN G, 991).In contrast stereospe-cific reactions of the dideuteromalate (217)gave a mixture of produc ts (Fig.80).Fumarasecatalyzed dehydration in tritiated water ster-eospecifically gave the mo nodeuterofum arate

(216).How ever, this can bind in the active sitein two different orientations, thus hydrationwas non-selective and a mixture of isotopom-ers (219) (220)were form ed (LE NZ t al., 1976;YOUN G, 991).

Malonate (89) Fig. 29) is propro chira l, ie. atleast two groups attached to the central carbonatom must be labeled to render it chiral andhence amenable to stereochemical studies ofdecarboxylation or homologation (FLOSS tal., 1984). This has been achieved in an ex-tremely neat way by the fumarase catalyzed

hydration of [1,4-13C,]-fumarate(221) o givelabeled malate (222),which can be stored forprolonged periods (Fig. 81). Rapid cleavagewith perma nganate, at p H 10, gives chiral mal-onate (223),with minimal opportunity for ex-change of the acidic hydrogens (H UA NG t al.,1986; JORDANt al., 1986). This m ethodologyhas been used in studies of decarboxylation(MICKLEFIELDt a l., 1995;HANDAt al., 1999),fatty ac id (JO RD AN t al., 1986; JORDANndSPENCER,991),6-methylsalicyclic acid (SPE N-CER and JORDAN,990;JORDAN and SPENCER,1990, SPEN CERnd JORDAN,992b) and orsel-




2H

4 213 [2,3-ZH2]-Fumarate

H2O

Fumarase

0

214 (2S,3s)-[2,3-2H2]-Malate

tFumarase

0

215 (2s)-[2,3,3-ZH3]-Malate

tI Malate dehydrogenase

(~R)-[~-’H, ]-NADH

r0 2

0

216 [3-ZH2]-Oxaloacetate

Fig. 79. Two routes to (2S,3S)-[2,3-”Hz]-malate214)(ENGLARDnd HANSON,969; CLIFFORDt al., 1972;YOUNG,991).

linic acid biosynthesis (SPENSERnd JORDAN,1992a).

Non-Natural SubstratesFumarases are highly specific for fumarate

(1%) and L-malate (197). A limited range ofother unsaturated dicarboxylic acids undergohydration, but with the exception of 2-fluoro-

fumaric acids and 2,3-difluorofumaric acid, the

0

21 7 (2S)-[3,3-’Hz]-Malate

Fumarase

2~~~

H ZH

H218 [2-2H,]-Fumarate

H

I I

219 220

Fig. 80.Stereospecific dehydration-hydration yieldsa mixture of isotopomers (219) (2u)) (LENZ t al.,

1976).

reaction rates are m uch slower. Pig liver fuma-rase catalyzed hydration of 2-chlorofumarate(224),yields enantiomerically pure L-threo-chloromalate (>99.5% ee) (225) (Fig. 82). Un-

fortunately, the reaction has an unfavorableequilibrium co nstant (Keq= 6.2), which limitsthe theoretical yield to 86%. However, using

PA N gel imm obilized fumarase, 63.7 g of ch lo-rofumaric acid was converted to 54.5 g (76%yield) of L-fhreo-chloromalicacid, in two reac-tiodisola tion cycles, over two weeks. The re-covered immobilized fumarase retained 60%of the initial activity (FINDE IS nd WH ITE-SIDES, 1987). Interestingly, the correspond ingfluorofumarate (228) reacts with the oppositeregiochemistry to give an unstable fluoroa lco-ho l (229),which spontaneously eliminates hy-drogen fluoride t o give the tritiated oxaloace-tate (230) Fig. 83). Th e stereoch emistry of the

tritium label was determined by reduction



106 2 Lyases

KMn04, pH 10,

H 20, OOC, 5 mins

1 13c0,00 2 I 221 [1,4-13C2]-Fumarate

-'H 22 2 (2S,3R)-[1,4-'3C,,3-2H]-Malate

-'H 22 3 (R) - [ -'3C,2-2H]rnalonate

Fig. 81. The synthesis of chiral malonate(HUANGt al., 1986).

with malate dehydrogenase and treatmentwith fumarase, which yielded unlabeled fuma r-ate (196).Consequently the tritium label musthave been located at the (3R)-position. (MA R-LETTA et al., 1982).The hydration of 2,3-difluo-rofumaric acid (233) also proceeds with elimi-nation of hydrog en fluoride, but in this case insitu redu ction yields L-threo-fluorom alic acid(235) which was isolated as the dimethyl esterin 49% yield (Fig. 84) (FINDEIS nd WHITE-SIDES, 1987). Fumarase binds the nitronates(237a,b) more tightly than fumarate (1%),

whereas the nitrocompounds (236a,b) arebarely inhibitory. Similarly the nitonates(237b,c,d) binds to aspartase, more stronglythan aspartate. Although , all of the analogs areinhibitors but not substrates, the fact that theybind so well indicates there are prospects fordesigning unnatural substrates (PORTER ndBRIGHT, 1980).

BiotechnologyFumaric acid has been m anufactured by fer-

mentation of glucose by Rhizopus nigricans

but now adays most is produced a s a byproduct

OH

c l 225 L-threo-Chloromalate= (2R,3X)-3-chloromalate

I

J Steps ti

226 D-(+)-erythro- 227 2-Deoxy-D-ribose

Fig.82. Fumarase catalyzed hydration of 2-chloro fu-marate (224).

Sphingosine

from D,L-malic acid m anufacture. M aleic anhy-dride (239) is produce d on a very large scale byoxidation of butane (238) or crude benzene or

naphthalene fractions (Fig. 86). The majorityof this is used in the pro duc tion of plastics an dpaints a nd detergents. Hydrolysis of maleic an-hydride gives maleic acid (240),fumaric acid(241) r D,L-malic acid (242).All of these acidsare used as acidulants (preservatives) forcanned fruit, preserves, soft drinks, cosmeticsand pharmaceuticals. However, there is con-siderable (and increasing) dem and for L-malicacid, which is the naturally occurring enan-tiomer (a pple acid).The curre nt world produc-tion of L-malic acid is about 500 tonnes per

year, most of which is used in food and phar-




F

228 2-Fluorofurnarate

Spontaneous

iF

Oxaloacetate

NADH + H@


NAD@

0

H 3H 231 (2S,3R)-

[3-3H,]-Malate1'-H

A 196Fumarate

Fig. 83.Fumarase catalyzed hydration of 2-fluorofu-marate (226).

F 232 2,3-Difluorofumarate

SpontaneousI- F

n

F H 234(35')-3-FlUOKO-

I oxaloacetate

NADH + H@


NAD @

L-rhreo-Fluoromalate

= (2R,35')-3-fluorornalate

Fig. 84.Fumarase catalyzed hydration of 2,3-difluor-ofumarate (232).

maceuticals. It has been used in organic syn-thesis as a chiral starting material (e.g., Clozy-lacon (24%) (Fig. 87), BUSER, t al., 1991;Pyr-rolam A (247) (Fig. 88), HUANGt al., 1999),

however, such applications are rare because itis less versatile than the tartaric acids, whichhave a G-axis of symmetry (GAWRONSKIndGAWRONSKI,999).

Most of this demand for L-malate (197) issatisfied by the fumarase catalyzed hydration

of fumarate (196) (Fig. 75). This has been ac-



108 2 Lyases

a C13CCH0, Hz S04 (81% yield);

b BH3.SMez,THF,

c aq NaHC03 (20-55% yield)

a X = Hb X = O H

d X = NH3@c X = N H Z

H 237

Fig. 85. Fumarase inhibitors (PORTERnd BRIGHT,1980).

238Butane

0 239Maleicanhydride

kHZ0rnHO,C C02H 240Maleic acid

H 0 2 6 OZH 241Fumaxic acid

HOZ& OzH 242 DL-Malic acid

Fig. 86.?he manufacture of C,-dicarboxylic acids.

a (CF3SO2),0, pyridine, CCI,;

b 2,6-Dimethylaniline, K 2C 03 (49%yield);

c MeOCH,COCl, DMF, PhMe (96%yield)1 ?Me

a R = H (88.6%ee);

b R = C1; Clozylacon

Fig. 87. Synthesis of the fungicide CGA 8oo00,Clo-zylacon (245b) (Ciba-Geigy) from L-malic acid (243)(BUSER,t al., 1991).

complished with immobilized whole cells ofBrevebacterium ammoniagenes or laterly B.flavum (CHIBATA,t al., 1987b; WANG, t al.,1998).Various imm oblized gels have been ap-

plied in the continuous process such as poly-acrylamide and rc-carrageenan (CHIBATA, tal., 1995,1987 a).Fum arase activity (unit mL -gel) and relative productivity (%) of cell im-mobilized Brevibacterium flavum increasedrespectively from 10.2 and 273 on polyacryl-amide to 15.0 and 897 on K-carrageenan.Theimmobilized B. flavum on either gel showedhigher fum arase activity and relative produc-tivity compared to B. ammoniagenes cells onpolyacrylamide gel (TOSA t al., 1982;TANAKAe t al., 1983a, 1983b, 1983c, 1984). Many other

organisms have been investigated for im-



4 Examples ofLyases 10 9

ate salts. Therrnus thermophilus fumaraseshowed optimum activity of 85 O C and over80% of th e activity rema ined a fter a 24-h incu-bation at 9 0°C . Furthermore, the enzyme

show ed good chemostability; 50% of th e initialactivity was detected in assay mixtures con-taining 0.8% M guanidine hydrochloride (MI-ZOBATA,et al., 1998).The fumC genes from an-other T therrnophilus strain (KOSAGE t al.,1998) and Sulfolobus solfataricus (COLOMBO,et al., 1994) have b een expressed in E. coli.Both retain full activity at 70 " Cand in the lat-ter case this is despite an 11 amino acid C -ter-minal deletion.

a AcCI;

c A cCl(93% yield);

d AcCl, EtOH, 5OoC,(91%yield)

b PMB-NH,;

COZH

0 246 (S)-a-Hydroxy-N-

@-methoxy benzy 1)malimide

Steps

t

H

Fig. 88. Synthesis of the pyrrolizidine alkaloid, (-)-(R)-pyrrolam A (247) from L-malic acid (243)(HUANGt al., 1999)using the method of LOUWRIERet al. (1996) for the preparation of the malimide

(246).

proved fumarase activity including Coryne-bacterium equi, Escherichia co li, Microbacteri-

urn flavurn, Proteus vulgaris, Pichia farinosa,Leuconostos brevis (MIALL, 1978; DUFEY,1980) and yeast (WA NG t al., 1998) but th ehighest levels of process intensification (high-er space-tim e yields) will requ ire the use of en-zymes (CHIBATA,t al., 1995).

B. flavurn immobilized in K-carrageenan andChinese gallotannin has a ope ration al half lifeof 310 days at 37 "C.Although the half life isexcellent by any measure, productivity couldbe increased by a higher working tempera-tures, mo reover this would be a n advantage for

poorly soluble magnesium or calcium fumar-

4.2.1.3 Aconitase (A conitateHydratase, CitrateDsocitrate

Hydro-Lyase, EC 4.2.1.3) and

Citrate D ehydratase (E C 4.2 .1.4)

Acon itase catalyzes the dehyd ration of cit-rate (248) to cis-aconitate (73) nd hydrationto isocitrate (249) (Fig. 89). The prostheticgroup is a [4Fe-4S] cluster an d hence the fun-dam ental chem istry, resembles that of class I

fumarases (BEINERTt al., 1996; G R U ER t al.,1997). Acon itase h as significant sequence ho-mology with iron regulatory factor o r iron-re-sponsive element binding protein and indeedthe major difference between them seems tobe the presence or absence of the [4Fe-4S]cluster (BEINERTnd KENNEDY, 993). Citratedehydratase was reported to dehydrate citrateto cis-aconitate, but n ot isocitrate to c is-aconi-trate.

The conversion of citrate (248) to isocitrate(249) catalyzed by aconitase is a key step in the

citric acid (Kre bs) cycle (Figs. 1, 89 ). Th e firststep is the trans-elimination of the hydroxylgroup and the (pro-R)-hydrogen of the (pro-R)-methylene carboxylate group of citrate(248) to give cis-aconitate (73). n the reversedirection, this can be reg arded as the additionof a proton to th e re-face of C-2 and a hydrox-yl group to t he re-face of C-3. Hy dratio n of cis-aconitate (73) proceeds by trans-addition ofhydroxyl to the re-face of C-2 and a proton tothe re-face of C-3. Thus each addition o r elimi-nation of a proto n o r a hydroxyl from the two

carbons centres, occurs on the same side for



110 2 Lyuses

pro-R

I [2-3Hl]-Citrate

HO

C O P 73 cis-Aconitate

Ser-642;: 4 conitase

249 (2R,3S)-

0

[3-3HI]-Isocitrate

Fig. 89. The a conitase catalyzed isomerization of cit-rate (248) o isocitrate (249). For the purposes ofbrevity, no distinction is ma de in the text, betweenlabeled and unlabeled citrate and isocitrate.

each carbon centre. Remarkably, the proton(tritium) a bstracted from C-2 of citrate (prob -ably by Serine-642) is delivered back to th e o p-

posite face of cis-acon itate to give H-3 of isoci-trate, whereas the hydroxyl group is ex-changed with solvent. Hen ce either cis-aconi-tate must flip in the active site so that the re-tained proton can be transferred to the oppo-site face or less likely the proton must betransferred from one face to the o ther. This isall the more extraordinary, because it occurswithout dissociation of cis-aconitate (73) fromaconitase.

Fluoroacetate (250) (Fig. 90) is one of themost toxic small molecules known (LDS0 at

0.2 mg kg-l). It is converted in vivo to

0 250 Fluoroacetate

Acetatethiokinase

AMP, HP0 4 CoASH

28 Oxaloacetate b Hydrolysis

t

Fluorocitrate

Aconitase

0 1bop 254 (4R)-4-hydroxy-

trans-aconitate

Fig. 90. The biosynthesis of fluorocitrate (252) fromfluoroacetate and oxaloacetate (28) and the lethalbiosynthesis of (4R)-4-hydroxy-truns-aconitate254)catalyzed by aconitase.




(2R,3R)-fluorocitrate (252),which acts as botha competitive and semi-irreversible inhibitorof aconitase. Aconitase catalyzed eliminationproceeds normally (253),but rehydration oc-

curs with S N2 ‘ displacem ent of fluoride, togive (4R)-4-hydroxy-truns-aconitate (254)

which binds very strongly. It cannot be dis-placed with a 20-fold molar excess of citrate(248), but with a lo6 fold molar excess of isoci-trate (249) lag kinetics are observed indicatingdisplacement. X-ray crystal structures havebeen determined for aconitase binding trans-

aconitate, nitrocitrate (LAUB LE t al., 1994)and (4R)-4-hydroxy-truns-aconitate (254)

(LAUB LE t al., 1996). Fur ther aspects of fluo-roacetate and fluorocitrate metabolism are

discussed in Chap ter 3, this volume: Halocom-pounds.

4.2.1.4 a$-Dihydroxy Acid

Dehydratase (2,3-Dihydroxy-Acid

Hydro-Lyase,EC 4.2.1.9)

a#-Dihydroxyacid dehydratase (D H A D )catalyzes the dehydration of (2R)-2,3-dihy-droxy-3-methylbutanoate (255a) (Fig. 91) t o 2-0x0-3-methylbutanoate (256a) and (2R,3R)-2,3-dihydroxy-3-methylpentanoate 25%) to(3S)-2-0~0-3-methylpentanoate256b),Theseare the penultimate intermediates in the bio-synthesis of L-valine an d L-isoleucine, respec-tively. Unusually the last four steps in thesepathways ar e all catalyzed by a comm on set ofenzymes. D H A D has a n absolute requirementfor the (2R)-configuration, but substrates withboth the (3R)- and (3S)-configurations ar e de-hydrated stereospecifically.Thus (2R,3S)-2,3-

dihydroxy-3-methylpentanoate ndergoes de-hydration to give the enantiom er of the “nor-mal” product (ent-256b),which is a precursorof L-alloleucine.

D H A D has been found in bacteria, yeast, al-gae and higher plants (PIRR UN Gt al., 1991).E.coli D H A D contains a [4Fe-4S] cluster pros-thetic group (FLINT t al., 1996), whereas spin-ach D H A D contains a [2Fe-2S] cluster (FLINTet al., 1993; FLINTnd EMFTAGE, 1988). Mostsubstrate studies have been performed withSalmonella typhimurium (by CROUT’S roup)

and spinach DH A D (by PIRRUNG’Sroup) and

H ,OH@02VO255a R = Hb R = C H 3

u,a-Dihydroxyaciddehydratase

HZO

n

@ 0 2&R H ‘* 256

a R = H

b R = C H 3

Fig. 91. The a$-dihydroxyacid dehydratase cata-lyzed dehydration of (2R)-2,3-dihydroxy-3-methyl-butanoate (2%) and (2P,3P)-2,3-dihydroxy-3-methylpentanoate (25%) to 2-0x0-3-methylbuta-noate (256a) and (3S)-2-oxo-3-methylpentanoate(256b), espectively.

in both cases it has been dem onstrated that thea-proton is lost to the solvent during thecourse of dehydration.

Substrate homolog rate study results (Tab.7) demonstrate that substituents larger thanmethyl are better accommodated a t the R’ o-sition than the R2 position (255e,f and g,h),however, if only one methyl group is present,rates of dehy dration are faster when the meth-yl grou p is loca ted in the R2 position than theR1 position (255ij).O ne interpretation, is thatthe binding site for R2 s tight and responsiblefor steering the stereochemistry of elimina-tion, whereas the R1 inding site is loose and

accommodates larger groups (AR MS TRO NGtal., 1985; cf. L IMB ERGnd THIEM, 996). Com-parable result have been obtained with spin-ach DH A D (PIRRUNGt al., 1991).

4.2.1.5 3-Dehydroquinate

Dehydratase (EC 4.2.1.10)

3-Dehydoquinate dehydratase (DHQD)catalyzes the dehydra tion of 3-dehydoquinate(257) to give 3-dehydroshikimate (258) (Fig.

92).This is the central step in th e catabolic qui-



112 2 Lyases

Tab. 7. Th e Salmonella typhimurium a,P-Dihy-droxyacid Dehydratase Catalyzed Dehydration of

Substrate Hom ologs (ARMSTRONGt al., 1985)

S r pro-S

R or pro-R

a$-Dihydroxyaciddehydratase

H2O

255 Substituents Relative

R' R2 rate %

a

b

de

f

ghi

k

C

j

MeEtMe

EtPrMeBuMeMeHH

MeMeEt

EtM ePrMeBuHMeH

100

-4844-46

19-207-9

2-31-2

0-17-10

45-47

3 4

nate pathway and the biosynthetic shikimate

pathway (MA", 1987, MA" et al, 1994).DHQDs ar e divided into two types. The type I

enzymes are heat labile, anabolic enzymeswhich catalyze syn-elimination, via a imine de-rived from lysine, whereas the type I1 enzymesare heat stable and catalyze an anri-elimina-tion via an unknow n m echanism (FLOROVAtal., 1998).

-OH 257

3-Dehydroquinatedehydratase

HZO

0 O 2 C

HO\"'op0OH 258

Fig. 92. The 3-dehydroquinate dehydratase cata-lyzed dehydration of 3-dehydoquinate (257) to give3-dehydroshikimate (258).

4.2.1.6 Enolase (2 -P ho sp ho -~ -

Glycerate H ydro-Lyase,EC 4.2.1.11)

Enolase catalyzes the dehydration of 2-phospho-D-glycerate (259) to phosphoenolpyr-uvate (PEP) (148) (Fig. 93). En olase lends itsnam e to a superfamily of enzymes which cata-lyze deprotonation adjacent to carboxylategroups. The active sites are located in P-barrel(TIM barrel) domains and contain Mg2+,bound to a conserved sequence. Examp les in-

clude yeast enolase, mucona te lactonising en-zyme, mandelate racemase, D-gluconate dehy-dratase from E. coli and D-glucarate dehydra-tases from several eubacteria. The enolasesuperfamily is divided into three subfamiles,ma ndelate racem ase, mu conate lactonizing en-zyme (WILLIAMSt al., 1992) an d enolase. Allmembers of the enolase subfamily of the eno-lase superfamily bear a conserved lysine,which abstracts protons from cen tres with ( R ) -stereochemistry, adjacent to a carboxylategroup (BABBITTt al., 1995,199 6).Thisnomen-

clature is discussed further in connection with




‘O& H 4”’H00

0 =\0 00 2592-Phospho-

D-glycerate

Enolase

H2O

0

& 00P

0” ‘0° 148 Phosphoenol-pyruvate

Fig. 93. The enolase catalyzed dehydration of 2-phospho-D-glycerate (259) to phosphoenolpyruvate(PEP) (148).

L-serine dehydratase

HZO

H,N0JL C02 261

NH4@Hzol0A G0 2 lPyruvate

Fig. 94. The L-serine dehydratase catalyzed dehydra-tion of L-serine (259 to 260) to pyruvate (1).

D-gluconate dehydratase and D-glucarate de-

hydratase (Sect. 4.3.1.1).

166L-Threonine4.2.1.7 L-Serine Dehydratase(EC 4.2.1.13), D-Serine H

Dehydratase (EC 4.2.1.14) and

L-Threonine Dehydratase

(EC 4.2.1.15)

L-Serine dehydratase catalyzes the dehydra-

tion of L-serine (259 to 260) o pyruvate (1)(Fig. 94) and L-threonine dehydratase catalyz-es the dehydration of L-threonine (166) o 2-oxobutanoate (262) (Fig. 95). Both enzymes al-so catalyze the other reaction more slowlythan with the “native” substrate. Similarly D-

serine dehydratase also has some activity withD-threonine. Most forms of these enzymes arePLP-dependent, but no bacterial L-serine de-hydratase has been shown to be PLP depen-dent. The L-serine dehydratase from Pepto-streptococcus asaccharolyticus serine dehydra-

tase contains a [4Fe-4S] cluster (HOFMEISTER

L-Thonine dehydratase

OACOF 262 2-Oxobutanoate

Fig. 95. The L-threonine dehydratase catalyzed d e-hydration of L-threonine (166) to 2-oxobutanoate(262).

et al., 1994;GRABOWSKIt al., 1993) and henceis related to fumarase and a,p-dihydroxyacid

dehydratase.



114 2 Lyases

4.2.1.8 Imidazoleglycerol

Phosphate Dehydratase

(EC 4.2.1.19)

Imidazoleglycerol phosphate dehydratasecatalyzes the dehydra tion of imidazoleglycerolphosphate (263) o imidazoleacetol phosphate(264) (Fig. 96), which is a la te s tep in the bio-synthesis of L-histidine in plants a nd m icroor-ganisms. L-H istidine is an essential nu trient foranimals, because they lack the requisite bio-synth etic pathway. The elimination is fairly un -usual because the proton lost is essentiallynon-acidic, moreover no co-factor has beenidentified other than Mn2+ ions. Elimination

proceeds with inversion of configuration atC-3 and the proto n is derived from the solvent(PARKERt al., 1995).

4.2.1.9 Maleate Hydratase

[ R)-Malate Hydro-Lyase,

EC 4.2.1.311

M aleate hydratase catalyzes the addition ofwater to m aleate (265) to give D-malate (266)(Fig. 97). 2-Methyl- and 2,3-dimethylmaleatear e also substrates. The stereochem istry of hy-

lmidazolegl cerolphosphate

dehydratase

0264

Fig. 96.The imidazoleglycerol phosphate dehydra-tase catalyzed dehydration of imidazoleglycerolphosphate (263) to imidazoleacetol phosphate

(264).

cop

b E 5 Maleate

Fig.W. tereochemistryof the hydration of maleate(265) to D-malate (266)catalyzed by m aleate hydra-

tase. For the purposes of brevity, no distinction ismade in the text, betwee n labeled and unlabeled D-

malate.

dration has remarkable similarities to that offumarase. Both enzym es catalyze the trans-ad-dition of the elements of water to the alkenebond, with the hydrogen introduced into the(3R)-position of malate. Consequently fuma-rase gives (2S) -~- m ala te197) Fig. 73) w here-as maleate dehydratase gives the enantiomer,(2 R) -~ -ma l a t e265 to 266).

Maleate hydratase has been purified fromrabbit kidneys, Arthrobacter sp. strainMC12612 a nd Pseudo mon as pseudoalcaligenes(VAN DER WER F t al., 1992).The rabbit kidneyand Arthrobacter sp. strain MCI2612 formscontain an iron-sulfur cluster (DR EY ER ,985),whereas tha t from Pse udom onas pseudoalcali-genes doe s not (VAN ER WE RF t al., 1993).

L-Malate (197) s a valuable bulk commod-ity (see Sect. 4.2.1.2, Fumarase), whereas D-mal-ate (266) as only found a few minor applica-

tions in organic synthesis. Nev ertheless p oten -tially, the most efficient and cheapest way toprepare it, is by maleate hydratase catalyzedhydration of maleate. Kinetic analysis of D-

malate production by permeabilized Pseudo-monaspseudoaligenes has shown that the reac-tion suffers from product inhibition and bio-catalyst inactivation. Optimal production oc-curred at p H 8 and 35 O C, but this was accom-panied by complete biocatalyst deactivationover 11hours (MICHIELSENt a l., 1998). Clear-ly, considerable development is required to

ma ke this a practical method.



4 Examples ofLyasm 115

OH4.3 Carbohydrate Lyases

(EC 4.2.X.Y)

The lyases are involved in the metabolism ofcarbohydrates may be separated into twoclasses: the hydro-lyases which catalyze theelimination of water and the dissacharide andhigher lyases which catalyze the the cleavageof glycosidic bonds

4.3.1 Carbohydrate Hydro-Lyases

(EC 4.2.1.X)

4.3.1.1 Monosaccharide CarboxylicAcid Dehydratases

The monosaccharide carboxylic acid dehy-dratases (Tab. 8), catalyze the a,p-eliminationof water to give enols which isomerize to thecorresponding ketone (e.g., Fig. 98). The stere-ochemistry at C-2 and C-3 is lost during thecourse of the elimination, comequently, D-al-tronate (274), D-mannonate (276), D-gluconate(280) and D-glucosaminate (279) all give thesame product;2-dehydro-3-deoxy-~-gluconate(275) (Fig. 99). In general enzyme catalyzedeliminations which involve a$-unsaturatedesters, proceed with anti-stereochemistry(MOHRIG t al., 1995; PALMERt al., 1997),hence the enol (289) derived from D-manno-nate (276),will be the (E)-rather than the (2) -

stereoisomer formed by the other substrates.Galactonate dehydratase is a member of the

enolase superfamily of enzymes, which in-cludes yeast enolase, muconate lactonking en-zyme and mandelate racemase.The active sites

are located in @barrel (TIM barrel) domainsand contain Mg2+,bound to a conserved se-quence. The prototype enzyme is mandelateracemase from Pseudomonas putida which hasbeen structurally characterized. Lysine-166acts as a (,!+selective base (LANDRO t al.,1994) and histidine-297 as a (R)-selective base.All members of the mandelate racemase sub-group of the enolase family have either one ofboth of these residues (BABBITT t al., 1995,1996).

C-2 of D-galactonate (272) has the (R)-con-

figuration (Fig. 100). Galactonate dehydratase

OH OH 270D-Arabinonate

D-h abino nate dehydratase

H2O

0

O OH

OH OH 288Enol

0

Oz OH

0 OH 271 2-Dehydro-3-deoxy-D-arabinonate

Fig. 98. D-Arabinonate dehydratase catalyzes thea,p-elimination of water from D -arabinonate (270)

to give an eno l(288 ) which tautom erizes to 2-dehy-dro-3-deoxy-~-arabinonate271).

from E. coli bears histidine-285 and aspartate-258, which act as the base and histidine-185which acts as a general acid catalyst to pro-mote elimination (WIECZOREKt al., 1999).Protonation of the enol, occurs such that hy-drogen is incorporated in the (3-pro-S)-posi-

tion (PALMERt al., 1997). Other members ofthis group include L-rhamnonate dehydratasefrom E. coli and D-glucarate dehydratasesfrom several eubacteria (HUBBARD t al.,1998).

Glucarate dehydratase from Pseudomonasputida acts on D-glucarate(281) and L-idarate(294) (epimers at C-5) to give exclusively 5-de-hydro-4-deoxy-~-glutarate297) (Fig. 101).The two substrates are interconverted via theenol (295), which partitions between epimer-ization (reprotonation) and dehydration in a

ratio of 1 4. Early experiments suggested that



116 2 Lyuses

Tab. 8. Monosaccharide CarboxylicAcid Dehydratases (EC 4.2.1.X)

X Substrate, Product Reference

5

6

7

8

25

26

39

OH OH 0 OH

270 D-h&inona te 271 2-Dehydro-3-deoxy-D-arabinonate

@ 0 2 - Ooz& OH

OH OH OH

272 D-Galactonate 273 2-Dehydro-3-deoxy-D-galactonate

OH OH

00, OH

OH OH OH

274 D-Altronate 275 2-Dehydo-3-deoxy-D-gluconate

OH OH

00,&yo,Q02GHO HO OH

276 D-Mannonate 275 2-Dehydo-3-deoxy- D-gluconate

OH OH 0 OH

277 L-Arabinonate 278 2-Dehydro-3-deoxy-L-arabinonate

279 D-Glucosaminate 275 2-Dehydo-3-deoxy- D-gluconate

OH OH

' O 2 W H @0 2 OH

HO HO OH

280 D-Gluconate 275 2-Dehydo-3-deoxy-D-gluconate

DAHMSt al., 1982;WIECZOREKt al., 1999

ROBERT-BAUDOUYt al.,1982;DREYER,987

ROBERT-BAUDOUYt al.,1982;DREYER,987

IWAMOTO t al., 99

GOTTSCHALKnd BENDER,1982




Tab. 8. (continued)

X Substrate, Product Reference

40

42

67

68

OH OH PALMERt al., 1998

@02+co,e

HO HO

281 D-Glucarate 282 5-Dehydo-4-deoxy-D-glucarate

283 D-Galactarate 282 5-Dehydro-4-deoxy-D-glucarate

284 D-Fuconate 285 2-Dehydro-3-deoxy-D-fuconate

ANDERSONnd HAIJSWALD,@ &982;VEIGAnd GIIIMARAES,0 2 0 2 1991-

OH OH OH

287 -Dehydro-3-deoxy-L-fuconate@F286 L-Fuconate~ _ _ _ _ _

All 1982 references in this table are taken from Methods in Enzymology

the deh ydration of D-glucarate (281)was onlypartially regioselective, based on the partialscrambling of a radiochemical label from C-1to C-6. Ho wever, epimerization of D-glucarateto D-idarate which has &-symmetry renders

C-1 and C-6 equivalent (PALMERnd GERLT,1996; PALM ERt al., 1998). Elimination of th e4-hydroxyl group gives an a-hydroxyacrylate(296) which is stereospecifically reprotonatedby water. The hydrogen is incorporated exclu-sively with (4-pro-S) stereochem istry (PAL ME Ret al., 1997). Given th at glutarate de hydrataseis capable of deprotonating both D-glucurate(281) and L-idarate (294), the model for theenolase family predicts that two bases shouldbe present. A n X-ray structure determinationenabled the assignment of the (S)-specificbase

as lysine-213 (which is activated by lysine-211)

and the (R)-specific base as histidine-297 andaspartate-270 (GULICK,t al., 1998).

4.3.1.2 3,6-Dideoxysugars

3,6-dideoxyhexoses are found, (with a fewexceptions) exclusively in the cell wall lipo-polysaccharides of gram-negative bacteria ofthe family, Enterobacteriaceae, where they actas the dominant O-antigenic determinants(JOHNSONnd LIU, 1998; KIRSC HNIN Gt al.,1997; LIU and THORSO N,994). There are five3,6-dideoxy sugars know n in na ture: D-abe-quose (298), L-ascarylose (299), L-colitose(300),D-paratose (301) and D-tyvelose (302)

(Fig.102).Many organisms produce most o r all

of these simu ltaneously and all except for col-



118 2 Lyuses

OH OH OH OH OH

0 2 OH ‘0 OH OH

OH OH HO HO HO HO

274 D-Altronate 276 D-Mannonate 280 D-Gluconate

GluconateH20f deqfy:z ~ H20dG%g ~ H20jehydratase

0

OH I)H

‘ OH

279 D-Gluconaminate 275 2-Dehydo-3-deoxy-D-gluconate

H2O

HO 290 Protonated enamine

Fig. 99.D-A ltronate (274) , D-mannonate (27 6) and D-glUCOnate (280) undergo cY,p-eliminationof water to

give an enol(28 9) which tautomerkes to 2-dehydro-3-deoxy-~-gluconate275). Dehyd ration of D-glucosam-inate (279) also gives the same product.

itose (W),re produced biosynthetically byvia a complex pathway commencing withCDP-D-glucose (30%). Colitose (300) isformed biosynthetically from CDP-D-man-nose.

The biosynthesis of CDP-ascarylose (299c)from D-glucose (303a) by Yersinia pseudotu-berculosis has been investigated in consider-able detail (Fig. 103; review HALLISnd LIU,1999b).The pathway employs two dehydratas-es (one with a “hidden” redox function), plustwo oxidoreductases and an epimerase to de-liver a product in which two stereocentreshave been retained, two inverted and two re-duced (Fig. 103). Three overlapping clonescontaining the entire gene cluster required forCDP-ascarylose biosynthesis have been iden-

tified (THORSONt al., 1994b). Each enzyme

has a trivial, but conveniently short designa-tion which is used in all the figures, except Fig.103 where the full name is shown for referencepurposes. The full enzyme commission desig-nation is shown in the text.

Biosynthesis commences with D-glucose(303a) which is converted sequentially into the1-phosphate (303b) and the 1-0-cytidylyl(30%) derivatives catalyzed by a-D-glucose-1-phosphate cytidylyltransferase (E,) (THORSON

et al., 1994a).CDP-D-glucose 4,6-dehydratase (Eod) [EC

4.2.1.451 catalyzes the transformation of CDP-D-glucose (30%) to CDP-6-deoxy-~-threo-~-glycero-4-hexulose(304c). 4-*Hl]-CDP-~-glu-cose (306)Fig. 104) has been used to investi-gate the pathway. It consists of three distinct

steps; oxidation of the 4-hydroxy group to a




00, 00,

-*11 OH 2 o r 5 -IIOHOH

OH OH 272 D-Galactonate

" j ; Ico: OH copO

D-Galactonate dehydratase

281 D-Glucarate 294 L- Idarate

@0 2 OH

OH 291Enol

I2H 20

f

'H OH 292

OH

293 (3S)-[3-*Hl]-2-Dehydro-3-

deoxy-D-galactonatehemiketal

Fig. 100. D-Galactonate dehydratase catalyzes thea,p-elimination of water from D-galactonate (272)to give an e nol (291) which tautomerizes to 2-dehy-dro-3-deoxy-D-galactonate (273) (Tab. 8).In deuteri-um oxide, (3S)-[3-2H,]-2-dehydro-3-deoxy-~-galac-

tona te (292) is form ed which cyclizes to the pyrano-syl hem iketal(293).

ketone (307) and reduction of NAD toNADH, deprotonation and P-elimination ofthe 6-hydroxyl group (308) and conjugate ad-

dition of the hydride ion which originated

OH OH 295

0

OH OH 296

' H a d 2H20

40

&0

0 2 -OH 0

297 (4S)-4-2Hl]-5-Dehydro-

4-deoxy-D-glucarate

Fig. 101.Glucurate dehydratase catalyzes the dehy-dration of D-ghCUrate (281) nd L-idarate (294).

from C-4 to C-6 to give the 6-deoxy-sugar

(309).The replacement of the 6-hydroxyl



120 2 Lyases

HO .,OH

OH298 D-Abequose 299 L-Ascarylose

300 L-Colletose 30 1 D-Paratose

302 D-Tyvelose

Fig. 102. Naturally occurring 3,6-dideoxyhexoses.

grou p by hydride occu rs with inversion of con-figuration (Y u et al., 1992; RUSSELL nd Yu ,1991). This stereochem ical cou rse is also fol-lowed by the GDP-D-mannose dehydratase

from th e soil organism (ATC C 19241) ( O m set al., 1990) and th e TDP-D-glucose 4,6-dehy-dratase from E. coli (WANG and GABRIEL,1970; SNIPES t al, 1977). Th e basic features ofthe mechanism are nicely confirmed by th e be-havior of the difluoro-analog substrate (310)

(Fig. 105). A fte r traversing th e reaction se-quence shown above (Fig. loo), the mo nofluo-ro-derivative (311) so formed undergoes a fur-ther p-elimination of fluoride to give the al-kene (308),which is the normal intermediatein th e sequence. However, there is no further

hydride t o donate and instead add ition of a nu-cleophile located on the protein backbone oc-curs to give an isolatable covalent derivative(312) (CHAN G t al., 1998).

The facial selectivity of hydride donationfrom NADH is rather difficult to determinefor this class of enzymes, because N A D H is on-ly presen t durin g the catalytic cycle. Ne verthe -less it has been demon strated that the product(304c) s reduc ed by [4S-’H,]-NADH with in-corporation of label when incubated w ith apo-E,, and n o label is transferred from [4R-*H,1-

a R = O H

HO

IE,, a-D-glucose-1 phosphate

cytidylyltransferase

Ed, DP-D-glucose 4.6-dehydratase

H2O +

E ,.CDP-6-deoxy-L-threo-D-glycero-4-hexulose-3-dehydre

H2OPMP 122, Pyridoxamine5’-phosphate

NADH E,, CDP-6-deoxy-L-threo-

Nm@ dehydrase reductase

PM P (CDP-6-deoxy-A3’-

H20$22 glucoseenreductase)

D-glycero-4-hexulose-3-

E..4NAD(P)H

Ered

NAD(P)

OH

HO

Q - q R299

R = OH; L-Ascarylose

c R = O C D P

Fig. 103. The biosynthesis of CDP-ascarylose ( 2 9 9 ~ )from D-glucose (303a) by Yersiniu pseudotuberculo-

sis (122, PMP,pyridoxamine 5 ’-phosphate).

NA DH hydride donation to the a,p-unsaturat-ed ketone (308)might occur from the otherface. bu t this is im dau sib le (HALLISnd LUI.

N A D H (Fig. 106). It is conceivable that 1998). ( pro - S) Hyd;ide donation has also been




demonstrated for TDP-D-glucose4,6-dehydrat-ase (WANG nd GABRIEL,970), L-myo-inosi-tol-1-phosphate synthase (BRYUNand JEN-NESS, 981;LOEWUSt al., 1980) and UDP-D-

glucose-4-epimerase (NELSESTUENnd KIRK-WOOD, 1971), which all utilize NAD and act onsugars.

Most of the 4,6-dehydratases belong to arare group of enzymes which contain a tightlybound NAD(H). As demonstrated above, thisis responsible for the redox process and is re-tained throughout the catalytic cycle as a pros-thetic group, rather than a co-substrate(THOMPSON, t al., 1992; NAUNDORF ndKLAFFKE,996). However, CDP-D-glucose 4,6-dehydratase is a dimer of identical sub-units

(42,500 Da), which only binds one equivalentof NAD, despite the presence of two bindingsites and there is a large anti-co-operative ef-fect for binding a second molecule of NAD.NADH is bound much more strongly and is re-leased when the substrate binds indicating thatthis may be the normal “resting” state of theenzyme in vivo (HE et al., 1996).

The deoxygenation at C-3 is catalyzed byCDP-6-deoxy-~-threo-~-gZycero-4-hexulose-3-dehydrase (El)pyridoxamine 5 ’-phosphate

(PMP)-linked, [2Fe-2S] containing protein andCDP-6-deoxy-~-threo-~-gZycero-4-hexulose-3-dehydrase reductase (formerly designatedCDP-6-deo~y-A~~~-glucoseeneductase) (E3) a[2Fe-2S]-containing flavoprotein (Lo et al.,1994). In the first step CDP-6-deoxy-~-threo-~-glycero-4-hexulose304c) Fig. 107) under-goes condensation with PMP (122) to form thecorresponding imine (313) which undergoes1,4-eliminationof water to give the protonated2-aza-1,4-diene (314).Both steps are rever-sible and the elimination is stereospecific; only

the (4-pro-S)-hydrogen of PMP undergoes ex-change (WEIGEL t al., 1992). The reactionstops at this stage, unless CDP-6-deoxy-~-threo-~-gZycero-4-hexulose-3-dehydrasee-ductase is present. This forms a tight complexwith the dehydrase (CHEN t al., 1996) and me-diates a reduction in which electrons are trans-ferred within the reductase from NADH toFAD to the [2Fe-2S]-center and from there tothe [2Fe-2S]-center of the dehydrase. The de-tails of the transfer of electrons from there tothe PMP-sugar adduct are somewhat specula-

tive. It is likely to involve stepwise addition of

309 308

Fig. 104. The isomerhation of [4-2H,]-CDP-D-glu-cose (306) o [6-ZH,]-CDP-6-deoxy-~-threo-~-glyce-ro-Chexulose(309) y CDP-D-glucose 4,6-dehydra-

tase (Eod) ro m Yersinia pseudotuberculosis.

HO%O

R R

310 I 311

R = OCDP

312 308

Fig. 105. Biotransformation of an affinity label by

CDP-D-glucose 4,6-dehydratase (Ed)rom Yersinia

pseudotuberculosis.

H O l L

‘1” .R

3304c R = OCDP 306

Fig. 106. Reduction of CDP-6-deoxy-~-threo-~-gZy-cero-4-hexulose(304c) y CDP-D-glucose 4,6-dehy-

dratase (Ed) rom Yersinia pseudotuberculosis.



122 2 Lyases

R = O C D P I # 122PM P

El-PMP4OmH

HO IR 316 cf . 30%)

Fig. 107. The El, CDP-6-deoxy-~-threo-~-glycero-4-hexulose-3-dehydrase catalyzed reduction (R =

OCDP).

single electrons and this is supported by thedetection of an E SR signal for a phenoxy radi-ca l (JOHNSONet al., 1996). O ne attractive pos-sibility is that e lectron transfer occurs to an u-

quinom ethide formed by protonation at C-3 ofthe sugar (PIEPE R t al., 1997).This occurs ster-eospecifically with incorp oration of a proto nfrom water exclusively at the C-3 equatorialposition (316) (PIEPERt al., 1995).

Reduction stepsCDP-3,6-dideoxy-~-glyceru-~-glyceru-4-

hexulose is the common intermedaite for D-

abequose (298) (LINDQUISTt al., 1994), L-as-carylose (299), -paratose (301) (HALL IS t al.,1998) and D-tyvelose (302) HALLIS nd LUI,

1999a) by redu ction an d/or epimerization (Fig.108). The gene encoding CD P-paratose syn-thase (rfbS) in Salmonella typhi has been iden-tified, sequence d, amplified (PC R) and clonedinto a PET-24(+) vector. Expression an d pur-ification of the enzyme e nabled it to b e fullycharactorized. This homodimeric oxidoreduc-tase transfers the pro-S hydrogen fromNADH, and has a 10-fold preference forNAD PH over NADH . (HALLIS t al.,1998).

4.3.2 Lyases which Act on

Polysaccharides (EC 4.2.2)

4.3.2.1 Pectin and Pectate Lyases

Pectin, pectins or pectic substances are ge-neric names used to describe a group of heter-ogenou s acidic polysaccharides found widelyin primary plant cell walls and intercellularlayers. They m ake up for example, 30% of the

dry weight of apples, but ar e usually present inmuch smaller amounts in woody tissues. Theprincipal sugar residue is D-galacturonic acidwhich may be partially esterified (pectinic ac-ids) or present a s the free acid or calcium salt(pectic acids) (Fig. 109).The pectinic acids arefreely soluble in water and are good gellingagents, which are used as pharmaceutical exci-pients and as gelling agents for fruit jellies(PILNIK nd VO RAGEN ,992). Pectic acids a remuch less soluble and usually require ligandsfor the counterions of the carboxylates (e.g.,

ED TA ) to solubilize them. The most common




pectins are heteropolysaccharides containingneutral and acidic sugars, but th ere ar e alsomo re ra re homopolysaccharides such as D-ga-lactans, D-galacturonans and L-arabinans

(KENNEDYnd WHITE,990).Pectinic an d pec-tic acids are cleaved by different classes of en-zymes (Tab. 9).

Pectolytic enzymes are essential in the foodindustry for the clarification of fruit juices an dwines. For example, in freshly pressed applejuice, pectins act as solubilising agents for o th-erwise insoluble cell debris. Hydro lysis of th epectin cau ses floculation of the insoluble com -ponents. Pectolytic enzymes are also useful fordehulling, extractive, pressing (sugar beet andolives) and ferm entation (cocoa) processes be-

cause they w eaken the cell walls and increaseyields (ALKORTA t al., 1998). Commercialpectolytic enzymes a re almost invariably mix-ture s which also contain cellulases, hem icellu-lases and protease s and only part of the pecto-lytic activity is due to lyases (NAIDU t al.,1998).

Enzymes that d egrade pectic substances arefrequently the first cell wall degrading en-zymes produced by plant pathogens. Depoly-merization of pectins not only provides nutri-

ents for the invading organidm, but also expos-es other cell wall com ponents to de gradation.Pectate and pectin lyase activities have beendetected in the w hite rot fungus Trametes tro-

gii (LEVIN nd FORCHIASSIN,998), Colletotri-

chum gloeosporiodes (ORTEGA,996),Botryiscinerea (MOVAHEDRInd HEALE, 990;KAPATet al., 1998a, b), Aspergillus niger (VAN DEN

HOMBERGHt al., 1997), Pseudocercosporellaherpotrichoides infecting wheat plants (MBWA-GA et al., 1997), Amycoluta sp. (BRUHLMANN,

HO

R

CDP-Abequose 317 CDP-D-AbequoseI R = OCDPsynthase

t

R OH 319

318 CDP-D-ParatoseNAD(P)H1 CDP-tyvelosepimerase F: A D ( p )

Hoqo q R

R OH

320 CDP-D-Tyvelose 299 c CDP-L-Ascarylose

Fig. 108.The El, CDP-6-deoxy-~-threo-~-glycero-4-hexulose-3-dehydrase catalyzed reduction (R =OCD P). C DP-D-tyvelose 2-epimerase (HALLISndLUI,1999a), CDP-paratose synthase (HA LLIS t al.,1998).

Tab. 9. Lyases (E C 4.2.2.X) A cting on Poly-D-Galacturonates(321)~

X Nam e Regioselectivity Occurrence Conditions

2 Pectate lyase endo bacteria and pathogenic fungi C az +,pH 9-10,long chains

6 Oligogalacturonide yase exo Erwinia carotovoransand -

E. aroideae9 Exopolygalacturonate exo rare, Clostridium,Streptomyces, Ca2+ ,pH 9-10

lyase Erwinia,Fusarium10 Pectin lyase endo common in fungi pH 5-6, long

chains



124 2 Lyases

I

“ W Y

] D-Galacturonic acid residue

R’=metal ons, eg. Naf3Cazo; Pectates

RO R = Me; Pectins

RO

D-Galacturonic

RO

R = H, acetyl, L-arabinose,L-fucose, D-galactose,D-glucose, D-xylose, I

araban, galactan

RO mu

Q-

acid residue

Fig. 109. Generalized structure for pectic substances.The distance between “branching” rhamnose residuesdepends on the source, e.g., in citrus pectin there are typically 25 intervening D-galaCtUrOniC acid residues and16 in tomato pectin.

1995),Pseudo mo nas marginalis N6031 (NIKAI-DO U et al., 1995) an d a variety of org anisms onrottern cocoyams (UGWUANYInd OBETA,1997)

4.3.2.2 Pectate Lyases [Poly(l,4-a-

D-Galacturonide)Lyase,PectateTranseliminase, EC 4.2.2.2, formerly

EC4.2.99.31

Pec tate lyases catalyzes the cleavage of pec-tate (321a) to give D-galacturonate (332a) and4-deoxy-cr-~-gluc-4-enuronate323a) termi-nated oligomers. The cleavage reaction is atrans-a,p-elimination, operates at pH 8.0-10and has an absolute requirement for Ca2+counterions. As with the pectin lyases, attack is

random, endo-selective and occurs preferen-

tially with long chain pectates. Care must betaken when using pectate lyases to get ob tainma terial free from esterases, which will en ab lecleavage of pectins.

Pectate lyases have bee n identified in m anybacteria a nd pathogenic fungi.The X-ray crys-tal structures for pectate lyase C (PelC) and

pectate lyase E (P elE) from Erwinia chrystha-nemi an d from Bacillus subtilis (BsPel) (PICK-ERSGILL et al., 1994) are very similar to eachother and that of A spergillus niger pectin lyase(MAYANS t al., 1997). The structures are com-posed of all parallel P-strands wound into anunusual right-handed parallel P-helix (JENKINS

et al., 1998; PICKERSGILLt al., 1994 and cf.1998).The activekalcium binding site is locat-ed in a cleft between this featu re and tw o T3-loops, which also contain s a highly basic regionwhich is presumably responsible for dep roto-

nation adjac aent to the carboxylate group.




The genes for Butyrivibrio fibrisolvens en-do-p-1,4-glucanase plus the Erwinia pectatelyase and polygalacturonase were each insert-ed between a yeast expression secretion cas-

sette and yeast gene terminator and clonedinto yeast centromeric shuttle vectors. Coex-pression in Saccharomyces cerevisiae success-fully resulted in secreted activity for all threeenzymes (VAN ENSBUR Gt al., 1994).

4.3.2.3 Pectin Lyases

[Poly(Methoxygalacturonide)Lyase,

EC 4.2,2.10]

Pectin lyase only acts on poly(methy1 D-gal-acturonate esters) (321b) (Fig. 110)and has noactivity with the correspond ing acids. The mainsource of this enzyme is Asperg illus sp. whichreleases pectin lyases A and B types whengrown on pectin (DELGADO ,t. al., 1993; VAN

DEN HOMBERGHt al., 1997). Pectin lyases ar eendo-selective for cleavage of poly(methy1 D-

galacturonate esters) (321b) at random posi-tions and show very little activity with di- a ndtrisaccharides. p H optima range from 5.5 to 8.6

and a re generally lower than those for the pec-tate lyases. However, pectin lyase A activitydrops above pH 7 due to conformationalchanges at the active site triggered by theproximity of asparta te residues (MA YANSt al.,1997).

4.4 Carbon-Nitrogen Lyases

(EC 4.3)

4.4.1 Ammonia Lyases(EC 4.3.1.X)

The ammonia lyases have achieved consid-erable industrial importance because of theiruse in the m anufacture of am ino acids. This ar-ea has been spurred by the development ofAspartame (325), (Nutrasweet@, -L-aspartyl-L-phenylalanine methyl ester, Fig. l l l ) , whichis a n artificial sweeten er that is some 200 timessweeter than sucrose. Unlike m any oth er a rti-

ficial sweetness its has no bitter after taste andit is used extensively in soft drinks, confection-ery and many o ther processed foods. The pat-en t for its use ran o ut in the 1980s and hencefor producers to remain competitive they mustdevelop new, cheaper and patentab le ways tomanufacture it (OYAM A,992).

101 L-Aspartate

I 1JIMT -

ROk

Rk

ROw

Fig. 110. The cleavage of pectic substances (321) bypectic lyases to give terminal 4-deoxy-a-D-gluc-4-

enuronosyl residues (323).

H f e

H O

324 L-Phenylalaninemethyl ester

325 Aspartame

Fig. 111. The manufacture of Aspartame (325) re -quires cost effective rout es to L -aspartate (101) and

L-phenylalanine methyl est er (324) (OYAMA,1992).



126 2 Lyases

4.4.1.1 Aspartase (L-Aspartase,

Aspartase Ammonia-Lyase;

EC 4.3.1.1)

Aspartase catalyzes the addition of ammo-nia to the alkene bond fumarate (1%) o giveL-aspartate (101) (Fig. 112).The reaction is forall practical purposes, absolutely specific forfumarate and L-aspartate (FALZONE t al.,1988), but hydroxylamine can substitute forammonia (KARSTEN nd VIOLA,1991). Theequilibrium constant for the reaction heavilyfavors aspartate formation over the reverseprocess (Keq 5 . DG” 3.2 kcal mol-l).Carbon-nitrogen bond cleavage is at least par-

tially rate limiting, but carbon-hydrogen bondcleavage is not. This can be rationalized as aconventional conjugate addition (Michael) re-action to give a stabilized carbanion which israpidly protonated. In the reverse reaction thisequates to efficient deprotonation adjacent tothe carboxylate group of L-aspartate followedby slow p-elimination (PORTER nd BRIGHT,1980; NUIRY t al., 1984)

Aspartase was discovered by Harden (1901)in Bacillus coli communis over one hundred

years ago, and in 1932 it became the first mi-crobial enzyme to be used in the synthesis ofan amino acid. It is widely distributed in bacte-ria as well as in fungi, higher plants and ani-mals but not in mammals. The enzyme hasbeen purified and characterized from microor-ganism includingBacillus fluorescens, B. subti-lis, B. strearothermophilus, Pseu domonas fluo-

rescens, r! trefolii, E scherichia feundii, E. coli,

0

ao,& “2 196 Fumarate

I 1 Asp-e

ONH,

101 L-(3-(+)-Aspartate

Fig. 112.The aspartase catalyzed the ad dition of am-monia to fumarate (1%)to g ive L-aspartate (101 ).

Bacterium cadaveris, Brevibacterium amm oni-agenes and Brevibacterium fZavum (DRAUZand WALDMANN,995). L-Aspartase is postu-lated to play a significant role in the mineral-

ization of soils, by the release of ammoniumions (SENWOnd TABATABAI,996,1999).

Aspartase is one of a number of related en-zymes which catalyze additions to the alkenebond of fumarate, such as class I1 fumarases(EC 4.3.1.2, water, WOODS t al., 1986), and theamidine lyases; arginosuccinase (EC 4.3.2.1,L-

arginine, COHEN-KUPEICt al., 1999; TURNERet al., 1997) and adenylsuccinase (EC 4.3.2.2,AMP, LEEet al, 1999).These enzymes have sig-nificant sequence homology (WOODS t al.,1988a) and catalyze the addition-elimination

reactions with identical stereochemistry, i.e.trans, with the hydrogen added to the (3-pro-R-)position (Fig. 113) (GAWRONnd FONDY,1959). The reaction is fully reversible andhence can be used to prepare hydrogen isotop-omers of aspartate (328) from, for examplepertritiated-aspartate (327) (Fig. 114) (PICCI-RILLI et al., 1987; cf. YOUNG, 991). Large scalesyntheses of of 15N-labeled L-aspartate(SOOKKHEOt al., 1998) and (2S,3S)-[2,3-’H2]-and (2S,3R)-[3-’H]-aspartate (LEEet al., 1989)

have been achieved using immobilized aspart-ase.Asparatase from E. coli (and most other

sources) is a tetramer of identical 50 kDa sub-units (50 kDa). It is in a pH dependent equilib-

$ 2 ~ c o ’ 96 Fumarate

4 1

,H

326 (2S,3R)-[3-’H,]-Aspartate

Fig. 113. The stereochemistry of the aspartase cata-lyzed addition of ammo nia in deuterium oxide to fu -marate (1%) to give (2S,3R)-[3-2H1]-aspartate326)

(GAWRONnd FONDY,959).




b HZO

3-ZHl,3Hl]-Aspartate

Fig. 114. Stereospecific exchange of the (3-pro-R)-tritium of (2 s) - [2-3H,3-3H2]-aspartate327) to give(2S,3S)-[2-3H,3-2H1,3H1]-aspartate328) PICCIRILLIet al., 1987).

rium between two forms. A t p H 7.5 and above,divalent metal ions (e.g., Mg 2+ ) are requiredfor activity, and in the amination direction alag time is observed which is eliminated by theaddition of L-aspartate (101)or structural ana-

logs. At lower pH values there is no require-me nt for metal ions and no lag time (KARSTENand VIOLA,1991). In contrast, the aspartasefrom BaciZZus sp. YM 55-1 is insensitive to mag-nesium ions over the complete pH range(KAWATA,t al., 1999) and th e aspartase fromH. alvei is activated by magnesium ions at allpH s (YOO N t al., 1995; cf. NUIRY t al., 1984).In contrast to the absolute substrate specific-ity, aspartase is activated by a range of metalions. Transition metal ions are bound betterthan alkali earth metals, but they ar e inhibito-

ry above 100 pM, whereas there is no signifi-cant inhibition by alkali earth metals. Electronparamagnetic resonance measurements indi-cated that at higher pH, Mn” is bound moretightly (FALZON Et al., 1988).

The X-ray crystal structure of aspa rtate am-monia-lyase from E. coli has been determinedto 2.8 A. A putative active site was identifiedbut the precise residues involved in catalysiscould not be assigned (SHIet al., l993,1997).Asuccession of candidate residues have beenpostulated as active site acids and bases, but

site specific mutagenic studies have excluded

all of these, thus fa r (JAYASEKERA t al., 1997).For example, L-aspartate P-semialdehyde(329),undergoes deamination by aspartase togive fumaric P-semialdehyde (330), which

reacts w ith cysteine-273 to give a catalyticallyinactive covalent adduct with the putativestructure (331) Fig. 115) (SCHINDL ERnd VI-

OLA, 1994). Inhibition of inactivation by fu-m arate and divalent metal ions, indicates thatinactivation by L-aspartate P-semialdehyde(329),occurs at th e active site. How ever, mu ta-genesis of Cysteine-27 3, yielded mu tants(C273A, C273S) which were still catalyticallyactive, but less sensitive to inactivation (GIO R-GIANNI et al., 1995). Similarly, treatment ofaspartase with N-ethylmaleimide caused loss

of activity du e to mo dification of cysteines-140or 430, but mutagenesis (C430W) increasedactivity (MURASE t al., 1991).Analysis of con -served residues and comparisons with thestructu re of fumarase, provided a furthe r list of

329 L-Aspartate1 P-semialdehyde

NH4@ /I Aspartase

A s p m e 11330 Fumaratesemialdehyde

IspartaseI

cys-273-s

4JH2

331

Fig. 115. Covalent mod ification of aspartase by fu-marate semi-aldehyde (330) formed by aspartasecatalyzed deam ination of L-aspartate semi-aldehyde

(329).



12 8 2 Lyases

candidates for mutagenesis (C389S, C389A,H123L), but these caused only minor reduc-tions in activity. How ever, these studie s result-ed in the identification of lysine-327 as a bin d-

ing site for one of the carboxylate groups ofthe substrate a nd the othe r is probably boundby A rginine-29 (SARIBASt al., 1994;JAYASEK-

ERA et al., 1997). In summ ary, the role of theactive site residues in aspartase is subtle andquite different to that of fumarase. Their fullidentification will probably require an X-raycrystal structure w ith a substrate analog.

BiotechnologyThe m ajor use for L-aspartate (101) is an im-

portant food additive. It is also used in many

medicines, as a precu rsor of th e artifical sweet-ner, Aspartame (325; Fig. 112), an d as a chiralsynthon. L-Glutamate (114) and L-aspartate(101) ar e the most imp ortant am ino donors intransamination in nature. L-Aspartate is themost common donor in the manufacture ofamino acids by transam ination (Fig. 116), butthere are a few transaminases which do not ac-cept it as an amino donor. However, by ex-ploiting a L-glutamate (114)la-ketoglutarate(334) couple, it can be used with virtually all

transaminases. Moreover the unfavorableequilibrium constant ( K e g= circa 1) can be dis-placed by decarboxylation of a-ketoglutarate(334) to pyruvate (1) (STIRLING,992).

L-Aspartate has been produced in batch cul-tur e since 1960, using free, intact E . coli cellswith h igh aspa rtase activity. In 1973, a co ntinu-ous process was introduced w hich used E . colicells entrap ped in polyacrylamide gel, packedinto a column. Such columns have a half-life of120 days at 37 O C. Immobilization in k-carra-geenan, hardened with glutaraldehyde and

hexam ethylene diamine extended th e half lifeto almost two years an d this system was intro-duced into produc tion in 1978. Using a 1,000Lcolumn packed with hardened k-carrageenanimmobilized E. coli some 3.4 tonnes of L-aspartate can be produced per day and 100tonnes per month. E. coli EAPc7 producesseven times higher aspartase activity than theparent strain ( E . coli AT CC 11303), but alsohas fum arase activity. This can be abolished byincubating the culture broth with 50 mM, L-aspartate at 45 " C for 1 hour. After immobil-

ization, the pro ductivity of L-aspartate treated

332 a-Ketoacid 333 L-Amino acid

Aminotransferase

114L-Glutamate 334 2-Ketoglutarate

28 Oxaloacetate 101L-Aspartate

Oxaloacetatedecarbox ylase

c02

1 Pyruvate

Fig. 116. L-aspartate (101) is an amino donor fortransaminationwith most a-ketoacids (332).When itcannot act as a direct donor, the amino group can bedonated via L-glutamate(114).The equilibrium con-stant is typically ab out 1,but may be displaced to-wards product (333) formation by in situ decarboxy-lation of oxaloacetate (28) to give pyruvate (l) ,

which is a less reactive substrate.




E. coli EAPc7 was some six times the parentstrain and the half life was 126 days. Thissystem was commercialized in 1982 (CHIBATA,et al., 1995).The activityof L-aspartase from E.

coli has also been enhanced 2 to 3 fold by site-directed mutagenesis (C430W) (MURASEetal., 1991), limited C-terminal proteolysis andacetylation of amino groups by acetic anhy-dride (MURASE nd YUMOTO,993).

The amination of fumarate to give aspartateis exothermic, which means that on a largescale, an expensive heat exchanger system isrequired to minimize thermal degradation ofthe enzyme. Consequently, there has beenmuch interest in aspartases from thermostableorganisms, which will tolerate operation at

higher temperatures. A thermostable aspart-ase has been purified from the thermophile,Bacillus sp. YM55-1, which has an activity 3 to4 times higher than those from Escherichia co-li and Pseudomonas fluorescens respectively at30 “C (KAWATA,t al., 1999).

4.4.1.2 P-Methylaspartase

(Methylaspartase Ammonia-Lyase,

EC 4.3.1.2)p-Methylaspartase catalyzes the addition of

ammonia to mesaconate ((E)-Zmethylfumar-ate) (335) to give ~-threo-(2S,3S)-3-methyl-aspartate (336) (Fig. 117). The stereochemistryof addition is anti. Ammonia adds to the si-faceof C-2 and hydrogen to the re-face of C-3.Thisreaction is the second step in the catabolism ofL-glutamate (114) nder anaerobic conditions(KATo and ASANO, 997). In the prior step, L-

glutamate (114)undergoes an extraordinary

rearrangement to give ~-threo-3-methylaspar-tate (336),which is catalyzed by coenzymeBIZ-dependent glutamate mutase (EC 5.4.99.1).

The p-methylaspartase from the obligateanaerobe Clostridium tetanomorphum hasbeen extensively investigated. Intimate detailsof the mechanism have been deduced in a longseries of studies (GANIet al., 1999) and thegene has been cloned, sequenced and ex-pressed in E. coli (GODA t al., 1992).The pre-liminary data on enzymes from other sources,suggests that they are fairly similar. p-methyl-

aspartases from the facultative anaerobes, Cit-

I 335 Mesaconate

4 1

NH4@-4HpPMethylaspartase

I 1

1 336 L-thre0-(2Ss.3S)-3-Methylaspartate

Fig. 117. The p-methylaspartase catalyzed additionof ammonia to mesaconate ((E)-2-methylfumarate)(335) to give ~-threo-(2S,3S)-3-methylaspartate

(336).

robacter sp. strain YG-0505 and Morganellamorganii strain YG-0601 (KATo and ASANO,1995a), Citrobacter amalonaticus strain YG-1002 (ASANO nd KATO, 1994a,b; KATO andASANO, 995b) have been purified and crystal-lized. The gene for p-methylaspartase in Citro-bacter amalonaticus strain YG-1002, has beencloned, sequenced and expressed in E. coli, butsynthetic applications have barely been ex-plored (&TO and ASANO, 998).

The remainder of this section will deal ex-clusively with the p-methylaspartase from theClostridium tetanomorphum, unless indicatedotherwise. The gene has been cloned, se-quenced and expressed in E. coli. The openreading frame, codes for a polypeptide of 413amino acids (45,539 Da), which would exist asa homodimer in solution (GODA t al., 1992).

The deduced sequence for Citrobacter amalo-naticus YG-1002 p-methylaspartase is identi-cal in length and has 62.5% identity (KATOand ASANO, 998).

It has long been assummed (on the basis ofsubstantial circumstantial evidence) that p-methylasparatase, histidine ammonia-lyase(HAL) and phenylalanine ammonia-lyase(PAL) all bear a dehydroalanine group at theactive site, which acts as an electrophilic “trap”for ammonia during the catalytic cycle (GUL-

ZAR et al., 1995). However, a recent X-ray

structure of catalytically active Pseudomonas



130 2 Lyases

putida HAL has revealed a 4-methylene-imid-azole-5-one group at the active site (cf. Fig.123;SCHWEDEt al., 1999). It is not clear at thetime of writing, if this is a general or unique

phenomenon. Nevertheless, the gross mecha-nisms of action of dehydroalanine and 4-meth-ylene-imidazole-5-one group, are sufficientlysimilar, that a discussion of the chemistry ofthe former, provides a fair approximation tothat of the latter. Both of the P-methylaspar-tases bear the Ser-Gly-Asp sequence (Tab. 10)which is conserved in all putative dehydroala-nine bearing proteins, however, none of theother conserved residues found in HAL'S orPAL'S are found in the p-methylaspartases. Al-beit that the conserved glycine in the -5 posi-tion may be a conservative replacement forproline.

p-Methylaspartase has attracted much at-tention because (unlike aspartase), it accepts afair range of alternative alkenic substrates(Tab. 11) and nucleophiles (Tab. 12) (BOTTINGet al, 1988a). Substitution of the methyl groupof mesaconate (335, ntry 2) by alkyl groups(entries 3-5) reduces the rate, but neverthelessadequate yields of products can be obtained.Amination of the halo-derivatives (entries

7-10), is complicated by enzyme deactivationand only the chloro-fumarate (entry 8) gives

usuable yields of products. P-Methylaspartaseis much more tolerant of variation of the nu-cleophile. Alkyl amines (entries 1-43), hydra-zines (9-13), hydroxylamines (15-17) and me-

thoxylamines (entries 18-20) are all acceptedto some extent. Overall, structural deviationfrom the natural substrates reduces the rate inan additive way. The cyclopropyl methyl- andbromo-derivatives (Mia,b) are a potent com-petive inhibitors (Ki 20 and 630 pmol L -' andrespectively) (Fig. 118; BADIANIt al., 1996). Itwould be of great interest to discover if theamino derivative (341c)would yield the corre-sponding cyclopropene and if this reactioncould be extended to higher cyclic homologs.

P-Methylaspartase catalyzes the exchangeof the C-3 proton of ~-threo-3-methylaspartate(336) with solvent, faster than deamination(BOTTING t al., 1987). This was originallyinterpreted as indicating a carbanion interme-diate, but 15N/14N-2H/1Hsotope fractionation

Fig. 118. Cyclopropyl inhibitors (341)of p-methyl-asparatase (BADIANIt al., 1996).

Tab. 10.Am ino Acid Sequences of E nzymes Bearing Dehydroalanine/4-Methylene-Imidazole-5-Onere-cursor Mo tifs (entry 1,GODA t al., 1992;entry 2, KATO and ASANO,998;entries 3-15,LANGERt al., 1994b)

Entry Enzyme Species Range Residues5 4 3 2 1 0 1 2 3 4 5 6

1 BMA Clostridium tetanomorph um 168-179 P V F A Q S G D D R Y D

2 BMA CitrobacteramalonaticusYG-1002 168-179 P L F G Q S G D D R Y I

3 HAL P pu t i da 138-149 G S V G A S G D L A P L

4 HAL S. griseus 142-153 G S L G C S G D L A P L5 HAL B. subtilis 137-148 G S L G A S G D L A P L

6 HAL R.norvegicus 250-261 G T V G A S G D L A P L

7 HAL M.musculus 250-251 G T V G A S G D L A P L

8 PAL R . toruloides 205-216 G T I S A S G D L S P L

9 PAL R.rubra 211-222 G T I S A S G D L S P L

10 PAL 0. sa t i va 184-195 G T I T A S G D L A P L

11 PAL P c r k p u m 197-208 G T I T A S G D L P V L

12 PAL G.max 194-205 G T I T A S G D L P V L

13 PAL I. batates 187-198 G T I T A S G D L P V L

14 PAL L.esculentum 204-315 G T I T A S G D L P V L

15 PAL P vulgaris 193-204 G T I T A S G D L P V L

BMA,p-methylaspartase;HAL,histidine am monia-lyase;PAL,phenylalanine ammon ia-lyase




Tab. 11.The P-MethylaspartaseCatalyzed Addition of Ammonia to Substitu-ted Fumarates (AKHTARt al., 1986,1987a,b)

ONH,

p-Methylaspartase

R R

338 339

Entry R Yield Rate, Comments%

12

3

456

789

10

HMeEt

"Pr

'Pr"BuFC1BrI

9061

6049

540

60

-

-

-

very fast, (1%+ 101)fast (335+ 336)moderately fastvery

slowvery slownonevery slowmoderately fastmoderately fast, inhibitornot a substrate

me asurem ents demonstrate tha t elimination isconcerted and that product release from theenzyme occurs in a slower step. The C-3 proton

exchange reaction has a primary deuterium ki-netic isotope of circa 1.6 in the range pH6.5-9.0, but th is is redu ced to 1at high and lowconcentrations of potassium ions. Magnesiumions ar e also essential for activity (BOTTING tal., 1988b, 1989;BOTTING nd G AN I, 992).

The elimination of amm onia from th e 3-epi-mer of the natural substrate i.e., ~-erythro-3-methylaspartate (342), ccurs at abou t 1% ofthe rate of the n atural substrate (Fig. 119).Iso-tope studies have demonstrated that epimer-ization doe s not occur prior to elimination a nd

hence the process must be a syn-elimination.In the addition direction, the threo-stereoiso-m er (336) is formed first and the n th e erythro-stereoisomer is formed very slowly (ARCHERet al., 1993a,b; ARCHERnd GAN I, 993;BEARet al., 1998).

The mechanism of action of p-methylaspart-ase has undergone a series of refinements asthe evidence has accumulated (Fig. 120). Thecurrent model for the catalytic cycle encom-passes 16 intermediates an d 23 kinetic param-eters. The resting sta te of th e enzyme contains

a conjugate addition adduct of the dehydroala-

nine residue (most likely with cysteine = B,)and no bound metal ions (343). Binding ofmagnesium ions (344), romotes addition of L-

fhreo-3-methylaspartate (336) and binding ofpotassium ions prom otes th e P -elimination ofB, (346) o reveal the de hydrhydro alanine res-idue (347). Conjugate addition of the aminogroup of ~-threo-3-methylaspartate (336),

gives the adduct (349) which undergoes con-

342L-erythro-(2S,3R).

3-Methylaspartate

p-Methylaspartase

002(+rcoF35Mesaconate

Fig. 119.The P-methylaspartase catalyzed syn-elimi-nation of ammonia from ~-erythro-(2S,3R)-3-meth-ylaspartate to give (342) mesaconate ((E)-2-methyl-



132 2 Lyases

Tab. 12.The P-MethylaspartaseCatalyzed Addition of Amines to SubstitutedFumarates (GULZARt al., 1994,1997;BOTTINGt al., 1988a)

RZ, ,R3NI

HH 340 R2, I ,R3

\ ?KO,B-Methylaspartase 0 2&coy

R' R'

Substituents

Entry Alkene 338 Amine 340 Conversion , YieldR' R2 R3 % %

1

23456789

1011121314

151617181920

H

MeEt"Pr'PrHMeHHMeEt"Pr'Prc1

HMeEtHMeEt

Me

MeMeMeMeMeMeEt

NH2NHzNH2NHZNH2NHz

OHOH

OHOMeOMeOMe

H

HHHHMeMeHHHHHH

H

HHH

HHH

55

5460NRNR70NR

58991909090NR

9090608070NR

45

4035-

-

38-

-

4241573133

2819123134

-

-

NR, no reaction

certed elimination (350).A sequenc e of protontransfers then promotes the p-elimination ofamm onia (352) and the readdition of B2 to the

dehydroalanine residue (354), which is fol-lowed by product (336) and cofactor release(POLLARDt al., 1999;GANI t al., 1999; BOT -TING et al., 1992).

4.4.1.3 Histidase (Histidine

Ammonia-Lyase,EC 4.3.1.3)

Histidase catalyzes the a ddition of amm oniato uroconate (368) o give L-histidine (137)

(Fig. 121). Unlike the P-methylasparatase and

phenylalanine ammonia-lyase, histidase is

found in mamm als (TAYLOR,t al., 1990) andin fact uroconate was first discovered in dogurine in 1874. Histidase is fou nd in plants a nd

numerous microorganisms, e.g., Pseudomonastestosteroni, Bacillus su btilis and Streptomycesgriseus (Wu, et al., 1992). Several organismsare capable of growth by utilising exogenoushistidine. Th e majority of resea rche rs havestudied the enzyme from Pseudomonas ATCC

11299, which has b een described a t varioustimes as l? put ida, l? jluorescens, but is nowclassified as l? acidovorans (CONSEVAGEndPHILLIPS,985). We w ill follow the consensususauge w hich is l? putida.

Histidase is the first enzyme in the major ca-

tabolism of histidine. A deficiency of histidase




343

Q 0 2 q

367

366

H " 0 fl

347

Fig.UO.



134 2 Lyases

Ser Gly Asp

Histidase

NH,@

137L-Histidine

Fig. 121.The histidase catalyzed addition of ammo-nia to urocanate (368) o give L-histidine (137).

activity in humans causes accumulation of his-tidine in body fluids (histidinemia, TAYLOR,tal., 1991). The amino acid sequence is 93%conserved with rat and mouse sequences in-cluding four N-glycosylation consensus sites.Histidase activity has been detected primarily

in the liver and epidermis as well as in kidney,adrenal gland and skin. (SUCHI t al., 1993; SA-NO, t al., 1997).

It has long been assummed that the catalyticactivity of histidase results from the presenceof a dehydroalanine residue (370), ormed byp-elimination of the hydroxyl group of a serineresidue (369) (Fig. 122). This would plausiblyact as a Michael acceptor for the conjugate ad-dition of ammonia and or the amino group ofhistidine. The conjugate addition adduct acts ineffect as an immobilization device, around

which the other components of the reactioncan be optimally orientated to promote cataly-sis (cf. Fig. 120). Putative dehydroalanine resi-dues are found in a conserved sequence;GXXXXSGDLXPL in histidase and phenylal-anine lyase, whereas P-methylaspartase onlyhas the key SGD motif (Tab. 10). There are noX-ray crystal structures of enzymes bearing adehydroalanine residue. Consequently iden-tification of such residues, hangs on the detec-tion of the conserved sequence and label-inghhibition reactions with nucleophiles such

as [14C]-cyanide (CONSEVAGEnd PHILLIPS,

Enz

J

0 0

EnzH

0

370

Fig. 122. Formation of dehydroalanine residues inpeptides.

1985), [14C]-nitromethane,cysteine (HERNAN-DEZ et al., 1993), phenyl hydrazine and sodiumbisulfite. Double inhibition experiments andprotection by L-histidine established that thesereagents react at the same catalytically signifi-cant site. The adduct formed with [3H]-sodiumborohydride and [‘“CI- cyanide were shown tobe [3-3H]-~-alanine nd [4-14C]-aspartate re-spectively (HANSON nd HAVIR, 972; CON-SEVAGE and PHILLIPS,985).

This work enabled the serine precursor ofthe putative dehydroalanine to be assigned asserine-143 in l? putida histidase (HERNANDEZand PHILLIPS,994; LANGER,t al., 1994a) andserine-254 in rat histidase (TAYLORand

MCINNES,994)Mutation of the key serine residue residueto alanine abolishes activity (TAYLOR ndMCINNES, 994; LANGER t al., 1994b; HER-NANDEZ and PHILLIPS, 994), whereas muta-tion to cysteine, yields active enzyme, suggest-ing that the catalysis of post-translational issufficiently robust to tolerate other nucleofug-es (LANGER t al., 1994a). There is also someevidence that the thiol group of a cysteine res-idue close to the putative dehydroalanine resi-due may undergo conjugate addition to give a

protected or resting state of the enzyme (CON-




lized. The presence of a 4-methylidene-imida-zole-5-one residue (372), ather than a dehy-droalanine residue (370),equally explains allof the chemistry previously attributed to the

latter (SCHWEDE,t al., 1999). In addition, itexplains the former puzzling observation thatEdman degradation stops two residues beforethe serine (HERNANDEZ t al., 1993). It remainsto be seen if the 4-methylidene-imidazole-5-one residue (372)of histidase is a unique ab-beration or the first member of a new family ofenzymes.

The mechanism of interconversion of L-his-tidine (137) and urocanate (368)provides anumber of puzzles. The most of pressing ofwhich is, how is the non-acidic proton removed

in the presence of an ammonium group. Thestereochemistry of addition-elimination isstereospecifically trans with exchange of the(3 ’-pro-@-hydrogen (YOUNG, 1991). Ex-change of the proton at C-5 with solvent pro-tons provides a clue to the mechanism. This oc-curs much more rapidly with labeled uroca-nate (388a,b to M a) than L-histidine (376to137a) and is not an obligatory step in themechanism. However, an analogous reactionof an electrophile with L-histidine (376), ro-

vides an imminium ion (377),which is ideallyplaced to facilitate loss of the 3‘-proton. Theenamine (378)so formed can then eject am-monia to give another imminium ion (379),which is restored to neutrality by loss of theoriginal electrophile. Alternatively, release ofthe electrophile from the enamine (378) ieldsurocanate (288a to 368) directly. This mecha-nism was originally postulated with a proton asan electrophile (FURUTA t al., 1990, 1992),then dehydroalanine (LANGERt al., 1995) andfinally with a 4-methylidene-imidazole-5-one

residue (372) (Fig. 126) (SCHWEDE, t al.,1999).

Mutants of histidase that lack the serine res-idue (e.g., S143A), which is the progenitor ofthe 4-methylidene-imidazole-5-one moiety(372) Fig. 123), retain a minute amount of ac-tivity with L-histidine (137)and K,,, is essen-tially unchanged. More surpisingly, the kineticparameters for 5-nitro-~-histidine 385to 383)are virtually identical for the wild type andmutant! This can be rationalized by assum-ming that the nitro group plays the role of the

prosthetic moiety in the enzyme. Thus the

SEVAGE and PHILLIPS, 990). The activity ofhistidase is increased by 2-10 fold by Mn2+,Fez+,Cd2+ and Zn2+ and this correlates withthe excellent coordinating properties of dehy-

droamino acids towards metal ions (JEZOWS-KA-BOJCZUCnd KOZLOWSKI991; 1994).Fur-thermore, theoretical calculations showed thatall dehydroamino acids except dehydroalaninetend to bent a peptide chain towards a turnconformation which has a strong impact on theco-ordination ability of the dehydropeptide li-gand (JEZOWSKA-BOJCZUC,t al., 1996).

Methyl- and phenyl-glyoxal inhibit histidaseby reaction with arginine residues. Alignmentof four histidase and twelve phenylalaninelyase sequences revealed a single completelyconserved arginine residue and another con-served in 15 of the sequences, but not l? putiduhistidase (WHITE nd KENDRICK,993).

The preceding edifice of data and deduc-tions has been shaken by the publication of theX-ray crystal structure of histidase, whichshows that dehydroalanine is not present atthe active site, but a previously unknown 4-methylidene-imidazole-5-one residue (372)(SCHWEDE,t al., 1999),which is formed by anintramolecular condensation and elimination

of water. The authors postulate intramolecularnucleophilic attack to give the amidine (371),followed by a,p-elimination. One objection tothis proposal, is that the nucleophile is anamidic nitrogen, which is only weakly nucle-ophilic. Conversely, elimination of water firstto give dehydroalanine (370), educes the con-jugation between the amidic nitrogen and thecarbonyl group and changes the conforma-tional preferences of the peptide chain,so as topromote condensation. The 4-methylidene-imidazole-5-one residue (372) has a curious

conformational feature. The substituent at-tached to the “amidic” nitrogen is out of theplane of the ring (0-C-N-C, is 54”) andhence the nitrogen lone pair is sp3-hydridized.This renders the carbonyl, ketone like, andhence more susceptible to nucleophilic addi-tion. In this conformation, the adduct is ableutilize canonical forms (373)and (374) Fig.124), but not the aromatic 6welectron canoni-cal form (375). It is conceivable that additiontriggers a conformational change which ren-ders the nitrogen substituent co-planar and

hence enable this canonical form to be uti-



136 2 Lyases

M a Ser Gly Asp- -A -Enz’ Enz

‘ H O H O

369 extended)

Enz’

Enz

Enz NH

3724-Methylidene-imidazole-5-one. En;

Fig. 123. Formation of 4-methylene-imidazole-5-oneesidues (372) in peptides.



4 Examples ofLyases 13 7

372

Nu>

375

137

377H

R = H

~ R = ~ H

Fig. 124.Canonical forms for the addition adducts of4-methylene-imidazole-5-oneesidues (372) in pep-tides. The stereochemistry of the nitrogen substitu-ent controls access to the aromatic form (375).

HSY

378H

Ecanonical form (384)s ideally set up for loss of

the 3'-proton, and the elimination of the am-monia is driven by the e nam ine (385) as before(LANGERt al., 1997a;LAN GER t al., 1995).

Histidase has not found large scale practicalapplications as yet, although it has been usedfor the preparation of (3R)-[3-3H1]-histidineby exchange with tritium oxide (YOUNG,1991). Urocan ic acid acts a s a hydroxyl radicalscavenger in skin and like histamine (138) t isan immunosuppressant (KAMM EYER t al,1999).

4.4.1.4 Phenylalanine Ammonia-

Lyase [PAL, EC 4.3.1.51

Phenylalanine ammonia-lyase (PAL) cata-lyzes the deamination of L-phenylalanine(388)o give trans-cinnamate (389) (Fig. 128).PAL is widely distributed in higher plants,some fungi, yeasts and a single prokaryote,Streptomyces,but is not found in true bacteriaor animals. PAL is of prodigous importance in

plants, because som e 3 0 4 5 % of all plant

ZN+ co:

GN

379E

N H , ~

coy

368

Fig. 125. Genera lized mechanism for the histidasecatalyzed deamination of labe led L-histidine (376 to

137a ) and C-5 proton exchange.



138 2 Lyases

J 137L-Histidine

381

382

NH3j72

Hh r "368 Urocanate

Fig. 126.Mechanism for the histidase catalyzed de-amination of L-histidine (376 to 137) to urocanate

(368).

organic material is derived from L-phenylala-nine (388) nd to a lesser extent L-tyrosinethrough the cinnamate pathway (JONE S,984).The amm onium ions released in the formation

of cinnamate are recycled by conversion ofglutamate t o glutamine catalyzed by glutaminesynthetase (RAZAL t al., 1996).The ultimateproducts of the cinnam ate pathway ar e lignins,lignans, flavanoids, benzoic acids, coumarinsand suberins (MA", 1987, MAN N t al, 1994)and phenols (NICHOLSONnd HAMMERSCH-MIDT, 992).PAL levels are increased in plants,in response t o environm ental stress and path o-gen infection (RE GLINS KI,t al., 1998;ORCZYKet al., 1996; KIM et al., 1996). In microorgan-isms, PAL acts as the first step in th e catab o-

lism of the exogenous phenylalanine as a car-bon source. The enzyme has been purifiedfrom numerous sources, including potato tu-bers, maize, sunflower hypocotyls (JORR IN tal., 1988), S.pararoseus and the "smut" fungalpathogens, Ustilago hordei (HANSONnd HA-VIR,1972) and U.maydis (KIMet al., 1996) andtwo forms from leaf mustard, Brmsica junceavar. integrifolia (LIMet al., 1997,1998).

L-Phenylalanine is an essential nutrient forhuman nutrition and is used widely in the food

and pharmaceutical industries. Currently, mostL-phenylalanine is m anufactured by ferm enta-tion using an overproducing strain of E. coli,cultivated on glucose.The dem and for L-phen-ylalanine m ethyl ester (324) or the manufac-ture of th e artifical sweetner Aspartam e (325)(Fig. 111) assures continuing efforts to developimproved methods of manufacture (OYAMA,1992; KINOSHITAnd NAKAYAMA,978; DUF-FY,1980).

PAL is a mem ber of the putative dehydroa-lanine fam ily, which also contains p -methy l-

aspartase an d histidase. It has be en shown re-cently that th e active site of histidase doe s notcontain dehydroalanine (370, ig. 122),but a 4-methylidene-imidazole-5-one(372) Fig. 123)moiety instead (SCHWEDE,t al., 1999). Thehigh sequence homology between PAL andhistidase (Tab. 10 ) suggests that it almost cer-tainly contains the same functional residue.Labeling studies indicated that serine-202 isthe progenitor residue in parsly PAL (Petro-selinum crispum L.). The PAL1 gene was ex-pressed in E. coli and w hen serine-202 was mu-

tated to alanine, virtually all activity was lost



4 Examplesof Lyasrs 139

383 5-Nitro-L-histidine 384 385 I

387 386

Fig. 127. The m echanism for deamination of 5-nitro-~-histidine(383),with histidase mu tants (e.g., S143A),which lack the 4-methylidene-imidazole-5-oneesidue (372).

L-Pheny lalanine

IPhenylalanineammonia-lyase

trans-Cinnamate

Fig. 128.The phenylalanine ammonia-lyase (PAL)catalyzed deamination of phenylalanine (388) ogive trans-cinnamate (389).

(SCHUSTERnd RETE Y, 995). However, activ-ity was retained with 4-nitro-phenylalanine,which parallels the results achieved with 5 4 -trohistidine (383)and histidase (LA NGE Rt al.,1997a, 1995). Mu tation to cysteine (which is

able to un dergo elimination), yielded a m utant

which had activity, identical t o the wild type(LANGERt al., 1997b).

Potato tuber PAL catalyzed deam ination ofL-phenylalanine(388),o give trans-cinnamate(389) is stereospecific. The (3-pro-S)-proton islost, hence elimination occurs with trans-stere-ochemistry (HAN SON ,t al., 1971; WIG HTM AN,et al., 1972) (Fig. 129). This is exactly as ob-served with /I-methylaspartase and histidase.Intriguingly, PAL from Rhodotorula glutinis(and potato tubers) also catalyzes the deami-nation of D-phenylalanine to trans-cinnamate(389),albeit at 0.02% of the rate of L-phenylal-anine (388).Rhodotorula glutinis PAL cata-

lyzed deamination of D-phenylalanine resultsin preferential (but not exclusive) loss of the(3-pro-R)-proton , which is con sistant withtram-stereochemistry for the overall elimina-tion (EASTONnd HUITON, 1994).

Deamination of [2H,]-~-phenylalaninedeu-terated in the aromatic ring shows a kineticisotop e effe ct of 1.09, relative to L-phen ylala-nine (388) Fig. 129) (GLO GE t al., 1998).Thiscan be rationalized by postulating a Frie-dels-craft type reaction with an electrophile[presumably 4-methylidene-imidazole-5-one

(372)] to give the conjugated carbonium ion



140 2 Lyuses

L-Phenylalanine

Phenylalanineammonia-I yase

390

H@-1

-1? NH3

trans-Cinnamate

Fig. 129. The mechanism for the phenylalanine am-

monia-lyase (PAL) catalyzed deamination of phenyl-

alanine (388a to 388) o give trans-cinnamate (389bto 389).

(390). Deprotonation yields the triene (391),

which undergoes 1 Celimination to urocanate

(389).L-tyrosine (141a) is a poor substrate for

PAL, whereas racemic 3-hydroxyphenylala-

nine (392) (D,L-rneta-tyrosine) s a slightly bet-

ter substrate than L-phenylalanine,despite the

presence of the D-enantiomer (Fig. 130).The 3-

HO ONH3 141a

L-Tyrosine

L-Cyclohexyl-alanine

0°C";NH3 394L-CVCIOOCta-

mc8' 395.-

a R = O Hb R = N H ,

Fig. 130. Alternative substrates and inhibitors of

phenylalanine ammonia-lyase (PAL).

hydroxy-group is capable of promoting the

Friedels-Craft reaction, whereas the 4-hydrox-yl group of L-tyrosine is not (SCHUSTERnd

RETEY,1995). Cyclohexylalanine (393) doesnot undergo deamination with PAL, but it is a

good inhibitor, which indicates binding is not

impaired, Cyclooctatetraenyl alanine (394)

would not be expected to be able to participate

in the Friedels-Craft reaction, but both bind-

ing and turnover were impaired and hence it is

not possible to make a wholly electronic argu-

ment for the lack of reactivity. Substrates in

which the amino substituent has been replacedor substituted (395) are good inhibitors (Mu-

NIER and BOMPEIX,985; JONES nd NORTH-

COTE, 1984).




The 2-, 3- and 4-pyridylalanines (397) (Fig.131) were p repa red by P AL catalyzed amina-tion of the corresponding cinnamates (3%) us-ing a high concen tration of a queous am monia,

partially neutralized with carbon dioxide. Inthe reve rse direction, th e 2-, 3- and 4-pyridylal-anines (397)are also good substrates for PAL,with higher KM’sand similar or slightly higherVmax’shan L-phenylalanine. From the stand-point of classical solutio n phase chem istry, hisresult is unexpected. Pyridines undergo Frie-del-Craft reactions m ore slowly than benzene.But this is because unde r solution phase condi-tions they are present as the protonated form,in which the p yridinium grou p acts as an elec-tron withdrawing group. The enzyme catalyzed

reaction occurs without protonation, andhence under these circumstances the pyridylnitrogen acts as an activating group (GLO GE tal., 1998).

The p H optima for PA L is circa 8.7 (de pend-ing on source). Activity drops off quickly atlower pH, but higher pH is sometimes benefi-cial. Most PALS are deactivated by thiols, butonly a few by metals ions, e.g., A g+ , C u2 +,Hg2+ and Zn2+ (KIM,et al., 1996; LIM ,et al.,1997; 1998).The thermodynamically preferred

mode of action of PAL is deamination ofL-

phenylalanine (388) to give trans-cinnamate(389).However, in the presence of high con-

Pheny alanineammonia-lyase

5M NHJICO2

2-Pyridyl, 63% yield3-Pyridyl, 59% yield4-Pyridyl, 75%yield

Fig. l31. Pyridyl cinnamates (395) are good sub-

strates for phenylalanine ammonia-lyase (PAL) cat-

alyzed amination.

centrations of am monium salts the reverse re-action is also feasible, particularly if the am inoacid precipitates from solution. This is the di-rection of interest for the vast majority of in-

dustrial processes. Ammonia concentrationsup to 8 M have been utilized for L-phenyala-nine production (TAKAC, t al., 1995), but 5 Msuffices for most purposes (YAM ADA t al.,1981;EVANS, t al., 1987; GLO GE, 998).

trans-Cinnam ic acid is very cheap becau se itis easily prepared by aldol condensation ofbenzaldehyde an d a host of acetic acid eno lateequivalents (e.g., diethyl malonate). Un fortu-nately, trans-cinnamate (389) is a competitiveinhibitor of of most forms of PAL (Ustilagomaydis PAL is a n exception, KIMet a l., 1996),

which limits the reactor concentration andhence productivity. P-cyclodextrin (398) (Fig.132) has been advocated as a sequestrant fortrans-cinnamate in the deam ination direction,but it remains t o be seen if it is beneficial foramination (EASTONt al., 1995).

Despite the widespread abundance of PAL,industrial processes have almost exclusivelyused P AL from Rhodotorula glutinis, becausethe organism is easily cultivated, PAL levelsare high and the enzyme is easily purified in

high yield (K AN E ndFISKE,

985; D’CUN HA tal., 1996b). Sporidiobolus pararoseus, Rhod o-sporidium toruloides (WA TANAB Et al., 199 2),Rhodotorula rubra (EVAN S t al., 1987), Rho-dotorula graminis and Cladosporium dado-sporioides are unproven alternatives (TAKAC,et al., 1995, DRAU Z nd WALDM ANN,995).

The prod uction of L-phenylalanine (388) bywho le Rhodotorula g lutinis cells has been o pti-mized at p H 10.5,30 c , 8 M ammonia and aloading of 2 g of substrate per g of dry cells.Using a fed batch operation, the maximum

concentration of L-phenylalanine reached12.57 g L - l (70.19 mM ) from 14.8 g L-’ (100mM) cinnamate (389). Sodium glutamate a ndpenicillin increased PA L activity (TAKA Ct al.,1995).

Thus far the unfavorable equilibrium and in-hibition by cinnamate have m ade the produc-tion of L-phenylalanine using PAL, uncompet-itive compared to manufacture by fermenta-tion. However, PA L also catalyzes the amina-tion of trans-methyl cinnamate (which ischeaper than cinnamic acid) at 90% of the rate

of the natural su bstrate. The p roduct L -phenyl-



142 2 Lyases

398 P-Cyclodextin

Q

399 P-Cyclodextin complexwith trans-cinnamate 389

Fig. 132.P-Cyclodextrin is a cyclic hepta-saccharide ,which preferentailly binds trans-cinnamate (389) overL-phenylalanine (388), o give the complex (399) (EASTONt al., 1995).

alanine methyl ester is more valuable than L-

phenylalanine, because it is a direct precursorof Aspartame (325). The amination was runwith w hole R hodotorula glutinis cells, in a two

phase heptane: water system to overcome thelow solubility of trans-me thyl cinn am ate in wa-ter. The conditions were optimized a t pH 9.0,30' C, 16-18 h, 0.1 M substrate and 1M ammo-nium sulfate and 70% conversion wasachieved (D'CU NH A t al., 1994).But the activ-ity of th e cells declined rapidly during the bio-conversion and they were not resuable. Byadding small amounts of magnesium ions and10% glycerol, the cells could be recycled up tonine times with a total yield of 92g L-' (D'-

CU NH A t al., 1996a). PAL from the red basi-

domyces yeast Rhodosporidium toruloides hasbeen purified (REES and JONES, 996) andused in a single phase octanol: water mixture.At least 2% water was required for activity.The best result was obtained with freshly lyo-philized PA L, in a 96.4 :3.6, v/v, mix ture whichretained, up to 17% of the activity in aque ousmedia (REES nd JONES,997). The use of en -zymes in organic solvents offers the prospectof using hydrophobic substrates a t higher con-centrations than would otherwise be possiblein aqueous m edia (GU PTA, 992; TRA MP ER,t

al., 1992)

4.5 Other Carbon-Nitrogen Lyases

The gram-negative bacterium DSM 9103 isable to grow on ethylenediaminetetra-acetic

acid (EDTA) (400) (Fig. 133) as the solesource of carbon and nitrogen. The initial stepin the degradation is stepwise removal of theacetate groups as glyoxylate by an oxidativeprocess (WITSCHELt al., 1997).The same spe-cies also grows on (S,S)-ethylenediaminedisuc-cinate (401)and cell free extracts, without ad d-ed cofactors, degrade this substrate. An en -zyme was purified 41-fold that catalyzed rev er-sible formation of fumarate (1%) and W ( 2 -

aminoethyl) aspartic acid (402). (S,S)- an d(S,R)-ethylenediaminedisuccinate were ac-

cepted as substrates, but the (R,R)-stereoiso-mer was unchanged in both cell free and puri-fied enzyme assays (WITSCHEL nd EGLI,

1997). This d ata clearly indicates a stereospe-cific lyase system.

4.6 Carbon-Halide Lyases

(EC 4.5.1.X)

This subclass was originally conceived forthe enzyme w hich catalyzes the elimination of

hydrogen chloride from DDT (DDT-dehy-



5 References 14 3

me rcaptide (GUPT A t al., 1999). Plants an dmicroorganisms are extraordinarily adept athandling materials that a re toxic to higher or-ganisms (BALDI,1997). Toxic alkyl mercury

compounds are transformed either by alkylmercury-lyase or by a redu ctase which produc-es metallic mercury. The re a re c onsiderableprospects for the use of such microorganismsand enzymes in organometallic chem istry, butat present the m ain use of these enzymes is forland rem ediation (G HOS H t al., 1996;HEATONet al., 1998).

@o,& 400 EDTA

196Fumarate f

Fig. 133. The degradation of (S,S)-ethylenediami-nedisuccinate (401).

drochlorinase, EC 4.5.1.1). The other earlymember of this class is 3-chl oro -~- alan ine e-hydrochlorinase (EC 4.5.1.2), which is a pyri-doxal 5 ' -phosphate dependent protein, thatalso catalyzes p-replacement reactions. Fur-ther aspects of these enzymes are covered indetail in Chapter 3, this volume a nd in reviews

(FETZNER, 998; FETZNER nd LINGRENS,1994).

4.7 O th er Lyases

4.7.1 Alk yl Mercury-Ly ase(EC 4.99.1.2)

Alkylmercury lyase catalyzes the cleavageof alkylmercury bon ds in the presence of a thi-

5 ReferencesABELYAN,. A. (1999), Simultaneous production of

D-aSpartiC acid and L-alanine,Ap pl. Biochem. Mi-crobiol. 35,152-156.

ABELYAN,. A., AFRIKYAN,. G. (1997), On e s tepsynthesis of L-alanine from fumaric acid, Appl .Biochem. Microbiol. 33,554-556.

A B E L Y A N , ~ .., BAGDASARY AN,.N., AFRIYAN,. G.

(1991), Intracellular L-aspartate-0-decarboxylaseof Pseudomonas sp. and Alcalignes sp. and its im-mobilization, Biochemistry (Moscow) 56, 899-905.

AKHTAR, ., COHEN,M. A., GA NI,D. (1986), Enzy-matic syn thesis of 3-halogenoaspartic acids usingPmethylaspartase - nhibition by 3-brom oaspar-tic acid,J. Chem. Soc., Chem. Com mun. 1290-1291.

AKHTAR, ., COHEN,M. A., GA NI,D. (1987a), Ster-eochemical course of the enzymatic amination of

chlorofumaric acid and bromo-fumaric acid by 3-methylaspartate ammonia-lyase, TetrahedronLett. 28,2413-2416.

AKHTAR,M., BOTIING, N. F COHEN,M. A., GA NI,D.(1987b), Enantiospecific synthesis of 3-subs titut-ed a spartic acids via enzymic amination of subs ti-tuted fumaric acids, Tetrahedron43,5899-5908.

ALAM,N., SAFFERSON,. D. (1998), Malonyl-CoAand the regulation of fatty acid oxidation in sole-us muscle, Biochem. J . 334,235241,

ALBER,B. E., COLANGELO,. M., DONG,., STAL-HANDSKE,. M.V., BA IRD,T.T., t al. (1999), Ki-netic and spec troscopic characteriza tion of the 'y-carbonic anhydrase from the methanoarchaeonMethanosarcina thermophila, Biochemistry 38,

ALBERT,A., HA NA RA J ,~ . ,ENSCHEL,., KHAN,G.

L., RAMJEE, . K. et al. (1998), Crystal structureof aspartate decarboxylase at 2.2 A resolutionprovides evidence for an e ster in protein self pro-

13 119-13128.

01 to give a hydrocarbo n an d mercu ry (11) as a cessing,Nature Struct. Biol. 5,289-293.



14 4 2 Lyases

ALBERTY,.A., BENDER,. (1959), Proton magneticresonance spectra of malic acid and its salts indeuterium oxide,J. Am. Chem. SOC.1,542-546.

ALLERT, ., KJELLSTRAND,. , BROO, ., NILSSON,A., BALTZER,. (1998), A designed folded poly-peptide model system that catalyzes the decar-boxylation of oxaloacetate, J. Chem. SOC., erkinTrans II, 2271-2274.

ANE T, . A . L. (1960),The configuration of deuterio-L-malic acid produced enzymatically.Synthesis ofthre0-3-deuterio-~~-maliccid. J . Am. Chem. SOC.82,994-995.

ALEXANDER,. W., SANDMEIER, ., MEHTA, . K.,CHRISTEN,. (1994), Evolutionary relationshipsamong pyridoxal-5 -phosphate - dependent en-zymes - regiospecific a-family, P-family and y-

family, Eur. J . Biochem . 219,953-960.ALKORTA,., GARBISU,. LLAMA, . J., SERRA,. L.

(1998), Industrial applications of pectic enzymes:a review, Process Biochem . 33,21-28.

ALVAREZ,. J., ERMER,., HUBNER , ., SCHELLEN-BERGER, A. , SCHOWEN,. L. (1995), The linkageof catalysis and regulation in enzym e action, sol-vent isotope effects as probes of protonic sites inthe yeast pyruvate decarboxylase mechanism, J .

Am . Chem. SOC. 17,1678-1683.ANDERSON,. L., HAUSW ALD,. L. (1982), D-galac-

tonateb-fuconate dehydratase, Methods Enzy-mol.90,299-302.

ANDERSON,. A., MOO-YOUNG,., LEGGE, . L.(1987a), Optimization of growth cond itions forthe induction of tyrosine decarboxylase in Strep-tococcus faecalis, Biotechnol. Lett. 9,685-690.

ANDERSON,. A ., MOO -YOUNG, ., LEGGE, . L.(1992a), Developmen t of a multienzyme reactorfor dopamine synthesis.1.Enzym ology and kinet-ics, Biotechno l. Bioeng. 39,781-789.

ANDERSON,. A., MOO-YOUNG,., LEG GE, . L.(1992b), Developm ent of a multienzyme reactorfor dopamine synthesis. 2. Reactor engineeringand simulation , Biotechnol. Bioeng. 40,388-395.

Anonymous (1984, 1992), Enzyme Nomenclature:Recommendationsof the Nomenclature Commit-

tee of the International Union of Biochemistry o nthe Nomenclature an d Classification of Enzy me-Catalyzed Reactions. Orlando, FL: AcademicPress.

ANT SON,A. ., STROKOPYTOV,.V., MUR SHUDOV ,.N., ISUPOV, . N., HARUTYUNYAN,. H. et al.(1992), The polypeptide-cha in fold in tyrosinephenol-lyase, a pyridoxal-5’-phosphate-depen-de nt enzym e, FE BS Lett. 302,256-260.

ANTSON, . A., DEMIDKINA,. V., GOLLNICK,.,DAUTER, ., VONTERSCH,. et al. (1993) 3-Di-mensional structure of tyrosine phenol-lyase,Biochemistry 32,4195-4206 .

ARCHER,. H., GANI, . (1993), Kinetics and mech-anism of syn-elimination of ammonia from

(2S,3R)-3-methylaspartic acid by methylaspar-tase,J. Chem. SOC.,Chem. Commun. 14C142.

ARCHER,. H., THOMAS,. R., GA NI, . (1993a),Syntheses of (2S,3R)-3-methylaspartic and(2S,3R)[3-H2]-3-methylasparticcids - slow sub-strates for a syn-elimination reaction catalyzed bymethylaspartase, Tetrahedron Asymmetry 4,

ARCHER,. H.,THOMAS,. R., GANI , . (1993b),Ki-

netics and mechanism of syn-e limination of am-monia from (2S,3R)-3-methylaspartic acid bymethylaspartase,J. Chem. Soc., Chem. Commun.,140-142.

ARMSTRONG,.B., LIPSCOMB ,. L., CROUT, . H.G.,MORGAN,. J. (1985), Structure-activity studieswith the a$-dihydroxyacid dehydratase of Sal-monella typhimurium, J. Chem. SOC.erkin Trans.I,691-696.

ARMSTRONG,F.., HEDGE COCK,. J. R .,REARY,. B.,WHITEHOUSE,., CROUT, . H. G. (1974), Stereo-chemistry of the reductoisomerase and a,P-dihy-droxyacid dehydatase steps in valine and isoleu-cine biosynthesis,J. Chem. SOC.,Chem. Commun.,

ARMSTRONG,. B., LIPSCOMB,. L., CROUT,. H. G.,MITCHELL,. B., PRAKASH,. R. (1983) Biosyn-

thesis of valine and isoleucine: synthes is and bio-logical activity of (2S)-a-acetolactic acid (2-hy-droxy-2-methyl-3-oxobutanoic cid) and (2R)-and (3S)-a-acetohydroxybutyric acid (2-ethyl-2-hydroxy-3-oxobutanoic acid), J. Chem. SOC.,Per-kin. Trans.l,ll97-1201.

ARJ UNAN,., UMLAND,., DYDA,., SWAMINATHAN,

S.,FUREY, . et al. (1996), Crystal struc ture of thethiamin diphosphate-dependent enzyme pyru-vate d ecarboxylase from the yeast Saccharomyc-es cerevisiae at 2. 3 angstrom resolution, J. Mol.Biol. 256,590400.

ASAD A, ., TANIZAW A,., SAWADA,., SUZUKI,.,MISINO , ., SODA,K. (1981), Stereochemistry ofrneso-a,cdiaminopimelate decarboxylase reac-tion - The 1st evidence for pyridoxal 5’-phos-phate dependent decarboxylation with inversion

of stereochemistry,Biochemistry 20,6881-6886 .ASAN O, ., KATO,Y. (1994a), C rystalline 3-methyl-aspartase from a facultative anaerobe, Escheri-chia coli strain YG1002, FEM S M icrobiol. Lett.118,255-258.

ASANO, ., KATO,Y. (1994b), Occurrence of 3-m eth-ylaspartase ammonia lyase in facultative an-aerobes and their application to synthesis of 3-substituted (S)-aspartic acids, Biosci. Biotech. Bi-ochem. 58,223-224.

AXELSSON,. S.,BJURLING,., MATSSON,., LANG-STROM, B. (1992), 11C/14C ine tic isotope effects inenzyme mechanism studies, J. Am. Chem. SOC.

AXELSSON,. S., O’TOOLE, . J., SPENCER,. A , ,

1141-1152.

351-352.

114,1502-1503.



5 References 145

YOUNG , . W. (1994), Versa tile synthesis of ster-eospecifically labeled D-amino acids via labeledaziridines - Prep aratio n of (2R, 3S)-[3-H-’H1]-se-rine and (2R, 3R)-[2, 3-2H2]-serine; 2S, 2‘S, 3S,3’S-[3, 3’-’H2]- and (2S, 2’S, 3R, 3’ R)-[ 2,2’ , 3,3’-’&]-cystine and; (2S, 3S)-[3-’H1]- and (2S,3R)-[2,3-’H,]-pchloroalanine, . Chem. SOC.,Per-kin TransI., 07-815.

BABBITT,. C., MRACHK O,.T., HA SSO N, . S., HU IS-MA N , G. W., KO LTE R, . et al. (1995), Functionallydiverse enzyme superfamily that abstracts the a -protons of carbox ylic acids,Science 267, No. 5201,

BABBIIT, P. C., H AS SO N,M. S., WEDEKIND,. E.,PALMER,. R. J., BA RRE TT, . C. et al. (1996),Theenolase family: a general strategy for enzy me cat-alyzed abstraction of the a-protons of carboxylicacids,Biochemistry 35,16489-16501.

BABURINA,., GAO,Y. H., Hu, Z. X., JORDAN,F.,

HOHMANN,., FUREY ,W. (1994), Subs trate activa-tion of brewers yeast pyruvate decarboxylase isabolished by mutation of cysteine-221 to serine,Biochemistry 33,5630-5635.

BABURINA,., MOORE,D. J., VO LKO V, ., KA HY AO -GLU, A., JORDA N, ., MENDELSO HN,. (19%),Three of four cysteines, ncluding that responsiblefor substrate activation, are ionized at p H 6. 0 inyeast pyruvate decarboxylase: Evidence fromFourier transform infrared and isoelectric focus-ing studies,Biochemistry 35,10249-10255.

BABURINA,., LI, H. J., BENNION,., FUREY,., JOR-DAN, F. (1998a), Interdom ain inform ation transferduring sub strate activation of yeast pyruvate de-carboxylase: The interaction between cysteine221 and histidine 92, Biochemistry 37,1235-1244.

BABURINA,., DIKDA N, ., Guo, F. S., Tous, G. I.,ROOT,B., JORDAN, . (1998b), Reactivity at thesubstrate activation site of yeast pyruvate decar-boxylase: Inhibition by distortion of domaininteractions, Biochemistry 37,1245-1255.

BACH,R . D., CA NEP A, . (1996), Electronic factorsinfluencing the d ecarboxylation of pke toa cid s.Amodel enzym e study, . Org. Chem .61,6346-6353.

BACH,R . D., C ANEP A, . (1997), Theoretical m odelfor pyruvoyl dependent enzymatic decarboxyla-tion of a-amino acids, J. Am. Chem. SOC. 119,

BACH,R. D., CANEPA,C., GLUKHOVTSEV, . N.(1999), Influence of electrostatic effects on acti-vation barriers in enzy matic reactions: Pyridoxal5 ’-phosphate dependent decarboxylation of a-

amino acids, J . Am . Chem. SOC .121,6542-6555.BADIANI,., LIGHTFOOT,., GANI,D. (1996), Synthe-

sis and mode of action of 1-substituted trans-cy-clopropane 1,2-dicarboxylic acids: inhibitors ofthe methylaspartase reaction, J. Chem. Soc.,Chem. Commun. 675-677.

BAETZ , . L., ALLISON , . J. (1992), Localization of

1159-1 161.

11725-11733.

oxalyl-coenzyme A decarboxylase and formyl-co-enzyme A transferase in Oxalobacter formigenescells,Systemat. Ap pl. Microbiol. 15,167-171,

BAI,Q., SOMERVILLE,. L. (1998), Integration hostfactor and cyclic AMP receptor protein are re-quired for ’QrR-mediated activation of TPL inCitrobacter freundii,J. Bacteriol. 180,61734186.

BALDI, . (1997),Microbial transform ation of mercu-r y species and their importance in the biogeo-chemical cycle of mercury, Metal Ions Biol.Systems 34,213-257.

BARBAS11, C. F., HEN, A., ZHONG, i. F., HOW-M A ” , T., GRAMATIKOVA,. et al. (1997), Immuneversus natural selection: Antibody aldolases withenzymatic rates but broader scope, Science 278,

BARROWMAN,. M., FEWSON, . A. (1985), Phenyl-glyoxylate decarboxylase and phenylpyruvate de-carboylase from Acinetobacter calcoaceticus,Curr. Microbiol. 12,235-239.

BATTAT, ., PELEG, ., BERCOW TZ, ., RO KEM, . S.,GOLDBERG,. (1991), Op timization of L-malic ac-id production by Aspergillus fravus in a stirredfermenter, Biotechnol. Bioeng . 37,1108-1116.

BATTERSBY,. R. (1985), Enzym ic synthesis of la-belled chiral substances, in: Enzymes in OrganicSynthesis, Ciba Foundation Sympos ium 111 (POR-TER, R., CLARK, ., E ds ), pp. 22-30. London: Pit-man Press.

BAITERSBY,A. R . , N I C O L E ~ I ,., STAUNTON,.,VLEGGAAR,. (1980), Studies of enzym e mediat-ed reactions. Part 13. Stereochemical course ofthe formation of histamine by deca rboxy lation of(2s)-histidine with enzymes from Clostridiumwelchii and Lactobacillus 30a, J. Chem. SOC.,Per-kin Trans.I . ,43-51.

(1982a), Studies of enzyme mediated reactions.14. Stereochem ical course of the fo rmation of ca-daverine by decarb oxylation of (2S)-lysine withlysine decarbox ylase (EC 4.1.1.18) from Bacilluscadaveris,J. Chem. SOC., erkin I , 449-453.

BAITERSBY, . R ., STAUNTON,., TIPPET T,. (1982b),Studies of enzyme mediated reactions. 15. Stereo-chemical course of the formation of y -amino bu-tyric acid (GABA) by decarboxylation of (2s)-

glutamic acid with glutamate decarboxylase fromEscherichia coli, J . Chem. SOC.,Perkin I , 455-459.

BAITERSBY, . R., SCOIT, A., STAUNTO N,. (1990),Studies of enzyme m ediated reactions. 16. Stereo-chemical course of the formation of 5-hydroxy-tryptamine (serotonin) by decarboxylation of(2S)-S-hydroxytryptophan ith the aromatic L-

amino acid decarbox ylase (EC 4.1.1.28) from hogkidney, Tetrahedron46,4685-4696.

BAU,R.,BREWER,.,CHIANG , .Y., FUJITA, ., HOFF-MAN, J. et al. (1983), Absolute configu ration of achiral CH D g roup via ne utron diffraction - con-

NO.5346,2085-2092.

BATTERSBY,. R., MURPHY, R., ST,4UNTON, J.



146 2 Lyases

formation of the absolute stereochemistry of the

enzymatic formation of malic acid, Biophys. Res.Cornmun. 115,1048-1052.

BEAR,M. M., MONNE, ., ROBIE,D., CAMPION,.,L A N G L O I S , ~ .t al. (1998),Synthesis and polymer-

ization of benzyl(3R,4R)-3-methylmalolactonatevia enzymatic preparation of the chiral precursor,

Chiralily 10,727-733.BEARNE, . L., WOLFENDEN, . (1995), Enzymatic

hydration of an olefin: The burden borne by fu-

marase,J.Am . Chem. SOC.117,9588-9589.BEECKMAN,.,VAN DRIESSCHE,. (1998), Pig heart

fumarase contains two distinct substrate binding

sites differing in affinity, J . Biol. Chem. 273,31661-31669.

BEHAL, , T., OLIV ER, . J. (1997), Biochemical and

molecular characterization of fumarase from

plants: purification and characterization of the

enzyme cloning, sequencing, and expression of

the gene,Arch. Biochem. Biophys. 348,65-74.BEINERT, ., KENNEDY, . C. (1993), Aconitase, a

two faced protein-enzyme and iron regulatory

factor, FASEB J.,7,1442-1449.BEINERT, ., KENNEDY, . C., STOUT,C. D. (1996)

Aconitase as iron-sulfur protein, enzyme and

iron regulatory protein, Chem. Rev. %,2335-2373.BELL, . J.,ANDREWS,. C., SIVA K, . N., GUEST,. R.

(1989),Nucleotide sequence of the FNR egulat-

ed fumarase gene (fumB) of Escherichia coli K-12,J. Bacteriol. 171,3494-3503.

BELLEAU,., BURBA,. (1960),The stereochemistryof the enzymatic decarboxylation of amino acids,

J . Am . Chem. SOC. 2,5751-5752.BELLEAU,., FANG, ., BURB A,., MORAN,. (1960),

The absolute optical specificity of monoamine

oxidase,J.Am . Chem.SOC.82,5752-5753.BEN ITEZ,T., ASENT-RAMIREZ,. M, CASTREJON ,.,

CODON,A. . (1996),Development of new (yeast)

strains for the food industry, Biotechnol. Progr.

BENNER,.A., MORTON,T.. (1981),Analogs for ac-

etoacetate as an enzyme substrate. Stereochemi-

cal preferences, J . Am . Chem. SOC. 103,991-993.

BENNER,. A., ROZZELL,. D., MO RTO N,T. . (1981),Stereospecificity and stereochemical infidelity of

acetolactate decarboxylase (AAD),J.Am . Chem.

BENNEIT,G. N., RUDOLPH,. B. (1995), The central

metabolic pathway from acetyl-CoA to butyryl-

CoA in Clostridium acetobutylicum, FEMS Mi-crobiol. Rev. 17,241-249.

BENTLEY,., THIESSEN,. P. (1957), Biosynthesis of

itaconic acid in Aspergillus terreus. 111. The prop-

erties and reaction mechanism of cis-aconitic acid

decarboxylase, . Biol. Chem. 226,703-.BERG,M., HILBI,H., DIMROTH, . (1996), The acyl

carrier protein of malonate decarboxylase ofMalonomonas rubra contains 2’-(5 ”-phospho-

12,149-163.

SOC.103,993-994.

ribosyl)-3’-dephosphocoenzymeA as a prosthet-

ic group, Biochemistry 35,46894696.BERG,M., HILBI,H., DIMROTH,. (1997), Sequence

of a gene cluster from Malonomonas rubra en-

coding components of the malonate decarboxy-

lase Na+ pump and evidence for their function,

Eur.J . Biochem. 245,103-115.BERMEJO,. L., WELK ER, . E., PAPOUTSAKIS,. T.

(1998),Expression of Clostridium acetobutylicumATCC 824 genes in Escherichia Cali for acetone

production and acetate detoxification,App l. En-viron. Microbiol. 64,1079-1085.

BERING , . L., KUH NS,. J., ROWL ETT,. (1998),Pur-ification of bovine carbonic anhydrase by affinity

chromatography- an undergraduate experiment,

J. Chem. Ed. 75,1021-1024.BERRY,M. D., JUR IO,A. V., LI, X. M., BOULTON,.

A., (1996),Aromatic L-amino acid decarboxylase:

a neglected and misunderstood enzyme, Neuro-chem. Res .21,1075-1087.

BERTOLDI,., DOMINICI,., MOORE,. S., MAR AS, .,VOLTATTORNI,. B. (1998),reaction of DOPA de-

carboxylase leads to an oxidative deamination

producing 3,4-dihydroxyphenylacetone, Bio-chemistry 37,6552-6561.

BERTOLDI, . , CARBONE,., VOLTATTORNI,. B.(1999), Ornithine and glutamate decarboxylases

catalyze an oxidative deamination of their a-

methyl substrates,Biochem. J. 342,509-512.BJORNESTEDT,., ZHON G, . F., LERNER, . A., BAR-

BAS,c.F. (1996),Copying nature’s mechanism for

the decarboxylation of P-keto acids into catalytic

antibodies by reactive immunization, J. A m .Chem. SOC.118,11720-11724.

BODE,R., BIRNBAUM ,. (1991), Enzymatic produc-

tion of indolepyruvate and some of its methyl and

fluoro-derivatives,Acta Biotechnol. 11,387-393.BOEKER, .A., FISCHER,. H. (1983), Lysine decar-

boxylase (Escherichia c oli-b ),Methods Enzym ol.94,180-184.

BONNARME,., GILLET, ., SEPUL CHR E,A.., ROLE,C., BELOEIL,. C., DU CRO CQ,. (1995), Itaconate

biosynthesis in Aspergillus terreus, J . Bacteriol.

BORNEMANN,., CROUT,D. H. G., DALTON,H.,HUTCHINSON,. W., DE AN ,G. et al. (1993), Ster-

eochemistry of the formation of lactaldehyde and

acetoin produced by the pyruvate decarboxylases

of yeast (Saccharomyces sp.) and Zymomonasmobilis - different Boltzmann distribuons be-

tween bound forms of the electrophile, acetalde-

hyde, in the 2 enzymatic-reactions, . Chem. SOC.,

Perkin Trans.1,309-311.BORNEMANN,., CROUT,D. H. G., DA LTO N, .,

KRE N, ., LOBEL L, . et al. (1996), Stereospecificformation of R-aromatic acyloins by Zyrnomonas

mobilis pyruvate decarboxylase, J . Chem. Soc.,Perkin Trans.1,425-430.

177,3573-3578.



5 References 147

signals modulate the expression of an indole-3-acetic acid biosynthetic gene in Erwinia herbico-la,Mol. Plant Microbe Interact. 10,499-505.

BRANDL, .T., LINDOW,. E. (1998), Contribution ofindole-3-acetic acid production to the epiphyticfitness of Erwinia herbicola, Appl . Environ. Mi-crobiol. 64,32563263.

BRESSLER,., BRUAN,. (1996), Use of gel entrap-ment techniques for the resolution of itaconic ac-id acid biosynthetic pathway in Aspergillus terre-us,J.Sol Gel S ci. Technol.7,129-133.

BRIGANTI, ., MANGANI,., SCOZZAFAVA,., VER-NAGLIONE, G., SUPURAN,.T. (1999), Carbonicanhydrase catalyzes cyanamide hydration tourea: is it mimicking the physiological reaction, J.

Biol. Inorg. Chem .4,52%536.BROEK,A. V., LAMBRECHT,., EGGERMONT,,,

VANDERLEYDEN,. (1999), Auxins upregulate ex-pression of the indole-3-pyruvate decarboxylasegene in Azospirillum brasilense,J . Bacteriol. 181,1338-1342.

BRUHLMANN,. (1995), Purification and character-ization of an extracellular pectate lyase from anAmycolata sp., Appl. Environ. Microbiol. 61,3580-3585.

BRYUN,.M., JENNESS,. (1981), StereospecificityofL-myo-inositol-1 phosphate synthase for nicotin-amide adenine dinucleotide, Biochemistry 20,5174-5177.

BUNCH, . (1998), Nitriles, in:Biotechnology Vol. 8a,Biotransformations (REHM, H.-J., REED, G.,

PUHLER, ., STADLER,. J. W., Eds.), pp. 277-326.Weinheim: WILEY-VCH.

BUSER,H. P., PUGIN,B., SPINDLER,., SUITER,M.(1991), livo enantioselective syntheses of a pre-cursor of the biologically most active isomer ofCGA 8oooO (Clozylacon), Tetrahedron 47,5709-5716.

CALIK,G., VURAL,H., CALIK, ., OZDAMAR,. H.(1997), Bioprocess parameters and oxygen trans-fer effects in the growth of Pseudomonas dacun-hue for L-alanine production, Chem. Eng. J. 65,

CALIK,G., SAVASCI,., CALIK, ., OZDAMAR,. H.(1999), Growth and -carrageenan immobilizationof Pseudomonas dacunhae for L-alanine produc-tion, Enz yme Microb. Technol.24,67-74.

CANDY, . M., DUGGLEBY,. G. (1998), Structure andproperties of pyruvate decarboxylase and site-di-rected mutagenesis of the Zymomoizas mobilisenzyme,Biochim. Biophys. Acta. 1385,323-338.

CARDILLO,., SERVI, .,TINTI, . (1991), Biotrans-formation of unsaturated aldehydes by microor-ganisms with pyruvate decarboxylase activity,Appl . Microbiol. Biotechnol. 36,300-303.

CARROLL,. M., Ross, R. P., KELLY, . M., PRICE,N.C., SHEEHAN,., COGAN,T.M. (1999), Character-ization of recombinant acetolactate synthase

109-116.

BOTRE, . MAZZEI, . (1999), Interactions betweencarbonic anhydrase and some decarboxylatingenzymes as studied by a new bioelectrochemicalapproach, Bioelectrochem. Bioenerget. 48, 463-467.

BOTTING,. P., AKHTAR, ., COHEN,M. A., GANI,D.(1987),Mechanism of the elimination of ammoniafrom 3-substituted aspartic acids by 3-methylaspar-tase,J. Chem.Soc., Chem. Commun. 1371-1373.

BOTTING,. P., AKHTAR, ., COHEN,M. A., GANI,D.(1988a), Substrate specificity of the 3-methyl-aspartate ammonia-lyase reaction - Observationof differential relative reaction rates for substrateproduct pairs,Biochemistry 27,2953-2955.

BOTTING, . P., COHEN,M. A., AKHTAR, ., GANI,D.(1988b), Primary deuterium isotope effects forthe 3-methylaspartase catalyzed deamination of(2s)-aspartic acid, (2S,3S)-3methylaspartic acidand (2S,3S)-3-ethylasparticacid,Biochemistry 27,29562959.

BOTTING,. l?, JACKSON,A.A., ANI,D.1989), 15N-isotope and double isotope fractionation studiesof the mechanism of 3-methylaspartase-concert-ed elimination of ammonia from (2S,3S)-3-meth-ylaspartic acid, J . Chem. Soc., Chern. Comm un.

BOTTING,. P.,GANI,D. (1992), Mechanism of C-3hydrogen-exchange and the elimination of am-monia in the 3-methylaspartate ammonia-lyasereaction, Biochemistry 31,1509-1520.

BOTTING,. P., AKHTAR, ., ARCHER, . H., COHEN,M. A., THOMAS,. R. et al. (1992), in: MolecularRecognition: Chemical and Biochemical Prob-lems I I (ROBERTS,. M., Ed.) Special PublicationNo. 111.Cambridge: RSC.

BOUMERDASSI,., DESMAZEAUD,., MONNET, .,BOQUIEN, . Y., CORRIEU, . (1996), Improve-ment of diacetyl production by Lactococcus lactisssp lactis CNRZ 483 through oxygen control, J .

Dairy Sci. 79,775-781.BOURGEN,., CHRETIEN,., POGGI-BACH,., Doo-

N A N , s.,RABIER, . et a]. (1994), Mutation of thefumarase gene in two siblings with progressiveencephalopathy and fumarase deficiency, . Clin.Invest.93,2514-2518.

BOYNTON,. L., BENNETT,G. N., RUDOLPH,. B.(1996), Cloning sequencing and expression ofclustered genes encoding P-hydroxybutyryl-co-enzyme A (CoA) dehydrogenase, crotonase andbutyryl-CoA dehydrogenase from Clostridiumacetobutylicum ATCC 824, J . Bacteriol. 178,

BRANDL,M. T., LINDOW,. E. (1996), Cloning andcharacterization of a locus encoding an indolepy-ruvate decarboxylase involved in indole-3-aceticacid synthesis in Erwinia herbicola, Ap pl. Envi-ron. Microbiol. 62,4121-4128.

BRANDL, .T., LINDOW,. E. (1997), Environmental

1583-1585.

3015-3024.



148 2 Lyases

from Leuconostoc lactis, Enzyme Microb. Tech-nol. 25,6147.

CASTILLO,. M., MIZUGUCHI,., DHANARAJ,~.,L-

BERT,A., LUNDELL,T.., MURZIN,A. . (1999) A

six stranded double psi pb ar re l is shared by sev-

eral protein superfamilies,Structure with Folding& Design 7,227-236.

CHANG, . C., LAGHAI, ., O’LEARY,M. H., FLOSS,

H. G. (1982), Some stereochemical features ofaspartate P-decarboxylase, J. Biol. Chem. 257,

3564-3569.

CHANG, .-W. T., CHEN,X. H., LIU, H.-W. (1998),

CDP-6-deoxy-6,6-difluoro-~-glucose:A mecha-

nism based inhibitor for CDP-D-glucose 4,6-de-

hydratase,J. Am . Chem. SOC.120,9698-9699.

CHAITERJEE, . (1998), Lysine production by Brev-ebacterium linenes and its mutants: Activities and

regulation of enzymes of the lysine biosynthetic

pathway,Folia Microbiol. 43,141-146.

CHEN,. S. (1995), Alcohol dehydrogenase -multi-

plicity and relatedness in the solvent producing

Clostridia, FEMS M icrobiol. Rev. 17,263-273.

CHEN,C. K., BLASCHEK,. P. (1999a), Effect of ace-

tate on molecular and physiological aspects of

Clostridium beijerinckii NCIMB 8052, solvent

production and strain degeneracy,Ap pl. Environ.Microbiol. 65,499-505.

CHEN,C. K., BLASCHEK,. P. (1999b), Examination

of the physiological and molecular factors in-

volved in the enhanced solvent production by

Clostridium beijerinckii BAlO1, Appl. Environ.Microbiol. 65,2269-2271.

CHEN,H., PHILLIPS, . S. (1993), Binding of phenol

and analogs to alanine complexes of tyrosine phe-

nol-lyase from Citrobacter freundii: Implications

for the mechanism of a$-elimination and alanine

racemization, Biochemistry 32,11591-11599.

CHEN,X. M. H., PLOUX,O., LIU, H.-W. (1996) Bio-

synthesis of 3,6-dideoxyhexoses:Zn vivo and in vi-tro evidence for protein-protein interaction

between CDP-6-deoxy-~-threo-~-glycero-4-hexu-lose 3-dehydratase (E(1)) and its reductase

(E(3)), Biochemistry 35,16412-16420.

CHIBATA,., TOSA,., SATO, ., TANAKA,. (1987a),

Immobilization of cells in carragenan, MethodsEnzymol . 135,189-198.

CHIBATA,., TOSA,T., YAMAMOTO,., TANAKA,.

(1987b), Production of L-malic acid by immobi-

lized microbial cells,Methods Enzym ol. 136,455-

463.

CHIBATA,,TosA,T., SHIBATANI,T.1995),The indus-

trial production of optically active compounds by

immobilized biocatalysis, in: Chirality in Zndustry(COLLINS, . N., SHELDRAKE,. N., CROSBY.,Eds.), pp. 352-370. Chichester: John Wiley &Sons.

CHIPMAN,., BARAK, ., SCHLOSS,. V. (1998), Bio-

synthesisof 2-aceto-2-hydroxy-acids:cetolactate

synthases and acetohydroxyacid synthases, Bio-chim. Biophys. Ac ta. 1385,401-419.

CHOHNAN,.,KURUSU,Y., ISHIHARA,., TAKAMU-

RA, Y. (1999), Cloning and characterization of

mdc gnes encoding malonate decarboxylase of

Pseudomonus putida, FEMS Microbiol. Lett. 174,

CHOW, . S., SHIA,H. S., ADESINA, . A., ROGERS,?

L. (1995),A kinetic model for the deactivation of

pyruvate decarboxylase (pdc) by benzaldehyde,

Biotechnol. Lett. 17,1201-1206.

CLIFFORD, . H., CORNFORTH,.W., DONNINGER,.,

MALLABY, . (1972), Stereochemical course of

the decarboxylation of S-malate on malic en-

zyme,Eur.J. Biochem. 26,401-406.

COHEN-KUPEIC,., KUPIEC,M., SANDBECK,.,

LEIGH, J. A. (1999), Functional conservation

between the argininosuccinate lyase of the ar-

chaeon Methanococcus maripaludis and the cor-

responding bacterial and eukaryal genes, FEMSMicrob iol. Lett. 173,231-238.

COLOMBO,., GRISA,M., TORTORA,., VANONI,M.

(1994), Molecular cloning, nucleotide sequence

and expression of a Sulfolobus solfataricus gene

encoding a class I1 fumarase, FEBS Lett. 337,

93-98 (correction 1994,340,151-153).

CONSEVAGE,. W., PHILLIPS,. T. (1985), Presence

and quantity of dehydroalanine in histidine am-

monia-lyase from Pseudomonas pu tida, Biochem-istry 24,301-308.

CONSEVAGE,. W., PHILLIPS,. T. (1990), Sequence

analysis of the huth-gene encoding histidine am-

monia-lyase in Pseudomonas putida, J. Bacteriol.172,2224-2229.

CORNFORTH,. W., REDMOND,. W., EGGERER, .,

BUCKEL,W., GUTSCHOW,. (1969), Asymmetric

methyl groups and the mechanism of malate syn-thase, Nature 221,1212-1213.

CORNICK,. A., ALLISON, . J. (1996), Assimilation

of oxalate, acetate and C 02 by Oxalobacter for -migenes, Can J. Microbiol. 42,1081-1086.

CORNICK,. A., YAN,B., BANK, ., ALLISON, . J.

(1996), Biosynthesis of amino acids by Oxalobact-er formigenes. Analysis using C-13-NMR, Can J.

Biochem. 42,1219-1224.

COTON,E., ROLLAN,G. C., LONVAUD-FUNEL,.

(1998), Histidine decarboxylase of Leuconostocoenos 9204: Purification, kinetic properties, clon-

ing and nucleotide sequence of the hd c gene, J.

Ap pl. Microbiol. 84,143-151.

COUGHLIN,. M., CHRISTENSEN,., KUNZ, ? L.,KRISHNAMOORTHY,. S., WALKER,^ et al. (1998),

Molecular analysis and prenatal diagnosis of hu-

man fumarase deficiency,Mol. Genet. Metab. 63,

254-262.CROUT,D. H. G., HEDGECOCK,. J. R. (1979), The

base catalyzed rearrangement of a-acetolactate

(2-hydroxy-2-methyl-3-oxobutanoate): novel

311-319.



5 References 149

carbox ylate ion migra tion in a tertiary ketol rear-rangementJ Chem.SOC.,Perkin. Trans11982-1989.

CROUT, . H. G., RATHBONE,. L. (1988), Biotrans-formations w ith acetolactate decarboxylase: un-usual conversions of both substrate enatiomersinto products of high optical purity,J. Chem. SOC.,

Chem. Commun. 98-99.CROUT, . H. G., LITIZECHILD,., MITCHELL,. B.,

MORREY,. M. (1984) Stereochemistry of th e de-carboxylation of a-acetolactate (2-hydroxy-2-methyl-3-oxobutanoate) by the acetolactate de-carboxylase of Klebsiella aerogenes,J. Chem .SOC.,

Perkin . Trans.12271-2276.CROUT,D. H. G., LITTLECHILD,., MORREY,. M.

(1986) Acetoin metabolism: stereochemistry ofthe acetoin produced by the pyruvate decarboxy-lase of wheat germ and by the acetolactate decar-boxylase of Klebsiella aerogenes, J . Chem. SOC.,Perkin. Trans.l,105-108.

CROUT,. H.G., LEE,E. R., RATHBONE,. L. (199O),Absolute-con figuration of the product of the ace-tolactate synthase reaction by a novel method ofanalysis using acetolactate decarboxylase, J.

Chem. SOC.,Perkin Trans.11367-1369.CROUT,. H. G., MCINTYRE,. R., ALCOCK,. W.

(1991a), Stereoelectronic control of the tertiaryketol rearrangement - mplications for the mech-anism of th e reac tion catalyzed by the en zym es ofbranched-chain amino-acid-metabolism, reduc-toisomerase and acetolactate decarboxylase, J.

Chem. SOC.,Perkin Trans. 2,53-62.CROUT, . H. G., DALTON, ., HUTCHINSON,. W.,

MIYAGOSHI, . (1991b), Studies on pyruvate de-carboxylase - acyloin formation from aliphatic,aromatic and heterocyclic aldehydes, J. Chem.SOC.,Perkin Trans. 1,1329-1334.

CROUT, . H. G, DAVIES,., HEATH,. J., MILES, .O., RATTIBONE,. R.et al. (1994),Applications of

hydrolytic and decarboxylating enzymes in bio-transformations,Biocatalysis 9,l-30.

DAHMS,. S., DONALD,. (1982), D-xylo-Aldona tedehydratase, Methods Enzy mol. 90,302-305.

DAHMS,. S., SIBLEY,., HUISMAN,., DONALD,.(1982), D-Galactonate dehydratase, Methods En-zymol. 90,294-298.

D’CUNHA, . B., SATYANARAYAN,., NAIR, . M.(1994), Novel direct synthesis of L-phenylalan inemethyl ester by using Rhodotorula glutinis phen-ylalanine ammonia lyase in an organic-aqueousbiphasic system, Enzyme Microb. Technol. 16 ,318-322.

D’CUNHA, . B., SATYANARAYAN,., NAIR, . M.(1996a), Stabilization of phenylalanine ammonialyase containing Rhodotorula glutinis cells for th econtinuous synthesis of L-phenylalanine methylester, Enzym e M icrob. Technol.19,421427.

D’CWHA,G. B., SATYANARAYAN,., NAIR,P. M.(1996b), Purification of phenylalanine ammonia

lyase from Rhodotorula glutinis, Phytochernistry

DE BOER, . s.,MARSHALL, ., BROADMEADOW,.,HAZELDEN,. (1993),Toxicological evaluation ofacetolactate decarboxylase,J. Food. Protect. 56,510-517.

DEHNING,., SCHINK,. (1989),Malonomonas rubragen. nov. sp. nov., a microaerotoleran t bacteriumgrowing by decarboxylation of malonate, Arch.Microbiol, 151,427433.

DEHNING,., SCHINK,. (1994), Ana erobic degrada-tion of ma laonate via maloyl-C oA by Sporomusama bn ica, Klebsiella oxytoca and Rhododbactercapsulatus, A . Van Leeuwenhoek Int.f. Gen. Mol.Microbiol. 66,343-350.

DELGADO,., TREJO,. A., HUITRON,. AGUILAR,G. (1993), Pectin lyase from Aspergillus sp. CH-Y-1043.Ap pl. Microbiol. Biotechnol. 39,515-519.

DENG, . D., COLEMAN,. R. (1999), Ethano l synthe-sis by gene tic engineering in cyanobac teria,Appl .Environ. M icrobiol. 65,523-528.

DE VIVO1993), The expan ding clinical spectrum ofmitochondria1diseases,Brain Dev. l5,1-22.

DEWICK,.M. (1998),Th e biosynthesis of shikim atemetabolites,Nut. Prod. R ep. 15,17-58.

DI GANGI,. J., SEYZADEH,., DAVIS, R . H. (1987),Ornithine decarboxylase from Neurospora cras-sa , purification, characterization and regulationby inactivation,J. Biol. Chem. 262,7889-7893.

DIMROTH,., 1994) Bacterial sodium ion coupledenergetics A . Van Leeuwenhoek Znt. J . Gen. Mol.Microbiol. 65,381-395.

DIMROTH,., HILBI, . (1997), Enzymatic and genet-ic basis for bacterial growth on malonate, Mol.Microbiol.25,3-10.

DIMROTH,., SCHINK,. (1998), Energ y con servationin the decarboxylation of dicarboxylic acids byfermenting bacteria, Arch. M icrobiol. 170,69-77.

DIMROTH,., THOMER,A.1993), O n the mechanismof sodium ion translocation by oxaloacetate de-carboxylase of Klebsiella pneumoniae, Biochem-istry 32,1734-1739.

DIRMAIER,. J., GARCIA, . A., KOZARICH,. W.,KENYON,G. L., (1986), Inhibition of benzy lfor-mate decarboxylase by [p-(bromomethy1)ben-zoyllformate - enzyme catalyzed mo dification ofthiamine pyrophosphate by halide eliminationand tautomerizationJ Am . Chem.SOC.108,3149-3150.

DOBRITZSCH,.,KONIG, ., SCHNEIDER,., Lu, G. G.(1998), High resolution crystal structure of pyru-vate decarboxylase from Zymomonas mobilis -implications for substrate activation in pyruvatedecarboxylases, . Biol. Chem. 273,20196-20204.

DOELLE,. W., KIRK,L., CRIITENDEN,., TOH,H.,DOELLE, .B. (1993), Zymo monas mobilis - sci-ence and industrial application, Crit Rev. Biotech-nol. 13,57-98.

42,17-20.



150 2 Lyases

DOMINICI,., MOORE,P. S., VOLTAITORNI,. B.(1993), Modified purification of L -aromatic ami-no acid decarboxylase from pig kidney, ProteinExpr. Purq . 4,345-347.

DOYON,. B., JAIN,A. (1999), Th e pattern of fluorinesubstitution affects binding affinity in a small li-brary of fluoroaromatic inhibitors for carbonicanhydrase, Org. Lett. 1,183-185.

DRAUZ, ., WALDMANN,. (1995), Enzyme Cataly-sis in O rganic Synthesis, A Comprehensive H and-book ,pp. 4814 83,10 50. Weinheim: WILEY-VCH.

DRAKE,. F.,SILIGARDI, ., CROUT,. H. G., RATH-

BONE, D. L. (1987),Applications of vibrational in-frared circular dichoism to biological problems:Stereochemistry of the proton exchange in ace-toin (3-hydroxybutan-2-one) catalyzed by acetol-actate dehydrogenase,J. Chem. SOC.,Chem. Com-mun. 1834-1835.

DREYER,. L. (1985), Isolation and biochemicalcharacterization of maleic-acid dehydratase, aniron containing hydro-lyase, Eur. J . Biochem. 150,

DREYER,. L. (1987), Th e role of iron in t he activa-tion of mannonic and altronic acid hydratases,two Fe-requiring hydro-lyases, Eur. J . Biochem.166,623-630.

DUCROCQ,., FRAISSE,., TABET,. C.,AZERAD,.(1984), An enzymatic method for the determina-tion of enantiomeric composition and absoluteconfiguration of deuterated or tritiated succinicacid, Anal. Biochem. 141,418-422.

DUERRE,. 1998), New insights and novel dev elop-ments in Clostridial acetonelbutanollisopropanolfermentation, Appl. Microbiol. Biotechnol. 49,639-648.

DUERRE,., BAHL, ., GOITSCHALK,. (1992), Ace-tone-butanol fermentation - Basis of a modernbiotechnological process, Chem. Ing. Tech. 64,491498 .

DUERRE,., FISCHER,. J., KUHN,A., LORENZ,.,SCHREIBER,. et al. (1995), Solventogenic genesof Clostridium acetobutylicum - atalytic prop er-ties, genetic organization a nd transcriptional reg-ulation, FEMS M icrobiol. Rev. 17,251-262.

DUFFY,. I. (1980), Chemicals by enzym atic and mi-crobial processes, Chemical Technol. Rev. 161,

DUSCH,., PUEHLER,., KALINOWSKI,. (1999), Ex -pression of the Corynebacterium glutamicumpanD gene encoding L-aspartate-a-decarboxy-lase leads to panto thenate overproduction in Es-cherichia coli, Ap pl. Environ. Microbiol. 65,153&1539.

DYDA, F., FUREY,W., SWAMINATHAN,., SAX,M.,FARRENKOPF,., JORDAN, F. (1990), Preliminarycrystallographic dat a for th e thiamin diphosphate-dependent enzyme pyruvate decarboxylase frombrewers’ yeast,J. Biol. Chem. 265,17413-17415.

145-154.

107-109.

DYDA, F., FUREY,W., SWAMINATHAN,., SAX,M.,FARRENKOPF,., JORDAN,F. (1993), Catalytic cen-ters in the thiamin diphosphate dependent en-zyme, pyruvate decarboxylase at 2.4-angstromresolution, Biochemistry 32,61656170.

EARNHARDT,. N., Tu C., SILVERMAN,. N. (1999),Intermo lecular proto n transfer in catalysis by car-bonic anhydrase, Can.J . Chem.77,726732.

EASTON, . J., HUITON,C. A. (1994), Synthesis ofeach stereoisomer of [3-ZH,]phenylalanine andevaluation of the stereochem ical course of the re-action of (R)-phenylalanine with (S)-phenylala-nine ammonia lyase, J . Chem. SOC.,Perkin Trans.

EASTON,. J., HARPER,. B., LINCOLN,. (1995), Useof cyclodextrins to limit product inhibition of (S)-phenylalanine amm onia lyase, J. Chem. SOC.,Per-kin Trans.1,2525-2526.

EDENS,W. A. A., URBAUER,. L., CLELAND,. W.

(1997) Determ ination of the chemical mechanismof malic enzyme by isotope effects, Biochemistry

EFFENBERGER,. (1999), Enzyme catalyzed prepa ra-tion and synthetic applications of optically activecyanohydrins, Chimica 53,3-10.

EHRLICH,. I. , HWANG, .-H., COOK,? F., BLAN-

C H A R D , J.s.(1999), Eviden ce fo r a stepwise mech-anism for OMP decarboxylase, J. Am . Chem. SOC.

121,6966-6967.EMANUELE,. J. FITZPATRICK,. F. (1995), Mechanis-

tic studies of the flavoprotein tryptophan 2-monooxygenase. 1.Kinetic mechanism, Biochem-istry 34,3710-3715.

ENGLARD,., HANSON,. R. (1969), Stereospecifi-cally labeled citric acid ccyle intermed iates, Meth-ods Enzymol . 13,567-601.

EREZ,T., GDALEVSKY,. Y., TORCHINSKY,. M.,PHILLIPS,. S.,PAROLA,A.. (1998), Cold inacti-vation a nd dissociation into dimers of Escherichiacoli tryptophanase and its W330F mutant form,Biochim. Biophys. Ac ta. 1384,365-372.

EVANS,. T., HANNA, ., PAYNE, ., CONRAD,.W.,MISAWA,M. (1987), Biotransformation of transcinnamic acid to L-phenylalanine: Optimizationof reaction co nditions using whole yeast cells,En-zym e Microb. Technol.9 , 4 1 7 4 2 1 .

FACCHINI,. J. (1998), Temp oral co rrelation of tyra-mine metabolism with alkaloids and amide bio-synthesis in elicited opium p oppy cultures, Phyto-chemistry 49,481-490.

FALEEV,. G., SPIRINA,. N., PERYSHKOVA,.E.,SAPOROVSKAYA,. B., TSYRYAPKIN,. A., BELI-KOV, V. M. (1995 ), T he synth esis of L-tyros ine andits aromatic ring substituted analogs by microbialcells containing tyrosine phenol-lyase, Appl . Bio-chem. Microbiol. 31,155-160.

FALEEV,. G.,SPIRINA,. N., IVOILOV,. S., DEMID-

KINA, T. V., PHILLIPS,. S. (1996), Th e catalytic

1,3545-3548.

36,1141-1147.



5 References 151

FLINT,D. H., EMPTAGE,M. H. (1988), Dihydroxy ac-

id dehydratase from spinach contains a [2Fe-2S]

cluster,J. Biol. Chem. 263,3558-3564.FLINT, D. H, MCKAY, . G. (1994) Escherichia coli fu-

marase A catalyzed transfer of 0-18 from C-2and H-2 from C-3 of malate to acetylene-dicar-

boxylate to form 0-18-labeled and H-2-labeled

oxalacetate,J.Am. Chem. SOC.16,5534-5539.

FLINT, . H., NUDELMAN,., CALABRESE,. C.,

G O ~ I E B ,. E. (1992a), Enol oxalacetic acid ex-

ists in the Z-form in the crystalline state and in so-

lution, J . Org. Chem. 57,7270-7274.

FLINT, . H., EMFTAGE, M. H., GUEST,. R. (1992b),

Fumarase A from Escherichia coli: Purification

and characterization as an iron-sulfur cluster con-

taining enzyme, Biochemistry 31,10331-10337.FLINT, D. H., TUMINELLO,. F., EMPTAGE,M. H.

(1993),The inactivation of Fe-S cluster containing

hydro-lyases by superoxide, J . Biol. Chem. 268,

FLINT,D. H., TUMINELLO,. F., MILLER, . J. (1996),

Studies on the synthesis of the Fe-S cluster of di-

hydroxy-acid dehydratase in Escherichia coli

crude extract - solation of 0-acetylserine sulfhy-

drylases A and B and P-cystathione based on

their ability to mobilize sulfur from cysteine and

to participate in Fe-S cluster synthesis, J . Biol.

Chem. 271,16053-16067.

FLOROVA,., DENOYA,. D., MORGENSTERN,. R.,SKINNER,. D., REYNOLDS,. A. (1998), Cloning,

expression, and characterization of a Type I1 3-

dehydroquinate dehydratase gene from Strepto-

myces hygroscopicus, Arch. Biochem. Biophys.

350,298-306.

FLOSS,H. G., TSAI,M. D., WOODARD, . W. (1984),

Stereochemistry of biological reactions at pro-

prochiral centres, Topics Stereochem. 15,253-321.

FOOR, ., MORIN, ., BOSTIAN,. A. (1993), Produc-

tion of L-dihydroxyphenylalanine n Escherichia

coli with the tyrosine phenol-lyase gene clonedfrom Erwinia herbicola, Appl. Environ. Microbi-

01.59,3070-3075.

FRIEDEMANN,., NEEF,H. (1998), Theoretical stud-

ies on the electronic and energetic properties of

the aminopyrimidine part of thiamin diphos-

phate, Biochim. Biophys. Acta. 1385,245-250.

Fu, D.X., MALONEY,.C. (1998), Structure-function

relationships in OxlT, the oxalate-formate trans-

porter of Oxalobacter formigenes - topological

features of transmembrane helix 11as visualized

by site directed fluorescent labeling, J. Biol.

Chem. 273,17962-17967.FUKUYAMA,., MATIOSHI, ., IWASAKI, ., TAKI-

ZAWA, ., MNAZAKI,M. et al. (1999), Preparative

scale synthesis of (R)-a-fluorophenylaceticacid,

Biosci. Biotech. Biochem. 63,1664-1666.

FUREY,W., ARJUNAN,., CHEN, ., SAX,M., Guo, F.,

JORDAN, . (1998), Structure-function relation-

22369-22376.

mechanism of tyrosine phenol-lyase from Erwi-

nia herbicola: The effect of substrate structure on

pH-dependence of kinetic parameters in the re-

actions with ring-substituted tyrosines, Zt. Natur-forsch. C 51,363-370.

FALZONE,. J., KARSTEN, . E., CONLEY,. D., VIO-

LA , R.E. (1988). L-Aspartase from Escherichiaco-li: Substrate specificity and role of divalent metal

ions, Biochemistry 27,9089-9093.

FARRAR,. C., GUTOWSKY,. S., ALBERTY,R. A.,

MILLER,W. G. (1957),The mechanism of the ster-

eospecific enzymatic hydration of fumarate to L-

malate,J. Am. Chem. SOC.9,3978-3980.

FARRENKOPF,. C., JORDAN,. (1992), Resolution of

brewers' yeast pyruvate decarboxylase into mul-

tiple isoforms with similar subunit structure and

activity using high-performance liquid-chroma-

tography, Protein Expr. P u r q 3,101-107.

FELL,L. M., FRANCIS,. T., HOLMES,. L., TERLOUW,

J. K. (1997), The intriguing behavior of (ionized)

oxalacetic acid investigated by tandem mass spec-

trometry, Znt.J . Mass Spectrom. 165,179-194.

FESSNER,. D. (1998), Enzyme mediated C-C bond

formation, Curr. Opin. Chem. Biol. 2,85-97.

FESSNER, .-D., WALTER, . (1997) Enzymatic C-C

bond formation in asymmetric synthesis. Topics

Curr. Chem. 184,97-194.

FETZNER,. (1998), Bacterial dehalogenation, Appl.

Microbiol. Biotechnol. 50,633457.

FETZNER,., LINGRENS,(1994), Bacteriol dehydro-

genases - biochemistry, genetics and biotechno-logical applications, Microbiol. Rev. 58,641485.

FINDEIS, . A., WHITESIDES,. M. (1987) Fumarase

catalyzed synthesis of L-threo-chloromalic acid

and its conversion to 2-deoxy-~-ribose nd D-

erythro-sphingosine, . Org. Chem. 52,2838-2848.

FISHER, . L.,FRIEDEN, ., MCKEE, . S. M., ALBER-

TY, R. A. (1955), Concerning the stereospecificity

of the fumarase reaction and the demonstration

of a new intermediate,J. Am. Chem. SOC.7,4436.FISHER, . J., HELMS,., DUERRE,. (1993), Cloning,

sequencing and molecular analysis of the sol op-

eron of Clostridium acetobutylicum,a chromoso-

mal locus involved in solventogenesiqJ. Bacteriol.

175,69594969.

FITZPATRICK,. R., HILL,A., TOLMIE,. L., THOR-

BURN,. R., CHRISTODOULOU,. (1999),The mo-

lecular basis of malonyl-CoA decarboxylase defi-

ciency, Am. J. Human Genet. 65,318-326.

FLINT,D. H. (1993), Escherichia coli fumarase A cat-

alyzes the isomerization of enol and keto oxala-

cetic acid, Biochemistry 32,799-805.FLINT,D. H. (1994), Initial kinetic and mechanistic

characterization of Escherichia coli fumarase A,

Arch. Biochem. Biophys. 311,509-516.FLINT, D. H., ALLEN,R. M. (1996), Iron-sulfur pro-

teins with non-redox functions, Chem. Rev. %,23 15-2334.



152 2 Lyases

ships and flexible tetramer assembly in py ruvatedecarboxylase revealed by analysis of crystalstructures, Biochim. Biophys. Acta. 1385, 253-270.

FURUKAWA,.,KOGA,J., ADACHI,.,KISHI,K., SYO-

NO , K. (1996), Efficient conversion of L-trypto-phan to indole-3-acetic acid an d o r tryptophol bysome species of Rhizocton ia, Plant Cell Physiol.

FURUTA,., TAKAHASHI,., KASUYA,.,(1990), Evi-dence for a carbanion intermediate in the elimi-nation of amm onia from L-histidine catalyzed byhistidine am monia-lyase,J . Am . Chem. SOC. 112,

FURUTA,T.,AKAHASHI,.,SHIBASAKI,., KASUYA,Y.(1992), Reversible stepwise mechanism involv-ing a carbanion intermediate in the elimination ofammonia from L-histidine catalyzed by histidineammonia-lyase, . Biol. Chem. 267,12600-12605.

GALLAGHER,., ROZWARSKI,. A., ERNST,. R.,

HACKETT,M. L. (1993), Refined structure of thepyruvoyl dependen t decarboxy lase from Lactob-acillus 30a,J. Mol. Biol. 230,516-528.

GANI,D. (1991),A structural and mechanistic com-parison of pyridox al5 ’-phosphate depen dent de-carboxylase and transaminase enzymes, Phil.Trans.Royal SOC.B . 332,131-139.

GANI,D., YOUNG, . W. (1983) Stereochemistry ofthe dihydouracil dehydrogenase reaction in themetabolism of uracil to palanine, J. Chem. Soc.,

Chem. Com mun.,576-578.GANI,D., RCHER,. H., B O ~ N G ,. P.,POLLARD,.

R. (1999), The 3-methylapartase reaction p robedusing H-2 and N-15 isotope effects for three sub-strates: A flip from a co ncerted to a carbocationicmechanism upon changing the C-3 stereochemis-try in the substrate from R to S, Bioorg. Med.Chem.7,977-990.

GAO, ., O H , D. H., BROADBENT,. R., JOHNSON,M.E., WEIMER,. C, STEELE,. L. (1997), Aromaticamino acid catabolism by Lactococci, Lait 77,

GAO,G., NARA, ., NAKAJIMA-SHIMADA,., AOKI, .

(1999a), Novel organization an d sequences of fivegenes encoding all six enzymes for de novo pyrim-idine biosynthesis in Trypanosoma cruzi, J . Mol.Biol. 285,149-161.

GAO, . M., WABER, ., BENNETT, . J., GIBSON,.M., COHEN,. C. (1999b), Cloning and mutationalanalysis of human malonyl-coenzyme A decar-boxylase,J. Lipid Rex 40,178-182.

GAWRON,.,FONDY,T.. (1959), Stereoch emistry ofthe fumarase and aspartase catalyzed reactionsand of the K rebs cycle from fum aric acid to d-iso-

citric acid,J. Am . Chem. SOC.81,63334334,GAWRON,.,GLAID, . J., FONDY,. P. (1961), Ster-

eochemistry of K rebs’ cycle hydra tions and relat-ed reactions,J.Am. Chem. SOC.83,3634-3640.

37,899-905.

3633-3636.

371-381.

GAWRONSKI,., GAWRONSKI1999), Tartaric andMalic Ac ids in Organic Synthesis:A Source Boo kOf uilding Blocks, Ligands, Auxilaries and Re-solving Agents. Chichester: John Wiley & Sons.

GELFMAN,. M., COPELAND,. C., ROBERTUS,. D.

(1991), Site directed a lteration of 4 active site res-idues of a pyruvoyl-dependent histidine decar-boxylase, Biochemistry 30,1057-1062.

GERISCHNER,., DUERRE,. (1990), Cloning, se-quencing, and molecular analysis of the acetoace-tate decarboxylase gene region from Clostridiumacetobutylicum, J . Bacteriol. 172,6907-6918.

GERMAN,A.1992), Use of a-acetolactate decarbox-ylase from a recombinant strain of Bacillus subti-lis for the production of beer and alcohol, Bull.Aca d. Natl. Med. 176,1175-1176.

GHOSH, ., SADHUKHAN,. C., CHAUDHURI,.,

GHOSH, . K., MANDAL,A. (1996), Volatilizationof mercury by immobilized mercury resistantbacterial cells,J . Ap pl. Bacteriol. 81,104-108.

GIORGIANNI,F.,ERANOVA,.,WESDEMIOTIS,.,VI-

OLA, R. E. (1995), Elimination of the sensitivityofL-aspartase to activate site directed inactivationwithout alteration of catalytic activity.Biochemis-try 34,3529-3535.

GIRBAL,., SAINT-AMANS,. V. S. , SOUCAILLE,.(1995), How neutral red modified carbon andelectron flow in Clostridium acetobutylicumgrown in chemostats at neutrel p H, FEMS M icro-biol. Lett. 16,151-162.

GIRBAL,., SOUCAILLE,. 1998), Regulation of sol-vent production in Clostridium acetobutylicum,

GISH,G., SMYTH,T., KLUGER,R. (1988), Thiamindiphosphate catalysis. Mechanistic divergence asa probe of subs trate activation of pyruvate decar-boxylase,J .Am . Chem. Soc. 110,623M234.

GLOGE, ., LANGER,., POPPE,., RETEY,. (1998),The behavior of substrate analogs and secondarydeuterium isotope effects in the phenylalanineammo nia-lyase reaction, Arch. Biochim. Biophys.

GODA,. K., MINTON,. P., BOITING,. P., GANI, .

(1992), Cloning, sequencing, and expression inEscherichia coli of the Clostridium tetanomor-phum gene encoding P-methylaspartase andcharacterization of the recombinant protein, Bio-chemistry 31,10747-10756.

GOODWIN,.W., TAEGTMEYER,., (1999), Regula-tion of fatty acid oxidation of the hea rt by MCD

and ACC during co ntractile stimulation, A m . J.Physiol. Endocrinol. Metabol. 277, E772-E777.

GONZALEZ,. H., ANDREO,. S. (1988), Carboxyla-tion and dephosphorylation of phosphoenol-3-fluoropyruvate by maize leaf phosphoenolpyru-vate carboxylase,Biochem. J. 253,217-222.

GOPALAN,. (1999), Kine tics of decarboxylation ofethylmalonic acid: A hundred fold reactivity of

TZBS 16,ll-16.

359,l-7.



5 References 153

mentation technology - Zymomonas mobilis,Curr. Sci. 77,5648.

GUITA, M. N. (1992), Enzyme function in organicsolvents. Eur.J. Biochem. 203,25-32.

GuITA,A., PHUNG,. T., CHAKRAVARTY,., SILVER,S. (1999), Mercury resistance in Bacillus cereusRC607: Transcriptional organization and two newopen reading frames,J.Bacteriol. 181.7080-7086.

GUTHRIE, . D., JENCKS, . P. (1989), IUPAC recom-mendations for the representation of reactionmechanisms,Acc . Chem . Res. 22,343-349.

GYAMERAH,. H. (1995), Oxygen requirement andenergy relations of itaconic acid fermentation byAspergillus terreus NRRL 1960,Ap pl. Microbiol.Biotechnol.44,20-26.

HALLIS,. M., LIU,H. W. (1998), Stereospecificity ofhydride transfer for the catalytically recycledNAD+ in CDP-D-glucose 4,6-dehydratase, Tetra-hedron 54,15975-15982.

HALLIS,T.M., LIU,H.-W. (1999a), Mechanistic stud-ies of the biosynthesis of tyvelose: Purificationand characterization of CDP-D-tyvelose 2-epime-rase,J.Am . Chem.SOC.121,676556766.

HALLIS,T.M., LIU,H.-W. (1999b), Learning nature’sstrategy for making deoxy sugars: Pathways,mechanisms and combinatorial applications,Acc.Chem. Res. 32,579-598.

HALLIS,. M., LEI,Y., QUE,N. L. S., LIU,H., (1998),Mechanistic studies of the biosynthesis of para-tose: purification and characterization of CDP-paratose synthase,Biochemistry 37,49354945.

HAMILTON,.A., WESTHEIMER,. H. (1959),On themechanism of the enzymatic decarboxylation ofacetoacetate,J. Am . Chem.SOC.81,6332-6333.

HANDA,. , Koo, J. H., KIM,Y.S., FLOSS,H. G. (1999),Stereochemical course of biotin-indepenent mal-onate decarboxylase catalysis,Arch. Biochem. Bi-ophys.370,93-96.

HANSON, . R., HAVIR,E. A. (1972), The enzymicelimination of ammonia, in: The Enzymes 3rdEdn. Vol. 7 (BOYER,. D., Ed.), pp. 75-166. NewYork Academic Press.

HANSON, . R., WIGHTMAN,. H., STAUNTON,.,BATERSBY,A. R. (1971), Stereochemical courseof the elimination catalyzed by L-phenylalanineammonia-lyase and the configuration of 2-ben-zamidocinnamic azlactone, J . Chem. Soc., Chem.Commun. ,185-186.

HARDEN, . (1901), The chemical action of Bacilluscoli communis and similar organisms on carbohy-drates and allied compounds. J. C h t m SOC. 9,611-628.

HASSON,M. S., MUSCATE, ., HENEHAN, . T. M.,GUIDINGER,. F., PETSKO,G. A. et al. (1995),Pur-ification and crystallization of benzoylformatedecarboxylase,Protein Sci. 4,955-959.

HASSON, . S.,MUSCATE,A., CLEISH,M. J.,POLOV-NIKOVA, L. s.,GERLT,. A. et al. (1998),The crystal

the acid molecule over its mono-anion, Asian J.Chem.11,1474-1482.

GOITSCHALK, ., BENDER,R. (1982), D-glUCOnatedehydratase from Clostridium pasteurianum,Methods Enzym ol. 90,283-287.

GRABER,R., KASPER, ., MALASHKEVICH,. N.,STROP, ., GEHRING, . et al. (1999), Conversionof aspartate aminotransferase into L-aspartatep-decarboxylase by a triple active site mutation, J.Biol. Chem. 274,31203-31208.

GRABOWSKI,., HOFMEISTER,A.. M., BUCKEL, .(1993), Bacterial L-serine dehydratases - a newfamily of enzymes containing iron-sulfur clusters,Biochem. Sci. 18,297-300.

GRIGG,R., HENDERSON,., HUDSON, . J. (1989)Decarboxylation of a-amino acids by pyruvic ac-id and its derivatives: evidence for azomethineylides in in vitro analogsof pyruvoyl enzymic pro-cesses, Tetrahedron L ett.30,2841-2844.

GRISHIN, . V., OSTERMAN,. L., BROOKS, . B.,PHILLIPS, . A., GOLDSMITH,. J. (1999), X-raystructure of ornithine decarboxylase from Trypa-nosoma brucei: The native structure and thestructure in complex with a-difluoromethylorni-thine, Biochemistry 38,15174-15184.

GRUER,M. J., ARTYMIUK,. J., GUEST, . R. (1997),The aconitase family: Three structural variationson a common theme,. Trends. Biochem. Sci. 22,3-6.

GRUPE,H., GOITSCHALK, . (1992), Physiologicalevents in Clostridium acetobutylicum during theshift from acidogenesis to solventogenesis in con-tinuous culture and presentation of a model forshift induction, Appl. Environ. Microbiol. 58,3896-3902.

GULICK, . M., PALMER, . R. J., B A B B I ~ ,. C.,GERLT,.A., RAYMENT,. (1998), Evolution of en-zymatic activities in the enolase superfamily:Crystal structure of D-glucarate dehydratase fromPseudomonas putida, Biochemistry 37, 14358-14368.

GULZAR, . S ., AKHTAR,M., GANI,D. (1994), Enan-tiospecific conjugate addition of N-nucleophilesto substituted fumaric acids using methylaspar-tase, J . Chem. Soc., Chem. Commun. 1994,1601-1602.

GULZAR, . S.,MORRIS, . B., GANI,D.1995), Con-trol of the regioselectivity of N-nucleophile addi-tion to N - carbonyl protected dehydroalanines- amodel for the ammonialyase enzymes, J . Chem.Soc., Chem. Commu n. 1995,1061-1062.

GULZAR, . S., AKHTAR, ., GANI,D. (1997), Prep-aration of N-substituted aspartic acids via enan-tiospecific conjugate addition of N-nucleophilesto fumaric acids using methylaspartase: Syntheticutility and mechanistic implications, J. Chem.Soc., Perkin Trans.l.,649-655.

GUNASEKARAN,., RAJ,K. C. (1999), Ethanol fer-



154 2 Lyases

structure of benzoylformate decarboxylase at 1.6

angstrom resolution: Diversity of catalytic resi-

dues in thiamin diphosphate-dependent enzymes,

Biochemistry 37,9918-9930.HARRIS,. K., WASHABAUGH,. W. (1995a), Distri-

bution of the thiamin diphosphate C-2-proton

during catalysis of acetaldehyde formation bybrewers-yeast pyruvate decarboxylase,Biochem-istry 34,13994-14000.

HARRIS,. K., WASHABAUGH,. W. (1995b), Sol-vent-derived protons in catalysis by brewers-

yeast pyruvate decarboxylase, Biochemistry 34 ,

HASSETT,D. J., HOWELL,M. L., SOKOL,. A., VASIL,M. L., DEAN, . E. (1997), Fumarase C activity is

elevated in response to iron deprivation and in

mucoid, alginate producing Pseudomonas aerugi-nosa: Cloning and characterization of fumC andpurification of native FumC, J . Bacteriol. 179,

HAYASHI,. (1995), F'yndoxal enzymes-mechanisticdiversity and uniformityJ Biochem. 118,463-473.

HAYASHI,. , KAMEJI,T.1983), Ornithine decarboxy-lase (rat-liver),Methods Enzymol. 94,154-158.

HE,X. M, THORSON,. S., LIU,H.W. (1996), Probingthe coenzyme and substrate binding events of

CDP-D-glucose 4,6-dehydratase: Mechanistic im-plications,Biochemistry 35,47214731.

HEATON,A.. P., RUGH, .L.,WANG,N. J., MEAGHER,R. B. (1998), Phytoremediation of mercury and

methymercury polluted soils using genetically en-gineeered plants, J . Soil Contamin.7,497-509.

HEBEL, ., FURLANO,D..,PHILLIFS,R. S.,KOUSHIK,S.,CREVELING,. R., KIRK,K. L. (1992),An enzy-matic-synthesis of 2-azido-~-tyrosine, Bioorg.Med. Chem. Lett. 2 , 4 1 4 .

HERBERT,. B., KNAGGS, . R. (1992), Biosynthesisof the antibiotic obafluorin from para-amino-

phenylalanine and glycine (glyoxylate),J . Chem.SOC.,Perkin Trans. I , 109-113.

HERNANDEZ,., PHILLIPS,. T. (1994), Ser-143 is an

essential active-site residue in histidine ammonia-lyase of Pseudomonas putida. Biochem. Biophys.

Res. Comm un. 201,1433-1438.HERNANDEZ,., STROH,. G., PHILLIPS,. T. (1993),Identification of Ser'43 as the site of modificationin the active-site of histidine ammonia-lyase.Arch. Biochem.. Bioph ys.307,126-132.

HIGHBARGER,. A., GERLT,. A., KENYON,G. L.(1996), Mechanism of the reaction catalyzed by

acetoacetate decarboxylase. Importance of lysine116 in determining the pK, of active site lysine115,Biochemistry 35,4146.

HILL,C. M., PANG, . S., DUGGLEBY,. G. (1997),Purification of Escherichia coli acetohydroxyacidsynthase isoenzyme I1 and reconstitution of ac-tive enzyme from its individual pure subunits,Bi-ochem.J . 327,891-898.

14001-14011.

1442-1451.

HIRAHARA,., HORINOUCHI,., BEPPU,T. (1993),Cloning, nucleotide-sequence, and overexpres-

sion in Escherichia coli of the P-tyrosinase gene

from an obligately symbiotic hermophile,Symbi-obacterium thermophilum, Appl. Microbiol. Bio-technol. 39,341-346.

HIRSCH, ., SCHAGGER,., FUCHS, . (1998), Phenylglyoxylate :NAD+ oxidoreducatse (CoA benzoy-lating), a new enzyme of anaerobic phenylalaninemetabolism in the denitrifying bacterium, Azoar-cus evansii, Eur.J. Biochem. 251,907-915.

HIRSCHBEIN,. L., MAZENOD,. P., WHITESIDES,.M. (1982), Synthesis of phosphoenolpyruvate andits use in adenosine triphosphate cofactor regen-

eration,J . Org. Chem. 41,3765-3766.

HOENKE,., DIMROTH,. 1999), Formation of cata-lytically active S-acetyl(ma1onatedecarboxylase)requires malonyl-coenzyme Aacyl carrier pro-

tein-SH transacylase as auxilary enzyme, Eur J .

Biochem. 259, 181-187 (correction 1999, 259,961).

HOENKE, ., SCHMID, ., DIMROTH,. (1997), Se-quence of a gene cluster from Klebsiella pneu-moniae encoding malonate decarboxylase and

expression of the enzyme in Escherichia coli, Eur.J . Biochem. 246,530-538.

HOFMEISTER,. E. F., ALBRACHT,.P. J., BUCKEL,,W. (1994) Iron sulfur containing L-serine dehy-

dratase from Peptostreptococcus asaccharolyti-cus:correlation of the cluster type with enzymaticactivity,FEBS Lett. 351,416-418.

HOHENESTER, ., KELLER,. W., JANSONIUS,. N.(1994),An alkali metal ion size dependent switchin the active site structure of dialkylglycine decar-boxylase, Biochemistry 33,13561-13570.

HOHMANN,. (1991a), PDC6, a weakly expressed py-

ruvate decarboxylase from yeast, is activated

when fused spontaneously under the control ofthe P D C l promoter, Curr. Genet. 1991,20,373-

378.HOHMANN,. (1991b), Characterization of P D C 6 , a

3rd structural gene for pyruvate decarboxylase in

Saccharomyces cerevisiae, J . Bacteriol. 173,7963-

7969.HOHMANN,. (1993), Characterization of PDC2, agene necessary for high-level expression of pyru-vate decarboxylase structural genes in saccharo-myces-cerevisiae, Mol. Gen. Genet. 241, 657-666.

HUANG, .,BEALE,. M., KELLER, J., FLOSS, . G.

(1986), Synthesis of (R)-[1-13C,,2-2Hl)]malonateand (S)-[1-'3C,,2-2H,)]malonatend its stereo-chemical analysis by NMR spectroscopy,J . Am.

Chem. SOC .108,1100-1101.HUANG,. L., WU,C. C., LIEN,M. H. (1997),Ab initio

studies of decarboxylations of P-keto carboxylicacids, XCOCH2COOH(X = H, OH, and CH3), .

Phys. Chem. A 101,7867-7873.



5 References 155

SON,G.G.,DEMENTIEVA,.S. et al. (1998), Crystalstructure of tryptophanase, J . Mol. Biol. 276,603-623.

IWAMOTO,., TANIKI,H., KOISHI, ., NAKURA,.(1995), D-Glucosamine aldolase activity of D-glu-cosaminate dehydratase from Pseudomonas fluo-

rescens and its requirement for MnZ+,Biosci. Bi-otech. Bioche m. 59,408-411.

JAKLITSCH, . M., KUBICEK,. P., SCRUTTON, . C.(1991), The subcellular organization of itaconatebiosynthesis in Aspergillus terreus,J. Gen. Micro-biol.W7,533-539.

JANSONIUS,. N. (1998), Structure, evolution and ac-tion of vitamin B,-dependent enzymes, Curr.Opin. Struct. Biol. 8,759-769.

JANSSEN,. H., HARFOOT,C. G. (1992), Anaerobicmalonate decarboxylation by Citrobacter divs.,Arch. M icrobiol. 157,471-474.

JAYASEKERA,. M. K., SHI,W. X.,FARBER, . K., VI-OLA, R. E. (1997b), Evaluation of functionally im-portant amino acids in L-aspartate ammonia-lyase from Escherichia coli, Biochemistry 36 ,

JEBAI,F., HANOUN, ., HAMON,M., THIBAULT,.,PELTRE,G. et al. (1997), Expression, purificationand characterization of rat aromatic r.-amino aciddecarboxylase in Escherichia coli, Protein Expr.Purif. 11,185-194.

JENKINS,., MAYANS, ., PICKERSGILL,. (1998),Structure and evolution of parallel P-helix pro-teins, J. Struct. Biol. 122,236-246.

JETEN, M. S. M., SINSKEY,. J. (1995), Purificationand properties of oxaloacetate decarboxylasefromCorynebacterium glutamicum, A. Van Leeuwen-hoek Int.J . Gen. Mol. M icrobiol. 67,221-227.

JEZOWSKA-BOJCZUK,., KOZLOWSKI,. (1991), Co-ordination of copper(I1) ions by a$-dehydrodi-peptides. Potentiometric and spectroscopic study,Polyhedron 10,2331-2335.

JEZOWSKA-BOJCZUK,., KOZLOWSKI,. (1994), Un-usual binding ability of a$-dehydrodipeptidestowards metal ions,Polyhedron 13,2683-2687.

JEZOWSKA-BOJCZUK,., VARNAGY,., SOVAGO,.,PIETRZYNSKI,., DYBA, . et al. (1996), Co-ordi-nation of copper(I1) ions by prolyl-a$-dehydroa-mino acids: comparative studies and general con-siderations, J . Chem. SOC., Dalton Trans. 3265-3268.

JOHN,R. A. (1995) Pyridoxal phosphate dependentenzymes,Biochim. B iophys. Acta. l248,81-96.

JOHNSON,. A. CHEN, . S. (1995), Taxonomic rela-tionships among strains of Clostridium acetobu-tylicum and other phenotypically related organ-isms,FEMS Microbiol. Rev. 17,233-240.

JOHNSON,. A., LIU,H.-W. (1998) Mechanisms andpathways from recent deoxysugar biosynthesisresearch, Curr. Opin. Chem . Biol. 2,642.

9145-9150.

HUANG,., XUE, Y., SAUER-ERIKSSON,., CHIRICA,L., LINDSKOG,., JONSSON,.-H. (1998), Crystalstructure of carbonic anhydrase from Neisseriagonorrhoeae and its complex with the inhibitoracetazolamide, . Mol. Biol. 283,301-310.

HUANG,. Q.,CHEN,Q.F.,CHEN,C.L., ZHANG, . K.(1999), Asymmetric synthesis of (-)-(R)-pyrro-lam A starting from (S)-malic acid, Tetrahedron:Asymmetry 10,3827-3832.

HUBBARD,. K., KOCH,M., PALMER, . R. J., BAB-BIIT, P. C., GERLT,. A. (1998) Evolution of enzy-matic activities in the enolase superfamily: Char-acterization of D-glucarate/galactarate catabolicpathway in Escherichia coti, Biochemistry 37,14369-14375.

HUBNER, ., TIITMANN,., KILLENBERG-JABS,.,SCHAFFNER,., SPINKA, . et al. (1998),Activationof thiamin diphosphate in enzymes, Biochim. Bi-ophys. Acta. 1385,221-228.

HUGENHOLTZ,. (1993), Citrate metabolism in lacticacid bacteria, FEMS M icrobiol. Rev. 12,165-178.

HUHTA,D. W., HECKENTHALER,., ALVAREZ,. J.,ERMER,., HUBNER,. et al. (1992),The catalyticpower of pyruvate decarboxylase - a stochasticmodel for the molecular evolution of enzymes,Acta Chem. Scand. 46,778-788.

IDING,H., SIEGERT,., MESCH,K., POHL,M. (1998),Application of a-keto acid decarboxylases in bio-transformations, Biochim. Biophys. Acta W85,

IKEMOTO, ., SASAKI, ., HARADAHIRA,., YADA,T., OMURA, . et al. (1999), Synthesis of L-[P-"Clamino acids using immobilized enzymes,Ap pl. Rad. Isotopes 50,715-721.

IKUSHIRO,., HAYASHI,., KAWATA, ~.,AGAMIYA-MA,H. (1998). Analysisof the pH and ligand in-duced spectral transitions of tryptophanase: Acti-vation of the coenzyme at the early stages of thecatalytic cycle,Biochemistry 37,3043-3052.

IMPERIALI, ., MCDONNELL,K. A., SHOGREN-KNAAK,M. (1999), Design and construction ofnovel peptides and proteins by tailored incorpo-ration of coenzyme functionality, Topics Curr.

Chem. 202,l-38.INGRAM,. O., ALDRICH, . C., BORGES, . C. C.,CAUSEY,T.., MARTINEZ,A.t al. (1999), Entericbacterial catalysts for fuel ethanol production,Bi-otechnol. Prog. 15,855-866.

IRAQUI,., VISSERS,.,ANDRE, ., URRESTARAZU,A.(1999); Transcriptional induction by aromaticamino acids in Saccharomyces cerevisiae, Mol.Cell. Biol. 19,3360-3371.

ISHII,S., MIZUGUCHI,., NISHINO,., HAYASHI, .,KAGAMIYAMA,. (1996), Functionally importantresidues of aromatic L-amino acid decarboxylaseprobed by sequence alignment and site directedmutagenesis.J . Biochem. 120,369-376.

307-322. '

ISUPOV, .N., ANTSON,A. A,, DODSON,. J.,DOD- JOHNSON,. A., GASSNER,G.T., BAN DAR IAN,~ ,u-



156 2 Lyases

ZICKA, F. J., BALLOU, . F? et al. (1996), Kinetic

characterization of an organic radical in the as-

carylose biosynthetic pathway, Biochemistry 35,

15846-15856.

JOHNSON,. L., To m, J., SANTIWATANAKUL,., CHEN,

J. S. (1997), Cultures of “Clostridium acetobutyli-cum” from various culture collections comprise

Clostridium acetobutylicum, Clostridium beije-rinckii and two other distinct types based onDNA-DNA reassociation, J. Systemat. B acteriol.47,420-424.

JOHNSON,K., ALLEMANN,. K., WIDMER, ., BEN-

NER, .A. (1993), Synthesis, structure and activity

of artifical, rationally designed catalytic polypep-

tides,Nature 365, No. 6446,530-532.JONES,D. H. (1984), Phenylalanine ammonia-lyase:

Regulation of its induction, and its role in plant

development,Phytochemistry 23,1349-1359.

JONES,D. T., KEIS,S. (1995), Origins and relation-

ships of industrial solvent producing clostridial

strains,FEMS M icrobiol. Rev . 17,223-232.

JONES,D. H., NORTHCOTE,. H. (1984), Stabilityofthe complex formed between French bean Phase-olus vulgaris) phenylalanine lyase and its transi-

tion state analog, Arch. Biochem. Biophys. 235,

JORDAN,. (1999), Interplay of organic and biologi-

cal chemistry in understanding coenzyme mecha-

nisms: example of thiamin diphosphate-depen-

dent decarboxylations of 2-0x0 acids,FEBS Lett.457,29&301.

JORDAN,. M., SPENCER,. B. (1990), Stereospecific

manipulation of hydrogen atoms with opposite

absolute orientations during the biosynthesis of

the polyketide 6-methylsalicyclicacid from chiral

malonates in Penicillin patulum, J . Chem. SOC.,

Chem. Commun.,238-242.

JORDAN,. M., SPENCER,.B. (1991),Mechanistic and

stereochemical investigation of fatty acid and

polyketide biosynthesis using chiral malonates,

Tetrahedron47,60154028.

JORDAN,? M., SPENCER,. B., CORINA, . L. (1986),

The synthesisof chiral malonates and determina-

tion of their chirality using fatty acid synthase and

mass spectrometry, J. Chem. Soc., Chem . Com-mun.,911-913.

JORDAN,.N., O’DONOGHUE, ., CONDON,., Co-

GAN, . M. (1996). Formation of diacetyl by cell-

free extracts of Leuconostoc lactis, FEMS Micro-biol. Le tt.143,291-297.

JORDAN,., NEMERIA, ., Guo, F. S., BABURINA,.,GAO,Y. H. et al. (1998), Regulation of thiamin

diphosphate-dependent 2-0x0 acid decarboxylas-

es by substrate and thiamin diphosphate. Mg(I1)-

evidence for tertiary and quaternary interactions,

Biochim. Biophys. Ac ta. 1385,287-306.

JORRIN,., LOPEZ-VALBUENA,., TENA,M. (1988),

Purification and properties of phenylalanine am-

167-177.

monia-lyase from sunflower (Helianthus annuusL.) hypocotyls,Biochim. Biophy s.Acta 964,73432.

KAMATH,. V., VAALER, . L., SNELL, . E. (1991),

Pyridoxal phosphate dependent histidine decar-

boxylases - Cloning sequencing and expression

of genes from Klebsiella planticola and Entero-bacter aerogenes and properties of the overex-

pressed enzymes, . Biol. Che m.266,9432-9437.

KAMMEYER,, EGGELTE, . A., Bos, J. D., TEUNIS-

SEN, M. B. M. (1999), Urocanic acid isomers are

good hydroxyl radical scavengers: a comparative

study with structural analogs and with uric acid,

Biochim. Biophys. Ac ta. 1428,117-120.

KANE,. F., FISKE,M. J. (1985), Regulation of pheny-

lalanine ammonia lyase in Rhodotorula glutinis, J.Bacteriol. 161,963-966.

KAPAT,A., ZIMAND, ., ELAD,Y. (1998), Effect oftwo isolates of Trichoderma harzianum on the ac-

tivity of hydrolytic enzymes produced by Botrytiscinerea, Physiol. M ol. Plant Pathol. 52,127-137.

KAPAT,A.,ZIMAND, ., EL AD,^. (1998), Biosynthe-

sis of pathogenicity hydrolytic enzymes by Botry-tis cinerea during infection of bean leaves and invitro, Mycol. Rex 102,1017-1024.

KARSTEN,W. E., VIOLA, . E. (1991), Kinetic studies

of L-aspartase from Escherichia-coli:pH depen-

dent activity changes. Arch. Biochem. Biophys.287,6047.

KATAOKA,., IKEMI,M., MORIKAWA,~.,IYOSHI,T.,

NISHI,K. et al. (1997), Isolation and characteriza-

tion of D-threonine aldolase, a pyridoxal-5 phos-

phate dependent enzyme from Arthrobacter sp.

DK-38, Eur. J . Biochem. 248,385-393.

KATO,Y., ASANO, . (1995a), Purification and prop-

erties of crystalline 3-methylaspartase from two

facultative anaerobes, Citrobacter sp. strain YG-

0505 and Morganella morganii strain YG-0601,

Biosci. Biotech. Biochem. 59,93-99.

KATO,~.,SANO,~.1995b), 3-Methylaspartase am-

monia-lyase from a facultative anaerobe strain

YG-1002, Ap pl. Microbiol. Biotechnol. 43, 901-

907.

KATO,Y., ASANO, . (1997), 3-Methylaspartase am-

monia-lyase as a marker enzyme of the mesaco-

nate pathway for (S)-glutamate fermentation in

Enterobacteriaceae, Arch. Microbiol. 168, 457-

463.

KATO,Y., ASANO, . (1998), Cloning, nucleotide se-

quencing and expression of the 3-methylaspar-

tase ammonia lyase gene from Citrobacter amalo-naticus strain YG-1002,Ap pl. M icrobiol. Biotech-nol. 50,468-474.

KAWAGUCHI,., SYONO, . (1996), The excessive

production of indole-3-acetic acid and its signifi-

cance in studies of the biosynthesis of this regula-

tor of plant growth and development, Plant CellPhysiol. 37,1043-1048.

KAWASAKI,., WATANABE,., OHTA, H. (1995), A



5 References 157

tion of the pyruvate decarboxylase from a hap-

loid strain of Saccharomyces cerevisiae, Biol.Chem. Hoppe-Seyler 377,313-317.

KILLENBERG-JABS,M., KONIG, ., EBERHARDT,.,HOHMANN,., HUBNER, . (1997), Role of Glu51for cofactor binding and catalytic activity in pyru-

vate decarboxylase from yeast studied by site-di-

rected mutagenesis,Biochemistry 36,1900-1905.KIM,Y. S., BYUN, . S. (1994) , Purification and prop-

erties of a novel type of malonate decarboxylase

from Acinetobacter calcoaceticus,J . Biol. Chem.269,29636-29641.

KIM,K., COLE, . A. (1999), Synthesisof (2S,3R)-P-methyltyrosine catalyzed by tyrosine phenol-

lyase,Bioorg. Med. Ch em. Lett,9,1205-1208.KIM,Y. S., K O L A ~ K U D Y ,. E., Boos , A. (1979),

Malonyl-CoA decarboxylase from Mycobacteri-um tuberculosis and Pseudomonas fluorescens,Arch. Biochem. Biophys. 1%, 543-551.

KIM, . H., KRONSTAD,. W., ELLIS, . E. (:1996),Pur-

ification and characterization of phenylalanine

ammonia-lyase from Ustilago maydis, Phyto-chemistry 43,351-357.

KIM,M. S.,LEE,W. K. , KIM,H. Y., KIM,C.,RYU,Y. W.(1998), Effect of environmental factors on flavo-

no1 glycoside production and phenylalanine am-

monia-lyase activity in cell suspension cultures of

Ginkgo biloba. J. Microb. Biotechnol. 8, 237-244.

KINOSHITA,.,NAKAYAMA,. (1978). Amino acids,

in: Economic Microbiology: Primary Products of

Metabolism Vol. 2 (ROSE,A. H ., Ed.),pp. 210-256.London: Academic Press.

KIRIMURA, ., SATO,T., AKANISHI,., TERA DA, .,USAMI,. (1997), Breeding of starch utilizing and

itaconic acid producing koji molds by interspecif-

ic protoplast fusion between Aspergillus terreusand Aspergillus usamii, Ap pl. M icrobiol. Biotech-nol. 47,127-131.

KIRSCHNING,., BECHTHOLD,. F. W., ROHR, .

(1997), Chemical and biochemical aspects of de-

oxysugars and deoxysugar oligosaccharides,Top-ics Curr. Chem . 1 8 8 , 1 4 4 .

KLUGER, ., LAM, . F., PEZACKI,. P., YANG,C.-H.

(1995), Diverting thiamin from catalysis to de-

struction. Mechanism of fragmentation of N(1' ) -methyl-2-(l-hydroxybenzyl)thiamin, J. A m .Chem. SOC .117,11383-11389.

KNIGHT, . L.,LEEPER,. J. (1998), Comparison of

chiral thiazolium and triazolium salts as asym-

metric catalysts for the benzoin condensation, J.

Chem . SOC., erkin Trans.1 .,1891-1893.KNOX,c., ASS, E., NEUPERT, ., PINES,0. 1998),

Import into mitochondria, folding and retrograde

movement of fumarase in yeast, J . Biol. Chem .

KOBAYASHI,., YAMANISHI,., TUBOI,S. (1981),Physiochemical, catalytic and immunochemical

273,25587-25593.

novel enzymatic decarboxylation proceeds via a

thiol ester intermediate, Bull Chem. SOC.Jpn. 68,

KAWASAKI,., HORIMAI,., OHTA,H. (1996a), Onthe conformation of the substrate binding to the

active site during the course of enzymatic decar-

boxylation,Bull. Chem. SOC.Jpn. 69,3591-3594.KAWASAKI,., YOKOTA,., TOMITA,. (1996b), L-

Tryptophan production by a pyruvic acid produc-

ing, Escherichia coli strain carrying the Entero-bacter aerogenes tryptophanase gene, J . Ferment.Bioeng. 82,604-606.

KAWASAKI,., SAIT O, ., OHTA,H. (1997),The mode

of substrate recognition of arylmalonate decar-

boxylases, Chem. Lett. 351-352.KAWATA,, TAM URA, ., YANO,S., MIZOBATA,.,

NAGAI,. et al. (1999), Purification and character-

ization of thermostable aspartase from Bacillussp.YM55-1,Arch. Biochem. Biophy s. 3 6 6 , 4 0 4 6 .

KEHAYOVA,P. D., BOKINSKY,. E., HUBER, . D.,

J A I N , A .1999), A caged hydrophobic inhibitor of

carbonic anyhydrase, Org. Lett.,1,187-188.KELLARD, . G.,PALCIC, . M., PICKARD, .A.,VED-

ERAS, J. C. (1985), Stereochemistry of lysine for-

mation by meso-diaminopimelate decarboxylase

from wheat-gem, Biochemistry 24,3263-3267.KENNEDY,. F., WH ITE, . A. (1990) The plant, algal,

and microbial polysaccharides, in: CarbohydrateChemistry (KENNEDY,. F.,Ed.), pp. 22Ck261. Ox-

ford: Oxford University Press

KERN,D., KERN,G., NEEF,H., TITMANN,, KILLEN-BERG-JABS, M. et al. (1997), How thiamine diph-

osphate is activated in enzymes, Science 275,

KERN,A. D., OLIVEIRA, . A., COFFINO,., HACK-ERT, M. L. (1999), Structure of mammalian orni-

thine decarboxylase at 1.6 8, resolution: stereo-

chemical implications of PLP-dependent amino

acid decarboxylases,Structure with Folding & D e-sign 7,567-581.

KERUCHENKO,. S., KERUCHENKO,. D., GLADILIN,K. L., ZAITSE V,. N., CHIRGADZE,. Y. (1992),Purification, characterization and preliminary X-ray study of fumarase from Saccharomyces cerev-isiae, Biochem. Biophys. Acta 1122,85-92.

KHOMUTOV, . M., FALEEV,. G.,BELYANKIN,. V.,

KHURS,E. N., KHOMUTOV,. R. et al. (1997),l-Amino-2-(4-hydroxyphenyl)ethyl phosphinic

acid - a novel substrate of tyrosine phenol lyase,

Bioorganicheskaya Khimiya 23,919-921.KHRISTOFOROV,. R., SUKHAREVA,. S. ,DIXON,.

B. F., SPAR KES,. J., KRASNOV,. F?, BUKRINA,.M. (1995),The interaction of glutamate decarbox-

ylase from Escherichia coli with substrate analogs

modified at C-3 and C-4, Bioch em. M ol. B iol. Znt.36,77-85.

KILLENBERG-JABS,., KONIG,S., HOHMANN,.,HUBNER, . (1996), Purification and characteriza-

2017-2020.

67-70.



158 2 Lyases

prop erties of fumarases crystallized separatelyfrom mitoch ondria1 and cystolic fractions of ratliver,J. Biochem. 89,1923-1931.

KOGA, . (1995), Structure and function of indolepy-ruvate decarboxylase, a key enzym e in indole-3-acetic-acid biosynthesis, Biochim. Biophys. Acta.

KOGA,J., ADACHI,., HIDAKA, . (1991), Molecularcloning of the g ene for indolepyruvate decarbox-ylase from Enterobacter cloacae, Mol. Gen. Gen et.22410-16.

KOGA,J., ADACHI,., HIDAKA, . (1992), Purifica-tion and characterization of indolepyruvate de-carboxylase- novel enzym e for indole-3-acetic-acid biosyn thesis in Enterobacter cloacae, J. Biol.Chem. 267,15823-15828.

KOGA, J., SYONO, ., ICHIKAWA,., ADACHI,.

(1994), Involvement of L-tryptophan aminotrans-ferase in indole-3-acetic-acid biosyn thesis in En-terobacter cloacae, Biochim. Biophys. Acta. 1209,

KOK,. (1996) Inducible gene expression and envi-ronmentally regulated genes in lactic acid bacte-ria, A . Van Leeuwenhoek Znt. J . Gen. Mol. Micro-biol. 70,129-145.

K O L A ~ K U D Y ,. E., POULOSE, . J., KIM, Y. S.(1981), Malonyl-CoA from avian, mammalianand microbial sources, Methods Enzymol. 71,150-163.

KOLB,S., SEELINGER,., SPRINGER,., LUDWIG,.,SCHINK,. (1998), The fermen ting bacteriumMalonomonas rubra is phylogenetically relatedto th e sulfur reducing bacteria and contains a c-type cytochrome similar to those of sulfur andsulfate reducers, System. Appl. Microbiol. 21,

340-345..KONIG,S. (1998), Subunit structure, function and or-

ganization of pyruvate decarboxylases from vari-ous organisms, Biochim. Biophys. Acta. l385,

KONIG,S., SVERGUN,. I. , VOLKOV,. V., FEIGIN,.A ,,KOCH,M. H.J.(1998), Sm all-angle X-ray solu-tion-scattering studies on ligand-induced subunit

interactions of the thiamine diphosphate de pen-den t enzyme pyruvate decarboxylase from differ-ent organisms,Biochemistry 37,5329-5334.

KOSUGE,T.,MEHARA,., HOSHINO,T.1998), Clon -ing and sequence analysis of fumarase and super-oxide dismutase genes from an extreme thermo-phile Thermus thermophilus HB27,J. Ferment.Bi-oeng. 86,125-129.

KRAUTWURST,., BAZAES,., GONZALAE,. D.,JA-

BALQUINTO, A. M., FREY,P. A., CARDEMIL,.(1998), The strongly conserved lysine of Saccha-romyces cerevisiae phosphoenolpyruvate carbox-ykinase is essential for phosphoryl transfer, Bio-chemistry 37,6295-6302.

KREN,V., CROUT,D. H. G., DALTON, ., HUTCHIN-

l249,l-13.

241-247.

271-286.

SON, D.W., KONIG,W. et al. (1993), Pyruvate de-carboxylase - a new enzyme for the production ofacyloins by biotransformation, J. Chem. SOC.,

Chem. Comm un.,1993,341-343.KYRIAKIDIS,. A., PANAGIOTIDIS,. A., GEORGAT-

SOS, J. G. (1983), Ornithine decarboxylase (germi-nated barley-seeds), Methods Enzym ol. 94, 162-166.

LABROU,. E., CLONIS,. D. (1995), Biomimeticdye-ligands for oxalate recognizing enzymes -studies with oxlate oxidase and oxalate decarbox-ylase,J. Biotechnol. 40,59-70.

LABROU,. E., CLONIS,. D. (1999), Oxaloacetatedecarboxylase from Pseudomonas stutter? Purifi-cation and characterization, Arch. Biochem. Bi-ophys.365,17-24.

LAI,M. C., TRAXLER,. W. (1994), A coupled twostage continuous fermentation for solvent pro-duction by Clostridium acetobutylicurn, En zym eMicrob. Technol. 16,1021-1025.

LANDRO,. A., GERLT,.A.,KOZARICH,.W., Koo,C.W., SHAH, . (1994), The role of lysine 166 in themechanism of m andelate racem ase from Pseudo-monas putida - mechanistic and crystallographicevidence for stereospecific alkylation by (R)-a-

phenylglycidateBiochemistry 33,635-643.LANGER,., LIEBER,A.,ETEY, . (1994a), Histidine

ammo nia-lyase mutant-S143C s post-translation-ally converted into fully active wild-type enzyme- evidence for ~ e r i n e l ~ ~o be the precursor of ac-tive-site dehydroalanine, Biochemistry 33,14034-1 4038.

LANGER,., RECK,G., REED,., RETEY,. (1994b),Identification of serin el43 a s the most likely pre-cursor of dehydroalan ine in the active-site of his-tidine ammonia-lyase - a study of the overex-pressed en zyme by site-directed m utagenesis,Bi-

ochemistry 33,6462-6467.LANGER,., PAULING,., RETEY,. (1995), The role

of dehydroalan ine in catalysis by histidine amm o-nia lyase,Ange w. Chem. Znt. Edn. E ngl. 34,1464-1465.

LANGER,., STARCK,., LANGER, ., RETEY,.

(1997a), Formation of the Michaelis complexwithout involvement of the prosthetic grou p de-hydroalanine of histidine ammonialyase, Bioorg.Med. Chem. Lett. 7,1077-1082.

LANGER,B.,ROTHER,D.,RETEY,. (1997b), Identifi-cation of essential amino acids in phenylalanineammonia-lyaseby site-directed mutagenesis,Bio-chemistry 36,10867-10871.

LAUBLE,.,KENNEDY,M. C., BEINERT,.,STOUT,C.D. (1994), Crystal structures of aconitase withtrans-aconitate and nitrocitrate bound, J. Mol.Biol. 237,437451.

LAUBLE, .,KENNEDY,M. C., EMFTAGE, . H.,BEI-

NERT, H., STOUT,c . D. (1996),The reaction of flu-orocitrate with aconitase and the crystal structure



5 References 159

tryptophan 412 in the substrate activation path-way of yeast pyruvate decarboxylase,Biochemis-try 38,10004-10012.

LIM,H. W., PARK,. S., LIM,C.J. (1997), Purification

and properties of phenylalanine lyase from leaf

mustard,Molecules Cells 7,715-720.LIM,H. W., SA,J. H., PARK, . S., LIM,C. J. (1998), A

second form of phenylalanine ammonia lyasefrom leaf mustard,Molecules Cells 8,343-349.

LIMBERG, ., THIEM,. (1996), Synthesis of modifiedaldonic acids and studies of their substrate effi-ciency for dihydroxy acid dehydratase, Aust. J.Chem.49,349-356.

LINDQUIST,., SCHNEIDER,. (1993), Thiamin diph-osphate dependent enzymes - transketolase, py-ruvate oxidase and pyruvate decarboxylase,Curr.Opin. Struct. Biol. 3,896-901.

LINDQUIST,., SCHWEDA,. H., REEVES,? R., LIND-

BERG, A. A. (1994), In vitro synthesis of CDP-D-abequose using Salmonella enzymes of cloned rfbgenes: Production of CDP-6-deoxy-~-xylo-4-hex-ulose, CDP-3,6-dideoxy-~-xylo-4-hexulosendCDP-3,6-dideoxy-~-galactose and isolation byHPLC, Eur. J . Biochem. 225,863-872.

LINDSKOG,. (1997), Structure and mechanism ofcarbonic anhydrase,Pharmacol. Ther. 74,l-20.

LIU, H. W., THORSON,. S. (1994), Pathways andmechanisms in the biogenesis of novel deoxysug-

ars by bacteria,Ann . Rev. Microbiol. 48,223-256.

LLOYD-GEORGE,., CHANG,T. . S. (1993), Free andmicroencapsulatedEnvinia herbicola for the pro-

duction of tyrosine, Biom at. Artif: Cells Immobil.Biotechnol. 21,323-333.

LLOYD-GEORGE,., CHANG,. M. S. (1995), Charac-terization of free and alginate-polylysine-alginatemicroencapsulatedEnvinia herbicola for the con-version of ammonia, pyruvate, and phenol into L-

tyrosine,Biotechnol. Bioeng. 48,706-714.Lo, S . F., MILLER,. ?,LEI,Y. ., THORSON,. S., LIU,

H.-W., SCHOTTEL,. L. (1994), CDP-6-deo~y-A~.~-

glucoseen reductase from Yersinia pseudo tuber-culosis,J . Bacteriol.176,460-468.

LOBELL, ., CROUT, . H. G. (19%a), New insightinto the pyruvate decarboxylase-catalyzed orma-

tion of lactaldehyde from H-D exchange experi-ments: A “water proof” active site,J. Chem. SOC.,

Perkin Trans.1,1577-1581.LOBELL,M., CROUT,D. H. G. (1996b), Pyruvate de-

carboxy1ase:A molecular modeling study of pyru-vate decarboxylation and acyloin formation, J.Am . Chem. SOC.118,1867-1873.

LOEWUS, . W., LOEWUS,. A., BRILLINGER,. U.,OTSUKA, ., FLOSS, .G. (1980), Stereochemistryof the L-myo-inositol-1-phosphateynthase reac-tion,J. Biol. Chem.255,1710-1712.

LOUWRIER,. , OSTENDORF,., BOOM.A., HIEM-STRA, ., SPECKAMP,. N. (19%), Studies towardsthe synthesis of (+)-ptilomycalinA, stereoselec-

of the enzyme-inhibitor complex,Proc. Natl. A c-ad . Sci.U S A .93,13699-13703.

LEE,K.-L., RAMALINGAM,., SON,.-K., WOODARD,

R. W. (1989), A highly efficient and large scalesynthesis of (2S,3S)-[2,3-ZH2]-and 2S,3R)-[3-*H]-

aspartic acids via an immobilized aspartase con-

taining microbial cell system, J . Org. Chem. 54,

LEE,S.G.,Ro,H. S.,HONG,.P.,KIM,E.H.,SUNG,M.H. (1996a), Production of L-DOPA by thermo-stable tyrosine phenol-lyase of a thermophilic

Symbiobacterium species overexpressed in re-combinantEscherichiacoli, . Microbiol. Biotech-nol. 6,98-102.

LEE,S.-G., HONG, .P., SUNG,M. H. (1996b), Remov-al and bioconversion of phenol in wastewater by

a thermostable p-tyrosinase, Enzyme Microb.Technol. 19,374-377.

LEE, S.-G., HONG, .-P., (301, Y.-H., CHUNG, .-J.,SUNG,M.-H. (1997), Thermostable tyrosine phe-nol-lyase of Symbiobacterium sp. SC-1: gene clon-

ing, structure determination and overproductionin Escherichia coli, Protein Expr. Purif: 11,

LEE,S.-G., HONG, .-F’., SUNG,M.-H. (1999a), Devel-opment of an enzymatic system for the produc-

tion of dopamine from catechol, pyruvate andammonia,Enz yme Microb. Technol.25,298-302.

LEE,S.-G., HONG, .-P.,KWAK, . S.,ESAKI, ., SUNG,M.-H. (1999b), Characterization of thermostabletyrosine phenol-lyasefrom an symbiotic obligato-

ry thermophile,Syrnbiobacteriumsp. SC-1,J . Bio-chem. Mol. Biol. 32,480485.

LEE,T.T.,WORBY, ., BAO,Z. Q.,DIXON,. E.,COLE-MAN, R. F. (1999c), Hisrn and His141 are criticalcontributors to the intersubunit catalytic site ofadenylosuccinate lyase of Bacillus subtilis, Bio-chemistry 38,22-32.

LEHOU,. E., MITRA,B. (1999), (S)-Mandelate dehy-drogenase from Pseudomonas putida: Mechanis-

tic studies with alternative substrates and pH and

kinetic isotope effects, Biochemistry 38,

5836-5848.

LEMONNIER,., LANE,D. (1998), Expression of thesecond lysine decarboxylase gene of Escherichiacoli , Microbiology (U K) 144,751-760.

LENZ,H., WUNDERWALD,., EGGERER, . (1976),Partial purification and some properties of oxa-lacetase,Eur. J . Biochem. 65,225-236.

LEPINIEC,., VIDAL,.,CHOLLET,., GADAL,., CRE-TIN,C. (1994), Phosphoenolpyruvate carboxylase- Structure, regulation and evolution,Plant Sci.

LEVIN, ., FORCHIASSIN,. (1998), Culture conditionsfor the production of pectinolytic enzymes by thewhite-rot fungus Trametes trogii on a laboratory

scale,Acta Biotechnol. 18,157-166.LI, H., JORDAN,. (1999), Effects of substitution of

3195-3198.

263-270.

99,111-124.



160 2 Lyases

tive N-acyliminium ion coupling reactions toenantiopure C-2 substituted lactams, Tetrahedron

Lu, G. G., DOBRITZSCH,., KONIG,., SCHNEIDER,.(1997), Novel tetramer assembly of pyruvate de-carboxylase from brewers' yeast observed in anew crystal form, FEBS Lett. 403,249-253.

LUNG, .Y., BAETZ, . L., PECK,A. B. (1994), Molec-ular cloning, DNA sequence and gene expressionof the oxalyl-coenzyme A decarboxylase gene,oxc from the bacterium Oxalobacter formigenes,J.Bacteriol. 176,2468-2472.

LUTHY,., RETEY,.,ARIGONI, . (1969). Preparationand detection of chiral methyl groups,Nature 221,

LYNSTRAD,., GRASDALEN,. (1993), A new air liftbioreactor used in P-31-NMR studies of itaconicacid producingAspergillus terreus,J.Biochem. Bi-ophys. M ethods 27,105-116.

MALASHKEVICH,. N., BURKHARD,., DOMINICI,?,MOORE, . S.,VOLTATTORNI,. B., JANSONIUS,. N.(1999), Preliminary X-ray analysis of a new crystalform of recombinant pig kidney DOPA decarbox-ylase,Acta CrystallographicaD 55,568-570.

MALASHKEVICH,. N., STROP,?,KELLER,. W., JAN-SONIUS,. N., TONEY, . D. (1999), Crystal struc-tures of dialkylglycine decarboxylase inhibitorcomplexes,J.Mol. Biol.294,193-200.

MA", J. (1987),Secondary Metabolism,pp. 173-190.Oxford: Oxford Science Publications.

MA", J., DAVIDSON,. S., HOBBS,. B., BANTHORPE,D.V., HARBOURNE,. B. (1994),Natural Products:Their Chemistry and Biological Significance,pp.372-375. Longman Scientific and Technical.

MANULIS,., HAVIVCHESNER,., BRANDL,M. T.,LINDOW,. E., BARASH,. (1998), Differential in-volvement of indole-3-acetic acid biosyntheticpathways in pathogenicity and epiphytic fitnessof Erwinia herbicola pv, gypsophilae, Mol. PlantMicrobe Interact. 11,634-642.

MARLETTA, . A., CHEUNG,. F., WALSH,C. (1982),Stereochemical studies on the hydration ofmonofluorofumarate and 2, 3-difluorofumarateby fumarase,Biochemistry 21,2637-2644.

MAYANS,.,SCOTT,M., CONNERTON,., GRAVESEN,T., BENEN,. et al. (1997),%0 crystal structures ofpectin lyase A from Aspergillus reveal a pH driv-en conformational change and striking diver-gence in the substrate-binding clefts of pectin andpectate lyases,Structure 5,677-689.

MBWAGA,. M., MENKE, ., GROSSMANN,. (1997),Investigationson the activity of cell wall-degrad-ing enzymes in young wheat plants after infectionwith Pseudocercosporella herpotrichoides (Fron)Deighton,J. Phytopathol. 145,123-130.

MCCANN,? P., BACCHI, . J., HANSON, . L., NA-THAN, H. C., HUTNER, S. H., SJOERDSMA, .(1983), Methods for the study of the treatment of

52,2603-2628.

1213-1215.

protozoan diseases by inhibitors of ornithine de-carboxylase,Methods Enzy mol. 94,209-213.

MERMEISTEIN,. D., PAPOUTSAKIS,. T., PETERSEN,D. J., BENNEIT,G. N. (1993), Metabolic engineer-ing of Clostridium acetobutylicum ATCC-824 forincreased solvent production by enhancement ofacetone formation enzyme activities using a syn-thetic acetone operon, Biotechnol. Bioeng. 42,

MIALL,L. M., (1978), Organic acids, in: EconomicMicrobiology. Primary Products of M etabolism (-ROSE,A. H., Ed.), pp. 1-55. London: AcademicPress.

MICHIELSEN,. J. F., MEIJER, .A., WIJFFELS,R. H.,TRAMPER,., BEEFTINK,. H. (1998), KineticsofD-malate production by permeabilized Pseudo-monas pseudoaligenes, Enzyme Microb. Technol.22,621-628.

MICKLEFIELD,., HARRIS, . J., GROGER,., MOCEK,U., HILBI,H. et al. (1995), Stereochemical courseof malonate decarboxylation in Malonomonasrubra,J. Am . Chem. SOC.117,1153-1154.

MILLER, . G., TRAUT,T.W., WOLFENDEN,. (1998),Effects of substrate binding determinants in thetransition state for orotidine 5 -monophosphatedecarboxylase,Bioorg. Chem .26,283-288.

MINAMISAWA,., OGAWA, .,FUKUHARA,., KOGA,J. (1996), Indolepyruvate pathway for indole-3-acetic acid biosynthesis in Bradyrhizobium elka-nii, Plant Cell Physiol. 37,449-453.

MITCHELL,. L. A. (1983), Ornithine decarboxylaseand the ornithine decarboxylase-modifying pro-tein of Physarum polycephalum, Methods E nzy -mol. 94,140-146.

MIYAMOTO,., OHTA,H. (1990), Enzyme mediatedasymmetric decarboxylation of disubstituted mal-onic acids, J.Am . Chem. SOC.112,40774078.

MIYAMOTO,., OHTA,H. (1991), Asymmetric decar-boxylation of disubstituted malonic acid by Alcal-igenes bronch isepticus KU 1201, Biocatalysis 5,49-60.

MIYAMOTO,., OHTA, H. (1992), Purification andproperties of a novel arylmalonate decarboxylasefromAlcaligenes bronch isepticus KU 1201,Eur.J.

Biochem. 210,475481.MIYAMOTO,., TSUCHIYA,., OHTA,H. (1992), Mi-

crobial asymmetric decarboxylation of fluorinecontaining arylmalonic acid derivatives, J. Fluo-rine Chem. 59,225-232.

MIZOBATA,T.,LJJIOKA,T.,AMASAKI,., HIDAKA,M.(1998), Purification and characterizationof a ther-mostable class I1 fumarase from Thermus thermo-philus, Arch. B iochem. Biophys. 35549-55.

MOCHIZUKI,., HIRAMATSU,., SUGAI,T., HTA,H.,MORITA, ., ITOKAWA,. (1995), Reductive con-version of carbonyl compounds by yeast. 2. Im-proved conditions for the production and charac-terization of l-arylpropane-l,2-diols and related

1053-1060.



5 References 161

MURASE,., TAKAGI,. S., HIGA SHI, ., IMAISHI,.,YUMO TO, ., TOKUSHIGE1991), Activation ofaspartase by side-directed mutagenesis,Biochem.Biophys. Res. Commun. 177,414-419.

MURASE,.,YUMO TO, . (1993), Cha racterization of3 types of aspartase activated by site- directedmutagenesis, limited proteolysis, and acetylation.J . Biochem. 114,735-739.

MUSSONS, . L., RAPOSO, .,DE LA TORRE,M. F.,MORAN, . R., CAB ALL ERO , . C. (1999), Di-benz[c,h]acridine receptors for dibutylmalonicacid, Tetrahedron,55,40774094.

NAGAI, ., HAMADA, ., KAI, N., TANO UE, ., N A-GAYAMA, F. (1996), Distribution of aromatic L-amino acid decarboxylase in tissues of skipjacktuna using L-DOPA as the substrate,J. Fish. Biol.48,1014-1017.

NAGASAWA,T.,TAGAWA,T.,OTO,.,KIM,C. J., TA-NI,Y.et al. (1981), Syn theses of L-tyrosine-relatedamino-acids by tyrosine phenol- lyase of Citro-bacter intermedius, Eur.J . Biochem. 117 ,3340 .

NAIDU,G. S. N., PANDA,. (1998), Production of pec-tolytic enzym es- a review,Bioproc. Eng. 19,355-361.

NAIR,R.V., GREEN , . M., WATSON,. E., BENNETT,G. N., PAPOUTSAKIS,. T. (1999), Regulation ofthe sol locus genes for butanol and acetone for-mation in Clostridium acetobutylicum ATCC 824by a putative transcriptional represssor,J. Bacteri-

NAUNDORF,., KLAFFKE,W. (1996), Subs trate speci-ficity of native dTDP-~-glucose-4,6-dehydratase:Chem o-enzymatic syntheses of artificial and nat-urally occurring deoxy sugars, Curb. Res. 285,

NELSESTUEN,. L., KIRKWOOD,. (1971),The mech-anism of action of the enzyme uridine diphos-phogiucose 4-epimerase,J . Biol. Chem. 246,7533.

NICHOLSON,. L., HAMMERSCHMIDT,. (1992), Phe-nolic compou nds and th eir role in disease resis-tance, Ann. Rev. Phytop athol. 30,369-389.

NIEDEN D, . R., PA RKER , . R ., STIERL E,D. B.,STIERLE, . A. (1999), First fungal aromatic L-amino acid decarboxylase from a paclitaxel pro-ducing Penicillium raistrickii, Mycologia 91,619-626.

NIKAIDOU,., NAGANUMA,., KAMIO,Y., IZAKI,.(1995), Production, purification, and properties ofa pectin lyase from Pseudomonas marginalisN6301, Biosci. Biotechnol. Biochem. 59,323-324.

NIKOLOVA,., WARD,0.P. (1991), Production of L-

phenylacetyl carbinol by biotransformation -

product and by-product formation and activitiesof the key enzymes in wild-type and A D H isoen-zyme mutants of Saccharomyces cerevisiae, Bio-technol. Bioeng. 38,493-498.

NUIRY,. I., HERME S,. D., WEISS,P. M., CHEN ,C-Y.,COOK, . F. (1984), Kinetic mechanism and loca-

01. 181,319-330.

141-150.

compounds, Biosci. Biotechnol. Biochem. 59,

MOHRIG,. R., MOERKE, . A., CLOUTIER,D. L.,LA NE , . D., PERSON, . C., ONA SCH,T.. (1995),Importance of historical contingency in the ster-eochemistry of hydratase-dehydratase enzymes,Science 269, No. 5223,527-529.

MOMANY,., ERNST,S., GHOSH, ., CHANG,N. L.,HACKERT, . L. (1995), Crystallographic struc-ture of a PLP depe ndent ornithine decarboxylasefrom Lactobacillus 30a to 3.0 A resolutionJ Mol.Biol. 252,643-655.

MONN ET, ., CORR IEU,G. (1998), Selection andproperties of Lactobacillus mutants producing a-

acetolactate, J . Dairy Sci. 81,2096-2102.MON NET, , SCHMITT,, DIVIES,C. (1994), Diace tyl

production in milk by an a-acetolactic acid accu-mulating strain of Lactococcus lactk ssp. lactis b i-ovar diacetylactis,J. Dairy Sci. 77,2916-2924.

MONTGOMERY,. A., ROSIERS, . D., BRUNENGRA-BER,H. (1992), Biosynthesis and characterizationof 3-hydroxyaikan-2 -ones and 2,3-a lkanedio ls -potential products of aldehyde metabolism, Biol.Mass. Spectr. 21,242-248.

MOORE,P. S., DOMINICI,., VOLTATTORNI,. B.(1996) Cloning and exp ression of pig kidney D O -PA decarboxylase: Comp arison of the naturallyoccumng and recombinant enzymes, Biochem. J .

315,249-256.MORGAN, . M. L. (1999), Polyamines, an overview,

Mol. Biotechnol. 11,229-250.MORENO-ARRIBAS,., LONVAUD-FUNEL,. (1999),

Tyrosine decarboxylase activity of Lactobacillusbrevis IOEB -9809 isolated from w ine and L. bre-vis ATCC-367,FEMS M icrobiol. Lett. 180,5540.

MORRIS, . R., BOEKER, . A. (1983), Biosyntheticand biodegradative ornithine and arginine decar-boxylases from Escherichia coli, Methods Enzy-mol. 94,125-134.

MOVAHEDI,., HEALE,J. B. (1990), The roles ofaspartic proteinase and endo-pectin lyase en-zymes in the p rimary stages of infection and pa th-ogenesis of various host tissues by different iso-lates of Botrytis cinerea per s ex per s, Physiol. MolPlant. Pathol. 36,303-324.

MUGURUMA,. (1999),Ab initio MO study on the cat-alytic mechanism in the active site of carbon ic an-hydrase,THEOCHEM,J.Mol. Struct.462,439-452.

MULLER,E. H., RICHARDS, . J., NORBECK,.,BYR NE, . L., KARLSSON ,. A . et al. (1999),Thia-mine repression and pyruvate decarboxylase au-toregulation independently control the expres-sion of the Saccharomyces cerevbiae PDC5 gene,FEBS Lett. 449,245-250.

MUNIER,. L., BOM PEIX,. (1985), Inhibition of theL-phenylalanine ammonia-lyase from Rhodoto-rula glutinis with N-substituted-phenylalanines,

C. R. I11 300,203-206.

2282-2291.



162 2 Lyases

tion of rate-determinig steps for aspartase fromHafnia a lvei, Biochemistry 23,5168-5175.

OBA,K., URITANI,. (1982), Pyruvate decarboxylasefrom sweet-potato roots, Methods Enzym ol. 90,528-532.

OGAWA, ., FUJIOKA,M. (1999), Is threonine aldo-lase identical to serine hydroxymethyltransfe-rase?, Seikagaki 71,1145-1152.

OGINO, ., TAMIYA,., KIMURA,~.,ZUMA, ., TA-GAKI,W. (1996), cis-Diamines as active catalystsfro the decarboxylation of oxalacetate, J . Chem.Soc., Perkin Trans I I , 979-984.

O’HARE,M. C., DOONAN,. (1985), Purification andstructural comparisons of the cytosolic and mito-chondrial isoenzymes of fumarase from pig liver,Biochim. Bioph ys. Acta 827,127-134.

OHBA,H., YASUDA,., HIROSUE, ., YAMASAKI,.(1995), Improvement of the thermostability ofpyruvate decarboxylase by modification with anamylose derivative,Biosci. Biotech. Biochem. 59,1581-1583.

OHTA,H. (1997), Enzyme mediated enantioselectiveprotonation of enolates in aqueous medium, Bull.Chem. SOC. pn. 10,2895-2911.

OKABE,M., OHTA,N., PARK,^. S. (1993), Itaconic ac-id production in an airlift bioreactor using a mod-ified draft tube, J . Ferment. Bioeng. 76,117-122.

OLIVER,A. L., RODDICK,. A., ANDERSON,. N.(1997), Cleaner production of phenylacetylcarbi-no1 by yeast through productivity improvementsand waste minimization, Pure Appl. Chem. 69,

ORCZYK,A., HIPSKIND,., DE NEERGAARD,.,GOLDBROUGH,., NICHOLSON1996), Stimulationof phenylalanine ammonia lyase in Sorghum inresponse to innoculation with Bipolaris maydis,Physiol. M ol. Plant. Pathol.4 8 , 5 5 4 .

ORR,G. R., GOULD, . J. (1982), Stereochemistry ofthe bacterial ornithine, lysine and arginine decar-boxylase reactions, Tetrahedron Lett. 23, 3139-3142.

ORTEGA,. (1996), Pectolytic enzymes produced bythe phytopathogenic fungus Colletotrichum gloe -osporioides, Texas J. Sci. 48,123-128.

OSTERMAN,. L., BROOKS, . B., RIZO, ., PHILLIPS,M. A. (1997), Role of Arg-277 in the binding ofpyridoxal 5‘-phosphate to Trypanosoma bruceiornithine decarboxylase, Biochemistry 36, 4558-4567.

OSTERMAN,. L., BROOKS, . B., JACKSON,., AB-BOIT, . J., PHILLIPS, . A. (1999), Lysine-69 playsa key role in catalysis by ornithine decarboxylasethrough acceleration of the Schiff base formation,decarboxylation and product release steps, Bio-chemistry 38,11814-11826.

OTHS,. J.,MAYER, . M.,FLoss,H. G. (1990) Stereo-chemistry and mechanism of the GDP-mannosedehydratase reaction, Curb. Res. 198,91-100.

2371-2385.

OYAMA, . (1992), The industrial production ofaspartame, in: Chirality in Industry (COLLINS,.N., SHELDRAKE,. N., CROSBY., Eds), pp. 237-247. Chichester: John Wiley & Sons.

PALMER, . R. J., GERLT, . A. (1996), Evolution ofenzymatic activities: multiple pathways for gener-ating and partitioning a common enolic interme-daite by glucarate dehydratase from Pseudomo-nusputida,J. Am . Chem. SOC .118,10323-10324.

PALMER, . R. J., WIECZOREK,. J., HUBBARD,. K.,MRACHKO,. T., GERLT,. A. (1997), Importanceof mechanistic imperatives in enzyme-catalyzedj3-elimination reactions: Stereochemical conse-quences of the dehydration reactions catalyzedby D-galactonate dehydratase from Escherichiacoli and D-glUCarate dehydratase from Pseudo-monas putida ,J . Am . Chem. SOC.119,9580-9581.

PALMER,D. R. J., HUBBARD,. K., GERLT, . A.(1998), Evolution of enzymatic activities in theenolase superfamily: Partitioning of reactiveintermediates by (D)-glucarate dehydratase fromPseudomonas putida, Biochemistry 37, 1435&14357.

PARK,. S., ITIDA, ., OHTA,N., OKABE,M. (1994),Itaconic acid production using an airlift reactor inrepeated batch culture of Aspergillus terreus, J .

Ferment. Bioeng. 71,329-333.PARK, . H., XING,R. Y., WHITMAN,. (1995), Non-

enzymatic acetolactate oxidation to diacetyl byflavin, nicotinamide and quinone coenzymes,Bi-ochim. Biophys. Ac ta. 1245,366370.

PARK, .S., LEE, . Y., KIM,H.S. (1998) Production ofL-DOPA (3-(3,4-dihydroxyphenyl)-~-alanine)from benzene using a hybrid pathway,Biotechnol.Bioeng. 58,339-343.

PARKER, . R., MOORE,. A., SCHWAB,. M., DAVIS-SON,V. J. (1999, Escherichia coli imidazoleglyce-rol phosphate dehydratase: Spectroscopic charac-terization of the enzyme product and the stericcourse of the reaction, J . Am. Chem. SOC.117,10605-10613.

PARKS,E. .,ERNsT, .R.,HAMLIN,., XUONG,N. .(1985), Structure determination of histidine de-carboxylase from Lactobacillus 30a at 3.0 A reso-lution, J . Mol. Biol. 182,455465.

PESCHKE,~.., SACHS,M. M. (1993), Multiple pyru-vate decarboxylase genes in maize are induced byhypoxia,Mol. Ge n. Genet. 240,26212.

PETERSEN,. J., BENNEIT,G. N. (1990), Purificationof acetoacetate decarboxylase from Clostridiumacetobutylicium ATCC-824 and cloning of the ac-etoacetate decarboxylase gene in Escherichia co -li, Ap pl. E nviron. Microbiol. 56,3491-3498.

PETERSEN,. J., VAN DER LEYDEN,., BENNEIT,G.N.(1993), Sequence and arrangement of genes en-coding the enzymes of the acetone productionpathway of Clostridium acetobutylicum ATCC-824, Gene l23,93-97.



5 References 163

F'ETRUCCIOLI,M., PmqV., FEDERICI,. (1999),Itacon-ic acid production by Aspergillus terreus on rawstarchy materials,Lett.App l. M icrobiol.28,309-312.

PHALIP, ., MONNET, ., SCHMI'IT,., RENAULT,.,

GODON,. J., DIVES,C. (1994), Purification andproperties of the cr-acetolactate decarboxylasefrom Lactococcus lactis ssp. lactis NCDO-2118,FEBS Lett. 351,95-99.

PHILLIPS,. S. (1989), Mechanism of tryptophan in-dole-lyase: Insights from pre-stady state kineticsand substrate and solvent isotope effects, J . Am.Chem. SOC.111,727-730.

PHILLIPS, . S., VO N TERSCH,R. L., SECUNDO,.(1997), Effects of tyrosine ring fluorination onrates an d equilibria of formation of intermediatesin the reactions of carbon-carbon lyases, Eur. J .

Biochem. 244,658-663.PIATIGORSKY,

.,WISTOW,

. (1991) The recruitmentof crystallins: new functions preced e gen e dupli-cation, Science 252,107S1079.

PICCIRILLI,. A., ROZZELL,r. , J. D., BENNER,. A.(1987), The stereospecificity of oxaloacetate de-carboxylase: A stereochemical imperative?, J.Am. Chem. SOC.109,8084-8085.

PICKERSGILL,., JENKINS,., HARRIS, ., NASSER, .,

ROBERT-BAUDOUY,. (1994),The structure of Ba -cillus subtilis pectate lyase in complex with cal-cium Struct. Biol. l,717-723.

PICKERSGILL,.,SMITH,D.,WORBOYS,K.,ENKINS,.(1998), Crystal structure of polygalacturonase

from Erwinia carotovora ssp. carotovora,J .

Biol.Chem 273,24660-24664.PIEPER,. A., G U O ,Z., LIU,W.-W. (1995),Mechanis-

tic studies of t he biosynthesis of 3,6-dideoxy sug-ars: stereochemical analysis of C-3 deoxygena-tion, J. Am. Chem. SOC.117,5158-5159.

PIEPER,. A., YANG,D. Y., ZHOU,H. Q., LIU,H.-W.(1997), 3-deoxy-3-fluoropyridoxamine -phos-phate: Synthesis and chemical and biologicalproperties of a coenzyme B6analog,J. Am. Chem.

PILNIK, ., VORAGEN,. G. J. (1992), Gelling agents(pectins) from plants for the food industry, in:Ad-

vances in Plant Cell Biochemistry and Biotech nol-ogy Vol. 1, (MORRISON,., Ed.),pp. 219-270. Lon-don: JAI Press.

PIRRUNG, . C., HOLMES, . P., HOROWITZ,. M.,NU", D. S. (1991),Mechanism an d stereochemis-try of a,P-dihydroxyacid dehydratase, J. Am.Chem. SOC.113,1020-1025.

PISHKO,.J.,ROBERTUS,.D. (1993), Site directed al-teration of 3 active site residues of a pyruvoyl-de-pendent histidine decarboxylase, Biochemistry32,4943-4948.

PLETNEV,. V., ANTSON, . A., SINITSYNA,. I. ,DAUTER,., ISUPOV, . N. et al. (1997),Crystallo-

graphic study of tyrosine phen ol-lyase from Erwi-nia herbicola, Crystallogr. Rep. 42,809-819.

SOC.119,1809-1817.

PLETNEV,. V., ISUPOV, . N., DAUTER,., WILSON,

K. S., FALEEV,. G. e t al. (1996),Purification andcrystals of tyrosine phenol-lyase from Erwiniaherbicola, Biochem. Mol. Biol. Int. 3 8 , 3 7 4 2 ,

POCHLAUER,. (1998),Syntheses of homochiral cya-nohydrins in an industrial environment: hydroxynitrile lyases offer new options, Chimica Oggi(Chemistry Today) 16,15-19.

POHL,M., MESCH, ., RODENBROCK,A.,ULA,M. R.(1995), Stability investigations on the pyruvatedecarboxylase from Zymomonas mobilis, Bio-technol. Ap pl. Biochem. 22,95-105.

POLAK,., BRZESKI,. (1990), Isolation of the tyro-sine phenol-lyase gene from Citrobacter freundii,Biotechnol. Lett. 12,805-810.

POLLARD,. R., RICHARDSON,., AKHTAR, ., LAS-

RY , P., NEAL,T. et al. (1999), Mechanism of 3-methylaspartase probed using deuterium and sol-vent isotope effects and active-site directed re-agents: Identification of an essential cysteine res-idue, Bioorg. Med. Chem .7,949-975.

PORTER,. T. J., BRIGHT, . J. (1980), 3-Carbanionicsubstrate analogs bind very tightly to fumaraseand aspartase, J. Biol. Chem. 255,47724780.

POUPON, ., JEBAI, ., LABESSE, ., GROS,F., THI-

BAULT, J. et al. (1999),Structure modeling and sitedirected mutagenesis of the rat arom atic L-aminoacid pyridoxal 5 -phosphate dependent decar-boxylase: A functional study, Proteins Struct.Funct. Genet. 37,191-203.

PRONK,. T., STEENSMA,. Y., VA N DIJKEN,. P.(1996), Pyruvate metabolism in Saccharomycescerevisiae, Yeast 12,1607-1633.

PROSEN,., WARD,0.P. (1994), Optimization of thereaction conditions for production OF (S)-(-)-2-hydroxypropiophenone by Acinetobacter calcoa-ceticus, . Ind. Microbiol. 13,287-291.

QIAN,M. Z., EARNHARDT,.N., WADHWA,. R., Tu,C. K., LAIPIS, . J., SILVERMAN,. N. (1999), Pro-ton transfer to residues of basic pK, during catal-ysis by carbonic anhydrase, Biochim. Biophys. A c-ta 1434,l-5.

Qu, K. B., MARTIN, . L., LAWRENCE,. E. (1998),Motifs and str uctu ral fold of the cofactor bindingsite of human glutamate decarboxylase, ProteinSci. 7,1092-1105.

RAMJEE,M. K., GENSCHEL,.,ABELL, ., SMITH, .G. (1997), Escherichia coli L-aspartate-a-decar-boxylase: preprotein processing and observationof reaction intermedaites by electrospray massspectrometry, Biochem. J. 323,661-669.

RANZI,B. M., RONCHETTI,., Russo, G., TOMA, .(1981),Stereochemistry of the introduction of th ehydrogen atom at C-2 of cis-aconitic acid duringits transformation into itaconic acid in Aspergil-lus terreus:a 2H N. M. R. approach,J. Chem. SOC. ,

Chem. Commun.,1981,1050-1051.RAZAL, . A., ELLIS,S., SINGH, ., LEWIST, . G.,



164 2 Lyases

TOWERS, . H. N. (1996), Nitrogen recycling inphenylpropanoid metabolism,Phytochemistry 41,

REID,M. F., FEWSON,. A. (1994), Molecular charac-terization of microbial alcohol dehydrogenases,Crit. Rev. M icrobiol. 20,13-56.

REES,D. G., JONES,. H. (1996), Stability of L-phen-ylalanine ammonia-lyase in aqueous solution andas the solid state in air and organic solvents,En-zym e Microb. Technol.19,282-288.

REES,D. G., JONES,. H. (1997), Activity of L-phen-ylalanine ammonia-lyase in organic solvents,Bio-chem. Biophys. Acta 1338,121-126.

REGLINSKI,., STAVELY,. J. L., TAYLOR,. L. (1998),Induction of phenylalanine ammonia lyase activ-ity and control of Sphaeropsis sapinea infection inPinus radiata by 5-chlorosalicylic acid. Eur. J . Fo-

restry Pathol. 28,153-158.REHMAN, ., MCFADDEN,. A. (1996), The conse-

quences of replacing histidine 356 in isocitratelyase from Escherichia coli, Arch. Biochem. B i-

RENAULT,., CALERO,., DELORME,., DROUAULT,S., GOUPIL-FEUILLERAT,. et al. (1998) From ge-nome to industrial application,Lnit 78,39-52.

REYNOLDS,. J., GARCIA,G. A., KOZARICH,. W.,KENYON,. L. (1988), Differential reactivity inthe processing of p-[(halomethyl)benzoyl]for-mates by benzoylformate decarboxylase, a thia-min pyrophosphate dependent enzyme,Biochem-istry 27,5530-5538.

RHEE, . K., FUCHS,. (1999), Phenyl-CoA:acceptoroxidoreductase, a membrane bound molybde-num-iron-sulfur enzyme involved in the anaero-bic metabolism of phenylalanine in the denitriy-ing bacterium, Thauera aromatica, Eur. J . Bio-chem.262,507-515.

RICHARDS,. C., SPENSER,. D. (1983), 2H NMRspectroscopy as a probe of the stereochemistry ofenzymic reactions at prochiral centres, Tetrahe-dron 39,3549-3568.

RIKIN, ., SCHWARTZBACH,.D. (1989), Translation-al regulation of the synthesis of Euglena fumaraseby light and ethanol, Plant Physiol. 90,63-69.

ROBERT-BAUDOW,., JIMENO-ABENDANO,., STOE-BER, F. (1982), D-Mannonate and D-akronate de-hydratases of Escherichia coli-K-12,Methods En-zymol . 90,288-294.

ROBINSON,. B., SINGH,M., SRERE,P. A. (1976),Structure of the prosthetic group of Klebsiellaaerogenes citrate (pro-3S)-lyase,Proc. Natl. Aca d.Sci. USA 73,1872-1876.

RONDAGS,., GERMAIN,., MARC, . (1997), Quan-tification of extracellular a-acetolactate oxidativedecarboxylation in diacetyl production by an a-

acetolactate overproducing strain of Lactococcuslactis ssp. lactis bv. diacetylactis, Biotechnol. Lett.

31-35.

OphyS. 336,309-315.

21,303-307.

RONDAGS,., GERMAIN,., MARC,., (1998), Kineticstudies on a-acetolactic acid extra and intracellu-lar oxidative decarboxylation to diacetyl by Lac-tococcus lactis ssp. lactis var. diacetylactis SD 933,Lait 78,135-143.

ROSE,A. H. (1997), Restructuring the active site offumarase for the fumarate to malate reaction, Bi-ochemistry 36,12346-12354.

ROSE,A. H. (1998), How fumarase recycles after themalate fumarate reaction. Insights into the the re-action mechanism,Biochemistry 37,17651-17658.

ROZZELL r., J. D., BENNER, . A. (1984), Stereo-chemical imperative in enzymatic decarboxyla-tions. Stereochemical course of the decarboxyla-tion catalyzed by acetoacetate decarboxylase, J.

Am . Chem.SOC.106,4937-4941.RUAN, . S.,ANANTHARAM,., CRAWFORD,.T., AM-

BUDKAR,s.v,RHEE,s.Y. et al. (1992), Identifica-tion, purification and reconstitution of OxlT, theoxalate-formate antiport protein of Oxalobacterformigenes,J . Biol. Chem.267,10537-10543.

RUSSELL, . N., Yu, Y. (1991), Stereochemical andmechanistic studies of the CDP-D-glucoseoxidor-eductase isolated from Yersiniapseudotuberculo-sis, J .Am . Chem. SOC.ll3,7777-7778.

SACCHETTINI,. C., MEININGER,., RODERICK,.,BANAZAK,. J. (1986), Purification, crystalliza-tion, and preliminary X-ray data for porcine fu-marase,J . Biol. Chem. 261,15183-15185.

SACHS,M. M., SUBBAIAH,. C., SAAB,. N. (1996),Anaerobic gene expression and flooding toler-ance in maize,J. Exp. Bot. 47,l-15.

SACKSTEDER,. A., MORELL,. C., WANDERS,. J.A., MATALON,., GOULD, . J. (1999), MCD en-codes peroxisomal and cytoplasmic forms of mal-onyl-CoA decarboxylase and is mutated in malo-nyl-CoA decarboxylase deficiency, . Biol. Chem.

SALVARREY,. S., CAZZULO,. J., CANNATA,. J. B.(1995), Effects of divalent cations and nucleotideson the (C0,)-'4C-oxaloacetate exchange cata-lyzed by the phosphoenol pyruvate carboxyki-nase from the moderate halophile, Vibrio costico-la, Biochem. M ol. Biol. Int. 36,1225-1234.

SANO, ,TADA,T,MORIYAMA,., OGAWA,., ASAI,K. et al. (1997), Isolation of a rat histidase cDNAsequence and expression in Escherichia coli - Ev-idence of extrahepatic/epidermal distribution,Eur.J. Biochem. 250,212-221.

SANTANGELO,. D., DUERRE, . (1996), Microbialproduction of acetone and butanol: Can historybe repeated? Chimica Oggi (Chemistry Today)

SANTA-MARIA,. R., COLL,R., FUENTESPINA,,(1993), Comparative study of two commericalkits for determining oxalate in urine, Clin. Bio-chem.26,93-96.

SANTANIELLO,., MANZOCCHI,., BIONDI, . A.

274,24461-24468.

14,29-35.



5 References 165

(1981), O n the stereochemistry of the decarboxy-lation of (2s)-histidine catalyzed by histidine de-carboxylase from Clostridium welchii (E C4.1.1.22), . Chem. SOC.,Perkin Trans. I , 307-309.

SARIBAS,. S., SCHINDLER,. F., VIOL A, . E. (1994),Mutagenic investigation of conserved functionalamino acids in Escherichia coli L-aspartase. J .

Biol. Che m.269,6313-6319.SAUER, .,DUERRE,. 1995),Differential induction

of genes related to solvent formation during theshift from acidogenesis to solventogenesis in con-tinuous culture of Clostridium acetobutylicum,FEMS M icrobiol. Lett. 125,115-120.

SCAPIN, ., BLANCHARD,. S. (1998),Enzymology ofbacterial lysine biosynthesis, Adv. Enzymol . 72,

SCHELLENBERGER,. (1998), Sixty years of thiamindiphosp hate biochemistry, Biochim. Biophys. Ac -ta. 1358,177-186.

SCHINDLER,. F., VIOLA,R. E. (1994), Mechanism-based inactivation of L-aspartase fro m E. coli, Bi-ochemistry 33,9365-9370.

SCHMID,M., BERG,M., HILBI,H., DIMROTH,.(1996), Malonate decarboxylase of Klebsiellapneumoniae catalyzes the turnover of acetyl andmalonyl thioester residues on a coenzyme-A likeprosthetic group, Eur. J . Biochem. 237,221-228.

SCHNEIDER,., MOHAMED, . E., FUCHS,G. (1997),Anaerobic metabolism of L-phenylalanine viabenzoyl-CoA in the denitrifying bacteriumThauera aromatica, Arch. Microbiol. 168,310-320.

SCHORKEN,.,SPRENGER,. A . (1998),Thiamin-de-penden t enzym es as catalysts in chemoenzymaticsyntheses, Biochim. Biophys. Acta. 1385,229-243.

SCHUSTER,., RETEY,. (1994),Serine-202 is the pu-tative precursor of the active-site dehydroalanineof phenylalanine ammonia-lyase - Site directedmutagenesis studies on t he enzym e from parsley(Petroselinum crispum L.), FEBS Lett. 349,252-254.

SCHUSTER,., RETEY,. (1995), The mechanism ofaction of phenylalanine amm onia-lyase - he role

of prosthetic dehydroalanine, Proc. Natl. Aca d.Sci. US A 92,8433-8437.SCHWARTZ,. L., CHANDONIA,. M., KASH,S.F., KA-

NAANI, J., TUNNELL,. et al. (1999), High resolu-tion auto reactive epitope m apping and structuralmodeling of the 65 kDa form of human glutamicacid decarboxylase,J. Mol. Biol. 287,983-999.

SCHWEDE,. F., RETEY, ., SCHULZ, . E. (1999),Crystal structure of histidine ammonia lyase re-vealing a novel polypeptide modification as thecatalytic electrophile, Biochemistry 38, 5355-5361.

SCOPES,. K. (1997),Eth ano l from biomass:The po-tential use of thermophilic organisms in fermen-tation, Australasian Biotechnol. 7,296-299.

279-327.

SEELY,. E. , PEGG,A.E. (1983),Ornith ine decarbox-ylase (mouse kidney), Methods Enzymol. 94,158-161.

SEELY,. E., POSO,H., PEGG, . E . (1983), Labelingand quantitation of ornithine decarboxylase pro-tein by reaction with a-[5-'4C]difluoromethylor-nithine, Methods Enzym ol. 94,206-209.

SENUMA, ., OTSUKI,.,SAKATA, ., FURUI, ., To-SA, T. (1989) Industrial production of D-asparticacid an d L-alanine from DL-aspartic acid using apressurized column reactor containing immobi-lized Pseudomonas dacunhae cells,J . Ferment. Bi-oeng. 67,233-237.

SENWO, Z. N., TABATABAI,. A . (1996), Aspartaseactivity of soils.Soil Sci. Soc. A m . J . 60,1416-1422.

SENWO, . N., TABATABAI,. A . (1999), Aspartaseactivity in soils: effects of trace elem ents and rela-tionships to othe r am idohydrolases, Soil Biol. Bi-ochem. 31,213-219.

SEREBRENNIKOV,. M., KISRIEVA,. S., ZAGUSTINA,N. A., SEMINIKHINA,. F., BEZBORODOV, A. M.(1999),Th e ability of diacetyl-producing lactic ac-id bacteria of the genus Lactococcus to release a-

acetolactic acid into the medium, Ap pl. Biochem.Microbiol. 35,612-620.

SHARMA,., NATH,R., THIND,. K. (1993), Recentadvances in the m easurem ent of o xalate in bio-logical materials, Scanning Microscopy 7,431-441.

SHELP, . J., B O W , A. W., MCLEAN,M. D . (1999),Metabolism and functions of y-aminobutyric ac-id , Trends Plant Sci. 4,446-452.

SHEN, . W., HENNIG, ., HOHENESTER,., JANSON-

IUS, J. N., SCHIRMER,T.1998),Crystal structure ofhuman recombinant ornithine aminotransferase,J . Mol. Biol. 277,81-102.

SHERIDAN,. P., ALLEN,. C. (1981),The active siteelectrostatic potential of human carbonic anhy-drase, J . Am. Chem. SOC.103,1544-1550.

SHI,W., KIDD,R., GIORGIANNI,., SCHIND LER,. F.,VIOLA,R. E., FARBER, . K. (1993), Crystalliza-tion an d preliminary X -ray studies of i--aspartasefrom Escherichia coli, J . Mol. Biol. 234, 1248-1249.

SHI,W. X., DUNBAR,., JAYASEKERA,M. M. K., VIO-LA, R. E.,FARBER, .K. (1997),The structure of L-

aspartate ammonia-lyase from Escherichia coli,Biochemistry 36,9136-9144.

SHIBATA,., GARDINER,. E. , SCHWARTZBACH,. D.(1985), Purification and characterization and im-munological properties of fumarase form Eugle-na gracilis var. bacillaris, J . Bacteriol. 164, 762-768.

SHIN,H. S., ROGERS,. L . (1996a), Production of L-

phenylacetylcarbinol (L-PAC) from benzalde-hyde using partially purified pyruvate decarboxy-lase (PDC),Biotechnol. Bioeng. 4 9 , 5 2 4 2 .

SHIN,H . S. ,ROGERS,. L. (1996b),Kinetic evaluation



166 2 Lyases

of biotransformation of benzaldehyde to L-phen-ylacetylcarbinol by immobilized pyruvate decar-boxylase from Candida utilis, Biotechnol. Bioeng.49,429-436.

SHUKLA,. B., KULKARNI,. R. (1999), Downstreamprocessing of biotransformation broth for recov-ery and purification of L-phenyl acetyl carbinol(L-PAC),~ci. Ind. Res. 58,591-593.

SIDDIQUI, . A., FUJIWARA,., IMANAKA,. (1997),Indolepyruvate ferredoxin oxidoreductase fromPyrococcus sp. KODl possesses a mosaic: Struc-ture showing features of various oxidoreductases,Mol. Gen. Genet. 254,433439.

SIDDIQUI, . A., FUJIWARA,., TAKAGI,M., IMANA-KA,T. (1998), In vitro heat effect on heterooligo-meric subunit assembly of thermostable indole-pyruvate ferredoxin oxidoreductase, FEBS Lett.

SIDHU, ., ALLISON, ., PECK,A. B. (1997), Identifi-cation and classification of Oxalobacter form i-genes strains using oligonucleotide probes andprimers,J. Clin. Microbiol. 35,35&353.

SIDHU,H., HOPPE,B., HESSE,A., TENBROCK, .,BROMME,. et al. (1998), Absence of Oxalobacterformigenes in cystic fibrosis patients: a risk factorfor hyperoxaluria,Lancet 352, No. 9133,1026-1029.

SIDHU,H., SCHMIDT, . E., CORNELIUS,. G., THA-MILSELVAN,., KHAN,S. R. et al. (1999), Directcorrelation between hyperoxaluria/oxalate stonedisease and the absence oif the gastrointestinaltract dwelling bacterium Oxalobacter formige nes,J . Am. . SOC.Nephrol. 10, S334-S340.

SIMPSON,., BATEMAN,., DRIESSEN, ., LINDLEY,P., Moss, D. et al. (1994), The structure of avianeye lens Scrystallin reveals a new fold for asuperfamily of oligomeric enzymes,Nature Struct.

Biol. 1,724-733.SLOAN, . J., PHILLIPS,. S. (1996), Effects of a-deu-

teration and of aza and thia analogs of L-trypto-phan on formation of intermedaites in the reac-tion of Escherichia coli tryptophan indole lyase,Biochemistry 35,16165-16173.

SMITH,D. M., THOMAS, . R., GANI,D. (1991), Acomparison of the pyridoxal 5'-phosphate de-pendent decarboxylase and transaminase en-zymes at a molecular level,Experientia 47,ll-12.

SONG,Y. ., NIEDERER,., LANEBELL,. M., LAM,L.K. P., CRAWLEY,.et al. (1994), Stereospecific syn-thesis of phosphonate analogs of diaminopimelicacid (DAP) their interaction with DAP enzymesand antibacterial activity of peptide derivatives, .

Org. Chem .59,57865793.SOOKKHEO,., PHUTAKUL,., CHEN, .T., WANG,K.

T. (1998), Aspartase catalyzed preparative scalesynthesis of "N-aspartic acid,J . Chin. Chem. SOC.

SNELL, . E. (1986), Pyruvoyl-dependent histidine-decarboxylase from Lactobacillus-30a- purifica-

434,372-376.

45,525-528.

tion and properties, Methods Enzym ol. 122,128-135.

SNELL, . E. (1993), From bacterial nutrition to en-zyme structure - A personal odyssey, Ann. Rev.Biochem. 62,l-27.

SNELL, .E., GUIRARD,. M. (1986), Pyridoxal phos-phate-dependent histidine decarboxylase fromMorganella am-15,Methods Enzym ol. 122, 139-143.

SNELL, . E., HUYNH,Q. K. (1986), Prohistidine de-carboxylase from Lactobacillus-30a, M ethods En-zymol. 122,135-138.

SNIPES, . E., BRILLINGER,.-U., SELLERS,., MAS-CARO,L., FLOSS, . G. (1977), Stereochemistry ofthe dTDP-glucose oxidoreductase reaction, J .

Biol. Chem. 252,8113-8117.SOURKES,. L. (1987),Aromatic-~-amino cid decar-

boxylase,Methods Enzym ol. 142,170-178.SPAMPINATO,. P., ANDREO, . S. (1995), Kinetic

mechanism of NADP-malic enzyme from maizeleaves,Photosynth. Res. 43,l-9.

SPENCER,. B., JORDAN,. M. (1990), Use of chiralmalonates to determine the absolute configura-tion of the hydrogen atoms eliminated during theformation of 6-methylsalicyclic acid by 6-methyl-salicyclic acid synthase from Penicillin patulum , J .

Chem. SOC.,Chem. Comm un.,1704-1706.SPENCER,. B., JORDAN,.M. (1992a), Stereospecific

elimination of hydrogen atoms with opposite ab-solute orientations during the biosynthesis of or-sellinic acid from chiral malonates in Penicillincyclopium, J . Chem. Soc., Chem. Comm un.,646648.

SPENCER,. B., JORDAN,.M. (1992b), Investigationof the mechanism and steric course of the reac-tion catalyzed by 6-methylsalicylic acid synthasefrom Penicillium patulu m using (R)-[l-C-13,2-H-2lmalonate and (S)-[l-C-13,2-H-2]malonate,io -chemistry 31,9107-91 16.

SPRENGER,. A., POHL,M. (1999), Synthetic poten-tial of thiamin diphosphate-dependent enzymes,J . Mol. Catal. B Enzym atic 6,145-159.

STIM-HERNDON,. P., PETERSEN,. J., BENNETT, .N. (1995), Characterization of an acetyl-CoA C-acetyltransferase (thiolase) gene from Clostridi-um acetobutylicum ATCC-824,Gene 154,8145.

STIM-HERNDON,. P., NAIR,R., PAPOUTSAKIS,. T.,BENNETT,G. N. (1996), Analysis of degeneratevariants of Clostridium acetobutylicum ATCC-824,Anaerobe 2,ll-18.

STIRLING,. I. (1992), The use of aminotransferasesfor the production of chiral amino acids andamines, in: Chirality in Industry (COLLINS,. N.,SHELDRAKE,. N., CROSBY., Eds.), pp. 208-222.Chichester: John Wiley & Sons.

Su, P. H., LIN,C. H. (1996), Metabolic responses ofluffa roots to long-term flooding,J. Plant Physiol.148,735-740.



5 References 167

lalanine by L-phenylalanine ammonia-lyase of

Rhodotorula glutinis: parameters and kinetics,Enzym e M icrob. Technol.17,445452.

TAKAHASHI,., KAWASAKI,., SONE,H., YAMANO,

S. (1995), Genetic modification of brewers’ yeastto produce acetolactate decarboxylase and the

safety aspects of the beer brewed by the trans-formed yeast,ACS Symp. Sex 605,171-180.

TAKAMURA,., KITAYAMA,. (1981), Purification

and some properties of the malonate decarboxy-lase from Pseudomonas ovalis: an oligomeric en-

zyme with bifunctional properties, Biochem. Int.3,483491.

TAKATSUKA,., ONODA, ., SUGNAMA,T.,OMITA,T.,KAMmo,Y. (1999a), Novel characteristicsof Selen-omonas ruminantiurn lysine decarboxylase ca-

pable of decarboxylatingboth L-lysine and L-orni-thine,Biosci. Biotech. Biochem. 63,1063-1069.

TAKATSUKA,., TOMITA, ., KAMINO, Y (1999b),

Identification of the amino acids residues confer-

ring specificity upon Selenomonas ruminantiurnlysine decarboxylase, Biosci. Biotech. Biochem.

TANAKA,., TOSA, ., CHIBATA,. (1983a), Stabiliza-tion of fumarase activity of Brevibacterium fla-

vu m cells by immobilization with K-carrageenan,Ap pl. Biochem. Biotechnol. 8,31-38.

TANAKA,., TOSA, ., CHIBATA,. (1983b), Reasonsfor the high-stability of fumarase activity of Brev-ibacterium flavu m cells immobilized with K-carra-

geenan gel,Ap pl. Biochem . Biotechnol.

8,39-54.TANAKA,., TOSA, ., CHIBATA,. (1983c), Effect of

growth-phase on stability of fumarase activity of

Brevibacterium jZavum cells immobilized with K-

carrageenan,Agric. Biol. Chem . 47,1289-1296.

TANAKA,., TosA,T., CHIBATA,. (1984), Stability offumarase activity of Brevibacterium flavum im-mobilized with K-carrageenan and Chinese gallo-tannin,Ap pl. Microbiol. Biotechnol. 19,85-90.

TANAKA,.,LERNER,. A., BARBAS,.F. (1999),Thi-

azolium-dependentcatalytic antibodies producedusing a covalent modification strategy, Chem.Commun. 15,1383-1384.

TARI,L. W., MATTE, ., PUGAZHENTHI,.,

GOLDIE,H., DELBAERE,.T. J. (1996), Snapshot of an en-zyme reaction intermediate in the structure ofthe

ATP-Mg2+-oxalate ternary complex of Escheri-chia coli PEP carboxykinase,Nature S truct. Bio l.

TAYLOR, . G., LAMBERT,. A., SEXSMXH,.,SAD-LER, . J., RAY,P. N. et al. (1990), Cloning and ex-pression of rat histidase, J. Biol. Chem. 265,

TAYLOR,. G., LEVY,H. L., MCINNES,. R. (1991),Histidase and histidinemia- clinical and molecu-lar considerations,Mol. Biol. Med. 8,101-116.

TAYLOR, . G., MCINNES, . R. (1994), Site-directedmutagenesis of conserved serines in rat histidase-

63,1843-1846.

3,355-363.

18192-181 99.

SUCHI,M., HARADA,., WADA,~ .,AKAGI,~.1993),Molecular cloning of a cDNA-encoding humanhistidase,Biochem. Biophys. Acta 1216,293-295.

SUN, . X.,DUGGLEBY,. G., SCHOWEN,R.L.1995),Linkage of catalysis and regulation in enzyme ac-tion - carbon-isotope effects, solvent isotope ef-fects, and proton inventories for the unregulated

pyruvate decarboxylase of Zymom onas m obilis,J. Am . Chem. SOC.117,7317-7322.

SUN, . X., SMITH, . S., O’LEARY, . H., SCHOWEN,R. L. (1997), The linkage of catalysis and regula-tion in enzyme action. Fluoropyruvate as a probeof regulation in pyruvate decarboxylases,J. A m .Chem.SOC.119,1507-1515.

SUN, .X., ZABINSKI,.F.,TONEY, . D. (1998a), Re-actions of alternate substrates demonstrate ster-

eoelectronic control of reactivity in dialkylglycinedecarboxylase,Biochemistry 37,3865-3875.

SUN,S. X., BAGDASSARIAN,. K., TONEY,M. D.(1998b), Pre-steady-state kinetic analysis of thereactions of alternative substrates with dialkyl-

glycine decarboxylase, Biochemistry 37, 3876-3885.

SUNDARARAJU,., ANTSON,A. A., PHILLIPS, . S.,DEMIDKINA,.V., BARBOLINA,. V. et al. (1997),The crystal structure of Citrobacter freundii tyro-sine phenol-lyase complexed with 3-(4 ’-hydroxy-pheny1)propionicacid, together with site directed

mutagenesis and kinetic analysis, demonstratesthat arginine 381 is required for substrate speci-

ficity, Biochemistry 36,6502-6510.SUTTON,C. R., KING,H. G (1962), Inhibition of leu-

cine decarboxylase by thiol-binding reagents,

Arch. Biochem . Biophys. 96,360-370.SUZUKI,T., SATO, M., YOSHIDA,., TUBOI, S . (1989),

Rat liver mitochondria1 and cystolic fumaraseswith identical amino acids sequences are encodedfrom a single gene, J. Biol. Chem. 264,2581-2586.

SUZUKI,., HIRAHARA,., SHIM,. K., HORINOUCHI,

S . , BEPPU,T.1992), Purification and properties ofthermostable P-tyrosinase from an obligately

symbiotic thermophile,Symbiobacterium thermo-

ph ilum , Biosci. Biotech. Biochem. 56,84-89.SWANSON,., BROOKS,H. B., OSTERMAN,. L.,

O’LEARY,M. H., PHILLIPS,. A. (1998), Carbon-13 isotope effect studies of Trypanosoma bruceiornithine decarboxylase,Biochemistry 37,14943-14947.

TABIB,A., ACHRACH,. (1999), Role of polyamines

in mediating malignant transformation and onco-gene expression, Int. J. Biochem. Cell Biol. 31,

TADEGE,M., DUPUIS, ., KUHLEMEIER,. (1999),Ethanolic fermentation: new functions for an old

pathway, Trends Plant Sci. 4,32Q-325.

TAKAC,., AKAY, ULENT,OZDAMAR,. H. (1995),Bioconversion of trans-cinnamic acid to L-pheny-

1289-1295.



168 2 Lyases

identification of serine-254 as an essential active-

site residue,J . Biol. Chem. 269,27473-27477.

THATCHER,. J.,DOHERTY,A..,ORVISKY,.,MAR-

TIN, B. M., HENKIN, . I. (1998), Gustin from hu-

man parotid saliva is carbonic anhydrase VI, Bio-chem. Biophys. Res. Cornmun.250,635-641.

THOMPSON,. W., STROHL, .R., LOSS,.G. (1992)

Purification and characterization of TDP-D-glU-

cose 4,6-dehydratase from anthracycline-produc-

ing streptomycetes, J . Gen. Microbiol. 138,779-

786.

THORSON,. S., KELLY,T. M., LIU, H.-W. (1994a),

Cloning, sequencing and overexpression in Es-cherichia coli of the a-D-glucose-1-phosphatecy-

tidylyltransferase gene isolated from Yersiniapseudotuberculosis, J . Bacteriol. 176,1840-1849.

THORSON,. S.,Lo, . F.,PLOUX, .,HE,X. M., Lnr, H.

W. (1994b), Studiesof the biosynthesis of 3,bdide-

oxyhexoses-molecular cloning and characteriza-

tion of the ASC (ascrylose) region from Yersiniapseudotuberculosis, . Bacteriol. 176,5483-5493.

TILLEY, ., AKHTAR,M., GANI,D. (1992) The stereo-

chemical course of reactions catalyzed by Esche-richia coli glutamic acid decarboxylase,J . Chem.Soc., Chem. Comm un.,68-71.

TILLEY, ., AKHTAR, ., GANI,D. (1994),The stereo-

chemical course of decarboxylation, transamina-

tion and elimination reactions catalyzed by Es-cherichia coli glutamic acid decarboxylase, J .

Chem . Soc., Perkin. Trans.11994,3079-3087.

TOH,H., KONDO, ., TANABE,. (1993), Molecular

evolution of biotin-dependent carboxylases,Eur.J . Biochem. 215,687496,

TONEY,. D., HOHENESTER,., KELLER,. W., JAN-

SONIUS, J. N. (1995), Structural and mechanistic

analysis of 2 refined crystal structures of the pyri-

doxal phosphate dependent enzyme, dialkylgly-

cine decarboxylase,J . Mol. Biol.245,151-179.

TORSELL, . B. G. (1983),Natural Product Chemis-try: A Mechanistic and B iosynthetic Appro ach toSecondary Metabolism, pp. 80-88. Chichester:

John Wiley & Sons.

TOSA,T., TANAKA,., CHIBATA,. (1982), Stabiliza-

tion of fumarase activity of Brevibacterium flu-vu m cells by immobilization with K-carrageenan

and polyethyleneimine,Enzym e Eng. 6,237-238.

TOTH,C., COFFINO,? (1999), Regulated degradation

of yeast ornithine decarboxylase, J . Biol. Chem.274,25921-25926.

TOTH,., ISMAIEL, . A., CHEN,. S. (1999), The ald

gene, encoding a coenzyme A-acylating aldehyde

dehydrogenase, distinguishes Clostridium beije-rinckii and tw o other solvent producing clostridia

from Clostridium acetobutylicum, Ap pl. E nviron.Microbiol. 65,4973-4980.

TRAMPER,. VERMUE,M. H., BEEFTINK,H. H., VO N

STOCKAR,. (Eds.) (1992), Biocatalysis in Non-

conventional Med ia.Amsterdam: Elsevier.

TRIPATHI,. M., AGARWAL,. C., BASU, . K. (1997),

Production of L-phenylacetylcarbinolby fermen-

tation,J. Ferment. Bioeng. 84,487492.

TSENG, . P. (1997), Regulation of fumarase (furnB)gene expression in Escherichia coli in response to

oxygen, iron and heme availability: role of the ar-cA,fur,and hemA gene products,FEMS Microbi-01. Lett .157,67-72.

Tsou, A. Y., RANSOM,. C., GERLT,.A., BEUCHTER,

D. D., BABBITT,. C., KENYON,. L. (1990), Man-

delate pathway of Pseudomonas putida - se-

quence relationships involving mandelate race-

mase, (S)-mandelate dehydrogenase and benzyl-

formate decarboxylase and expression of benzyl-

formate decarboxylase in Escherichia coli, Bio-chemistry 29,9856-9862.

TUBOI, . , SUZUKI,., SATO,M., YOSHIDA,. (1990),

Rat liver mitochondria1 and cytosolic fumarases

with identical amino acid sequences are encoded

from a single mRNA with two alternative in-

phase AUG initiation sites, Adv. En zyme Regul.

TURNER, . A., SIMPSON,., MCINNES, . R., How-

ELL, €! L. (1997) Human arginosuccinate lyase: a

structural basis for intragenic combination, Proc.Natl. Aca d. Sci. US A.94,9063-9068.

TYAGI, . K., TABOR, .W., TABOR, . (1983), Orni-

thine decarboxylase (Saccharomyces cerevisiae),Methods Enzymol. 94,135-139.

UEDA,Y., YUMOTO, ., TOKUSHIGE,., FUKUI, .,

OHYA-NISHIGUCHI,. (1991), Purification and

characterization of two types of fumarases from

Escherichia coli, J . Biochem. 109,728-733.

UGWUANYI,. O., OBETA,. A. N. (1997), Some pec-

tinolytic and cellulolytic enzyme activities of fun-

gi causing rots of cocoyams, J . Sci. Food Agr. 73,

432436.

URBAN,. J., VO N TERSCH, . L. (1999), A computa-

tional study of charge delocalization and ring flu-

or0 substituent effects in 4-fluoromethylphenox-

ides,J. Org. Chem. 64,3409-3416.

VACCARO,. A., CRANE, . J.,HARRIS,T. ., WASHA-

BAUGH, . W. (1995), Mechanism of reconstitu-

tion of brewers’ yeast pyruvate decarboxylase

with thiamin diphosphate and magnesium, Bio-chemistry 34,12636-12644.

VALLEE,.,TURNER, . A., LINDLEY,? L.,HOWELL,

P.L. (1999), Crystal structure of an inactive duck

8-11-crystallin mutant with bound arginosucci-

nate, Biochemistry 38,2425-2434.

VAN DEN HOMBERGH,. P.T. w., FRAISSINET-TACHET,

L., VAN DE VONDERVOORT,! J. I., VISSER, .

(1997), Production of the homologous pectin

lyase B protein in six genetically defined protease

deficient Aspergillus niger mutant strains, Curr.Genet. 32,7341.

VAN DER WERF,M., VA N DEN TWEEL,W. J. J., HART-

MANS, S. (1992), Screening for microorganisms

30,289-304.



5 References 169

WATA NABE, . K., HERN AND EZ-VELAZ CO,.,ITURBE-CHINAS,., LOPEZ-MUNGIA,. (1992),Phenylalanine ammonia lyase from Sporidiobo-lus pararoseus and Rhodosporidium toruloides:Application for phenylalanine and tyrosine de-amination, World J . Microbiol. Biotechnol. 8,406-410.

WEAVER,., BANASZAK,. (1996), Crystallographicstudies of the catalytic and a second site in fuma-rase C from Escherichia c oli, Biochemistry, 35,

WEAVER,., LEES,M., BANASZAK,. (1997) Muta-tions of fum arase that distinguish between th e ac-tive site and a nearby dicarboxylic acid bindingsite, Protein Sci. 6,834-842.

WEA VER,T. ., LEV^, D. G., DONN ELLY,. I., STE-VENS, P. P.W., BA NA SZA K,., (1995) The multisub -unit active-site of fumarase-C from Escherichiacoli, Nature S truct. Biol. 2,654-662.

WEAVER,., LEES, M., ZAITSEV , ., ZAITSEV A,.,DUKE, . et al. (1998) Crystal structu res of nativeand recombinant yeast fumarase, J. Mol. Biol.280,431442.

WEIGEL,T.M, MILLER,^. P.,LIU,H.-W. (1992), Me-chanistic and stereochemical studies of a uniquedehydration catalyzed by CDP-4-keto-6-deoxy-~-glucose-3-dehydrase - a pyridoxamine 5 -

phosphate dependent enzyme isolated from Yer-sinia pseudotuberculosis, Biochemistry 31, 2140-2147.

WEISS,! M., G ARC IA, . A., K ENYO N, . L., CLE-LAND, W. W., COOK, P. F. (1988), Kinetics andmechanism of benzoylformate decarboxylase us-ing C-13 and solvent deuterium effects on benzyl-formate an d benzylformate analogs,Biochemistry

WESTHEIMER,. H. (1995), Coincidences,decarboxy-lation and electrostatic effects, Tetrahedron 51,3-20.

WHITE, . J., KENDRICK,. E . (1993), Inactivation ofhistidine ammo nia-lyase from Streptomyces gri-seus by dicarbonyl reagents, Biochinz. Biophys.Acta 1163,273-279.

WIECZOREK,. J., KALIDOVA, . A., CLIFTON,. G.,RINGE,D., PE TSKO , . A., GERLT,.A. (1999), Ev-olution of enzy me activities in the enolase super-family: Identification of a “new” general acid cat-alyst in the active site of D-galactonate dehyd ra-tase from Escherichia col i,J. Am . Chem. SOC. 21,4540-4541.

WIGHTMAN,. H., STAUNTON,., BAITERSBY,. R .(1972), Studies of enzym e-med iated reactions.Part 1.Syntheses of deuterium or tritium-labeled(3s)- and (3R)-phenylalanines: Stereochemicalcourse of the elimination catalyzed by L-pheny-lalanine ammonia-lyase, J . Chem. SOC.,PerkinTrans.I , 2355-2364.

WILCOCKS,., WARD ,0.P. (1992), Fasctors affecting

13955-13965.

27,2197-2205.

producing D-malate from m aleate,Ap pl. Environ.Microbiol. 58,2854-2860.

MANS, S. (1993), Purification and characteriza tionof maleate hydratase from Pseudomonas pseu do-alcaligenes, Ap pl. Environ. Micro biol. 59, 2823-2829.

(1994), Expression of the Butyrivibrio fibrisol-vens endo-P-1,4-glucanasegene together with theErwinia pectate lyase and polygalacturonasegenes in Saccharomyces cerevisiae, Curr. Genet.

VAN WAARDE,A. (1991) Alcoholic fermentation inmulticellularorganisms,Physiol. Zool.64,895-920.

Various (1998), Biochim. Biophys. Acta 1385, 175-419.

VEIGA,L. A., GUIMARAES,. F. (1991), L-Fuconatedehydratase - purification and properties of theenzyme from Pullularia pullulans, Arquivos deBiologia e Tecnologia 34,536-553.

VOLTATTORNI,. B., GIARTOSIO, ., TWRANO , .(1987), Aromatic-L-amino acid decarboxylasefrom pig-kidney,Methods Enzyrnol. 142,179-187.

VO N TERSCH, . L., SECUNDO,., PHILLIPS, . S . ,

NEWTON,M. G. (1996), Prep aration of fluorinatedamino acids with tyrosine phenol lyase - Effectsof fluorination on reaction kinetics and mecha-nism of tyrosine phenol lyase and tyrosine pro-tein kinase Csk,AC S Sym p. Ser. 639,95-104.

WAAGEPETERSEN,. S., SONNEWALD,., SCHOUS-BOE, A. (1999), The GA BA paradox: Multipleroles as metabolite, neurotransmitter and ne uro-differentitation agent, J. Neurochem. 73, 1335-1342.

WADA,M., SAKAMOTO,., KATAOKA, ., LIU ,J. Q.,YAMA DA, ., SHIMIZU,. (1998), Distribution ofthreonine aldolase activity with different stereo-specificities in aerobic bacteria, Biosci. Biotech.Biochem. 62,1586-1588.

WAGNER, . G., HASSLACHER,., GRIENGL,K.,SCHWAB , ., KRATKY,. (1996), Mechanism ofcyanogenesis: The crystal structure of hydroxyni-trile lyase from Hevea brasiliensis, Structure 4,

WAJANT,H., EFFENBERGER,. (1996), Hydroxyni-trile lyases of higher plants, Biol. Chem. Hop pe-Seyler 377,611617.

WANG, .-F.,GABRIEL, . (1970), Biological mech a-nisms involved in the formation of deoxysugars,J .Biol. Chem. 245,8-14.

WANG, . H., GO NG , . S.,TSAO , .T. (1998), Produc-tion of L-malic acid during biocatalysis employingwild-type and respiratory deficient yeast, Appl .Biochem. Biotechnol. 70-2,845-852.

WARD,0 . ., (1995), Application of bakers’ yeast inbioorganic synthesis, Can. J . Bot. 73, S1043-S1048.

VA N DER WERF, M. J., VA N DEN TWEEL,w. J.J., HART-

VAN RENSBURG,., VANZYL, w. H., PRETORIUS,1. s.

27,17-22.

811-822.



170 2 Lyases

2-hydroxypropiophenone fromation by benzoylformate decarboxylase from Pseudomonas puti-da , Biotechnol. Bioeng. 39,1058-1063.

WILCOCKS,.,WARD, 0.P., COLLINS,., DEWDNEY,N. J., H o N G , ~ .., PRO SEN , . (1992), Acyloin for-mation by benzylformate decarboxylase fromPseudomonas pu tida, Ap pl. Environ. Microbiol.

WILLIAMS,. M., YUAN,C. G. (1992), Asymm etricsynthes is of 2,6-diam inopim elic acids, J. Org.Chem. 57,6519-6527.

WILLIAMS,. E., WOO LRIDGE,. M., RANSOM,. C.,LANDRO,. A., BABBITT, . C., KOZARICH,. W.(1992), 3-Carboxy-cis,cis-muconate lactonizingenzyme from Pseudomonas putida is homologousto the class-I1 fumarase family: A new reaction inthe evolution of a mechanistic motif, Biochemis-try 31,9768-9776.

WILLIAMSON,. M. (1985),L-Aspartate-a-decarboxy-lase, Methods Enzym ol. 113,589-595.

WITSCH EL, ., EG LI,T. 1997), Purification and char-acterization of a lyase from the EDT A-degradingbacterial strain D SM 9103 that catalyzes the split-ting of [S,S]-ethylenediaminedisuccinate, struc-tural isomer of EDTA, Biodegradation 8,4 19428.

WITSCHEL,., NAG EL, .,EGLI,T. (1997), Iden tifica-tion and characterization of the two enzymesystem catalysing oxidation of E DT A in the E D -TA-degrading bacterial strain D SM 9103,J . Bac-teriol.

179,6937-6943.WOEHL,A., D u” , M. F. (1999), Mechanisms ofmonovalent cation action in enzyme catalysis:The tryptophan synthase a+, and @reactions,Biochemistry 38,7131-7141.

WOODRELL,. D., ICEHAYOVA,. D., JAIN ,A. 1999),Photochemically-triggered decarboxylatiodde-amination of o-nitrodimethoxyphenylglycine,

Org. Lett. 1,619-621.WOODS, . R. (1995),The genetic engineering of mi-

crobial solvent production, Trends Biotechnol. W,

WOODS, .A., GUEST ,. R. (1987), Differen tial roles

of the Escherichia coli fumarases and FNR-de-pend ent expression of fumarase B and aspartase.FEMS Micro biol. Lett. 48,219-224.

WOODS, . A., MILES,. S., ROBERTS,. E., GUEST ,.R. (1986), Structural and func tional relationshipsbetween fumarase and aspartase nucleotide se-quences of the fum arase (FumC) and aspartase (-Asp A) gen es of Escherichia coli-K12,Biochem. J .

WOO DS, . A., M ILES,. S., GUEST,. R. (1988a), Se-quence homologies between argininosuccinase,aspartase and fumarase - family of structurally-related enzymes, FEMS Microbiol. Lett. 51,

WOODS,S., SCHWARTZBACH,. D., GUEST, . R.

58,1699-1704.

29-264.

237,547-557.

181-186.

(1988b), Two biochem ically distinct classes of fu-marase in Escherichia coli, Biochem. B iophys. Ac -ta . 954,14-26.

WU, M., TZAGOLOFF,. (1987), Mitochondria1 andcytoplasmic fumarases in Saccharomyces cerevi-siae are encoded by a single nuclear gene FUM 1,J. Biol. Chem.262,12275-12282.

Wu, P. C., KRO ENING,. A., W HITE,P. J., KEND RICK,K. E. (1992), Histidine ammonia-lyase fromStreptomyces griseus, Gene 115,19-25.

Wu, C.-Y., LEE ,H.-W., W u, S.-H., CHE N, .-T., CH I-OU,S.-H., CH AN G, .-G. (1998), Chem ical mecha-nism of the endogenous argininosuccinate lyaseactivity of duck lens B -crystallin, Biochem. J. 333,

YA GASHI, ., OZAKI,A. (1998), Industrial biotrans-forma tions for the production of D-amino acids,J.Mol. Catal. B 4,l-11.

YAHIRO, ., TAKAHAMA,., PARK, . S., OKABE,M.(1995), Breeding of Aspergillus terreus TN-484for itaconic acid produ ction with high yield,J. Fer-ment. Bioeng. 79,506-508.

YAHIRO, ., SHIBATA,.,JIA,S. R., PA RK , ., OKABE,M. (1997), Efficient itaconic acid produc tion fromraw corn starch,J. Fermet. Bioeng . 84,375-377.

YAMADA, ., NABE,K., Izuo , N., NAKAMICHI,.,CHIBATA,. (1981), Produ ction of L-phenylala-nine from trans-cinnamic acid w ith Rhodotorulaglutinis: containing L-phenylalanine ammonia-lyase activity, Appl. Environ. Microbiol. 42,

Y A M A M O T O ,., MIWA,Y., M IYOSHI, ., FURU YA MA ,J., OHILIORI,. (1997), The Esch erichia coli ZdcCgene encodes ano ther lysine decarboxylase,prob-ably a constitutive enzyme,Gene s Genet. Systems

YAMANO,., TOMIZUKA,., TANAKA,., INOUE,.(1994), High-level expression of a-acetolactatedecarboxylase gene from Acetobacter aceti ssp.xylinum in brewing,J. Biotechnol. 37 ,4548 .

YAMANO,, TOMIZUKA,, SONE,H, IMURA, , TA-KEUCHI,T.t al. (1995),Brew ing perform ance of abrewers’ yeast having a- cetolactate decarboxy-

lase from Acetobacter aceti ssp. xylinum,J .

Bio-technol. 39,21-26.YANG, . P., TSAO,G. T. (1994), Mathm atical mo del-

ling of inhibition kinetics in acetone-bu tanol fe r-mentation by Clostridium acetobutylicum, Bio-technol. Prog. 10,532-538.

YANO,M., % M A , K., UMIJI,K., IZUI,K. (1995),Catalytic role of a n arginine residue in the highlyconserved and unique sequence of phosphoen ol-pyruvate carboxylase,J.Biochem. 117,1196-1200.

YOON,M. Y., THAYER-COOK,. A,, BERDIS, . J.,KARSTEN, . E., SCHNACK ERZ,. D., COOK, . F.(1995), Acid -base chem ical mechanism of aspar-

tase from Hafnia alvei, Arch. B iochem. Biophys.

327-334.

773-778.

72,167-172.

320,115-122.



5 References 171

the activation and inactivation of brewers' yeast

pyruvate decarboxylase investigated with a

PDCI-PDC6 fusion protein, Biochemistry 32,

2704-2709.

Z H O U , ~ ..,TONEY,M. . (1999),pH studies on the

mechanism of the pyridoxal phosphate depen-

dent dialkylglycine decarboxylase, Biochemistry38,311-320.

ZHU,M. Y., Juo~10,A.V.1995),Aromatic amino ac-

id decarboxylase - Biological characterization

and functional role, Gen. Pharmacol. 26,681496.

ZIMMER,W., HUNDESHAGEN,., NIEDERAU, .

(1994), Demonstration of the indolepyruvate de-

carboxylase gene homolog in different auxin-pro-

ducing species of the Enterobacteriaceae, Ca n J .

Microbiol. 40,1072-1076.

ZIMMER,W., WESCHE,M., TIMMERMANS,. (1998),

Identification and isolation of the indole-3-pyru-

vate decarboxylase gene from Azospirillum bras-ilense Sp7: Sequencing and functional analysis of

the gene locus,Curr. Microbiol. 36,327-331.

YOUNG, . W. (1991), Use of enzymes in the synthe-

sis of stereospecifically labeled compounds, in:

Isotopes in the Physical and Biomedical Sciences,Volume I , Labeled Compounds (Part B) (BUN-

CEL,E., JONES,. R., Eds.), pp. 341427. Amster-

dam: Elsevier.

Yu,Y., RUSSELL,.N., THORSON,. S., LIU,L. D.,LIU,

H.W. (1992), Mechanistic studies of the biosyn-

thesis of 3,6-dideoxyhexoses in Yersinia pseudo -tuberculosis - purification and stereochemical

analysis of CDP-D-glucose oxidoreductase, J .

Biol. Chem. 267,5868-5875.

YUMOTO, ., TOKUSHIGE,. (1988), Characteriza-

tion of multiple fumarase proteins in Escherichiacoli, Biochem. Biophys. Res. Commu n. 153,1236

1243.

ZAITSEV, . M., GOVORUKHINA,. I., LASKOVNEVA,

0.V., TROTSENKO,.A.

(1993), Properties of the

new obligatory oxalotrophic bacterium Bacillusoxalophilus, Microb iology 62,378-382.

Z E N G , ~ ..,FARRENKOPF,., HoHMANN,S.,DYDA,F.,

FUREY, ., JORDAN,. (1993), Role of cysteines in



3 Halocompounds

GARYK. ROBINSON

SIMON .JACKMAN

JANE STRATFORDCanterbury, UK

1 Introduction 1751.1 Introductory Rem arks 175

1.2 Distribution and Function 1751.3 Reason s for Synthesis by the Host 178

2.1 Haloge nation 1792.2 Dehalogenation 181

2 Synthetic Importanc e of Haloco mpo und Transforming Enzymes 179

3 Current Developments in Dehalogenase and Haloperoxidase Biochemistry and Genetics 1814 Mechanistic Aspects of Biohalogenation 184

4.1 Natural Organ ohalogens 1844.1.1 Chlorine and Brom ine 1844.1.2 Fluo rine 1844.1.3 Fluo roac etate 190

4.1.4 wF luorof atty Acids 1904.1.5 Fluoroaceton e 1904.1.6 Nucleocidin 1904.1.7 Fluorothreonine 1904.1.8 Iodine 190

5 Enzy mology of Biohalog enation 1925.1 Halope roxidases 192

5.1.1 Basic Mechanism 1925.1.2 Na tural Sources of Halope roxidases 1925.1.3 Halopero xidase Classes 1935.1.4 Enzyme Mechanisms 193

5.1.4.1 He me Haloperoxidase 193

5.1.4.2 Vanadium Ha lope roxid ase 1935.1.4.3 Non-Hem e Non-M etal Haloperoxidase 1955.1.4.4 Flav idH em e Halope roxidase 196





174 3 Halocompounds -Biosynthesis and Biotransformation

5.1.5 Reactions of the Haloperoxidases 1965.1.5.1 Phenols, Anilines, and other Aromatics 1965.1.5.2 Alkenes 1965.1.5.3 Alkynes 197

5.1.5.4 Cyclopropanes 1975.1.5.5 P-Diketones 1975.1.5.6 Nitrogen Atoms 1975.1.5.7 Sulfur Atoms 1975.1.5.8 Inorganic Halogens 1985.1.5.9 Stereo- and Regioselectivity 198

5.2 Halide Methyltransferase 1985.3 The Enzymes of Fluorine Metabolism 1995.4 Advances Towards the Commercial Application of Biohalogenation 201

6 Dehalogenations 2026.1 Introduction 2026.2 Aliphatic Dehalogenation 203

6.2.1 Reductive Dehalogenation 2036.2.2 Oxidative Dehalogenation 2036.2.3 Hydrolytic Dehalogenation 2046.2.4 Dehydrohalogenation 2056.2.5 Epoxide Formation 206

6.3.1 Reductive Dehalogenation 2076.3.2 Oxidative Dehalogenation 208

6.3 Aromatic Dehalogenation 207

6.3.2.1 Dioxygenase Catalyzed Dehalogenation 2086.3.2.2 Monooxygenase Catalyzed Dehalogenation 210

6.3.3 Hydrolytic Dehalogenation 210

6.3.4 Dehydrohalogenation 2116.4 Advances Towards the Commercial Application of Dehalogenases 2117 Finalcomments 2128 References 212



1 Introduction 175

1 Introduction

1.1 Introductory Remarks

Haloorganic compounds are ubiquitousthroughout all environmental matrices. Theirlevels range from ppb to saturation levels inwater, soil, and air. The anthropoge nic originof many of these co mp oun ds is not in questionand has been the subject of many reportswhich have influenced en vironme ntal legisla-tion throughou t the world. Aside from theseanthropoge nic inputs, it is becoming clear thatthe natural fo rmation of haloorganics plays amajor role in the global haloorganic balance.The burden of halocompound transformationand turnover falls on the major trophic level,the microorganisms. It has been suggested byPRIES t al. (1994) that the lack of halocom-pound biodegradation is due to biochemicalrather than thermodynamic constraints, i.e.,the fidelity of compound recognition by up-take proteins, regulatory proteins, or catabolicenzymes is insufficient for their transforma-tion and degradation. Indee d, it is known tha tboth oxidative conversion of chlorinated com-

pounds with oxygen as the electron acceptorand reductive degradation to m ethanes or al-kanes should yield sufficient energy to supportgrowth (DO LFINGt al., 1993).

Despite the apparent recalcitrance of manyhalogenated compounds it is clear that, pro-viding environmental conditions are favor-able, microbial populations may adapt to usethese compounds as novel carbon and energysources. The genetic mechanisms of adapta-tion, especially relating to ha loaromatic tran s-formation have recently been reviewed (VAN

DER MEER,1994). The autho r m akes th e cleardistinction between vertical evolution and hor-izontal evolution. The former is brought abo utby alterations in the DNA sequence whicharise due to the accumulation of mutations.These may arise due to th e presence of the ha-logenated sub strate itself o r be accelerated inthe laboratory by the battery of mutagenicprotocols available. Ho rizon tal evolution oc-curs when DNA sequences are exchangedbetween cells (intercellular movement) orbetween plasmid and chromosomal DNA

(intermolecular movement). Horizontal m ove-

men t of genetic information m ay be caused byrecombination or other mechanisms of geneexchange, e.g., conjugation, transduction, ortransformation. It is clear that both vertical

and horizontal evolution have a ro le to play inthe development of catalysts which may be of

use in organic syntheses but tech niques whichfacilitate this will not b e d iscussed here.

The present review will deal exclusivelywith the biosynthesis and biotransform ation ofhaloorganic compounds of potential interestto those practitioners developing new routesof o rgan ic synthesis. Ad ditiona lly, it will illus-trate the diversity of substrates able to betransformed by biological material, whethermicrobial, plant, or animal. It should be statedfrom the outset that microorganisms play acentral role in the transformation of halogen-ated compounds. As a consequence the m ajor-ity of currently available information and theobvious industry preference for microbialsystems dictate that halotransformation by mi-croorganisms forms the fou ndation of any re-view on the subject. However, where a ppropri-ate, halotransformation or synthesis by non-microb ial sources will be discussed.

1.2 Distribution and Function

Th e ubiquity of biological materials contain-ing a covalently bound halogen was thoughtunlikely only a quarter of a century ago. In-deed, FOWDEN1968) undertook the first pub-lished review of naturally occurring organo-halogens and stated that:

“present information suggests that organiccom pounds containing covalently bound halo-gens are found only infrequently in living or-

ganisms.”Today, approximately 2000 halogenated

chemicals, predominantly chlorinated, areknown to b e discharged into the biosphe re bybiological systems and natural processes

interesting to note tha t public perception andawareness of the incidence of halogenatedcompounds in the natural environment is of-ten co ncentrated, and rightly so,on anthropo-genic inputs and their associated recalcitrance.Such concerns are heightened by uninformed

statements such as that released by the Scien-

(GRIBBLE,992, 1994; VAN PEE, 1996). It is



176 3 Halocornpounds -Biosynthesis and Biotransforrnation

tific Advisory Board to the International JointCommission on the Great Lakes which statedthat:

“There is something nonbiological about

halogenated organics (excluding iodinatedcompounds). ...Chemicals [that] do not occurnaturally... are often persistent, since thereare often no natural biological processes tometabolize or deactivate them.”

The ignorance of this statement, made in1989, s clear when one considers the biologicalcontribution to atmospheric pollution by vola-tile hydrocarbons. Estimates of the industrialproduction of chloride and fluoride as halohy-drocarbons (1982-1984) put the release as2.28.10’* g and 0.273 lo1’ g, respectively, intothe environment. However, the biological con-tribution, as chloromethane from oceans andburning vegetation is put at l.4-3.5.1012 g(HARDMAN,991). The occurrence of organo-halogens in nature is usually quantified interms of the bulk parameter AOX (adsorbableorganic halogen). This parameter has been

shown to be at least 300 times greater than thatwhich can be accounted for purely by anthro-pogenic sources (ASPLUND nd GRIMVALL,1991). AOX production has been shown to

take place during leaf litter degradation andtherefore the organisms responsible for degra-dation of leaf litter, primarily basidiomycetes,have been targeted (ASPLUND,995). Basidi-omycetes produce a cross section of halometa-bolites including chloromethane (HARPER,1985) plus chlorinated anisoles (DE JONGetal., 1994), orcinol derivatives (1) (OKAMOTOet al., 1993), hydroquinones (2) (HAUTZELtal., 1990) and diphenyl ethers (3-8) (Fig. 1)(TAKAHASHI,993;OHTAet al., 1995). Recent-ly, it as been shown that the ability to produceAOX is fairly ubiquitous, with 25% of 191fungal strains examined showing moderate(0.5-5.0 mg L-’) to high (5-67 mg L-’) AOXproduction (VERHAGENt al., 1996).

Today, it is recognized that a diverse array of

organohalogens is produced by biologicalsystems. These range from the large volume

OR c1

OR’

1 2 M y c e n o n c1 Russuphelins D-F

a R = M e

b R = C H O

3 R = R’ = Me, R2 = H; D

4 R = Me, R1 = H, 2 = Me; E

5 R = H, R’ = R2 = Me; F

HO

OH

O M e c1b+&Hobcq:kLoHR = Me; RussuphelinR A

cpclH

7 R = H; Russuphelin B M e 0 8 R u s s u p h e l o l

Fig. l. Chlorinated aromatics from basidiomycetes.



1 Introduction 177

9 N u c l e o c i d i n 10 Blattellastanoside A

Fig. 2. Naturally occurring fluorinated nucleoside and cockroach aggregation pheromone.

but simple halogenated alkanes through to thecomplex secondary metabolites such as nucle-ocidin (9) and the cockroach aggregationpheromone (10) (Fig. 2). The simplest halogen-ated alkane, chloromethane, is produced by awide cross section of biological systems includ-ing marine algae (WUOSMAA nd HAGER,1990), wood rotting fungi (HARPE R, 1985),phytoplankton (GSCHWENDt al., 1985), andthe pencil cedar (ISIDOROV,990).As well asspecies diversity it has been shown that prod-

uct diversity can occur within one species.Asparagopsis taxiformis, an edible seaweed fa-vored by Hawaiians, produces nearly 100chlorinated, brominated, and iodinated com-pounds (MOORE, 977).

It is clear that naturally occurring organo-halogens, while diverse and somew hat ubiqui-tous, do not exert the same environmentalpressure as the man-made organohalogens.These products, which are often used in rela-tively large amounts, necessitate the evolutionof new detoxification m echanisms. In th e ma-

jority of cases, th e na ture, position, and num-ber of halosubstituents mirrors that found inthe natural organohalogens. How ever, it is clearthat manufactured organohalogens present adifferent set of problems to biosynthesizedorganohalogens.

The diversity of organohalogens which areroutinely used m ay best be illustrated by thoseclasses of compounds which are directly ap-plied to the environment, namely the pesti-cides. The c urrent Pesticide Manu al (TOMLIN,

1995) contains approxim ately 1500 pesticides

of which approximately 50% are halogenated.

The distribution of halogenation is outlined inFig. 3.

Although some of these pesticides havebeen superseded, the d ata illustrate two con-trasting requirements of interest in the presentreview:

(1) Haloorganics a re favored targets forchemosynthesis because they possess awide spec trum of biological activities.

(2 ) Haloorgan ics often possess increased

recalcitrance contributing to th eir in-creased persistence in th e environ-ment.

Microbial biodegradation is considered tobe th e most important route of breakdown for

Multi-

Halogenated

12%

Fluorinated

17%

lodinated

1%

Brominated ChlOrhlated

5% 65%

Fig. 3.

adapted fromTOMLIN,995).

Distributionof halogenated pesticides (data



17 8 3 Halocompounds - Biosynthesis and Biotransformation

the m ajority of organic chemicals dep osited insoil and water a nd the add ition of xenop horessuch as halosubstituents retards this process(HOWARDt al., 1992). In the pesticides out-

lined abo ve over 100 compounds possess twoo r m ore different halosubstituents. These com -pounds are obviously more recalcitrant thanthe non-halogenated homologs and requirethe development of specific detoxificationmechanisms.The pressure to biodeg rade pesti-cides is largely borne by m icroorganisms andthe likelihood of transformation may be in-creased if the level of application is high. Obvi-ously, abiotic mechanisms contribute to pesti-cide degradation but 52 of the halogenatedpesticides outlined above a re applied a t levels>5 0 t a- l in at least 1 EU member country(FIELDING,992) Consequently, the burden oftransformation of these com pounds falls uponbiological systems and the mechanisms oftransformation will be outlined herein.

1.3 Reasons for Synthesis

by the Host

Both chlorine and bromine are ubiquitouselements in terrestrial and marine environ-ments. The earth contains 0.031% C1 and0.00016% Br (w/w) (cf. 0.00005% I) in min-erals and 19gL-’ C1, 65mgL-’ Br, and0.05 mg L-’ I in seaw ater. The world’s oceanscompose over 70% of the earth’s surface andover 90% of the volume of its crust. Conse-quently, the richest source of halometabolitesare the marine algae which, in a survey byHAGER 1982), were shown to contain up to20% of extractable halogenated compounds.

The marine habitat is complex, as evidencedby th e wide variations in pressure, salinity, andtemperature. Therefore, marine microorgan-isms have developed unique metabolic andphysiological capabilities, offering the poten-tial for the production of metabolites un-known in terrestrial environments. Althoughthe present article stresses the synthesis ofhalometabolites, such synthetic versatility isshown in the p roduction of o ther, non-halog-enated com pounds and the reade r is directedto othe r reviews for further inform ation (FENI-CAL, 1993;JENSEN and FENICAL,994 ).

Whatever the nu mbe r and diversity of halo-compounds the question which needs to be ad-dressed is why invest metabolic energy in thesynthesis of halocompounds? NEIDLEMANnd

GEIGERT1986) speculate that two major rea-sons exist:

(1) Halointerm ediates ar e key elemen ts inbiosynthesis.NEIDLEMAN1975) stated that:“Microorganisms, . have discovered thefact, as have organic chemists, that halog-enated intermediates, on pathways to non-halogenated end products are useful de-vices.”

OH

11

a X = H; Corynecin I

b X = C1; Chloroamphenicol

0 s

M e 0

12 G r i s e o f u l v i n

HO 0 Ho 0 NH2

1 3

a X = H; etracycline

b X = C1; Chlorotetracycline

Fig. 4. Common antibiotics possessing a halosub-

stituent.



2 Synthetic Importance ofHalocompound Transforming Enzymes 179

Subsequent to this comment it has beenshown that halogena ted intermediates are syn-thesized by a variety of synthe tic pathways in-volving unsaturated carboxylic acids, cycliza-

tion of isoprenoids, ring contractions, andmethyl migration to yield rearranged terpen-oid derivatives (FENICA L,979,1982).Many ofthese can be rationalized by the interventionof halonium ion intermediates.

(2) Halogenation enhances the bioactive effi-

As stated at the beginning of this introduc-tion the presence of halogen substituents inpesticides is common place. Why this should beso is evidenced by the wide variety of natural

organohalogens which possess antibacterial,antifungal, and antitumor activities. Chloram-phenicol ( l l b ) , griseofulvin (U),nd chloro-tetracycline (13b) are all common antibioticsthat contain chlorine (Fig. 4) . However, it doesnot necessarily follow that the efficacy of thesecompounds is related t o the type, num ber, andposition of the halogen atoms. The bioactivityof brom o-analogs is usually very similar or lessthan their normally produced chlorinated

cacy of most compo unds.

counterparts. 7-Bromotetracycline has thesame level of activity as 7-chlorotetracycline(13b) (PE’ITY, 1961) whereas pyrrolnitrin (15,Fig. 5 ) is more efficacious as a chlorinated,

rather than a brominated derivative (VAN EEet al., 1983). The absence of the halogen atommay a lso cause different effects with respect t oantibiotic activity. Coryn ecin I (lla, ig. 4), hedechlorinated analog of chloramphenicol(llb), possesses only a fraction of the antibio-tic activity of its chlorinated congener (SU ZU KIet al., 1972) whereas actinobolin (14a) andbactinobolin (14b) have similar antibiotic ac-tivity (Fig.5).

Halocom pounds (e.g., fluoroacetate (19,Fig.13)) and halogenating enzymes are used for

defensive purposes by a variety of organisms.This subject is beyond the scope of the presentarticle but has been reviewed (NEIDLEMANand GEIGERT,986).

2 Synthetic Importance

of Halocompound

Transforming Enzymes0

14

a X = H; Actinobolin

b X = C1; Bactinobolin

”.

15 Pyrrolnitrin

Fig. 5. Microbial antibiotics possessing a halosub-

stituent.

2.1 Halogenation

A small num ber of functional groups occupya dominant position with regard to organicsynthesis. All of these groups, C = C (olefinic),C=O (carbonyl), C = N (cyano), C- OH (al-cohol), C O O H (carboxyl), NH 2 (amino), NO,(nitro), and C=C (acetylenic) will have beendiscussed within this volume as they are cen-

tral to organic syntheses due to their ease ofinterconversion. Halides are examples ofmuch less central or peripheral groups whichare requ ired in the target o r simply provide ac-tivation or control in the synthetic route. Theadvantages of biohalogenation are no differ-ent to those proposed for other enzyme cata-lyzed reactions:

EfSicient catalysts - usually expressed asturnover number, it has been shown to be ap-proximately lo5 or chloro- and bromoperoxi-dases (HA GER t al., 1966; MA NTH EYnd HA-

GER, 1981).



180 3 Halocornpounds - Biosynthesis and Biotransformation

Environmental acceptability - traditionalhalogenation and dehalogenation m ethodolo-gies use harsh reagents (Friedel-Crafts cata-lysts, Lewis acids, and heavy metal catalysts)

with high environm ental impact. Furtherm ore,unlike heavy metal catalysts, enzymes arecompletely b iodegradable.

Mild reaction conditions - enzymes catalyzereactions in the range, of pH 5-8, typically 5using haloperoxidases, and at tem peratures of2 0 4 0 "C.This minimizes problems of unde-sired side reactions, a major drawback to m anychemosynthetic methodologies.

Wide substrate specificity - enzymes may acton a broad range of substrates giving a rangeof variously substituted products. The centralgroup of enzymes assumed to catalyze halo-com pound synthesis, the haloperoxidases, areunu sual in that their primary synthetic utility isthe halide independent reactions (ZAKS ndDODDS,995).

Several general assay method s which permitthe rapid screening for haloperoxidases (themain halogenating enzyme) have been em-ployed including the use of monochlorodime-done (MCD) (16a) and phenol red (17a) assubstrates (Fig. 6). How ever, it is unlikely that

novel synthetic ac tivity will always be evidentwhen employing limited substrates for discov-ery. What are required (although it is difficultto see how they may be devised), are specificscreens aimed at elucidating enantiospecifichalogenations. When considering potentialsubstrates for halogenation there are fewstructural limits providing that they are rela-tively electron rich, e.g., pyridine and pyrimi-dine a re fairly unreactive towards haloperoxi-dase unless an electron dona ting substituent ispresent (ITOH t al., 1987b).Additionally, ole-

finic groups a re fairly unreactive unless cou-pled to an electron rich system, e.g., styrene.

The scope of reactions catalyzed by the ha-loperoxidases, exemplified largely by the chlo-roperoxidases, is broad and can be subdividedinto halide depende nt and halide indepen dentreactions. Halide independe nt asymmetric ox-idations were recently studied by ZAKS ndDODDS1995) who suggested the halide inde-pend ent reactions, epoxidation of alkenes, for-mation of sulfoxides, aldehydes an d nitrosocompounds and N-dealkylations, were cata-lyzed by a neutral form of the enzyme whereas

the halide dependent reactions, halogenationof P-diketones and halohydration, were cata-lyzed by an acidic form of the enzym e.

The stereoselective haloperoxidase cata-

lyzed epoxidation of alkenes can be highlyenantioselective (COLONNAt al., 1993; AL-LAIN et al., 1993) and does not require an allyl-ic alcohol (cf. Sharpless epoxidation). Al-though the abiotic enantioselective epoxida-tion of alkenes which do not bear an allylicalcohol group continues to be improved, thereaction is currently only highly stereoselec-tive with cyclic alken es (VELDE nd JACOBSEN,1995).

The oxidation of sulfides to sulfoxides orsulfones has been performe d with many typesof enzym e. With haloperox idases in the ab -sence of halide ion the reaction is enantiose-lective and an oxygen from hydrogen peroxideis incorporated into the sulfoxide. W hereas inthe presence of halide the ra te is increased, thereaction is not enantioselective, and there is nooxygen incorporation from peroxide. Thisclearly indicates that in the presence of halidea freely diffusing species such as hypohalous

16

a R = H; monochlorodimedone

b R = B r

0s037

a R = H; Phenol red

b R = Br; Bromophenol blue

Common substrates for screening haloper-ig. 6 .oxidases.



3 Current Develop ments in Dehalogenase and Haloperoxidase Biochemistry and Genetics 181

acid reacts nonenzym atically with the sulfide(or other substrate molecule) (PASTAet al.,1994).

2.2 Dehalogenation

Deha logenation is a process normally asso-ciated with the detoxification or am eliorationof a recalcitrant halocompound. There are sev-eral methods of dehalogenating but this re-view will concentrate on those catalyzed byspecific enzymes. Deha logenation which arisesdu e to facilitated abiotic m echanism s, e.g.,spontaneous elimination of a halide ion afterring cleavage of a haloaromatic compound,will not be considered. All of the enzymaticmechanisms require at least one halogenatom, which is removed and replaced by an-other atom , functional g roup, or is eliminated.In the case of reductive dehalogenation thiswill result in a molecule which is deactivatedand, unless selective removal was beingsought, would be of limited synthetic use. Allother dehalogenation mechanisms, oxygeno-lytic, hydrolytic, thiolytic, epoxide formation,dehydrohalogenation, and hydration result in

substitution of the halogen atom with a reac-tive carbonyl or alcohol group, with the excep-tion of deh ydrohalog enation, an eliminationreaction resulting in double bond formation.

3 Current Developments

in Dehalogenase

and Haloperoxidase

Biochemistry and Genetics

Biotransformations have come to the fore-front of biotechnology over the last 10 yearsdue to the legislative drive towards singleenantiomer preparations in the agrochemical,flavor and fragrance, and pharmaceutical m ar-kets (CANNARSA,996). The primary reasonsfor this have been the selectivity of enzymesand the potential to produce and manipulatethe whole cells and enzymes which catalyze

the reactions of interest. In the halogenation

and de halogenation field the impetus has beentwofold:

There has been a drive towards under-

standing the degrad ative pathways in-volved in the transforma tion and mine-ralization of organoh alogens in the en-vironment. This has led t o the develop-men t of biochemical and m oleculartechniques which have facilitated ou runderstanding of dehalogenation.There has been an attem pt to betterunderstand how single enantiom erhalocompounds are forme d in nature.It has been said that “b acterial non-hem e haloperoxidases are difficult to

isolate an d ar e available in small quan-tities” (WENGet al., 1991).Only now isprogress being made in this field but itstill remains in its infancy, especiallywhen ascribing function to haloper-oxidases in vivo.

Bo th synthesis and d egradation have reliedon the development of molecular techniquesand a slow increase in the amou nt of sequencedata, both DNA and protein, available for

manipulation and exploitation. Additionally,these approaches have caused us to re-evalu-ate the roles that enzym es such as haloperoxi-dase may play in halocompound synthesis.

In only a few cases has the genetics and bio-chemical mechanism for introducing chlorinebeen examined in detail. Indee d, following thestudies of chlorination in the fun gus Caldario-myces jkmago, which confirmed that caldari-omycin (18,Fig. 7) was chlorinated via a chlo-roperoxidase catalyzed reaction, it was antici-pated this would be a generic mechanism.

While haloperoxidase catalyzed incorporationof halide is still generally accepted as the pri-mary m ode of halogenation, it has still prov ed

18 Ca l d a r i o my ci n

Fig. 7. Caldariomycin.



182 3 Halocompounds-Biosynthesisand Biotransformation

difficult studying the biochemistry and genet-

ics of halogenated secondary metabolite

formation. The bacterial non heme-type ha-

loperoxidases differ greatly from the heme-

type and vanadium containing haloperoxi-

dases. Moreover, it has been shown that the

bacterial haloperoxidases form a totally dis-

tinct enzyme family for which the catalytic

mechanism is still poorly understood (see Sect.

5.1.4).

Chloramphenicol (llb,Fig.4),probably the

best known chlorinated antibiotic is believed

to be chlorinated due to the action of a chloro-

peroxidase (NEIDLEMANnd GEIGERT,986).

It is produced by Streptomyces venezuelue,

from which bromoperoxidases but not chloro-peroxidases have been isolated. It is well

known that chloroperoxidases will catalyze

the incorporation of both chloride and bro-

mide whereas bromoperoxidases will only, due

to thermodynamic constraints, catalyze bro-

mide incorporation.Presently, using molecular

techniques, there is no report of a chloroperox-

idase being probed and found in an organism

known to produce a chlorometabolite.To the

authors knowledge, sequence information is

only available on a very limited number of ha-loperoxidases as follows (information taken

from the European Bioinformatics Institute

WWW site,http://www. ebi.uc.uW):Chloroperoxidases - sequenced from Cul-

duriomyces fumugo (NUELL t al., 1988),Cur-vuluriu inequalis (SIMONSt al., 1995), Pseu-dom onas pyrrocina (WOLFFRAMMt al., 1993),

and Streptomyces lividuns (BANTLEONt al.,1994). Another sequence from Pseudomonasjluorexcens is available but unpublished (PEL-

LETIER et al., 1995).Bromoperoxidases - sequenced from Strep-tomyces venezuelue (FACEY t al., 1996) andStreptomyces uureofuciens (PFEIFERet al.,

1992; PELLETIERt al., 1994).Additionally, he role of the haloperoxidases

in the synthesis of halometabolites is not un-equivocally established, indeed it maybe bet-

ter argued that they have no role to play in the

synthesis of halometabolites. FACEYet al.

(1996) have recently shown that a bromoper-

oxidasexatalase gene from Streptomyces ve-nezuelue is not required for chlorination in

chloramphenicol (llb,Fig. 4) synthesis. How-

ever, the non-heme chloroperoxidase from

Pseudomonus pyrrocinu has been shown to be

involved in the synthesis of pyrrolnitrin (15,Fig. 5) (PELLETIERt al., 1995).

As mentioned previously, there are seven

mechanisms for dehalogenating compounds.

The biochemistry and genetics of these

systems have been thoroughly reviewed by

FETZNERnd LINGENS1994) and JANSSENt

al. (1994) and only a cursory glance and updatewill be provided here. The naming of dehalo-

genases, particularly those involved in haloali-

phatic metabolism, has been confused and this

has only started to be addressed. SLATER,

BULL,and HARDMAN1995) have attempted

to categorize the various dehalogenating en-

zymes on the basis of their mechanism and

substrate range with further subdivisions to

take into account molecular descriptors such

as protein and DNA sequence homology. Ulti-

mately this approach will lead to complete de-scriptions of the genetic and evolutionary rela-

tionships between classes and a classification

of the different enzyme mechanisms which are

undoubtedly involved in dehalogenation by

reference to significant tertiary structures.

Currently, the classification of dehalogenas-

es is largely dependent on mechanism and can

best be summarized as follows:

(1) Reductive dehalogenation- argely as-

sociated with anaerobic environments,but not well characterized (Fig.8)

(DOLFINGnd BUERSKENS,995).

(2) Oxygenolytic dehalogenation- arylhal-ide dehalogenation,usually, involvestemporary loss of aromaticity and labil-ization of the halogenated molecule bythe introduction of oxygen. This is not

cl+(c* + XH ”#”’ + 00 + x0

c1 c1 c1 c1

Fig. 8. Reductive dehalogenation.



3 Current Develop ments in Dehalogenase and Haloperoxidase Biochemistry and G enetics 183

Fig. 9. Oxygenolytic dehalogenation.

considered a tru e dehalogenase in thecontext of the present review but th egenetics have been extensively investi-gated and reviewed due t o their impor-tance in biodegradation (Fig.9) FETz-

NER and LINGENS,994).( 3 ) Dehalogenation via epoxide formation

- dehalogen ation of haloalcohols togive epoxides (Fig. 10) s frequently fol-lowed by hyd rolysis of th e epoxide t o adiol, catalyzed by a n epo xide hydro-lase. The dehalogen ation is catalyzedby lyases which have been variouslyterm ed halohydrin epoxidases (CAS-TRO and BAR TNICKI,968),haloalcoholdehalogen ases and haloalcohol halo-gen-halide lyases (VAN DEN WIJN-

GAARD e t al., 1989,1991), nd halohy-drin hydrogen-halide lyases (NAGA -SAWA et al., 1992).Recently, two halo-hydrin hyd rogen-halide lyases fromCorynebacterium sp. have b een cloned,sequence d, and expressed in E.coli(Y u e t al., 1994).

(4) Dehydrohalogenation- dehalogena-tion via dehydrohalogena tion is not acommon mechanism (Fig. 11). Themost o ften cited, and possibly uniqueexam ple, involves the elim ination of

hydrogen chloride from y-hexachloro-cyclohexane (lindane) (82), to give se-quentially: y-pentachlorocyclohexene

(83) and 1,3,4,6-tetrachloro-l,4-cyclo-

hexadiene (84).The enzyme, termeddechlorinase, is encod ed on the chro-mosome in Pseudomonas (or Sphingo-monus)paucimobilis.The gene ( l inA)has been cloned, sequence d, and ex-pressed in E. coli (see Fig. 32) IMAI tal., 1991).Am ino acids with good leaving groups

at the P-position such as pchlo roa la-

nine and serine-0-sulfate undergoelimination via imines form ed withpyridoxal phosphate, followed by con-jugate nu cleophilic addition an d hy-drolysis of the imine t o give overall asubstitution product (CONTESTABILE

and JOHN,996; AGASAWAnd YAMA-DA, 1986).

( 5 ) Hydrolytic dehalogenation- he mostcommon dehalogenation mechanismsare hydrolytic, and these ar e fairlyubiquitous througho ut the substraterange (Fig. 12).W hethe r thiolysis, hy-dration, or other hydrolytic reactions,they all fall within the classificationsystem put forward by SLATER, ULL,and HARDMAN1995).All enzymes

catalyzing this type of reaction may beclassified into 3 groups: hydrolytic d e-halogenases, haloalcohol dehalogenas-

x

Fig. 10. Dehalogenation via epoxide formation.

xFig. 11. Dehydrohalogenation.

R-X - -OH

Fig. 12. Hydrolytic dehalogenation.




es, and cofactor dep endent dehaloge-nases, which ar e furth er subd ividedinto 3 subgroups and 8 classes. Thebasis for th ese divisions is prima rilymechanistic but is supported andendorsed by extensive DNA and aminoacid sequence information.

4 Mechanistic A spects of

Biohalogenation

The development of our understanding of

the nature of biological halogenation has pro-gressed along traditional investigative lines.Beginning with the iden tification of halogenat-ed com pounds from different species and thepossible elucidation of their function, it hasprogressed to consider the biosynthetic path-ways involved. From this point, the ind ividualenzymes involved in each transformation ar eidentified, and it is then the ro le of the biotech-nologist to seek to apply these enzym e proc-esses to the produ ction of compounds of inter-

est and value. A plethora of literature is nowavailable concerning halogenated metabolitesidentified in bacterial and algal species. Whileit would app ear that most halogenations pro-ceed through a single ubiquitous route, namelythat catalyzed by the haloperoxidases, the de-velopment of standard assays for these en-zymes has failed to detec t novel catalysts. Onlythe identification of novel natural productscan give some early indication as t o their routeof biosynthesis.

In this discussion of biohalogenation, wewill therefore summarize the broad range oforganohalogens found in nature, drawing par-ticular attention t o those which have enhancedour understanding of the process and thosethat have led to technological advances in thisarea.This forms a very suitable backd rop fromwhich to enter into the m ore d etailed analysisof b iosynthetic routes and enzym ology. Onceagain, following the same th read of investiga-tion, this section will be rounded off with asummary of the develop ments of biohaloge na-

tion as a technology in its application to themanufacture of compounds that are of com-

mercial significance.

4.1 Natural Organohalogens

4.1.1 Chlorine and Bromine

Tab. 1 is a survey of natural organohalides,collated according to biogenic origins andstructural type. No attempt has been made togive an exhaustive list of organohalogens asnew metabolites are being discovered all thetime. Compounds of pharm aceutical or agro-chemical value are indicated, but the reader isreferred t o cited original references for moredetails on the structures of interest.

The most abundant halogenated organics

are alkane derivatives produced either throughnatural com bustive processes or as metabolicproducts in marine algae. Combustion ofplants, wood, soil, and minerals containingchloride ions leads to the formation of orga-nochlorine compo unds, but a larger contribu-tor to these sinks is the action of volcanoes,with halocarbons such as chloromethane beingproduced at th e rate of 5-106 a-' (RASMUS-SE N et al., 1980),cf. 26000 t from anthropogen-ic inputs (HARPER,985).Natural volcanic ac-

tivity has also been shown to produce morecomplex highly toxic organohalogens such aschlorofluorocarbons (CFCs), dioxins, and cer-tain polychlorinated biphenyl (PCB ) congen-ers. A considerable proportion of simple ha-logenated alkanes are produced in the oceansby m arine algae, together with smaller quan-tities of more complex halogenated metab-olites (MOORE,1977; WOOLARDt al., 1976;FENICAL,974).

4.1.2 Fluorine

Very few organofluorines have been isolat-ed from biological material to date, with noneof these coming from th e animal kingdom ormarine organisms. It is thought that the m ainreason for the scarcity of fluorine containingcompound s is the high heat of hydration of th efluoride ion w hich severely restricts its partici-pation in biochem ical processes (Tab.2). In ad-

dition, fluorine has a major electronic effectupon compounds in which it is incorporated,restricting their further metabolism.



T1N

uaOg

o

Cm

S

c

Po

e

h

D

as

R

ee

Tpn

- mx

cone

a

bomne

- one

so

c

as

a

ceag

- L

naa

Po

mum

m

nsp

- A

h

a

fe

na

eaesp

e

Sr

om

(A

aa

b

dom

eAm

aaoo

ge

ag

- N

m

sa

aa

C

m

ab

ba

- bomne

Nepn

-conea

bomne

-cone

P

u

m

sp

m

n

eag

- L

na

G

u

b

e

fu

- C

omumgo

um

te

e

apas- gRhm

agun

a

bu

ge

ag- N

o

n

a

m

nsp

-Hco

sp

- bomne

Amnadappd

-cone

m

nsp

- D

d

h

b

Sr

om

a

P

u

m

sp

Sgs

p

u

F

n

bu

ge

ag- A

n

sp

Akod

ra

eyc

anho

- one

te

e

apas- Snmnumauum

Meo

n

carod

m

co

g

sm

- Sr

om

sp

wd

a

om

a

p

y

onem

ep

sq

ep

- oo

ca

a

um

a

s

dep

a

aeaep

tep

whaeso

eu

o

y

1sq

ep

n

n

he

n

5

one

ay

q

ep

m

a

e

h

bcd

cm

bo-

m

a

c

lagv

eynu

n6c

ceh

som

ow

cae

aeb

o

nB

e

o

nI1 dcoo

y

ca

boc

4m

a

e

n

c

o

dna

au

q

shkm

e

d

v

v

15 d

dd

v

v

4p

a

o

2bom

ea

o

a

4

coom

h

m

a

e

nu

nd

dna

d

d

nc

dpo

C

=C

C

bem

n

o

amnada

p

d

m

ow

caep

e

a

b

eas

ycoo

v

n

4coohe

n

cdo

vc

cp

dpw

n

nca

nn

3

a

d

a

umna

a

umdn

2bsn

eakod

ca

m

nAa

B

1 1 1 1 1 1 1 1 1 1 1 2 1 1 1 1 1 1 1 1



T

1

N

uaOg

o

C

n

Cm

S

c

P

o

eoh

D

as

R

ee

Sed

raeyc

anho

-cone

pa

s-Wih

asom

eaa

oh

Fyadpaanapd

- one

e

beeysha

w

es

eysh

o

o

aT

eors

a

soo

e

Ca

uaavrds

fe

hw

eag

egOhom

d

c

s

oE

m

ah

m

u

od

m

nsp

- X

op

aa

P

roa

re

vye

oaagnm

nnue

b

ea- P

u

m

a

u

n

a

P

u

m

p

ona

Sr

om

sp

A

nm

sp

A

n

p

a

umvamn

um

m

nb

eum-Com

eumsp

m

nsp

egHmnad

A

as

rum

Hmnad

sp

A

ro

eaw

e

a

te

e

apasegPsumsavum

bu

ge

ag- H

op

o

n

s

sp

egB

z

asp

mco

Pnc

umcuoum

m

uk- Dch

sa

Mu

sp

a

nworm

Go

o

a

sp

re

ag- L

nabo

a

m

nP

u

m

sp

sp

- Peom

m

Ioroh

asp

sp

-bomne

Pye

- one

-bomne

Ine

- one

- bomne

w

h

d

egp

ao

a

aoaecm

(eg

a

o

oo

o

6

ayadcoo

n

8

n

conepoaa

ns

awha

um

a

v

e

c

oouopd

9

Odbom

a

9111ermoecad

s

eon

bom

ad

p

oue

n-

boc

p

on

n

p

oom

nBa

ao

c

ya

vp

oom

n

p

oom

nA

1

eaep

oom

n

p

y

omnep

oe

m

a

dbom

n

s

nc

anc

o

a

3

bom

oo

mdn

d

bom

en

4

coon

eeea

cb

cad

1cone

so

e

6

conem

a

ew

h

y

amnno

aa

t

oC4n

ep

o

3

6coon

em

a

e- v

yhgsuu

acm

e

y

V

a

P

pe

n

gd

v

v

2

dbom

n

e

4 p

y

om

n

e

6

b

om

ea

h

o

o

i

n

eayae

5

a

6bom

a

56dbom

y

amn

6

bom

n

1 1 1 1 1 1 3 1 1 1 ? 4 1 1 5 1 1 1 1 1 1 1 1 1 6 7



T1N

uaOg

o

C

n

Cm

S

c

Po

e

h

D

as

R

ee

tu

cea

sp

sp

a

eaem

nog

sm

bu

ge

ag

-Rvuaarm

cm

avyaencm

so

ou

o

g

ecb

e

bu

ge

ag-Hac

poa

b

nun

Cbe

- one

Cbm

- bomne

Indoo c

es

-

one

tu

ceegE

som

ga

u

m

co

a

em

ao

boh

eg

N

daa

o

o

gn

Qnn

-

one

Sr

om

nrop

u

- bomne

Fabnu

-one

- bomne

Ton

- one

pasegPeo

o

Necad

- one

A

nm

sp

- bomne

Misahee

- one

pasa

u

- bomne

y

c

no

s- Z

m

m

nby

- Furaoa

pa

- Gm

eahmca

sp

-Dn

asp

sp

- Ehn

cyumsp

pas

re

ag-Po

aa

raac

dm

cy

amn

a

bomne

bsn

e

2bom

y

o

1

6n

p

y

omnedn

e

11

bomne

n

eh

onc

cp

d

n

p

an

coo

a

e

3coo

b

e

1

2d

ee

bomnePcb

n

1

5ho

e

n

o23acb

e

eg

e

m

n-

c

a

vy

vam

n

gy

ym

a

e

bom

n

n

b

uam

hzn

2bom

ua

1 1 1 1 1 1 1

5cone

ho

1

2 coo

oan

n

bomnep

n

conex

h

iso

m

n

u

u

p

y

omnep

n

1 1



T1N

uaOg

o

C

n

P 0W

Cm

S

c

P

o

e

h

D

as

R

ee

Amccm

- onea

- one

b

dom

e(8g

a

b

dom

e

6g

a

Mssp

s

m

h“

euh

d

s

g

g

a

bomne

h

o

ea

e

c

oneasym

a

e

e

gdoo

nA

1

&

Pnsapnceh

I

1

5

bomne

s

a

z

oosuuapoen

0 9

m

nsp

m

x

bom

oo

p

eh

1

3

- one

so

Pnc

umsp

ra

odcoo

p

s

3 2

- bomne

a

nw

m

2

6dbom

d

e

m

sm

1

E a

sp

m

a

ed

v

ombomne

yon

1

- onea

s

og

sm

egm

ukB

nmuum ho

e

yon

eg3coo5bom

yon-

n

Ahn

- one

fu

- D

m

5

coo

m

uena

5coo

m

u

n

1

F

Donaraedcm

- one

so

a

w

emco

unh

ash

p

o

d

P

C

a

PC

omcoo

s

3

1G

B

(1

2KTA

WAea

1

3E

A

ea

1

4K

Hea

1

5WIL

AMa

FA

K

(1

6SA

a

M

N

K(1

7K

ea

1

8FEDea

1

9S

N

ea

1



4 Mechanistic Aspects ofBiohalogenation 189

Tab. 2. Com parison of the Physicochemical Prope rties of the Halides (data tak en from HARPERndO'HAGAN,994)

Bond Dissociation Bond Length Hydration Electronegativity

Energy CH,-X (C-X) Energy, X - (Pauling Scale)[kcal.mol-'] [A1 [kcal.mol- '1HalideF 11 0c1 85

Br 71I 57H 99

1.391.781.932.141.09

11 7 4.0

84 3.078 2.868 2.5

2.2

Of the ten known organofluorine metab-olites (Fig. 13), six are modified carboxylicacids (19-24) from the plant Dichapetalurntoxicarurn. While most of the compounds areelaborated by plants, the production of fluo-rothreonine (27) nd nucleocidin (9 ) fromstreptomycetes suggests that bacteria m ay alsobe good sources for fluorinated structures(MEYER nd O'HAGAN,992a).

F(CH2)n- CO2H

Althoug h this review deals exclusively with

organohalogens, it is of passing interest to notethe peculiarly high accumulation of the inor-ganic fluoride K,SiF, in th e ma rine organismHal i chondr i a m oor e i to 10% of dry weightfrom seawater containing 1.3ppm fluoride(GREGSONt al., 1979). This salt is a po ten tanti-inflammatory and may act as part of thedefense mechanism of the organism.

9

CO2H19 n = 1; Fluoroacetate

21 n = 13; 14-Fluoromyristic acid22 n = 15; 16-Fluoropalmitic acid 23 18-Fluorooleic acid

10 F2 0 n = 9; 10-Fluororcapric acid 1 8

24 ~ h r e o - 1 8 - F l u o r o - 9 , l O - 25 F l u o r o a c e t o n e 26 2R,3R-Fluoroc i t ra tedihydroxystear ic acid

OH 0 0

F @NH,

OH OH27 Fluorothreonine 9 N u c l e o c i d i n

Fig. 13. Naturally occurring organofluorine com pounds.




4.1.3 Fluoroace a e

Fluoroacetate (19) is the most ubiquitousfluorine metabolite (HARPER nd O’HAGAN,1994). It was first identified in the southernAfrican plant Dichapetalum cymosum, andsubsequently in over 40 other plant speciesfrom Africa, Australia, and South America.The richest single source is D. bruunii seedswhich contain up to 8000 ppm of dry weight,but no wfluorofatty acids (O’HAGAN t al.,1993).

Fluoroacetate (19) is also produced by thebacterium Streptomyces cattleya together withfluorothreonine (27). The mechanism of fluo-

roacetate biosynthesis has not been elucidat-ed, but some possible routes are described inSect. 5.3. Fluoroacetate is toxic to all cells, du eto the lethal synthesis of 2R,3R-fluorocitrate(26). This is a competitive inhibitor of aco-nitrase (citric acid cycle) and probably bindsirreversibly to mitochondria1 citrate transport-ers (MEYER nd O’HAGA N, 992a).

4.1.4 w-Fluorofatty Acids

Seeds from Dichapetalum toxicarum metab-olize fluoroacetate to wfluorofatty acids, pre-sumably as the starter unit fo r fatty acid syn-thetase(s). The major product is 18-fluorooleicacid (23), with minor quantities of the capric(20), myristic (21), nd palmitic (22) analogsplus the dihydroxylated stearate (24) (HARPERe t al., 1990). All of the wfluorofatty acids arevery toxic t o cells.

4.1.5 Fluoroacetone

Homogenates of plants such as Acaciageorginae convert inorganic fluoride into vola-tile organofluorine compounds, the majorcomp onent of which is fluoroacetone (25) (PE-TERS and SHORTHOUSE,971). It is believedtha t this pathway is an aber ration of fatty acidsynthesis. Fluoroace tone (25) has also beendetected in livers of rats perfused with fluo-

roacetate (MIU RA t al., 1956).

4.1.6 Nucleocidin

Nucleocidin (9) was isolated from Strepto-myces calvus, found in an Indian soil sample.Plausibly, it could be formed by oxidation ofthe ribose ring of adenosine to the correspond-ing lactol, followed by sub stitution of the lactolhydroxyl group by fluoride and sulfamoylation(MAGUIREt al., 1993).

4.1.7 Fluorothreonine

The second bacterial organofluoride,(2S,3S)-fluorothreonine (27) was discovered

during attempts to optimize the production ofthienomycin by Streptomyces cattleya. Fluoro-aceta te was also identified. The source of fluo-ride was traced to casein (containing 0.7% in-organic fluoride) used in the media. N o fluo-rothreonine was produced w hen it was absentfrom the media and production could be re-stored by 2 mM potassium fluoride supple-ments (SANADAt al., 1986). The stereochem -istry of fluorothreonine (27) is the same asnatural (2S,3R)-threonine (i.e., L-threonine)(AM IN t al., 1997), although the Cahn-Ingold-

Prelog assignment of C-3 is different becauseof the fluorine substitution.

4.1.8 Iodine

While iodine is less abundant than chlorineor brom ine, a significant numb er of iodinatedmetabolites have been isolated to date, al-though they are rare com pared to bromo andchloro natural products. However iodome-

thane has been detected in the oceans andmea suremen ts suggest up to 4 -106 a-’ is pro-duced by m arine algae and kelp. The tun icateDidemnum sp. produce s a n aro matic diiodide(28) to a concentration of 0.4% dry weight (Fig.14) (SESIN nd IRELAND, 984), and the marinesponge Geodia sp. produces a cyclic peptidecontaining either a 2-iodo- or 2-bromophenolmoiety (29, 0) CHANet al., 1987). The iodoanalog (29) is also produced by the spongePseudaxinyssa sp. together with the chloroana-log (31) plus the corresponding nor-methyl

compounds (32-34) DESILVA t al., 1990).



4 MechanisticAspects ofBiohalogenatiorr 191

OMe OMe

2 8

Geodiamolides A-F

R = Me

29 X = I; A

3 0 X = Br; B

3 1 X = C1; C

R = H

32 X = I; D

33 X = Br; E

3 4 X = C1; F

Fig. 14. Organoiodine compounds isolated from marine organisms.

The calicheamicins e.g. (35-41) are extreme-ly potent DNA cleaving agents (enediyne anti-biotics, SMITHnd NICOLAOU,996) which areproduced by Micromonospora echinosporassp. calichensis. During attempts to optimizethe fermentation conditions for the bromo-compounds (37,39), sodium iodide was addedto the media and the iodo compounds (35,36,38,40,41) were discovered (Fig. 15) (LEEet al.,1991,1992).

In higher organisms, thyroxine (42) withinthe thyroid of mammals is the most well-known and studied iodocompound. Details of

the action of the peroxidase will be dealt withbelow.

SSSMe

Calicheamicins a 4 OR1

Am =

Rh =

"5H

X R' R2 R 3 I

3 5 I H A m Et a

3 7 Br R h Am 'Pr BIB*

3 9 Br R h A m Et y l B r 1

3 6 I Rh H a

3 8 I Rh Am 'Pr p @ozc

4 0 I Rh A m Et y t

4 1 I R h Am M e 6,'

Fig. 15. Organoiodine compounds.

1

4 2 T h y r o x i n e



192 3 Halocompounds -Biosynthesis and Biotransforrnation

5 Enzymology of

Biohalogenation

5.1 Haloperoxidases

5.1.1 Basic Mechanism

Haloperoxidases catalyze halogenation byspecies which are “synthetically equivalent” tohalonium ions (X’). These are produced byoxidation of halide (X-) by a peroxide (Fig.16). Under normal aqueous conditions for anenzymatic reaction, the halonium ion will havehydroxyl as the c ounter ion and hence will be a

hypohalous acid (HO X). The halide can beiodide, bromide, o r chloride. The oxidant po-tential of the halides increase in the sequenc eI - < B r - < C l - < F - . Pero xid e is n ot able t oprovide the oxidation potential for fluorina-tion and, therefore, there are no fluoroper-oxidases.

The reac tion in Fig. 16 is shown with hydro-gen peroxide a s the oxidant, however any per-oxide (including on e bound at a n enzyme ac-tive site) can participate in the reac tion. If th e

hypohalous acid is bound to an active site (i.e.,Enz -OB r) it may react with organic substrateswith usefu l regiq- or stereoselectivity, howev-er, if it diffuses in to solution it will have identi-cal reactivity t o that of “chemically gene rated ”hypohalous acid. Th e majority of current evi-dence suggests that free hypohalous acid isproduced in most if not all cases, althoughresults from a vanadium bromoperoxidase(TSCHIRRET-GUTHnd BUTLER,1994) andsome non-heme haloperoxidases (BONGS ndVAN PEE, 1994; FRANSSEN ,994) indicate sub-

strate binding. Haloperoxidase reactions inwhich free hypohalous acid is produced have

H202 + X- + H+- OX + HzO

RX + H2O

Fig. 16. Haloperoxidase mechanism.

little advantage over conventional chemicalmethods.

The reactivity of a halonium ion (X’) in-creases as the stability of the counterion in-

creases. A s noted above, in wate r, this will usu-ally be hydroxyl to give a hypohalous acid(XO H). As the halide concentration increases,elemental halogen (X,) or trihalide (X;) willbe form ed which are more reactive. Un der thenormal conditions of haloperoxidase reactionssuch species will halogenate electron rich aro-matics or undergo addition to double bonds.Iodine has the lowest oxidation potential andhence iodination is the m ost favored reaction,with bromination and chlorination requiringincreasing levels of activation of the site for

halogenation. Consequently, each iodoperoxi-dase is only able to use iodide as substrate,whereas a bromoperoxidase can brominate oriodinate and a chloroperoxidase can use allthree halides. If h alide concentrations are t oolow to give an appreciable reaction with theperoxide, this may also react, e.g., with a lkenesto give epoxides.

Haloperoxidases are widespread amongplant, bacterial, and animal species with thelargest range having been found in algal and

bacterial isolates (VANPEE, 1996). Attemptsto find useful haloperoxidase activity havespawned ventures into exotic locations. Ateam from the Scripps Oceanographic In-stitute studied marine red and green algaeand found 50 bromoperoxidases (HEWSONand HAGER,1980). The Cetus Corporationsearched fungal populations in Death Valleyand found 80 chloroperoxidases (HU NTE R tal., 1984). The latter search yielded some po-tentially useful enzymes which will be dis-cussed in more detail below.

5.1.2 Natural S ources of

Haloperoxidases

Little is known of the role of haloperoxidas-es in nature, though it is thought that their pri-mary function may be as part of the defensemechanism of the cell (FRA NSSE Nnd VAN DE R

PLAS, 992). Seve ral algal metab olites ar e tox-ic to predatory organisms, in addition mam -

malian m yeloperoxidase (LI et al., 1994) and



5 Enzymology of Biohalogenation 19 3

eosinophil peroxidase both produce hypo-chlorous acid which kills target cells for leuko-cytes, including microorg anisms such as Staph-ylococcus aureus (MCCORMICKt al., 1994).

Lactoperoxidase is able to produce hypobro-mous acid which oxidizes thiocyanate to theantimicrobial hypothiocyanate. Having high-lighted this defensive role, there is such a di-versity of halogenated metabolites, especiallyin algal and bacterial species that other rolesfor these comp ounds in cellular processes can-not be ruled out.

5.1.3 Haloperoxidase Classes

The haloperoxidases may be divided intofour distinct categories according to the pros-thetic group. In order of their relative abun-dance in the organisms studied to date, theyare as follows:

(1) He me (heme-thiolate) containinghaloperoxidases;

(2) Vanadium haloperoxidases (VILTER ,1995; BUTLER nd WA LKER ,993);

(3) Non-h eme non-metal haloperoxidases;

(4) Flavin/heme haloperoxidases.

Early studies indicated that all haloperoxi-dases contained a hem e prosthetic group, withthe iron atom ligated to a cysteine thiolate inan arrangement similar to that of the cyto-chrom e P450s. In 1984 the first vanadium con -taining non-heme haloperoxidase was ob-tained from the brown alga, Ascophyllum no-

dosum (pigweed) (VILTER, 984) and subse-quently non-hem e non-m etal haloperoxidasesand flavidh em e haloperoxidases were discov-

ered.A summary of haloperoxidase producingorganisms and isolated haloperoxidases isshown in Tab. 3. The heme chloroperoxidase ofCaldariomyces fumag o (which produces cal-dariomycin (18,Fig. 7)) is the most extensivelystudied as well as the most stable haloperoxi-dase in use at the pre sent time. It can be pro-duced at concentrations up to 280 mg L-’ inbatch culture (CARMICHAELnd PICKARD,1989;PICUR D et al., 1991) but the enzyme dis-plays no significant stereoselectivity. Prelimi-nary X-ray crystal structure da ta has been re-

ported (SUNDARAM OORTHYt al., 1995).

5.1.4 Enzym e Mechanisms

5.1.4.1 Hem e Haloperoxidase

The basic mechanism can be sum marized asfollows. Hydro gen peroxide reacts with the na-tive enzyme producing “Compound I” inwhich the ferriprotoporphyrin IX prostheticgroup is two oxidizing equivalents above thenative sta te (Fig. 17, Step 1). Com pound I isunstable and is able to react with alkenes(epoxidation), sulfides, or halide ions to giveCompound EOX (Step 2).This is also very un-stable and decomposes to yield the oxidizedhalide (Step 3 ) which is the halogenating agent

for the organic substrate (Step 4). The halo-peroxidase is inactivated by reaction with hy-pohalous acid and consequently the reactionrate slows down as the reaction progresses(WAGENKNECHTnd WOGAN,997).

The first heme haloperoxidase to b e purifiedand characterized was the bromoperoxidase ofPenicillus capitatus, a hom odimer of 97 kD asubunits each containing one hem e (MA NTH EYand HAG ER, 981). Recently, a chloroperoxi-dase and a bromoperoxidase were purified

from Streptomyces toyocaensis (MARSHALLand WRIGHT, 996).

5.1.4.2 Vanadium Haloperoxidase

Non-h eme vanadium haloperoxidases (VIL-TER, 1995; BUTLERand WA LKER, 993) aregenerally more stable than heme enzymes tohigher temperatures and organic solventsystems (D E BOER et al., 1987) but the en-zymes isolated t o d ate generally have a lower

specific activity than the hem e enzymes.All vanadium bromoperoxidases a re similar

in structure and amino acid com position, beingacidic in nature with 65000 kDa subunits and0.4 vanadium atomshubunit. In the catalyticmechanism, hydrogen peroxide binds to thevanadium to form a vanadium-peroxo com-plex to which the halide ion binds. The result-ing complex breaks down with th e release ofthe oxidized halide species (HO X or XJ. Thisenzyme is also able to catalyze chlorinationalthough at a slower ra te than for bromination.

The detailed role of the vanadium as an elec-



+

T3H

o

o

d

P

o

nC

sa

oaeE

m

\o P

S

c

S

ae

N

e

R

ee

Co

d

C

d

on

un

A

1

5

a

19Woo

cn- hm

c

ann

p

eh

C

B

p

sC

B

a

n

C

B

p

a

ep

dn

C

n

ecb

C

B

p

p

oeC

B

nk

m

nB

o

od

B

p

B

in

eC

B

p

p

oeC

B

p

sC

B

Sr

on

a

aen- v

um

P

u

n

p

ona- v

um

N

on

u

o

u- a

d

m

C

uaan

s

E

n

p

o

d

Bmo

d

P

u

n

a

ouenA

40

A

unn

um- v

um

C

anpu

ea- v

um

Sr

on

p

hon

n

Pncucu

au(ge

ag

hm

P

oh

ech

op

un

g

n

- hm

R

a

C

oo

a

P

a

M

ks

v

a

e

Io

d

L

o

o

d

omb

nm

k-hm

p

o

op

on

n

p

sB

I

p

e

B

n

ecb

B

p

sB

p

s

e

o

I

H

asho

T

odp

o

d

T

npo

P

a

p

sI

[u

isoaee

m

omSgm

1

co

a

e

e

nEc

a

Sr

om

vd

2 E 3 I& 8’ 3 2 5 Q 2 tn s 8G

hgsa

yop

o

d

p

ea

c

aez

hm

g

co

Q 3 z =0 3

p

ee

m

- Sgm

C

bo

m

1

im

z

oS

b

- Sgm

im

z

whgu

o

d

o

e

m

- BoR

b

na

hm

cD

co

p

ee

m

- Sgm

C

bo

m

11 5 1 4

~~~~

~

1G

B

(1

~V

C

N

ea(1

3R

M

ea(1

4

SC

D

ea(1

5:V

P

(1



5 Enzymology of Biohalogenation 195

RH

Enz-Fe3+-PP

RxStep 3

HZO

PP = Protoporphyrin IX

Enz-Fe3+-PP-0-X

'Compound EOX

Enz-Fe4+-PPO+

'Compound I'

X-

Fig. 17. Mechanismof heme haloperoxidases.

tron transfer catalyst or Lewis base is notknown.

Of the non-heme vanadium-dependent ha-

loperoxidases, the chloroperoxidase from Cur-vularia inequalis has been purified (SIMONStal., 1995), and an X-ray structure for this en-zyme is now available (MESSERSCHMIDTt al.,1996).The bromoperoxidases from the brownalga Ascophyllum nodosum (WEYAND t al.,1996;WEVERet al., 1985)and the alga Coraf f i -napilulifera (ITOH t al., 1988)have also beenwell studied.

Less well studied but still well characterizedare the bromoperoxidases from Pseudornonuspyrrocinia (WIESNERt al., 1985).Pseudomon-as aureofaciens (VANPEEand LINGENS,985a),

Streptomyces phaeochromogenes (VAN PEEand LINGENS,1985b), Streptomyces aureo-faciens (VANPEEet al., 1987),Murex trunculis(JANNUN nd COE, 1987),Xanthoria parietina(PLATet al., 1987),Coraflina oficin afis (SHEF-FIELD et al., 1993),Pseudomonus putida (ITOH

et al., 1994), and the chloroperoxidase fromSerratia m arcescens (BURDet al., 1995).

5.1.4.3 Non-Heme Non-Metal

Haloperoxidase

The X-ray crystal structure of bromoperoxi-dase A2 from Streptomyces aureofaciens(ATCC 10762) shows a typical alp hydrolasefold with an active site containing a catalytictriad consisting of Ser 98, Asp 228,and His257

(HECHT t al., 1994).The non-heme non-metalhaloperoxidase have substantial sequence ho-mology with serine hydrolases, including theconsensus motif Gly-X-Ser-X-Gly (PELLETIERand ALTENBUCHNER,995).Given this similar-ity, it is perhaps not surprising to find that theesterase from Pseudomonus Juorescens alsoacts as a bromoperoxidase as demonstratedby the bromination of monochlorodimedone(MCD) (16a) and phenol red (17a) (see Fig.6).

Site directed mutagenesis and inhibitor studiesindicate the catalytic triad is essential for ha-logenation activity (PELLETIERt al., 1995).Inan alternative proposal for the mechanism, ithas been suggested that a polypeptide back-bone S-halogenated methionine, may be theactive halogenating agent (HAAG t al., 1991).




5.1.4.4 FlavinlHemeHaloperoxidase

This eight subunit enzyme was first isolatedfrom the marine worm Notomastus lobatus(CHENet al., 1991). The flavin moieties arecontained within four identical polypeptidechains and two pairs of nonidentical subunitscontain the heme groups.The ratio for optimalactivity is 1heme: 1 flavin.A lobatus containsa plethora of bromophenols, and it is thereforelikely that the enzyme prefers bromide overchloride.

5.1.5 Reactions of th eHaloperoxidases

There are comprehensive reviews of halo-peroxidase reactions (FRANSSEN,994; FRANS-SEN and VAN D ER PLAS, 1992; DRAUZ andWALDMANN,995) and detailed protocolsfor the halogenation of different substrates(NEIDLEMANnd GEIGERT, 986).

5.1.5.1 Phenols, Anilines,and other Aromatics

The major substrate requirement for suc-cessful haloperoxidase catalyzed halogenationof aromatic compounds is activation of thering structure. As a result, phenols, phenolethers, anilines, and electron rich heterocyclicaromatics make good substrates. The transfor-mation of phenol red (17a) to bromophenolblue (1%) (see Fig. 6) is often used as an indi-

cator of haloperoxidase activity.While the majority of aromatic halogena-

tions shows no regioselectivity, t is worth not-ing that Caldariomyces fumago chloroperoxi-dase displays a preference for para-bromina-tion of anisole (43, Fig. 18)(PICKARD t al.,1991). This enzyme has been used for largescale oxidation of indole (44) o oxindole (45)(SEELBACHt al., 1997). It has been claimedthat Pseudomonas pyroccinia chloroperoxi-dase catalyzed chlorination of indole (44)gives 7-chloroindole (WIESNER t al., 1986),

however reinvestigation (BONG, npublished

P 7 H

43 A n i s o l e 4 4 Indole

H

45 O x i n d o l eFig. 18. See text.

results) has shown that the product is 3-chloro-indole as would be expected on the basis ofelectron density.

5.1.5.2 Alkenes

Alkenes (46) re halogenated to cr,P-halo-hydrins (48) by haloperoxidases (Fig. 19). Ifthe halide concentration is sufficiently high,the intermediate cyclic halonium ion (47) willreact with halide ion to give a dihalide (49).Substrates ranging from simple alkenes to bi-

4 8*-R4 6 4 7

Y = O H , X , X f

X

4 9

Fig. 19. Mechanism of the halogenation of alkenes.




cyclic alkenes and steroids all undergo halog-enation (COUGHLINt al., 1993).The presenceof certain functional groups leads to differentproducts to the halohydrins, e.g., haloacetones

are produced from unsaturated carboxylicacids (GEIGERTt al., 1983a).

The enzymatic halogenation of propylenewas one of the few attempts made to utilizehaloperoxidases in a commercial synthesis.Propylene is converted to propylene halohy-drin by a haloperoxidase. Halohydrin epoxi-dase converts the propylene halohydrin topropylene oxide, regenerating the halide. Un-

fortunately this process was not economicaland the pressure to move to halogen indepen-dent manufacturing processes ensured the

method did not reach commercial reality(NEIDLEMAN,980).

5.1.5.3 Alkynes

Haloperoxidases catalyze the halogenationof alkynes (50, Fig. 20) to a-haloketones (53).As with alkenes, at higher halide ion concen-trations dihalides (54) can be formed and theuse of a mixture of different halide ions can

yield heterogeneous dihalides (GEIGERTt al.,1983b).

2 5 4 5 3

5.1.5.4 Cyclopropanes

The cyclopropanes are converted to a,-

halohydrins by haloperoxidases according to

the same principles as for alkenes. One exam-ple of this reaction is the conversion of methyl-cyclopropane (55, Fig. 21) to 4-chloro-2-hy-droxybutane (56) by C. furnugo haloperoxi-dase (GEIGERTt al., 1983b).

5.1S.5 P-Diketones

A wide range of diketones have been halog-enated using haloperoxidases, with the majordeterminant of reactivity being the enol con-tent of the substrate. The simple ketone 2-hep-tanone, with a low enol content, has low reac-tivity, whereas the diketone monochlorodime-done (MCD) (16a,Fig. 6), which is predomi-nantly enol, has a very high reactivity. This re-activity, together with a marked decrease inUV absorbance upon halogenation has led toits use as a substrate for quantifying halo-peroxidase activity.

5.1.5.6 Nitrogen Atom sThe reaction of haloperoxidases with amine

compounds yields haloamines, most of whichare unstable and rapidly decompose by deam-ination or decarboxylation, freeing the halide(GRISHAMt al., 1984).

5.1.5.7 Sulfur Atom s

The reaction of haloperoxidases with thiols

produces the sulfenyl halide (-SX) whichreacts with excess thiol to yield the disulfide(-S-S-) or with a nucleophile such as

OH

5 5 5 6

Fig. 21. Cyclopropane cleavage by Caldariomyces

fumago haloperoxidase.ig. 20. See text.




hydroxyl anion to yield the sulfenic acid(-SOH), which can be oxidized to the sulfinic(-S0,H) and sulphonic acid derivatives( -S0 ,H) . In common with other haloperox-

idase mediated reactions, the reduction or re-moval of halide ions favors competitive perox-idase activity with, for example, C. fumagochloroperoxidase oxidizing dimethylsulfoxideto its sulfone (GEIGERTt al., 1983~) .

5.1.5.8 Inorganic Halogens

The chloroperoxidase from C. fumago isable to oxidize iodine to iodate and to dismu-tate chlorine dioxide to chlorate (THOMASndHAGER,968).

5.1.5.9 Stereo- and Regioselectivity

Several groups have attempted to find re-gio- or stereoselective biotransformations cat-alyzed by haloperoxidases. Bromohydrin for-mation from five cinnamyl substrates by thechloroperoxidase from C. fumago, showed noenantioselectivity, but some subtle differences

in diastereoselectivity (COUGHLINt al., 1993).Similarly, the Corallinapilulifera bromoperox-idase (non-heme) gave the same regiospecifictransformations of anisoles as obtained withsodium hypobromite (NaOBr) and there wasno enantioselectivity for bromohydrin forma-tion from a variety of unsaturated substrates(ITOH t al., 1988).

C.fumago chloroperoxidase was used in theregioselective bromohydration of saccharideglycals (57, Fig. 22), to produce 2-deoxy-2-bromosaccharides (58) (F u et al., 1992; LIU

and WONG, 992). These have both biologicalactivity and potential uses as synthons. Themild, specific halogenation of both sugars andpeptides is a potentially rich source of applica-tions for these enzymes.

There remain some puzzles, however, inboth the enzymology and natural productschemistry. Most notably, the presence of multi-ple haloperoxidases in Strepto myc es aureofaci-ens, none of which appear to be present in thebiosynthetic pathway for 7-chlorotetracycline(13b,Fig.4) which is abundant in these cells. It

is an extraordinary fact, that despite the pleth-

HO

HOmsBr OH

5 8

Fig. 22. See text.

5 9 Aplysiaterpenoid A

Fig. 23. AplysiaterpenoidA.

ora of complex natural products containinghalogens, thus far no enzyme has been isolatedwhich is involved in their halogenation. It is in-conceivable that complex polyhalocompoundssuch as aplysiaterpenoid A (59,Fig. 23 ) (citedby FRANSSEN,994) could be produced by ha-logenation using currently known haloperoxi-dases.

5.2 Halide Methyltransferase

The study of the biosynthesis of chlorome-thane by the species Hymenochaetae, a familyof white-rot fungi, has demonstrated a novelbiohalogenation independent of that carriedout by haloperoxidases (HARPERnd HAMIL-TON, 1988).The mechanism appears to operatevia a novel methyltransferase system wherebyhalide ions are directly methylated by a meth-yl donor such as S-adenosylmethionine. Nor-mally, in the metabolism of these fungi, the

chloromethane goes on to methylate aromatic




acids such as benzoate and furoate, producingthe corresponding esters. This reaction is alsopossible when chlorine is replaced by bromineor iodine, but fluoromethane is neither a sub-

strate nor an inhibitor of the enzyme.Methyl halide formation can be uncoupled

from methylation under certain conditions,and it is then that the halomethane is emittedinto the atmosphere. Although these emissionswere originally thought to come mainly fromfungal sources, an in vivo halide methyltrans-ferase activity has been detected in a numberof species of higher plants (SAINI t al., 1995)and the enzyme responsible for this reactionhas been purified and characterized (ATTIEHet al., 1995). It is conceivable that these en-

zymes might be engineered to transfer alkylgroups more complex than methyl.

5.3 The Enzymes of Fluorine

Metabolism

The biosynthesis of the majority of naturalorganofluorine compounds can be rationa-lized as anabolites of fluoroacetate (19).Theterminally fluorinated capric (20), myristic

(21), nd palmitic acids (22) pparently resultfrom fatty acid biosynthesis initiated by fluo-roacetate. There is also presumably a 18-fluo-rostearic homolog which undergoes A9-desat-uration to give 18-fluorooleic acid (23),whichin turn undergoes cis-dihydroxylation to givethe dihydroxylated stearate (24) see Fig. 13)(HARPERt al., 1990).

Fluorocitrate (26) is formed by the action ofcitrate synthase on fluoroacetate (19) andoxaloacetate (60) n the “first step” of theKrebs cycle (Fig. 24). Removal of the pro-R

proton from fluoroacetate (19), ollowed by

attack of the si-face of oxaloacetate givesstereospecifically (2R,3R)-fluorocitrate (26).

Fluoroacetate (19) nd fluorothreonine (27)are both produced by Streptomyces cattleya

and labeling studies seem to indicate that theyhave a common origin. Feeding with [2-13C]-glycine (61) Fig. 25) gave fluoroacetate (62)containing 13C in both carbons and fluoro-threonine (63)with the label at C-3 and C-4.Feeding with [3-13C]-pyruvate 64) resulted inincorporation into the fluoromethyl groups ofboth metabolites (65,66). he results are ac-commodated by a pathway (Fig. 25) in whichC-2 of glycine (67) ontributes both C-2 andC-3 of pyruvate (69)which in turn is incorpo-rated intact into the fluorinated metabolites(19,27)HAMILTONt al., 1997).

The importance of a three carbon interme-diate such as pyruvate (69)was demonstratedby incorporation experiments with glycerol(Fig. 26) (TAMURAt al., 1995).When (2R)-[l-’H2]- and (2S)-[1-’H2] glycerol were fed inde-pendently to S. cattleya only the 2R-stereoiso-mer (70)yielded fluoroacetate (71) nd fluo-rothreonine (72) containing two deuteriumatoms. Glycerol kinase only phosphorylatesglycerol (73) at the pro-R-hydroxylmethyl

group, direct displacement of the phosphategroup (74) (or of a derivative) by fluoride isconsistent with the results (NIESCHALKt al.,1997). One derivative which might undergothis process is phosphoglycolate (76) o give 3-fluoropyruvate (77).This reaction has beendemonstrated in reverse in several pseudomo-nads where a haloacetate halidohydrolase en-zyme has been identified (WALKERnd LIEN,1981).The conversion of 3-fluoropyruvate (77)to fluoroacetate (19) also occurs in Dicha-petalum cymosum (MEYER and O’HAGAN,

1992b, c).

s i face attack

Citrate synthase

CO2H

60 Oxaloacetate 1 9 26 2R,3R-Fluoroci trate

Fig. 24. See text.




0 F ONH36 6

S. cattleya

2 )

6 4 6 5

0 HO 0 0

19 Fluoroacetate

27 F1 u or o t h re o n i n e) +oo""r,yo@q A 0 L

ONH3 ONH,

6 7 G l y c i n e 6 8 S er in e 6 9 P yru va te

Fig. 25. Biosynthesis of fluoroa cetate (19) and fluorothreonine (27).

O H 02H

S. cattleya

0DOH OH 0 F ONH3

7 0 ( 2 R ) - [ -2 H2]g ly cerol 7 1 7 2

OH-H Glycerol GH --Fk i n a s e -

F OH-P-0 OH

I0 0

OHn-H

7 3 G l y c e r o l 7 4 sn-G l y c e r 01- 3 - p h o s p h a t e 7 5

0 &C02H - F CO2H

FII0-7-0

@b7 6

Fig. 26. The role of glycerol in fluoroacetate biosynthesis (19).

7 7




The labeling data above do no t excludeelimination-addition mechanisms which havebeen discussed in detail by HARPERndO’HA GAN (1994). In one proposal, phos-

phoglycolic acid (76) forms an imine (79a)with pyridoxamine phosphate (78) Fig. 27).Loss of a proton (SOa) nd elimination of theterminal phosphate gives a substituted acry-late (81)which undergoes nucleophilic addi-tion of fluoride. Reversal of the steps (Sob,79b, 8)gives fluoropyruv ate (77).This mecha-nism, first prop osed by M EAD and SEGA L(1973), has bee n supported by several meta-bolic studies in D . cymosum, but it has yet tobe verified by in vivo studies of fluoride me-tabolism in these plants.

The caterpillar Sindris albimaculatus accu-mulates fluoroacetate and can degrade it tocarbon dioxide and fluoride. Attem pts to forcethis system to operate in reverse with theenzme generating fluoroacetate have provedunsuccessful.

The possibilities of ethylen e and chlorinatedorganics as precursors for fluoroacetate con-version have also been discussed by HARPERand O’HA GAN 1994), but the evidence forthese is sparse. In on e final suggestion, based

on studies by FLAVINt al. ( 1957),PETERS asproposed the presence of a fluorokinase, pos-sibly an existing pyru vate k inase, which cataly-zes the ATP- and C0,-dependent incorpora-tion of fluoride into phosphate (PETERS ndSHORTHOUSE,964). On ce again, he could not

confirm this hypothesis in homogenates of

plant cells and this lack of detailed metabolicproof leaves the whole question of the mecha-nism of fluorination open to the suggestions

summarized above.

5.4 Advances Towards the

Commercial Application of

Biohalogenation

Despite the lack of overall success in theapplication of enzymes to halogenation reac-tions, several biotechnological app roaches ar emaking some headway into developing en-zyme systems and applications that enablesmall steps in the development of processes.Certainly the advent of molecular biology hasled to th e cloning of chloroperoxidases fromhuman eosinophils, Caldariomyces furnugo,Curvularia inequalis, Pseudom onus pyrrociniu,and Streptomyces aureofaciens (Sect. 3). Withthe rapid advancement in expression technol-ogies these should enable the relatively cheapproduction of haloperoxidases from many dif-feren t species such tha t the cost of enzyme will

not limit its application. There has also beenmuch investigation of the stability of haloper-oxidases und er conditions of high tem pera tureand in the presence of organic solvents (DE

BO ER t al., 1987).Bo th of these conditions arepotentially useful for optimizing manufactur-

76 X = OP032-; Phosphoglycolic acid 78-81 R = C H Z O P O ~ ~ -

a x = 0 ~ 0 ~ ~ -

b X = F

QCOZH 77 X = F; Fluoropyruvate0

7 8 7 9 a b 8 0 a b 8 1

Fig. 27. See text.




ing processes as they may be more cost-effec-tive at higher tempe ratures, and the use of or -ganic solvents may also increase the substraterepertoire that is available to the enzymes. As

men tioned earlier (Sect. 5.1.1), the Cetus Cor-poration isolated several fungal chloroperoxi-dases from Death Valley, with particular em-phasis on those with good resistance to highconcentrations of hydrogen peroxide and hy-pochlorous acid. The chloroperoxidase fromCurvularia inequalis lost no activity after incu-bation for 25 h in the presence of 200mMhydrogen peroxide whereas that from Calda-riomycesfumago was inactivated w ithin 2 min.A similarly increased stability was observed inthe presence of hypochlorous acid.

O ne further technique applied to haloper-oxidases has been immobilization and chemi-cal modification. These techniques have beenused with many enzyme and whole cellsystems because of the advantages of separat-ing the catalyst from substrates and products,potential increases in stability of immobilizedenzymes, and also potential to use enzymes atwider p H a nd temperature ranges. ITOH t al.(1987a) studied the immobilization of bro-moperoxidase from Corallina pilulifera onto a

variety of matrices utilizing the techniques ofcovalent binding, adsorption, and entra pm ent.They determined that ionic adsorption toDEAE-cellulofine and encapsulation in ENT-2000 matrix were the immobilization systemsof choice for continuing process development.SHEITIELD t al. (1994) extended this workwith the enzyme from Corallina officinalisdem onstrating high therma l stability and toler-ance t o organic solvents when the enzyme wasimmobilized on cellulose acetate. There arecurrently two comm ercially available immobi-

lized enzyme prod ucts (see Tab. 3).Unfortunately, while the promised market

for enzym atic halogenation remains extremelygood with so many h alogenated organics beingused as pharmaceuticals, agrochemicals, andfine chemicals, the n atu re of the reaction proc-ess in all enzymes studied to date does notallow for selectivity and therefore we awaiteither the discovery of suitable enzymes fromas yet unexplored sources or further develop-ments in protein engineering such that speci-ficity can be conferred upon existing enzymes.

The lack of a binding pocket for organic sub-

strates makes the latter approa ch very difficultto conceive without either th e presence of ad-ditional molecules to protect certain groups orthe incorporation of binding doma ins through

protein engineering.

6 Dehalogenations

6.1 Introduction

Although organohalogens may arise natu-rally, many are anthropogenic inputs. Thesetend to be relatively resistant to degradationand, therefore, have become known as majorrecalcitrant pollutants. However, many ofthese compounds can be degraded or trans-formed in the environm ent by a range of bioticand abiotic processes. The ability of microor-ganisms to utilize halogenated compoundsmay have arisen by selective pressure and ad-aptation created by the presence of such com-pounds in their environm ent. Therefo re an im-portant relationship has been form ed betweenthe environmental exposure of microorgan-

isms and halogenated organics, especially interms of biodegradation, biotransformation,and biosynthesis. Many naturally occurring mi-croorganisms have been isolated that degradehalogenated compounds to some degree.Some of them have been shown to work inaxenic cultures but often, where m ore com plexmolecules or mixtures are involved, mixedcultures are required.

The removal of the halogen substituent isthe key step in the deg radation of halogenatedcompounds. This requires cleavage of the

carbon-halogen bond, a very difficult reactionto achieve as exemplified by the bond dissoci-ation energies given in Tab. 2.

During haloorganic degradation the deha-logenation step is key an d mayb e ra te limiting.It may occur early or late in metabolism.Thereare five well characterized mechanisms identi-fied for dehalogen ation: reductive, oxidative,dehydrogenative, epoxide formation, and hy-drolytic (encompassing thiolytic and hydrationreactions).



6 Dehabgenatioris 203

6.2 Aliphatic Dehalogenations

Haloaliphatics encompass the haloalcohols,haloacids, haloalkanes, and their unsaturated

derivatives. Chlorinated aliphatics are widelyused as solvents, herbicides, and a variety ofspecialist applications, e.g., tetra- and trichlo-roethane are used in dry cleaning. Many chlor-inated aliphatics are poorly degraded, mainlydue to biochemical, rather than thermodynam-ic factors (PRIES t al., 1994).Their biochemi-cal inertness is caused by an absence of meta-bolic routes from which energy and anabolicintermediates may be derived. In general, thedegree of recalcitrance increases as the degreeof chlorination increases and, therefore, deha-

logenation is a critical step.Twenty seven percent of the priority pollu-

tants listed by the US EPA are haloaliphatics.Of these, nearly two thirds are haloalkanesand the rest comprises unsaturated derivativesand activated derivatives. It should be statedthat metabolism of haloalkanes and haloali-phatic acids may overlap. If the halogen posi-tion on the alkane chain is not terminal, deha-logenation is likely to be preceded by oxida-tion of the terminal methyl group to yield the

2- or 3-haloaliphatic acids (YOKOTAt al.,1986).Therefore, the examples chosen to illus-trate specific dehalogenations are to guide thereader through the myriad of possible reac-tions without giving an exhaustive listof exam-ples.

Haloalkanes are formed by both natural andanthropogenic routes and are found in all en-vironmental matrices. Haloalkanes can bemetabolized in a variety of ways: as a carbonand energy source; cometabolised or partiallydegraded by microrganisms who are unable to

utilize them as a growth source. A range ofmono- and dichlorinated species ranging fromchloro- or dichloromethane up to halogenatedoctadecane, can be degraded by a limitedrange of microorganisms.An extensive reviewof haloalkane metabolism by microorganismsis available (BELKIN,992).

halogen as a halogenide ion and its replace-ment by hydrogen. Haloalkane reductionoccurs on a range of short chain ( C, and C ,)chlorinated alkanes by methanogenic, denitri-

fying, and sulfate reducing organisms in reduc-ing environments in nature (CURRAGHt al.,1994). The precise enzymatic nature of thetransformation is not well defined but there isevidence that reduced porphyrin and corrincomplexes are able to dehalogenate efficiently.The rate of reductive dechlorination, especial-ly in anaerobic environments with highlyhalogenated substrates, is slow and may takeseveral months.

Tri- or tetrachlorinated species may be re-ductively dehalogenated by strictly anaerobic

bacteria causing sequential chloride release toproduce dichloro- and monochloroderivatives,ultimately yielding methane. Alternatively an-aerobic substitutive dehalogenation may total-ly degrade tri- or tetrachloromethane to formcarbon dioxide, although this is possibly due toa nonenzymatic process. Both of these reac-tions have previously been shown in the sameorganism and the true catalyst has yet to beunequivocally verified (EGLI et al., 1988,1990). The list of haloaliphatic compounds

which have been shown to be reductively de-chlorinated is not extensive but has been re-viewed by Mom and TIEDJE1992) and FETZ-NE R and LINGENS1994). It is clear that apartfrom the haloaromatic transforming strain D e-sulfomonile tiedje DCB-1, the tetrachloro-ethylene transforming strain PER-K23 (seeFig. 8) is the only pure culture available forfurther studies into the energy production dur-ing reductive dehalogenation (HOLLIGERtal., 1993).

6.2.2 Oxidative Dehalogenation

Cleavage of a covalent carbon-halogenbond may proceed through polymerization offree radical intermediates (Fig. 28) or via oxi-

MonoxygenaseH3C-X * -(CHZO)-

.2.1 Reductive D ehalogenation0 2 , NADH

Reductive dehalogenation is a two-electron

transfer reaction whTch involves the release of Fig. 28. See text.




dative hydroxylation at the carbon a tom adja-cent to the halogen (Fig. 29). The formeroccurs due to th e action of peroxidase, whilethe latter is oxygenase mediated. Such oxygen-

ases are broadly distributed in nature and m aybe either m ono- or dioxygenases.The action ofoxygenase on haloalkanes leads to the form a-tion of the corresponding aldehyde, whilefurther dehydrogenation may yield the acid(Fig. 29). Such a reaction pathway has beenshown for 1-chlorobu tane grown Corynebacte-rium sp. strain m15-3 (YOK OTA t al., 1986 ,1987) and for the ammonia monooygenase(RASC HE t al., 1990).

Oxidative dehalogenation of haloalkanes isusually mediated by a relatively nonspecificmonooxygenase, which catalyzes the forma-tion of halohydrins or halogenated epoxides:these spontaneously decompose to aldehydes,releasing the halide ion (CUR RAG Ht al., 1994).Generally oxidative dehalogenation only oc-curs with short chain halocarbons comprisedof 1- 4 carbons: e.g., mono haloethanes and n-alkanes are dehalogenated by the ammoniamonooxygenase of the nitrifying bacteriumNitrosomonas europaea, and the methane oxy-genase found in methanogenic microorgan-

isms can dehalogenate C, and C compounds.In general, oxygenase catalyzed dehalogen -ation reactions of short chain haloalkanes aredue to multifunctional enzymes with broadsubstrate specificity (methane monooxygen-ase) or involve enzymes from a romatic degra-dative pathways (toluene dioxygenase). Oxi-dation can lead to dehalogenation as a resultof the formation of labile products, not truedehalogenation. There are no reports of oxy-

-R A X "202 Rx x

genolytic dehalogenases specific for halogen-ated alkanes (FETZNERnd LINGENS,994).

6.2.3 Hydrolytic Dehalogenation

Hydrolytic dehalogenation is perhaps thecommonest route for the dehalogenation andtransformation of haloaliphatic compounds.As outlined above, the hydrolytic transforma-tion of a haloalkane may subsequently giverise to a haloalcohol and a h aloacid before de-halogenation occurs.Therefore , it is possible tohave multiple dehalogenating activities in oneorganism. Several organisms able to aerobical-ly degrade dichloroethane have b een isolated(KEUNINGt al., 1985). Th e initial ste p is thehydrolytic rem oval of one chlorine to form thehaloalcohol, chloroethanol. The enzyme re-sponsible has since been cloned and se-quenced (JANSSENt al., 1989) and the struc-ture elucidated (FRANKENt al., 1991; VER -SCHUEREN et al., 1993a, b). Following the for-mation of th e chloroethano l, furthe r oxidationvia the aldehyde and acid occurs. It is then atthis stage that the second chlorine is removedto yield glycolate. Several enzymes able to

catalyze the dehalogenation of the haloacidhave been purified (KLAGE St al., 1983;MOTO-SUGI et al., 1982) and some have bee n clonedand sequenced (SCHNEIDERt al., 1991).

During hydrolytic dehalogenation of halo-alkanes nucleophilic substitution occurs, inwhich the halogen atom is replaced by a hy-droxyl ion to give the corresponding alcohol(Fig. 30). These reactions ar e catalyzed by ahalidohydrolase-type dehalogenase. The halo-alkane dehalogenases can be divided into twoclasses depending on their substrate specific-

ity. Those which have a fairly restricted rangeof substra te specificities ar e gram-nega tivestrains, best exem plified by Xanthobacter auto-

trophicus GJlO (JANSSENt al., 1989) for whicha detailed mechanism has been determinedfrom X-ray crystal structures. An aspartateresidue is alkylated by th e haloalka ne to give

Fig. 29. See text. Fig. 30. See text.



6 Dehalogenations 205

an ester and the displaced halide ion is boundto the ring NHs of two tryptophan residues. A

histidine/aspartate pair assist the hydrolysis ofthe aspartate ester to give the alcohol (VER-

SCHUEREN et al., 1993a, b). The second class ofhaloalkane dehalogenases is represented bythose enzymes isolated from a group of closelyrelated gram-positive organisms.Rhodococcuserythropolis strain Y2 (SALLISet al., 1990),Arthrobucter sp. strain HA1 (SCHOLTZt al.,1987), and Corynebacteriurn sp. strain m15-3(YOKATAt al., 1987) all demonstrate muchbroader substrate specificities and are suggest-ed to be grouped together using the system ofSLATER t al. (1995).

The haloalcohol dehalogenases are a dis-

tinct group of enzymes which result in the for-mation of epoxides. These are dealt with in aseparate section but are still hydrolytic en-zymes acting on haloaliphatic compounds (seeSect. 6.2.5).

Of greatest interest and by far the moststudied systems are those catalyzing the trans-formation of the haloaliphatic acids - thehaloalkanoic acids, the haloacids, and thehaloacetates. The enzymes catalyzing thetransformation are broadly divided into two

groups of halidohydrolases, namely the halo-acetate dehalogenases (EC 3.8.1.3) and the 2-haloacid dehalogenases (EC 3.8.1.2). The divi-sion of these enzymes was based on their sub-strate range, haloacetate dehalogenases actingexclusively on haloacetates and 2-haloacid de-halogenases acting on haloacetates and othershort chain fatty acids (&-C,).The two groupsof enzymes appear to have non-overlappingactivity spectra (JANSSEN et al., 1985). Bothtypes of halidohydrolases preferentially deha-logenate in the order F>Cl>I>Br. The

mechanism of displacement proceeds througha SN2-type reaction and has been found todiffer, depending on the enzyme. SLATER t al.(1995) categorized the enzymes into fourclasses, based on their enantiospecificity andability to invert the substrate-product config-uration. Class 1L and 1D catalyze the inver-sion of the product configuration comparedwith the original substrate and are D (Class 1D) or L (Class 1L) specific. Class 21 and 2R areable to dehalogenate both the D and L-enan-tiomers with either retention (Class 2R) or

inversion (21) of product configuration

compared with the original product configura-tion.

Within these classifications, 1L would ap-pear the most common, having been shown to

be present in strains of Pseudomonas (SCHNEI-

DER et al., 1991;KAWASAKIt al., 1994;NARDI-

DEIet al., 1994;JONESet al., 1992;MURDIYAT-

M O et al., 1992),Moraxellu (KAWASAKI t al.,1992) and Xunthobacter (VAN DE R PLOEG tal., 1991).

A final group of haloaliphatic transformingenzymes are best classified as those glutath-ione (GSH) dependent dehalogenases. STUCKIet al. (1981) isolated a Hyphomicrobium spe-cies able to dechlorinate dichloromethanewhich was GSH dependent. The mechanism of

these enzymes involves a nucleophilic substi-tution by GSH, yielding an S-chloromethylglutathione conjugate which is dehalogenatedwhen it rapidly undergoes hydrolysis in anaqueous environment. The thiohemiacetalthus formed is then in equilibrium with GSHand formaldehyde (KOHLER-STAUB and LEI-SINGER, 1985). The enzyme has only beenshown to work with dichloromethane and arestricted substrate range.

6.2.4 Dehydrohalogenation

Dehydrohalogenation occurs with simulta-neous removal of the halogen atom and thehydrogen atom on the neighboring carbon(Fig. 31). This results in the formation of al-kenes from simple halogenated hydrocarbons.The major substrates which appear to be trans-formed by this route are the lindane (82, Fig.32) pesticides, a halogenated cyclohexane, and

some steps in the transformation of dichloro-diphenyltrichloroethane (DDT), a haloaro-matic pesticide.

The transformation of lindane (y-hexachlo-rocyclohexane, y-HCH) (82) by a dehydro-

Fig. 31. Dehydrohalogenation.



82 Lindane 8 3

q-OH a

8 5 8 6

Ip o n t a n e o u sJ

a

8 7 8 8

Fig. 32. The metabolism of lindane (82) by Pseudornonas paucimobilis UT26.

chlorinase is probably the best studied of thislimited group of enzymes. Lindane (82) was

first shown to be aerobically transformed byTu (1976) and HAIDER1979). In Pseudomo-nas paucimobilis UT26, the proposed pathwayof lindane (82) degradation involves dehy-drochlorination to yield y-pentachlorocyclo-hexene (83) and 1,3,4,6-tetrachloro-1,4-cyclo-

hexadiene (84) which may decompose to thedead-end product 1,2,4-trichlorobenzene (87)or be acted upon by the hydrolytic dehaloge-nase LinB (Fig. 32) to give the dichlorodihy-droxycyclohexadiene (86). IMAIet al. (1991)have cloned and sequenced the gene encoding

the initial dehydrohalogenase LinA. It has nosequence homology with any bacterial oreukaryotic glutathione S-transferase and didnot show DDT dechlorinase activity in thepresence of glutathione.

Apart from the restricted range of dehy-drochlorinases which have been shown, it isalso clear that LinA from Pseudomonaspauci-mobilis UT26 has a very limited substraterange, only converting a-HCH, 7-HCH (82),GHCH, a-pentachlorocyclohexene, and 7-

pentachlorocyclohexene (83) (NAGATA t al.,1993). Other, less well characterized enzyme

systems responsible for the transformation oflindane (82) are suggested to be extrachromo-

somally located and present in Pseudomonasovalis CFT1 and Pseudo mon as tralucida CFT4(JOHRI t al., 1991).

A dehydrohalogenase or dechlorinase, iso-lated from the house fly (Musca domestica),(LIPKE nd KEARNS,959) is a glutathione S-transferase (ISHIDA, 968) which catalyzes themonodehydrochlorination of DDT.

P-Chloroalanine and serine-0-sulfate formimines (80)with pyridoxal phosphate (see Fig.27). These readily eliminate to give 2-aminoacrylates (81), which in turn undergo conju-

gate addition of nucleophiles. Hydrolysis ofthe imine yields amino acids with a new P-sub-stituent (CONTESTABILEnd JOHN,996). Us-ing tryptophan synthase or tyrosine phenollyase this has been developed into a powerfulmethod for the synthesis of p-substituted ala-nines (NAGASAWAnd YAMADA,986).

6.2.5 Epoxide Formation

Epoxidation results from the simultaneous

removal of the halogen atom and the hydro-




Fig. 33. See text.

gen atom from the adjacent hydroxyl group onthe neighboring carbon (Fig. 33).

Although the reaction was first shown near-ly 30 years ago (CASTROnd BARTNICKI,968)very few studies have extended this finding.VANDEN WIJNGAARDt al. (1989) have isolat-ed the strains AD1 (Pseudomonus sp.), AD2(Arthrobucter sp.) and AD3 (coryneform) andshown that a variety of vicinal haloalcohols

may be degraded. The formation of epoxidesfrom a range of haloalcohols (&-C3) has beenshown for the enzyme purified from AD2 (VANDEN WIJNGAARDt al., 1991). The enzyme re-quires no cofactors or oxygen and probablyacts as a simple acid-base catalysis. The halo-alcohol dehalogenases from Corynebacteriumsp. strain N-1074 are the best studied. In thewild-type strain, two haloalcohol transformingenzymes and two epichlorohydrin transform-ing enzymes have been isolated. The haloalco-

hol transforming enzymes catalyzed the trans-formation of halohydrins into epoxides andthe reverse reaction. However, they wereshown to differ in their enantioselectivity for1,3-dichloro-2-propanol conversion as well astheir substrate range, subunit composition, andimmunological reactivity (NAKAMURAt al.,1992). As outlined above for the other hydro-lytic enzymes, SLATERet al. (1995) haveattempted to classify the haloalcohol dehalo-genases on the basis of their substrate specific-ities. It is clear that the number of examples

limits this exercise but it may be significantthat the N-terminal amino acid sequence ofthe enzyme from Corynebacterium sp. strainN-1074 is similar to that of Arthrobacter strainAD2 (NAGASAWAt al., 1992).

6.3 Aromatic Dehalogenation

Chlorinated aromatic compounds havebeen widely used as solvents, herbicides, fungi-cides, and insecticides. The aromatic moiety

may range from the the monocyclic (chloro-

benzenes) to the polycyclic (polychlorinatedbiphenyls PCBs). Each may be modified togive a range of haloaromatic pesticides (e.g.,2,4-D, dichlobenil) and other anthropogenic

inputs such as dioxins. Natural haloaromaticsrange from the esoteric 1,2,3,4-tetrachloroben-zene (MILES t al., 1973) to the EPA prioritypollutant 2,4-dichlorophenol produced by aPenicillium sp. (ANDO t al., 1970).

As outlined previously for chlorinatedaliphatics (Sect. 6.2), the removal of the halo-gen substituent is a key step in the degradationof halogenated aromatic compounds. This mayoccur either as an early reaction, catalyzed bya true dehalogenase, or at a later stage in me-tabolism, often resulting from the chemical de-

composition of unstable primary products ofan unassociated enzyme reaction.

Generally, initial halogen removal occurs viareductive, oxygenolytic, or hydrolytic mecha-nisms while later removal may be a fortuitousevent caused by a separate abiotic reactionafter the loss of aromaticity.There is an exten-sive discussion of the microbial metabolismof haloaromatics (REINEKE nd KNACKMUSS,1988).

6.3.1 Reductive Dehalogenation

Under a variety of redox conditions the re-ductive dehalogenation of a wide cross sectionof haloaromatics has been demonstrated anddiscussed extensively (MOHN and TIEDJE,1992;WACKETTnd SCHANKE,992).Haloben-zenes (RAMANANDt al., 1993),halobenzoates(HAGGBLOMt al., 1993), haloanilines (KUHNet al., 1990), and halocatechols (NEILSONt al.,1987) are only a few of the compounds which

have been shown to be reductively dechlori-nated. Pentachlorophenol (89), which waswidely used as a herbicide and wood preserva-tive, undergoes reductive dehalogenation withsequential removal of the chloride substitu-tions to give 3-chlorophenol (94) as the finalproduct (Fig. 34).

It is also evident from the majority of studiesperformed to date that very fe w axenic cul-tures able to perform reductive dehalogena-tion have been studied. Desulfonomile tiedjeDCB-1 is one of the few organisms which have

been shown to perform reductive dechlorina-



208 3 Halocompounds -Biosynthesis and B iotransformation

c 1 c 1 c1

89 Pentachlorophenol

c 1 c 1

9 4 9 3 9 2

Fig. 34. Degradation of pentachlorophenol.

tion and has become a benchmark organismfor these studies. SHELTONnd TIEDJE1984)showed that this organism can grow and deriveenergy from the dechlorination of 3-chloro-

benzoate. The rate of dehalogenation dependsupon the number and position of the halogensubstitutions, since the presence of the carbo-nyl group favors removal at metu > ortho >para positions.

Due to the global environmental concernsmany studies have been made on the reductivedechlorination of polychlorinated biphenyls.Using Hudson River sediments BROWN t al.(1987) showed that there was an alteration inthe PCB congener distribution over a long peri-od of time; highly chlorinated congeners were

removed and there was an increase in the lowerchlorinated species. It has been shown that thepattern of dechlorination is specific for the par-ticular sediment from which it is derived withdechlorination occurring specifically at metuand para chloro substituents.A consortium de-void of sediment has never been shown to an-aerobically dechlorinate PCBs. It has beendemonstrated that as the degree of chlorinationincreases,so does the probability of dechlorina-tion, and as the number of mono- and dichlori-nated congeners increases,so the role of the an-aerobes decreases (ABRAMOWICZ,990).

6.3.2 Oxidative Dehalogenation

The commonest mechanisms in (ha1o)aro-matic degradation is the introduction of hy-

droxyl groups to facilitate ring cleavage. Bothmonooxygenases and dioxygenases may play arole in haloaromatic dehalogenation. In mono-cyclic aromatic metabolism, the key ring fis-sion substrates (which may be halogenated)are, in order of importance: catechol, proto-catechuate, and gentisate. Subsequently, thearomatic ring is cleaved and the halogenatedmolecule is labilized by the further incorpora-tion of oxygen. Halide removal is thereafterspontaneous and is associated with normalaromatic catabolism.

6.3.2.1 Dioxygenase Catalyzed

Dehalogenation

Oxygenolytic attack on halogenated aro-matic hydrocarbons may cause cleavage of thechlorine atom fortuitously when both atoms ofmolecular oxygen are incorporated into thearomatic nucleus. These reactions may be cata-lyzed by an aromatic dioxygenase and requiretwo adjacent carbons, only one of which is




halosubstituted. Of th e limited substrates anddioxygenases which fulfill these criteria themetabolism of 2-chlorobenzoate is the beststudied (FETZNERt al., 1989,1992 ).The 2-ha-

lobenzoate-1,2-dioxygenaserom Pseudomo-nu s cepacia 2CBS, which catalyzes the form a-tion of catechol from 2-halobenzoates, is amulticomponent enzyme which is structurallyrelated t o benzoa te 1,2-dioxygenase and th eTOL plasmid encoded toluate 1,Zdioxyge-nase. Despite this structu ral similarity, neithe rthe benzoate nor the toluate 1,2-dioxygenasesare able to convert the ortho substituted ben-zoates. Furthermore, although the enzymefrom f! cepacia 2CBS showed broad substratespecificity, it de mon strated a preference for 2-

halosubstituted benzoates, justifying its in-clusion and terminology as a dehalogenase.Other oxygenolytic haloaromatic dehaloge-nases have also bee n show n to be m ulticompo-nent enzymes. Fur ther 2-chlorobenzoate-1,2-dioxygenases have been identified in I! putidaCLB250 (ENGESSERnd SCHULTE,989) andI ! aeruginosa JB2 (HICKE Y nd FOCHT, 990)and shown to be plasmid encoded in strainPput ida P l l l (B RE NN ERt al., 1993). Dihy-droxylation of arom atics is cov ered in detail in

Chapter 10.Of significant environmental concern, andreliant upon oxygenolytic dehalogenation, isthe degradation of PCBs. The aerobic degrada-tion of lightly chlorinated PCBs has beenshown to involve an upper pathway in which,

e.g., biphenyl (95) is dihydroxylated (%), re-aromatized (97), cleaved to a ketoacid (98),

and then further cleaved to benzoic acid (99)

(Fig. 35).

This upper pathway has a broad substratespecificity and has been demo nstrated in a va-riety of microbial taxa including Achromo-bacter (AHM ED nd FOCHT,1973),Alcaligenes,and Acinetobacter (FURUKAWAt al., 1978).The action of the four enzym e system resultsin oxy genolytic cleavage of on e of the biphen-yl rings to produce (chlorinated) benzoates.Since the early work o n microbial degradationmany studies have been undertaken. Thesehave been summarized and som e general ob-servations have been made (ROBINSONnd

LENN , 994):

Deg radation decreases as the extent ofchlorination increases.Ortho substituted biphenyls have a lowerdegrada tion rate.Ring cleavage always occurs in the leastor non-chlorinated ring.Degrad ation decreases if both rings ar echlorinated (4 -substituted biphenylsresult in the forma tion of a yellow meta

cleavage product).

In summary, many dioxygenases ar e ab le tocatalyze the incorporation of dioxygen into a(ha1o)aromatic compound and m ay cause de-halogenation. We have attempted to highlight

ggOH\ ' O H

L g O H L H

4-,O H

HO

9 5 9 6 9 7 9 8 9 9

1 Biphenyl-2,3-dioxygenase

2 Dihydrodiol dehydrogenase

3 2,3-Dihydrobiphen yl- 1.2-dioxygenase

4 2-Hydroxy-6-oxo-6-phenylhexa-2,4-dienoiccid hydrolase

Fig. 35. PCB upper pathway.




some of those enzymes which cause dehalog-enation due to relaxed substrate specificityand not those enzymes simply associated withspontaneous elimination of halide after cleav-

age of the aromatic ring.

6.3.2.2 Monooxygenase Catalyzed

Dehalogenation

Despite having a high degree of chlorinationsome microorganisms have been shown to beable to degrade pentachlorophenol(89) underaerobic conditions (Fig. 36) (RADEHAUSndSCHMIDT,992; SCHENK t al., 1990; TOPPandHANSON,990). The initial reaction has beenshown to be an NADPH dependent oxygena-tion to form tetrachlorohydroquinone (100).The purified enzyme from Flavobacterium sp.strain ATCC 39723 was shown to catalyze theincorporation of 1 8 0 2 (XUNet al., 1992b). Pen-tachlorophenol-4-hydroxylasefrom Flavobac-terium sp. strain ATCC 39723, has been shownto be a flavoprotein monooxygenase catalyz-ing the para hydroxylation of a broad range ofsubstituted phenols. The requirement forNADPH was dependent upon the leaving

group with 1mole electron donating groups

OH

c1*c c1

c1

8 9

c1coclc1

100

Fig. 36. See text.

(H and NH,) and electron withdrawing groups(F, C1, Br, NO,, and CN) requiring 1mole or2 mole NADPH, respectively (XUN et al.,1992a). Although other isolates able to de-

grade halophenols have been isolated (Go-LOVLEVA et al., 1992; KIYOHARAt al., 1992),the monooxygenase catalyzed dehalogenationwould appear to be restricted to the metab-olism of more highly halogenated compounds,with the less halogenated aromatics being de-halogenated via the dioxygenase or post cleav-age routes outlined above.

6.3.3 Hydrolytic Dehalogenation

The replacement of a halogen with a hy-droxyl from water is rare in comparison withthe oxidative dehalogenations previously out-lined. The delocalization of the .rr-electronscontributes to the stability of the aromatic ringsystem but it does not preclude the direct hy-drolysis of the carbon-halogen bond. Variouss-triazines have been shown to be hydrolytical-ly transformed (COOK nd HUTTER, 986) butthe central focus of hydrolytic haloaromaticdehalogenation has been the initial step in the

transformation of 4-chlorobenzoate. It hasbeen suggested that 3-chlorobenzoate mayalso be metabolized by a similar route (JOHN-STON et al., 1972) but this is the only reportedexample of 3-chlorobenzoate transformationnot involving a dioxygenase.

The halidohydrolase responsible for catalyz-ing the dechlorination of 4-chlorobenzoate(101) has been extensively characterized inPseudomonas sp. strain CBS3 (Fig. 37). Themulticomponent enzyme system comprises aninitial 4-chlorobenzoate-CoA ligase which

adenylates the carboxyl group in an ATPdependent reaction. The AMP moiety is thenreplaced by coenzyme A resulting in the for-mation of a thioester (102).The second en-zyme, 4-chlorobenzoyl-CoA dehalogenase, isthen able to catalyze the nucleophilic attack atC-4, to yield 4-hydroxybenzoyl-CoA (103).

Finally, hydrolysis of the thioester yields 4-hy-droxybenzoate (104) (ELSNER t al., 1991;SCHOLTENt al., 1991; CHANGet al., 1992;LOFFLER t al., 1992).

Other analogous pathways have been

shown in Arthrobacter sp. strain SU (SCHMITZ




CO2H COS CoA COS CoA CO2H

Mg2+

Cl ATP A M P + c1 H 20 HC1 OH H 2 0 CoASH OH

PPi

10 1 0 2 103 1 0 4

Fig.37. Dechlorination of 4-chlorobenzoate to 4 -hydroxybenzoate by Pseudomo-nas sp. strain CBS3.

et al., 1992) and Acinetobacter strain 4-CB1(COPLEYnd CROOKS,992).

6.3.4 Dehydrohalogenation

Dehydrohalogenation of an aromatic yieldsa benzyne which rapidly undergoes additionreactions. Although this reaction can beachieved under extreme abiotic conditions(sodium hydroxide, 300"C), there is no knownenzyme catalyzed example. The correspondingreaction of the alicycle lindane (82,Fig. 32) andin the aliphatic moiety of DDT are well known

(see Sect. 6.2.4).

6.4 Advances Towards the

Commercial Application of

Dehalogenases

Dehalogenases have a key role to play in thedegradation of persistent haloorganic com-pounds. Compounds such as lindane (82, Fig.

32) DDT, and PCBs have been considered

above but it is well known that a large numberof commonly isolated microorganisms, such asPseudomonas and Alcaligenes, are able to de-halogenate a wide variety of halogenated com-pounds. Inevitably, the restricted range of se-lective conditions employed in the laboratoryhas, up to the present time, resulted in very fewwell characterized anaerobes being isolated.There can be little doubt that these complexconsortia are key catalysts in the turnover ofhaloaromatic compounds in the environment.

Microorganisms able to dehalogenate spe-

cific compounds are finding some uses in the

commercial sector. (S)-2-Chloropropionic acidis a key chiral synthon required for the synthe-sis of a range of pharmaceuticals and agro-

chemicals. Initial attempts to resolve racemic2-chloropropionic acid with hydrolytic en-zymes (CAMBOU nd KLIBANOV,984) weresuperseded by the use of an (R)-specific deha-logenase enzyme from Pseudomonas putida(Fig. 38). Thus from racemic chloropropionicacid (105, 106) the (R)-enantiomer (105) isconverted to (S)-lactate (107) with inversionof configuration leaving the (S)-chloropro-pionate (106)behind (ee 96--98%).This systemhas been commercialized by Zeneca BioMole-

cules with the enzyme being produced at>1000t a- ' at Chilton,Teeside.The bioremediation of halogenated organics

in effluent treatment has attracted much re-search interest. However, the application of abiocatalyst for the removal of halogenatedorganics in a product stream is novel and maynot require any significant alteration to themanufacturing process. Carbury Herne Ltd., acompany based in Canterbury and Cardiff, hascollaborated with Hercules Inc.. a multination-

105 107

( R ) - 2 - ~ h l o r o p r o p i o n a t e ( S ) - L a c t a t e

c1- 1 0 6

/ ~ C O , H ( S ) - 2 - ~ h l o r o p r o p i o n a t e

Fig. 38. See text.



212 3 Halocompounds -Biosynthesis and Biotra

a1 speciality chemical company based in Wilm-ington, Delaw are, USA, and they have success-fully developed such a process. Products likepaper towels, tissues, and packaging materials

need a measu re of wet strength if they a re toperform properly. This property is provided bya polymer which is produce d w hen epichloro-hydrin is reacted w ith an aqu eous solution of apoly(am inoam ide). Unfortunately, traces of thehaloalcohols 1,3-dichloropropan-2-ol (DCP)and 3-chloropropanediol (C PD ) are producedas by-products of this reaction. These com-pounds are contaminants and have no benefi-cial effects on the finished product. Using awell characterized microbial consortia, Agro-

bacter ium tumefaciens NCIMB 40313 and A r -

throbacter h i s t id inolovorans NCIMB 40 14,these molecules are mineralized via glycidoland glucose. This process, incorporating a de-halogenase step, has now been adopted at theproduction scale at two manufacturing sitesproducing several thousand tonnes polymerper annum.

7 Final CommentsIn this review we have attem pted to present

an up-to-date appraisal of halogenated mole-cules: their synthesis and transformation bybiological systems. Obviously, there will beomissions that m ay have enhan ced th e articlebut several facts are clear:

- The principal route to halocompoundsynthesis is that catalyzed by haloperox i-dases. Ho wever, this class of en zym esdoes not app ear to catalyze the selective(regio-, stereo-, and enantio-) inco rpora-tion of h alides which obviously occur innature. It is therefore likely that o ther,novel biohalogenation mechanisms re-main to be discovered.

- Of th e reactions that the haloperoxidasesare able to catalyze, it is the halide inde-pendent reactions that have demon strat-ed excellent synthe tic utility.

- A par t from a limited range of haloperox-

idases, no enzymes able to catalyze eitherthe form ation or dehalogenation of halo-

rnsformation

compounds are commercially available atthe current time.

It is clear that far more effort has been in-

vested in understanding the effects and persis-tence of halocompounds in the workplace andthe env ironment than elaborating novel routesfor their synthesis. Each of these activities areclearly overlapping and will require continuedresearch a nd discovery if funda mental u nder-standing and commercial exploitation are tooccur concurrently.

8 ReferencesABRAMOWICZ,. A. (1990),Aerobic and anaerobic

biodegradationof PCBs: A review, Crit. Rev. Bio-tech. 10,241-251.

AHMED,M., FOCHTD. D. (1973), Degradation of

polychlorinated biphenyls by two species of

Achromobacter, Can.J. Microbiol. 19,47-52.ALLAIN,. J.,HAGER,. P.,DENG,L.,ACOBSEN,.N.

(1993), Highly enantioselective epoxidation of

disubstituted alkenes with hydrogen peroxide

catalyzed by chloroperoxidase, . Am . Chem. Soc.1 1 5 , 4 4 1 5 4 1 6 .

AMIN,R. A., HARPER, . B., MOLONEY,. M., MUR-PHY, C. D., HOWARD,. K. A., O'HAGAN, . (1997),A short and stereoselectivesynthesis of the fluor-

inated natural product (2S,3S)-4-fluorothreonine,

J . C. S. Chem.Commun.1997,1471-1472.ANDO,K., KATO,A., SUZUKI,. (1970), Isolation of

2,4-dichlorophenol from a soil fungus and its bio-

logical significance, Biochem. Biophys. Res. Com-

mun.39,1104-1107.ASPLUND,. (1995), Origin and occurrence of halo-

genated matter in soil, in: Naturally-ProducedOrganohalogens (GRIMVALL,,, LEERE. W. B.,

Eds.), pp. 35-48. Dordrecht: Kluwer.

ASPLUND, ., GRIMVALL,. (1991), Organohalogensin nature, more widespread than previously as-sumed,Environ. Sci. Technol. 25,1346-1350.

ATTIEH,. M., HANSON, . D., SAINI,H. S. (1995),Purification and characterization of a novel

methyltransferase responsible for biosynthesis of

halomethanes and methane thiol in Brassica ole-racea,J. Biol. Che m. 270,9250-9257.

BANTLEON,., ALTENBUCHNER,., VAN PEE, K.-H.

(1994), Chloroperoxidase rom Streptomyces livi-duns - solation and characterization of the en-

zyme and the corresponding gene, J. Bacteriol.

BELKIN,. (1992), Biodegradation of haloalkanes,

176,2339-2347.

Biodegradation 3,299-313.



8 References 213

BONGS, .,VA N PEE,K. H. (1994), Enzymatic chlori-

nation using nonheme haloperoxidases, EnzymeMicrob. Technol.16,5340.

BRENNER,~.,ERNANDEZ,. S., FOCHT,D. D. (1993),

Variation in chlorobenzoate catabolism by Pseu-dornonasputida P l l l as a consequence of genetic

alteration,Ap pl. E nviron. Microbiol.59,2790-2794.

BROWN,. F.,WAGNER, . E., FENG,H., BEDARD, .

L., BRENNAN, . J., CARNAHAN,. C., RAY,R. J.

(1987), Environmental dechlorination of PCBs,

Environ. T oxicol. Chem . 6,579-593.

BURD,W., YOURKEVICH,.,VOSKOBOEV,. J., VAN

PEE,K. H. (1995), Purification and properties of a

nonheme chloroperoxidase from Serratia mar-cescens, FEMS Microb iol. Lett. 129,255-260.

BUTLER,A., ALKER,.V. (1993),Marine haloperox-

idases, Chem. Rev. 93,1937-1944.

CAMBOU,., KLIBANOV,. M. (1984), Lipase-cata-lyzed production of optically-activeacids via asym-

metric hydrolysis of esters - effect of the alcohol

moiety,Ap pl. Biochem. Biotechnol.9,255-260.

CANNARSA,. J. (1996), Single enantiomer drugs:

new strategies and directions,Chem. Ind., 374-378.

CARMICHAEL,. D., PICKARD, . A. (1989), Contin-

uous and batch-production of chloroperoxidase

by mycelial pellets of Caldariomyces furnago in

an airlift fermenter,Ap pl. Environ. Microbiol.55,

CASTRO,. E.,BARTNICKI,.W. (1968), Biodehalog-

enation. Epoxidation of halohydrins, epoxide

opening, and transdehalogenation by a Flavobac-terium species,Biochemistry 7,3213-3218.

CHAN,W. R.,TINTo,W. F.,MANCHAND,. S.,TODARO,

L. J. (1987), Stereostructures of geodiamolide-A

and geodiamolide-B, novel cyclodepsipeptides

from the marine sponge Geodia sp.,J. Org. Chem.

CHANG, . H., LIANG,? H., BECK,W., SCHOLTEN,

J. D., DUNAWAY-MARINO,. (1992), Isolation

and characterization of the three polypeptide

components of 4-chlorobenzoate dehalogenase

from Pseudomonas sp. CBS3, Biochemistry 31,

CHEN, . P., LINCOLN,. E., WOODIN,. A., LOVELL,C. R. (1991), Purification and properties of a

unique flavin-containing chloroperoxidase from

the capitellid polychaete Notomastus lobatus, J.

Biol. Chem. 266,23 909-23 915.

COLONNA,., GAGGERO,., CASELLA,., CARREA,

G., PASTA,. (1993), Enantioselective epoxidation

of styrene derivatives by chloroperoxidase cata-

lysts, Tetrahedr0n:Asymmetry 4,1325-1330.

CONTESTABILE,., JOHN,R. A. (1996), The mecha-

nism of high-yielding chiral syntheses catalyzed

by wild-type and mutant forms of aspartate-ami-

notransferase, Eur.J . Biochem. 240,15Ck155.

COOK,A. M., HUTTER,R. (1986), Ring dechlorina-tion of deethylsimazine by hydrolases from

17-20.

52,3091-3093.

5605-5610.

Rhodococcus corallinus, FEMS Microb iol. Lett.34,335-338.

COPLEY,.D., CROOKS,. P. (1992), Enzymatic deha-

logenation of 4-chlorobenzoyl coenzyme A in

Acinetobacter sp. strain 4-CBI, Appl. Environ.Microbiol. 58,138.5-1387.

COUGHLIN,?, ROBERTS,., RUSH,C., WILLETTS, .

(1993), Biotransformation of alkenes by haloper-

oxidases - regiospecific bromohydrin formation

from cinnamyl substrate, Biotechnol. Lett. 15,

CURRAGH,., FLY”, O.,LARKIN, . J.,STAFFORD,T.

M., HAMILTON,. T. G., HARPER,D. B. (1994),

Haloalkane degradation and assimilation by

Rhodococcus rhodochrous NCIMB 13064, Mi-crobiology 140,1433-1442.

DE BOER,E., PLAT,H.,TRoMP,M. G. M., WEVER, .,

FRANSSEN,. c . R.,VAN ER PLAS,H.

c.,MEIJER,

E. M., SCHOEMAKER,. E. (1987), Vanadium

containing bromoperoxidase - an example of an

oxidoreductase with high operational stability in

aqueous and organic media, Biotechnol. Bioeng.30,607610.

DE JONG,E., FIELD, . A., SPINNLER,. E., WIJNN-

BERG, J. B. P. A., D E BONT,. A. M. (1994), Signifi-

cant biogenesis of chlorinated aromatics by fungi

in natural environments, Appl. Environ. Mi-crobiol. 60,264-270.

DE SILVA,E. D., ANDERSEN,. J., ALLEN,T. M.

(1990), Geodiamolide-C to geodiamolide-F, new

cytotoxic cyclodepsipeptides from the marinesponge Pseudaxinyssa sp., Tetrahedron Lett. 31,

489492.

DOLFING,., BUERSKENS,. E. M. (1995), The micro-

bial logic and environmental significance of re-

ductive dehalogenation, Adv. Microb. Ecol. 14,

DOLFING,., VA N DER WIJNGAARD, A. J., JANSSEN,D.

B. (1993), Microbiological aspects of the removal

of chlorinated hydrocarbons from air, Biodegra-dation 4,261-282.

DRAUZ,K., WALDMANN,. (Eds.) (1995), EnzymeCatalysis in O rgan ic Synthesis,pp. 783-791. Wein-

heim: VCH.EGLI,C.,TSCHAN,T.,CHOLTZ,., COOK,A. M., LEI-

SINGER,. (1988), Transformation of tetrachlo-

romethane to dichloromethane and carbon diox-

ide by Acetobacterium w oodii,Ap pl. Environ. M i-

crobiol. 54,2819-2834.

EGLI,C., STROMEYER,., COOK,A.M., LEISINGER,.

(1990), Transformation of tetra- and trichlorome-

thane to CO, by anaerobic bacteria is a non-enzy-

matic process,FEMS M icrobiol. Lett. 68,207-212.

ELANDER,. P., MABE, . A., HAMILL, . H ., GOR-

MAN,M. (1968), Metabolism of tryptophans by

Pseudomonas aureofaciens.I1 Production of pyr-

rolnitrin by selected Pseudomonas species,Appl .Microbiol. 16,753-7.58.

907-912.

143-206.




ELSNER, ., MULLER, ., LINGENS,. (1991), Sepa-rate cloning and expression analysis of two pro-tein components of 4-chlorobenzoate dehalogen-ase from Pseudomonas sp. CBS3, J . Gen. Micro-

biol. 137,477481.ENGESSER,.-H., SCHULTE,. (1989), Degradation

of 2-bromo, 2-chloro and 2-fluorobenzoate byPseudomonas putida CLB250,FEMS Microbiol.Lett. 60,143-148.

FACEY,. J., GRO SS, ., VINING , . C., YANG ,K., VA N

PEE, K.-H. (1996), Cloning, sequencing and dis-ruption of a bromoperoxidase-catalase gene inStreptomyces venezuelae: evidence that it is notrequired for chlorination in chloramphenicol bio-synthesis,Microbiology 142,657-665.

FEN ICA L, . (1974), Polyhaloketones from the redseaweed Asparagopsis taxiformis, TetrahedronLett. 51/52 ,4463466.

FEN ICA L, . (1979), Mo lecular aspects of halogen-based biosyn thesis of marine natural produ cts,Recent Ad v. Phytochem. 13,219-239.

FEN ICA L, . (1982), Natural products chemistry inthe marine environment, Science 215,923-928.

FENICAL,. (1993), Chemical studies of marine bac-teria - developing a new resource, Chem. Rev. 93,

1673-1682.FETZNER,., LINGENS,. (1994), Bacterial dehalo-

genases: Biochemistry, genetics and biotechno-logical applications,Microbiol. Rev. 58,641-685.

FETZNER,.,MULLER, ., LINGENS,. (1989), Degra-dation of 2-chlorobenzoate by Pseudomonascepacia 2CBS, Biol. Chem. Hoppe-Seyler 370,1173-1 182.

FETZNER,., MULLER , ., LINGENS,. (1992), Purifi-cation and some properties of 2-halobenzoate 1,2-dioxygenase, a two-component enzyme systemfrom Pseudomonas cepacia 2CBS, J . Bacteriol.174,279-290.

FIELD,. A., VERHAGEN,. J. M., DE JONG,E. (1995),Natura l organohalogen production by basidiomy-cetes, Trends Biotechnol. 13,451456.

FIELD ING, . (1992), in: Water Pollution ResearchReports: Pesticides in Ground and Drinking WaterReport No. 27 (FIELDING,., Ed .), pp. 5-11 (pu b-lication of the Commission of the EuropeanCom munities, Directorate Gen eral for Science,Research and Development).

FLAVIN, . , CAST RO-~~EN DO ZA,., OCHO A,. (1957),Metabolism of propionic acid in animal tissues.I1 Propionyl coenzyme A carboxylation system,J . Biol. Chem. 229,981.

FOWDEN,. (1968), The occurrence and metabolismof carbon-halogen compounds, Proc. R. SOC.

London B 171,5-18.FRANKEN,.M., ROZEBOOM ,H.., KALK , . H., DIJK-

STRA, B. W. (1991), Crystal structure of halo-alkane dehalogenase: an enzyme to detoxify ha-logenated alkanes,EMBO J. 10,1297-12302.

FRANSSEN,. C. R. (1994), Haloperoxidases: usefulcatalysts for halogenation and oxidation reac-tions, Catalysis Today 22,441-457.

FRANSSEN,. c.R., VAN DE R RAS , H. c.B. (1992),

Haloperoxidases- heir properties and their use inorganic synthesis,Adv . App l. Microbiol. 37,41-99.

Fu, H., KON DO, ., ICHIKAWA,~.,OO K, . C., WO NG,C.-H. (1992), Chloroperoxidase-catalyzedasym-metric synthesis - enantioselective reactions ofchiral hydroperoxidase with sulfides and bromo-hydration of glycals,J . Org. Chem. 57,7265-7270.

FURUKAWA,., MATSUMURA,., TONOMURA,.(1978), Alcaligenes and Acinetobacter strainscapable of degrading polychlorinated biphenyls,Agric. Biol. Chem. 42,543-548.

GEIGERT,., NEIDLEMAN,. L., DALIETOS,. J., DE-WITT,S. K. (1983a), Haloperox idases - enzymaticsynthesis of a ,P-haloh ydrins from gaseous al-kenes,Ap pl. Environ. Microbiol. 45,366-314.

GEIGERT,., NEIDLEMAN,.L.,DALIETOS,.J. (1983b),Novel haloperoxidase substrates - alkynes andcyclopropanes,J. Biol. Chem. 258,2273-2277.

GEIGERT,., DEWITT, . K., NEIDLEMAN,. L., DALIE -TOS, D. J.,MORELAND,. (1983c),DMS O is a sub-strate for chloroperoxidase, Biochem. Biophys.Res. Comm un. 116,8245.

GOLOVLEVA,. A., ZABOFUNA,., F’ERTSOVA,.,BASKUNOV,., SCHURUK HIN,Y.,UZMIN,. (1992),Degradation of polychlorinated p henols by Strep-

tomyces rochei 303, Biodegradation 2,201-208.GREGSON,. P.,BALDO, . A.,THOM AS,. G., QU INN,

R. J., BERGQUIST,. R., STEPHENS,. R., HORNE,A.R. (1979), Fluorine is a major constituent of themarine sponge Halichondria moo rei, Science 206,110&1109.

GR IBBLE , . W. (1992), Naturally occurring organo-halogen compounds -A survey,J. Nat. Prod. 55,

GR IBB LE, . W. (1%4), Natural organohalogens,J . Chem. Educ. 71,907-911.

GR ISHA M, . B., JEFFERSON, . M., MELTO N, . F.,THO MA S, . L. (1984), Chlorination of endoge-nous amines by isolated neutrophils - ammonia-depen dent bactericidal, cyto-toxic, and cytolyticactivities of the chloramines, J . Biol. Chem. 259,10404-10413.

GSCHWEND,. M., MACFARLAND,. K., NEWMAN,.A. (1985), Volatile halogenated organic com-pounds released to seawater from tempe rate ma-rine macroalgae,Science 227,1033-1035.

HAAG,T., LINGENS,., VA N PEE, K.-H. (1991), A

metal-ion-independent and cofactor-independentenzymatic redox reaction - halogenation by bac-terial nonheme haloperoxidases, Angew. Chem.(Int. Ed n. E ngl.) 30,1487-1488.

HAGER,L. P. (1982), Mother nature likes somehalogenated compounds, Basic Life Sci. 19 ,4 15429.

1353-1395.



8 References 215

HAGER,.P., MORRIS, . R., BROWN,. S.,EBERWEIN,

H. (1966), Chloroperoxidase. I1 Utilization of

halogen anions,J. Biol. Chem. 241,1769-1776.

HAGGBLOM,. M., RIVERA,M. D., YOUNG, . Y.

(1993), Influence of alternative electron accep-tors on the anaerobic biodegradability of chlori-

nated phenols and benzoic acids,Ap pl. Environ.Microbiol. 59,1162-1167.

HAIDER,. (1979), Degradation and metabolization

of lindane and other hexachlorocyclohexane iso-

mers by anaerobic and aerobic soil microorga-

nisms,2.Naturforsch. (Sect. C ) 34,1066-1069.

HAMILTON,. T. G., AMIN,M. R., HARPER, . B.,

O’HAGAN, . (1997), Biosynthesis of fluoroace-

tate and 4-fluorothreonine by Streptomyces catt-leya. Glycine and pyruvate as precursors, J . C. S.Chem. Commun. 797-798.

HARDMAN,. J. (1991), Biotransformation of halo-genated compounds,Crit. Rev. Biotech. 11,140.

HARPER,. B. (1985), Halomethane from halide ion

- a highly efficient fungal conversion of environ-

mental significance,Nature 315,55-57.

HARPER, . B., HAMILTON,.T. G. (1988), Biosynthe-

sis of chloromethane in Phellinus pomaceus,J. Gen. M icrobiol. 134,2831-2839.

HARPER,. B., O’HAGAN, . (1994),The fluorinated

natural products,Nut. Prod. Rep. 11,123-133.

\RPER, D. B., HAMILTON,. T. G., O’HAGAN,D.(1990), Identification of threo-18-fluoro-9,lO-

dihydroxystearic acid - A novel *fluorinated

fatty-acid from Dichapetalum toxicarium seeds,Tetrahedron Lett. 31,7661-7662.

WTZEL, R., NKE,H., SHELDRICK, w. s. (1990),

Mycenon, a new metabolite from a Mycena spe-

cies TA-87202 (Basidomycetes) as an inhibitor of

isocitrate lvase. J.Antibiot. 43.1240-1244.

HECHT, . J., h E K , H., HAAG, ., PFEIFER, ., AN

PEE,K. H. (1994),The metal ion free oxidoreduct-

ase from Streptomyces aureofuciens has an alp

hydrolase fold,Nut. Struct. Biol. 1,523-537.

HEWSON,. D., HAGER, . I? (1980), Bromoperoxi-

dases and halogenated metabolites in marine

algae,J. Phycol. 16,340-345.

HICKEY, . J., FOCHT,D. D. (1990), Degradation ofmono-, di- and trihalogenated benzoic acids by

Pseudomonas aeruginosa JB2, Appl. Environ.Microbiol. 56,3842-3850.

HOLLIGER,., SCHRAA, ., STAMS,A. J. M., ZEHN-

DER, A. J. B. (1993), A highly purified enrichment

culture couples the reductive dechlorination of

tetrachloroethene to growth, Appl. Environ.Microbiol. 59,2991-2997.

HOWARD,? H., BOETHLING,. S., STITELER, . M.,

MEYLAN, . M., HUEBER, . E., BEAUMAN,. A.,

LAROSCHE, . E. (1992), Predictive model for

aerobic biodegradability developed from a file of

evaluated biodegradation data, Environ. Toxicol.Chem. 11,593403.

HUNTER,. C., FONDA, ., SOTOS,L.,Toso,B., BELT,

A. (1984), Ecological approaches to isolation,

Dev. Ind. Microbiol. 25,247-266.

IMAI,R., NAGATA, ., FUKADA,M., TAGAKI,M.,

YANO,M. (1991), Molecular cloning of a Pseudo-monas paucimobilis gene encoding a 17-kilodal-

ton polypeptide that eliminates HCl from y-hexa-

chlorocyclohexane, . Bacteriol. 173,6811-6819.

ISHIDA, . (1968), Comparative studies on BHC me-

tabolizing enzymes, DDT dehydrochlorinase and

glutathione S-transferase,Agric. Biol. Chem. 32,

947-955.

ISIDOROV,. A. (1990), Organic Chemistry of theEarths Atmosphere. Berlin: Springer-Verlag.

ITOH, N., CHENG,L. Y., IZUMI,Y., YAMADA, .

(1987a), Immobilized bromoperoxidase of Coral-linu pilulifera as a multifunctional halogenating

biocatalyst,J . Biotechnol. 5,29-38.ITOH, N., IZUMI,Y., YAMADA, . (1987b), Halo-

peroxidase-catalyzed halogenation of nitrogen-

containing aromatic heterocycles represented by

nucleic bases,Biochemistry 26,282-289.

ITOH,N., HASAN,A. . M. Q., IZUMI,Y.,YAMADA,.

(1988), Substrate-specificity, egiospecificity and

stereospecificity of halogenation reactions cata-

lyzed by non-heme-type bromoperoxidase of Co-rallina piluliferu, Eur. J. Biochem. 172,477-484.

ITOH, ., MORINAGA,., KONZAI,. (1994), Purifica-

tion and characterization of a novel metal-

containing nonheme bromoperoxidase from

Pseudomonas putidu, Biochim. Biophys. Acta1207,208-216.

JANNUN,., COE, E. L. (1987), Bromoperoxidase

from the marine snail Murex trunculis, Comp.Biochem. Physiol. 88,917-922.

JANSSEN,. B., SCNEPER, ., DIJKHUIZEN,., WIT-

HOLT, B. (1985), Degradation of halogenated

aliphatic compounds by Xanthobacter autotrophi-cus GJ10,Ap pl. Environ. Microbiol. 49,673-677.

JANSSEN,. B., FRIES, ., VA N DER PLOEG,., KAZE-

MIER, B., TEPSTRA,?,WITHOLT,. (1989), Cloning

of 1,2-dichloroethane degradation genes of Xan-thobucter autotrophicus GJlO and expression and

sequencing of the dhla gene, J . Bacteriol. 171,6791-6799.

JANSSEN,. B., PRIES,F.,VA N DER PLOEG,. R. (1994),

Genetics and biochemistry of dehalogenating

enzymes,Ann. Rev. M icrobiol. 48,163-191.

JENSEN,? R., FENICAL,. (1994), Strategies for di-

covery of secondary metabolites from marine

bacteria - Ecological perpectives,Ann . Rev. Mi-crobiol. 48,559-584.

JOHNSTON,. W., BRIGGS,G. G., ALEXANDER,.

(1972), Metabolism of 3-chlorobenzoic acid by a

pseudomonad, Soil Biol. Biochem. 4,187-190.

JOHRI, . , QUAZI,G. N., CHOPRA, . L. (1991), Evi-

dence of plasmid mediated dechlorinase activityin Pseudomonas sp.,J . Biotechnol. 20,73-82.



216 3 Halocompounds - Biosynthesis and Biotransformation

JONES,D. H.A., BARTH,! T., BYROM , ., THOMAS,.M. (1992), Nucleotide sequence of the structuralgene encoding a 2-haloalkanoic acid dehalogen-ase of Pseudomonas putida strain AJ1 an d

purification of the encoded protein, J. Gen.Microbiol. 138,675-683.

KAWASAKI,., TSUDA , ., MATSUHITA,.,TONOMU-RA, K. (1992), Lack of hom ology betwe en twohaloacetate dehalogenase genes encoded on aplasmid from Moraxella species strain B , J . Gen.

Microbiol. l38,1317-1323.KAWASAKI,., TOYAM A,., MAEDA, .,NISHINO, .,

T O N O M U RA ,. (1994), Cloning and sequenceanalysis of a plasmid-encoded 2-haloacid deha-logenase gene from Pseudomons putida no. 109,Biosci. Biotech. B iochern.58,160-163.

KEUNING,

s.,ANSSEN,. B.,WITHOLT, . (1985), Pur-

ification an d characterization of hydrolytic halo-alkane dehalogenase from Xanthobacter auto-trophicus GJ10, J. Bacteriol. 163,635-639.

KITAGAWA,., FUKUDA, ., TANIYAMA,., YOSHI-KAWA, M. (1 99 9, Chemical studies on crude drugprocessing. 8.O n the constituents of Rehmanniae

radix. 2. Absolu te stereostruc tures of rehmaglu-tin-C and glutinoside isolated from Chinese Reh-manniae radix, the dried root of Rehman nia gluti-nosa libosch, Chem. Pharm. Bull. 43,1096-1100.

KIYOHARA,., HAITA, T., OG AW A, ., KA KU DA , .,YOKOYAMA,., TAKIZAWA,. (1992), Isolation

of Pseudomonas pickettii strains that degrade2,4,6-trichlorophenol and their dechlorination ofchlorophenols, Appl. Environ. Microbiol. 58 ,1276-1283.

KLAGES, ., KRAUSS,., LINGENS,. (1983), 2-Halo-acid dehalogenase from a 4-chlorobenzoate-de-

grading Pseudomonas spec. CBS 3, Hoppe-Sey-ler’s Z . Physiol. Chem. 364,529-535.

KOBAYASHI,., KANDA, ., ISHIBASHI, ., SHIG EM OR I,H. (1991), Manzacidins AC, novel tetrahydro-pyrimidine alkaloids from the Okinawan marinesponge Hyrneniacidon sp., J. Org. Chem. 56,

45744576.KOHLER-STAUB,., LEISINGER,. (1985), Dichlo-romethane dehalogenase of Hyphomicrobiumspecies strain DM 2, J . Bacteriol. 162,676-681.

KONDO,K., NISHI,J., ISHIBASHI,., KOBAYASHI,.(1994),2 New tryptophan-derived alkaloids fromthe Okinawan m arine sponge Aplysina sp.,J.Nat.Prod. (Lloydia) 57,1008-1011.

KUHN,E. P., TOWNSEND,. T., SUFL ITA,. M. (1990),Effect of sulfate and organic carbon sup plemen tson reductive dehalogenation of chloroanilines inanaerobic aquifer slurries, Ap pl. Environ. Micro-biol. 56,2630-2637.

LEE,M. D., ELLESTAD,.A., BORDERS,. B. (1991),Calicheamicins - discovery, structure, chemistryand interaction with DNA, Acc. Chem. Res. 24 ,235-243.

LEE, M. D., DU NN E, . S., CHANG, . C., SIEGEL,M.M., MORT ON, .O., ELLESTAD,.A. , MCGAHREN,W. J., BORD ERS, . B. (1992), Calicheamicins, anovel family of an titumor antibiotics. 4. Structure

elucidation of calicheamicins PIBr , , ~ ‘ ,a2’, a31,PI1,yll, I , J.Am . Chem.SOC. 14,985-997.

LI , J. Z., SHA RM A, ., DILEEPA N, . N., SAW N,V. J.(1994), Polymorphonuclear leukocytes increaseglomerular albumin permeability via hypohalousacid, Kidney Znt. 46,1025-1030.

LIPKE, H., KEARNS,C. W. (1959), DDT dehy-drochlorinase. I Isolation. chemical DroDerties.and spectrophotometric assay,J . Biol. khdm. 234:

2123-2128.LIPKE,H.,KEAR NS, . W. (1959), D D T dehydroc hlor-

inase. I1 Substrate a nd cofactor specificity,J.Biol.Chem. 234,2129-2132.

LIU, K. K.-C., W ONG , C.-H. (1992), Enzym atichalohydration of glycals, J . Org. Chem. 57,3748-3750.

LOFFLER, ., MULLER, ., LINGENS,. (199 2), Purifi-cation and properties of 4-chlorobenzoate-coen-zyme A ligase from Pseudomonas sp. CBS3,Hoppe-Seyler’s Z . Physiol. Chem. 373,1001-1007.

MAGUIRE, . R., MENG,W.-D., ROBERT S,. M., WIL-LEmS, A. J. (1993), Synthetic approa ches towardsnucleocidin and selected analogs - anti-HIVactivity in 4 ‘-fluorinated nucleoside derivatives,J. Chem. SOC.Perkin Trans. 1,1795.

MANTHEY,. A., HAGER,L. P. (1981), Purificationand properties of bromopero xidase from Penicil-lus capitatus,J. Biol. Chem. 256,1232-1238.

MARSHALL,.C.,WR IGHT, . D. (1996), Purificationand characterization of 2 haloperoxidases from

the glycopeptide antibiotic producer Streptomy-

ces toyocaensis NRRL -15 009, Biochem. Biophys.Res. Comm un. 219,58&583.

MCCOR MICK, . L., RO EDER, . L., RAILSBACK,.A., BRITIGAN,B. E. (1994), Eosinophil per-oxidase-dependent hydroxyl radical generationby human eosinophils, J. Biol. Chem. 269,

27914-27919.MEAD,R. J., SEGAL,W. (1973), Form ation of P-

cyanoalanine and pyruvate by Accacia georginae,Phytochernistry 12,1977-1981.

MESSERSCHMIDT,., WEVE R,R. (1996), X-ray struc-ture of a vanadium-containing enzyme - chloro-peroxidase from the fungus Curvularia in-aequalis, Proc. Natl. Acad . Sci. US A 93,392-396.

MEYER,M., O’HAGA N, . (1992a), Ra re fluorinatednatural products, Chem. Brit. 28,785-788.

MEYER,. J. M., O’HAGAN, . (1992b), Conversion offluoropyruvate to fluoroacetate by Dichapetalumcymosum, Phytochemistry 31,499-501.

MEYER,. J. M., O’HAGAN, . (1992c), Conv ersion of3-fluoropyruvate to fluoroacetate by cell-free ex-tracts of Dichapetalum cymosum, Phytochemistry31,2699-2701.



8 References 217

MILES, . H ., MO DY, .V., MINYARD ,. P., HEDIN,P.A.(1973), Co nstituents of marsh grass. Survey of th eessential oils in Juncus raemerianus, Phytochem -

istry 12,1399.MI URA, ., OTSUKA,., H O N D A , . (1956),Abnormal

metabolism in animals poisoned with sodiumfluoroacetate, Bull. Agric. Chem. Soc. Jpn. 20 ,

MOHN,W. W., TIEDJE,. M. (199 2), Microbial red uc-tive dehalogenation, Microbiol. Rev. 56,482-507.

MOORE, . E. (1977), Volatile com pounds from ma-rine algae, Acc. Chem. Res. 10,40-47.

MOTOSU GI, . , ESAK I, . , SODA,K. (1982), Pu rifica-tion and properties of 2-haloacid dehalogenasefrom Pseudomonas putida , Agric. Biol. Chem. 46,

M UL LE R, . (1992), Bacterial deg radation of x eno-biotics, in: Microbial Control of Pollution, SGMSymposium 48 (FRY,. C., GAD D,G. M., HERBERT,R . A., JONES,C. W., WATSON-CRAIK,. A., E ds.),pp. 35-57. Cam bridge: Cam bridge U niversityPress.

MURDIYATMO,., ASMARA,W., TSANG, . S. H.,BAINES , . J.,BULL,A . T.,HARDMAN,. J. (1992),Molecular biology of the 2-haloacid halidohy-drolase from Pseudomonas cepacia MBA4, Bio-chem. J. 284,87-93.

NAGASAWA,T.,AM ADA , . (1986), Enzym atic trans-formations of 3-chloroalanine into useful amino-acids,Ap pl. Biochem. Biotechnol. 13,147-165.

NAGASAWA,., NAK AM URA,., Yu, F., WATAN ABE,.,YA MA DA , . (1992), Purification and character-ization of halohydrin hydrogen-halide lyase froma recombinant Escherichia coli containing thegene from a Corynebacteriurn species, Appl . Mi-crobiol. B iotech. 36,478482.

N A G A T A , ~ . ,ATTA,T.,MAI,R .,KIMBARA,. , FUK U-DA,M., YANO,K., TAK AG I, . (1993), P urificationand characterization of y-hexachlorocyclohexane(y-HCH) dehydrochlorinase (LinA) from Pseu-domonas paucimobilis, Biosci. Biotechnol. Bio-

chem. 57,1582-1583.N A K A M U R A ,., NAGASA WA,.,Yu, F., WA TANA BE,.,

YAM ADA , . (1992), Resolution and some pro-perties of enzymes involved in enantioselectivetransformation of 1,3-dichloro-2-propanol to (R)-

3-chloro-1,2-propanedioly Corynebacterium sp.strain N-1074, J . Bacteriol. 174,7613-7619.

NA RDI-D EI, ., KURIHA RA, .,OKAMURA, .,LIU,J.Q., KOSHIKAWA,. , OZAKI,H., TERASHIMA,.,ESA KI, ., SODA,K. (1994), Com parative studiesof genes encoding thermostable 1-2-halo acid de-halogenase from Pseudomonas species YL , otherdehalogenases, and two related hypothetical pro-teins from Escherichia co li,Ap pl. Environ. Micro-biol. 60,3375-3380.

NEIDLEMAN,. L. (1975), Microbial halogenation,CR C Crit. Rev. Microbiol. 5,333-358.

219-222.

837-838.

NEIDLEMAN,. L. (1980), Use of enzym es as catalystsfor alkene oxide production, Hydrocarbon Pro-cess. (Intl. Edn .) 59,135-138.

NEIDLEMAN,. L., GEIGER T,. (1986), Biohalogena-tion: Principles, Basic Roles and Applications.New Y ork: Halsted Press.

NEILSON,A . H., ALLARD,A . S., LINDGR EN, .,REMBERGER,. (1987 ), Transformation of chlo-roguaiacols, chloroveratroles and chlorocatecholsby stable consortia of anaerobic bacteria, Appl .Environ. Microbiol. 53,949-954.

NIESCHALK,., HAMILTON,. T. G., MURPHY, . D.,

HARPER, . B., O’HAG AN, . (1997), Biosynthesisof fluoroacetate and 4-fluorothreonine by Strep-

tomyces cattleya. The stereochemical processingof glycerol, J. C. S. Chem Commun. 799-800.

NUELL,M. J., FANG,G.-H., AXL EY,M. J.,KENISBERG,

P., HAG ER , . P. (1988 ), Isolation an d nu cleotide-sequence of the chloroperoxidase gene from Cal-dariomyces fumago,J. Bacteriol. 170,1007-1011.

O’HAGAN,D., PERRY,R., LOCK, . M., MEYER,. J.M., DASARADHI,. , HAMILTON,. T. G., HARPE,D. B. (1993), High levels of m ono fluoroac etatein Dichapetalum braunii, Phytochemistry 33 ,

OHTA,T., TAKAH ASH I,. , MATSUDA, . , KAMO,S.,AGATSUMA,T.,NDO,T.,NAZOE, . (1995),Russu-phelol, a novel optically active chlorohydroqui-none tetramer from the mushroom Russula sub-

nigricans, Tetrahedron Lett. 36,5223-5226,

OKAMOTO,., SHIMADA ,., SHIRA I, . , SAKAMOTO,H.,YOSHIDA,., OJIMA, ., ISHIG UR O,., SAK AI , .,KAWAGISHI,. (1993), Antimicrobial chlorinatedorcinol derivatives from mycelia of Hericium

erinaceum, Phytochemistry 34,1445-1446.PASTA, ., CA RR EA , ., CO LONN A,., GAGGERO , .

(1994), Effects of chloride o n th e kinetics andstereochemistry of chloroperoxidase catalyzedoxidation of sulfides, Biochim. Biophys. Acta

PELLETIER,., ALTENBUCHNER,. (1 99 3, A bacterialesterase is homologous with nonheme haloper-oxidases and displays brominating activity, M i-

crobiology 141,459468.PELLETIER,., PFEIFER,O., ALTENBUCHNER,., VAN

PEE , K.-H. (1994), Cloning of a second no nhe mebromo peroxidase gene from Streptomyces aureo-

faciens ATC C 10762. Sequence analysis, expres-sion in Streptomyces lividans and enzyme purific-ation, Microbiology 140,509-516.

PELLETIER,. , ALTENBUCHNER,.,MATTES, . (1 99 9,A catalylic triad is required by the no n-heme ha-loperoxidases to perform halogenation, Biochim.Biophys. Acta. 1250,149-157.

PETERS, . A., SHORTH OUSE,. (1964), Fluoride me-tabolism in p lants of Accacia georginae, Biochem.J. 93,P20.

1043-1 045.

1209,203-208.



218 3 Halocompounds -Biosynfhesis and Biofransformafion

PETERS, . A., SHORTHOUSE,. (1971), Identifica-tion of a volatile constituen t formed by homogen-ates of Accacia georginae exposed to fluoride,Nature 231,123-124.

P E T ~ Y , . A. (1961), An introduction to the originand biochemistry of microbial halometabolites,Bacteriol. Rev. 25,111-130.

PFEIFER,O., PELLETIER,., ALTENBUCHN ER,., VA N

PEE, K.-H. (1992), Molecular cloning an d se-quencing of a non-heme bromoperoxidase genefrom Streptomyces aureofaciens ATCC 10762, J .Gen . Microbiol. l38,1123-1131.

PICKARD, . A., K ADIMA ,T.., CARMICHEAL,. D.,(1991), Chloroperoxidase, a peroxidase with po-tential, J.Znd. Microbiol. 7,235-241.

PLAT, ., KREN,B. E., WE VER , . (1987),The bromo-

peroxidase from the lichen Xanthoria parietina isa novel vanadium enzyme, Biochem. J. 248,

277-279.PRIES,F.,VAN DER PLOEG, J. R., DOL FIN G,., JANSSEN ,

D. B. (1994), Degradation of halogena ted aliphat-ic compounds: The role of adaptation, FEMS Mi-

crobiol. Rev. 15,279-295.RADEHAUS,. M., SCHMIDT,. K. (1992), Cha racter-

ization of a novel Pseudomonas sp. that miner-alizes high concentrations of pentachlorophenol,Ap pl. Environ. Microbiol. 58,2879-2885.

RAMANAND,., BALBA,M. T., DU FFY ,. (1993), Re -ductive dehalogenation of chlorinated benzenes

and toluenes under methanogenic conditions,Ap pl. Environ. Microbiol. 59,3266-3272.

RASCHE, . E., HICKS, . E., HYM AN, . R.,ARP,D.J. (1990), Oxidation of m onohalogenated etha nesand N-ch lorinated alkanes by whole cells of Nitro-

somonas europaea, J.Bacteriol. 172,5368-5373.RASMUSSEN,. A., RASMUSSEN,. E., KHALI,M. A.

K., DALU GE, . W. (1980), Con centration distribu-tion of methyl chloride in the atmos phere,J. Geo-

phys. Res. 85,7350-7356.REIN EKE , ., KNACKMUSS,.-J. (1988),Microbial de-

gradation of haloaromatics, An n. Rev. Microbiol.

ROBINSON,., LENN ,M. J. (1994), The bioremedia-tion of polychlorinated biphenyls (PCBs): Prob-lems and p erspectives,Biotechnol. Gen . Eng. Rev.12,139-188.

ROMAN O, ., MELO ,C., BARALLE,.,DRI,P. (1992),

cDNA of myeloperoxidase (MPO) and eosino-phi1 peroxidase (EP O ) from hum an blood mon o-nuclear-cells differentiated into granulocyte pre-cursors in liquid cultures, J. Immunol. Methods154,265-267.

SAINI, .S.,ATTICH,. M.,HAN soN,A.D. (1995),Bio-synthes is of halom ethanes an d methanethiol by

higher plants via a novel methyltransferase reac-tion, Plant Cell Environ. 18,1027-1033.SALLIS,. J., ARMFIELD,. J., BULL,A. T.,HARDMAN,

D. J. (1990), Isolation and characterisation of a

42,263-287.

haloalkane halidohydrolase from Rhodococcus

erythropolis strain Y2, J. Gen. Microbiol. 136,

SANA DA, ., MIYAN O,., IWADARE,., WILLIAMSON ,

J. M.,AR ISON ,. H., SMITH,. L., DOUG LAS ,A. .,LIESCH,. M., INAMINE,. (1986), Biosynthesis offluorothreonine and fluoroacetic acid by thethienamycin producer, Streptomyces cattleya, J.

Antibiot. 39,259-265.SCHENK,T., ULLER, ., LINGENS,. (1990), Mecha-

nism of enzym atic dehalogenation of pentachlo-rophenol by Arthrobacter sp. strain ATC C 33 790,J . Bacteriol. 172,7272-7274.

SCHMITZ, ., GARTEMAN N,. H., FIELDER, .,GRUND, ., EICHENLAUB,. (1992), Cloning andsequence analysis of genes for dehalogenation of

4-chlorobenzoate from Arthrobacter sp. strainSU,Ap pl. Environ. Microbiol. 58,40684071.

SCHNEIDER,., MU LLER, ., FR AN K, ., LINGENS,.(1991), Com plete nucleotide sequences and com-parison of the structural genes of two 2-haloalka -noic acid dehalogenases from Pseudomonas sp.strain CBS 3,J. Bacteriol. 173,1530-1535.

SCHNEIDER,., BARTH,W., BOHME,H. J. (1996),Cloning and characterization of another ligninperoxidase gene from the white-rot fungus Pha-

naerochaete chrysosporium, Biol. Chem. Hoppe-

Seyler 371,399-402.SCHOLTEN,. D., CH AN G, . H., BABBIT,? C., CHA R-

EST, H., SYLVESTRE,., DUNAWAY-MARINO,.(1991), Novel enzymatic hydrolytic dehalogena-tion of a chlorinated aromatic, Science 253,182-185.

SCHOLTZ,., LEISINGER ,., SUTER, ., COO K,A. M.(1987), Characterization of 1-chlorohexanehalido-hydro lase, a deha logenase of wide subs trate rangefrom an Arthrobacter species, J. Bacteriol. 169,5016-5021.

Science Advisory Board of the International JointComm ission of the G rea t Lakes (1989), 1989 Re-port. Int. J. Com mission, Hamilton, O ntario, Oc-

tober 1989.SEARLE,. A,,MOLINSKI,T.. (1994), 5 New alkaloidsfrom the tropical ascidian, Lissoclinum sp. -lissoclinotoxin-A is chiral, J. Org. Chem. 59,6600-6605.

SEELBACH,., VAN DEURZEN, . P. L., VAN RANT-WIJK, F., SHELDON,. A., KRAG L,u. (1997), Im-provement of the total turnover number andspace-time yield for chloroperoxidase catalyzedoxidation,Biotechnol. Bioeng. 55,283-288.

SESIN, . F., IRELAND,. M. (1984), Iodinated prod-ucts from a d idemnid tunicate, Tetrahedron Lett.25,403-404.

SHEFFIELD, . J., HA RRY ,., SMITH, . J., ROGERS,.J. (1993), Purification and characterization of thevanadium bromoperoxidase from the macroalgaCorallina officinalis, Phytochemistry 32,21-26.

1 5-1 20.



8 References 219

SHEFFIELD,. J., HARRY,., SMITH, . J., ROGERS,.

J. (1994), Immobilization of bromoperoxidase

from Corallina officinalis, Biotechnol. Techniques

SHELTON,.

R., TIEDJE, . M. (1984), Isolation and

partial characterization of bacteria in an anaero-

bic consortium that mineralizes 3-chlorobenzoic

acid,Appl. Environ. Microbiol. 48,84&848.

SIMONS,. H., BARNETT,.,VOLLENBROEK,. G. M.,DEKKER,. L., MUIJSERS,A..,MESSERSCHMIDT,

A.,WEVER, . (1995),Primary structure and char-

acterization of the vanadium chloroperoxidase

from the fungus Curvularia ina equalis, Eur.J.Bio-

chem. 229,566-574.

SLATER,. H., BULL,A. T., HARDMAN,. J. (1995),

Microbial dehalogenation, Biodegradation 6,

SMITH, . L., NICOLAOU,. C. (1996), The enediyneantibiotics, . Med. Chem. 39,2103-2117.

SPINNLER,.-E., DE JONG, ., MAUVAIS,., SEMON,

E., LE QUERE,.-L. (1994), Production of halogen-

ated compounds by Bjerkandera adusta, A pp l.

Microbiol. Biotechnol. 42,212-221.

STUCKI, . R., GALLI, ., EBERSOLD,. R., LEISING-

ER, T. (1981), Dehalogenation of dichlorome-

thane by cell extracts of Hyphomicrobium DM2,

Arch. Microbiol. 130,881-886.

SUNDARAMOORTHY,., MAURO,. M., SULLIVAN,.

M., TERNER,., POULOS,. L. (1995), Preliminary

crystallographic analysis of chloroperoxidase

from Caldariomyces fumag o, Ac ta Crystallogr.51,

SUZUKI,., HONDA,., KATSUMATA,. (1972), Pro-

duction of antibacterial compounds analogous to

chloramphenicol by an n-paraffin-grown bacte-

rium,Agr.Biol. Chem. 36,2223-2228.

TAKAHASHI,., AGATSUMA,., OHTA,T., NUNO-

ZOWA,.,ENDO, ., NOZOE, . (1993), Russuphe-

lins B, C, D, E and F, new cytotoxic substances

from the mushroom Russula subnigricans HON-

GO, Chem . Pharm. Bull. 41,1726-1729.

TAMURA,.,WADA,M., ESAKI,N., SODA,K. (1995),

Synthesis of fluoroacetate from fluoride, glycerol,

and P-hydroxypyruvate by Streptomyces cattleya,J.Bacteriol. 177,2265-2269.

THOMAS,.A.,HAGER, . P. (1968), The peroxidation

of molecular iodine to iodate by chloroperoxi-

dase, Biochem. Biophys. Res. Commun. 32,

17&175.

IMLIN, C. (1995), The Pesticide Manual U K (10th

Edn.). R. SOC.Chem., Crop. Protection Publica-

tion (TOMLIN,., Ed.).

)PP,E., HANSON,. S. (1990), Degradation of pen-

tachlorophenol by a Flavobacterium species

grown in continuous culture under various nutri-

ent limitations, Appl. Environ. Microbiol. 56,

8,579-582.

181-189.

842-844.

54 1-544.

TSCHIRRET-GUTH,. A., BUTLER,A. (1994), Evi-

dence for organic substrate binding to vanadium

bromoperoxidase,J .Am . Chem.SOC.116,411-412.

Tu, C. M. (1976), Utilization and degradation of lin-

dane by soil microorganisms, Arch. Microbiol.108,259-263.

VAN DEN WIJNGAARD,A. J., JANSSEN,. B., WITHOLT,

B. (1989), Degradation of epichlorohydrin and

halohydrins by bacterial cultures isolated from

freshwater sediments, J. Gen. Microbiol. 135,

2199-2208.

VAN DE N WIJNGAARD, A. J., REUVEKAMP,. T. W.,

JANSSEN,. B. (1991), Purification and character-

ization of haloalcohol dehalogenase from Arthro-bacter sp. strain AD2, J. Bacteriol. 173,124-129.

VA N DER MEER, . R. (1994), Genetic adaptation of

bacteria to chlorinated aromatic compounds,

FEMS M icrobiol. Rev. 15,239-249.VAN DER PLOEG, . R., HALL,G., JANSSEN, . B.

(1991), Characterization of the haloacid deha-

logenase from Xanthobacter autotrophicus GJlO

and sequencing of the dhl B gene, J. Bacteriol.

173,7925-7933.

VAN PEE, K.-H. (1990), Bacterial haloperoxidases

and their role in secondary metabolism,Biotech-nol.Adv. 8,185-205.

VAN PEE,K.-H. (1996), Biosynthesis of halogenated

metabolites by bacteria, Ann . Rev. Microbiol. 50 ,375-399.

VAN PEE,K.-H., LINGENS,. (1985a), Purification of

bromoperoxidase from Pseudomonas aureofa-ciens,J. Bacteriol. 161,1171-1175.

VAN PEE,K.-H., LINGENS,. (1985b), Purification and

molecular and catalytic properties of bromoper-

oxidase from Streptomyces phaeochrom ogenes, J.

Gen. Microbiol. l31,1911-1916.

VAN PEE, K.-H., SALCHER, .,FISCHER,., BOKEL,

M., LINGENS,. (1983), The biosynthesis of brom-

inated pyrrolnitrin derivatives by Pseudomonasaureofaciens,J. Antibiot. 36,1735-1742.

VAN PEE,K.-H., SURY, G., LINGENS,. (1987), Purifi-

cation and properties of a non-heme bromoper-

oxidase from Streptomyces aureofaciens, Biol.

Chem. Hoppe-Seyler 368,1225-1232.

VANSCHIJNDEL,. W. F! M.,VOLLENBROEK,. G. M.,

WEVER, . (1993),The chloroperoxidase from the

fungus Curvularia inequalis - a novel vanadium

enzyme,Biochim. Biophys. Acta 1161,249-256.

VELDE, . L.V., JACOBSEN,. N. (1995), Kinetic reso-

lution of racemic chromenes via asymmetric

epoxidation - synthesis of (+ )-Teretifolione-B, .

Org. Chem. 60,5380-5381.

VERHAGEN, J. M., SWARTS, . J., KUYPER,. W.,WIJNBERG,. B. P. A., FIELD, . A. (1996), The

ubiquity of natural adsorbable organic halogen

production among basidiomycetes,Ap pl. Micro-biol. Biotechnol. 45,71&718.




VEKSCHUEREN,. H. G., FRANKEN,S. M., ROZE-

BOOM, H . J., KAL K,K. H., DIJKSTRA,.W. (1993a),Refined X-ray structures of haloalkane deh aloge-nase a t pH 6.2 and p H 8.2 and implications for the

reaction mechanism, J.Mol. Biol. 232,85&872.VERSCHUEREN,. H . G., FRANKEN,. M., ROZE-

BOOM, H . J., KALK,K. H.,DIJKST RA,. W. (1993b),Crystallographic analysis of the catalytic mec ha-nism of haloalkane dehalogenase, Nature 363,6 9 3 6 9 8 .

VILTER,H. (1984), Peroxidases from phaeophycae.A vanadium(V) dependent peroxidase fromAscophyllum nodosum (L) 5, Phytochemistry 23,

VILTER, . (1995), Vanadium depen dent haloperoxi-dases, Met. Ions Biol. Syst. 31,325-362.

WACKETT, . F!, SCHANKE,. A . (1992), Mechanismsof reductive dehalogenation by transition metalcofactors found in anaerobic bacteria, in: MetalIons in Biological Systems, Vol. 28: Degradationof Environmental Pollutants by Microorganismsand their Metalloenzymes (SIGEL,H., SIGEL,A.,

WAGENKNECHT,. A.,WOGAN,W. D. (1997), Iden-tification of intermediates in th e catalytic cycle ofchloroperoxidase, Chem. Biol. 4,367-312.

WALKER,. R. L., LIEN,B. C. (1981), Metabolism of

fluoroacetate by a soil Pseudomonas sp. andFusarium solani, Soil Biol. Biochem . 13,231-235.

WENG,M., PFEIFER, ., KRA USS,.,LINGENS,.,VANPEE, K. H. (1991), Purification, characterizationand comparison of 2 nonheme brom operoxidasesfrom Streptomyces aureofaciens ATCC 10762,J. Gen . Microbiol. 137,2539-2546.

WEVER,R . , PLAT,H., DE BO ER , . (1985), Isolationprocedure and some properties of the bromoper-oxidase from the seaweed Ascophyllum nodo-sum , Biochim. Biophys. Acta 830,181-186.

WE YA ND , ., HECHT,H. J.,VILTER,H., SCHOMBURG,D. (1996), Crystallization a nd preliminary X-rayanalysis of a vanadium dependent peroxidasefrom Ascophyllum nodosum ,Acta. Crystallogr.D52,864-865.

WIESNER,W., VAN PEE, K.-H., LINGEN S,. (1985),Purification and properties of bromoperoxidasefrom Pseudomonas pyrrocinia, Biol. Chem.Hoppe-Seyler 366,1085-1091.

1387-1390.

Eds .), pp. 329-356.

WIESNER,W.,VANEE ,K.-H., LINGE NS,. (1986), De -

tection of a new chloroperoxidase in Pseudomo-nus pyrrocinia , FEBS Lett. 209,321-324.

WILLIAMS,. H., FAULKNER,. J. (1996), N-Methyl-

ated ageliferins from the sponge Astrosclera wil-leyana from Pohnpei, Tetrahedron 52,5381-5390.

WOLFFRAMM,., LINGENS,., MUTZEL, ., VA N PEE,K.-H. (1993), Chloroperoxidase-encodinggenefrom Pseudomonas pyrrocina - equence , expres-sion in heterologou s hosts and purification of theenzyme, Gene 130,131-135.

WOOLARD,. X., MOORE, . E., ROLLE R,. P. (1976),Halogen ated acetamides, but-3-en-2-01s and iso-propanols from Asparagopsis taxiformis (Delile)Trev., Tetrahedron32,2843-2846.

WUOSMAA,A.., HAG ER, . P. (1990), Meth yl-chlo-ride transferase - a carbocation r oute for biosyn-thesis of halometabolites, Science 249,160.

XUN, L., TOPP, E., ORSER,C. S. (1992), Diverse sub-stra te range of a Flavobacterium pentachlorophe-no1 hydroxylase and reaction stoichiometries,J.Bacteriol. 174,2898-2902.

XUN,L., TOPP,E., ORSER,C. S. (1992), Confirmationof oxidative dehalogenation of pentachlorophe-no1 by a Flavobacterium pentachlorophenol hy-droxylase,J.Bacteriol. 174,5745-5747.

YOKOTA,., FUSE, . , OM ORI, .,MINODA,. (1986),Microbial dehalogenation of haloalkanes mediat-ed by oxygenase or halohydrolase, Agric. Biol.

Chem. 5 , 4 5 3 4 6 0 .YOKOTA,., OMORI, ., KODAM, . (1987), Purifica-

tion and properties of haloalkane dehalogenasefrom Corynebacterium sp. strain m15-3, J. Bacte-rial. 169,40494054.

Yu, F., NAKAMURA,., MIZUNAS HI, ., WATAN ABE,W. (1994), Clonin g of 2 halohyd rin hydrog en-halide lyase genes of Corynebacterium sp. strainN-1074 and structural comparison of the genesand gene-products, Biosci. Biotechnol. Biochem.

ZAKS,A., DODDS,D. R. (1995), Chloroperoxidase-catalyzed asymmetric oxidations: Substrate speci-ficity and mechanistic study, J. Am . Chem. SOC.

58,1451-1457.

117,10419-10424.



4 Phosphorylation

SIMONONESTempe, A Z, USA

1 Phosphoryl Transfer in Ge nera l 2221.1 Introduction 2221.2 Enzyme-C atalyzed Phosph ate Transfer 223

1.2.1 The Choice and Action of the Enzyme 2231.2.1.1 Ac etate Kinase 225

1.2.1.2 Glycerol Kinase 2251.2.1.3 Pyruva te Kinase 2251.2.1.4 Creatine Kinase an d Arginine Kinase 2251.2.1.5 Ad eny late Kinase 2261.2.1.6 Nucleoside Kinase 2261.2.1.7 Hexok inase 2261.2.1.8 Alkaline Phosph atase 2261.2.1.9 Phosph odiesterase 226

1.2.2 The Source of Phosph ate Gro up 2281.2.3 Cofac tor Rege neration 228

2 Formation of P -0 Bon ds 228

2.1 Inorganic Phosphate (Pi) as the Phosphate Source2.2 Nucleoside Triphosphates (NTPs) as the Phosphate Source 2282.2.1 Phosphoenolpyruvate/PyruvateKinase 2292.2.2 Acetyl Phosph ate/Acetate Kinase 2312.2.3 Methoxycarbonyl Phosph ate/Acetate Kinase or Carba mate Kinase 2312.2.4 Oth er Systems 232

2.3 Coupling to Form Phosphodiesters 2322.3.1 Cou pling with ATP 2322.3.2 Coup ling with CTP 2322.3.3 Coupling with U TP 2332.3.4 C oupling with Oth er Phosphate Sources 235

228

3 Cleavage of P -0 Bonds 235

3.1 Hydrolysis of Nucleic Acid Derivatives 2353.2 Hydrolysis of Substrates Oth er than Nucleic Acid Derivatives 2364 References 237





222 4 Phosphorylation

1 Phosphoryl Transfer

in General

1.1 Introduction

The formation and cleavage of the phos-phate bo nd is one of the most important reac-tions in biological systems. The release of en er-gy from the hydrolysis of a P -0 bond (602 kJmol-l; 144 kcal mo l-l) (BERKOWITZ,959) isan indication of the free energy available to"power" biological processes, i.e., to couple en-

dogonic processes to the exogon ic cleavage ofphosphate bonds. The m ost comm on way thisis achieved, is by transfer of a pyrophosphategroup from adenosine triphosphate (ATP) (1)

(Fig. 1) to an aliphatic alcohol. The alkyl pyro-phosphate so formed can then underg o nucle-ophilic displacement to form a new bond. Atypical exam ple is the coupling of dimethylal-lyl pyrophosphate and isopentenyl pyrophos-phate to give geranyl pyrophosphate, the key

w 2

H O O H

1Adenosine triphosphate (.4TP)

Fig. 1. Ade nosin e triphosphate (ATP) (1).

step in monoterpene biosynthesis. Using thelanguage of synthetic organ ic chemistry, for-mation of the pyrophosp hate converts the al-

cohol into a good leaving group (nucleofuge).ATP is in effect Nature's tosyl chloride!

The importance of nucleotide phosphatemetabolism can be judged from the dailyenergy requirements of an adult woman,1500-1800 kcal (6300-7500 kJ ). Th's corre-sponds to the energy of hydrolysis of 200 mol(ca. 150 kg) of ATP to ADP and phosphate,which must be generated by recycling less than70 g of ATP. The major pathway for the pro-duction of ATP is the exogonic oxidation of

glucose to carbon dioxide and water (VOETand VOET, 995).The key feature that distinguishes phos-

phate esters from other conceivable deriva-tives is that they are kinetically highly stable inaqueous solution. For example, the free energyof hydrolysis of acetic anhydride, acetyl phos-phate, and inorganic pyrophosphate are all ofthe same m agnitude, whereas the half lives inwater are a few seconds, several hours, andyears, respectively. Moreover the presence ofseveral oxygens is ideal for facilitating bindingto enzymes and for activation of cleavage byprotonation (Fig. 2) (KEIKINHEIMOt al.,1996). The predom inant counterion in vivo isMg2+ .

Phosphate esters have biological roles o therthan energy metabolism. For example, phos-phodiesters provide the backbone of bothDNA an d RNA, the carriers of the geneticcode, while phosphorylation of key enzymesand proteins regulates their activity (KREBSand BEAU, 979; GIBSO Nt al., 1990).

0 0 0 0 0

Fig. 2. pK, values for phosphate and pyrophosphate.



I Pho sph oryl Transfer in General 223

Since many important biological moleculesand drugs contain phosphate groups, there is aneed for efficient and specific synthetic meth-ods. Chemical synthesis often proceeds in high

yield, but currently there is a dearth of stereo-or enantioselective methods. Consequentlythis must be achieved indirectly via protectinggroups. The use of biological systems, whetherwhole cell or isolated enzymes, can allow thestereoselective and often enantioselective in-troduction of phosphate groups in comparableyields to those obtained by chemical methodsand without the need for protecting groups.

This chapter aims to provide an insight intothe use of biological systems for both the for-mation and cleavage of phosphate esters forsynthetic purposes. A range of examples ofphosphate forming reactions have been in-cluded in tabular form to illustrate the wideapplicability of the techniques (see Tab. 3).Other complex enzyme systems involvingphosphates in multi-sequence reactions, suchas aldolase reactions (ALLEN t al., 1992), arenot considered here.

1.2 Enzyme-Catalyzed Phosphate

Transfer

When we talk of an enzyme catalyzing thephosphorylation of an alcohol or similar nucle-ophilic substrate, we should bear in mind thatspecifically t is an enzyme catalyzed phospho-ryl transfer from a phosphate source or cofac-tor (such as ATP (1)) to the nucleophile.Therefore, there are two considerations to bemade when examining phosphoryl transfer,the enzyme and the source of phosphategroup.

1.2.1 The Choice and Action of the

Enzyme

The enzymes that catalyze phosphate trans-fer are categorized according to the bondsmade or cleaved. There are two general groupsof enzymes: those that accept phosphatemonoesters as substrates and those which ac-cept phosphate diesters or pyrophosphates(Fig. 3) (KNOWLES,980). DRAUZ nd WALD-

Group 1 - Phosphate Monoesters

(For simplicity, the cleavage of

y-phosphate groups of

triphosphates ar e covered here)

Phosphomu tases

Phosphorylases

Nucleotidases

Phosphatases

Phosphoki iases

Phosphotranferases

f 9 fRO- f -0- f - 0- - 0

o* o* $0

Phosphatases

Phosphokiiiases

Phosp ho ransferases

ATPases

Group 2 - Phosphate Diesters &

Pyrophosphates

0II

0I I

0I I

Nudeotidyl Tra nsf era ss Fyrophosphokinases

Nudeotidyl Cyclases

Triphosphohydrolases

Polyiiucleotide Synthetases

Phospholipases

Nucleases

Phosphodiesterases

Fig.3. The tw o groupsof phosphoryl transfer reac-tions and enzyme classes.

MANN (1995) correlated this classification withthe classes of enzymes specified by the No-menclature Committee of the International

Union of Biochemistry (1984) (Tab. 1).The




Tab. 1. Table of Enzymes that Fall into Gro ups 1and 2

Cleavage Functional IUB Class & NameSite Class

Role

@ Phosp hom utase 2.7.5. Phosph omuta se intramolecular transfer of a5.4.2. Intramolecular phosphotransferase phosphoryl grou p

0 Pho sphor ylase 2.4.1. Hexosyl transfera ses2.4.2. Pentosyl transferases

P-0 bond formation fromphosphorolytic C-hetero-atom cleavage

@ Nucleotidases 3.1.3. Phosphoric acid hydrolases trans fer of phosphoryl from3.1.4. Phosphoric diester hydrolases a nucleoside to w ater

@ Phosphatases 3.1.3. Phosphoric ester hydrolases trans fer of phosphoryl from3.6.1. Hyd rolases acting on acid anhydrides

@ in phosphorus containing anhydridesa m onoester to water

@ Phosphokinase 2.7.1. Phosphotransferases with an trans fer of phosphoryl froma nucleoside triphosphate

@ 2.1.2. Phosphotransferases with a to an acceptor other than

alcohol group as acceptor

carboxyl gro up as acceptor water

@ Phospho- 2.7.4. Phosphotransferases with a transfer of phosphoryl froma molecule other than a nu -cleoside triphosphate t o anacceptor other than water

@ ATPase 3.6.1.3. ATPases phosphatases which coupleATP cleavage with other

processes

transferase phosph ate group as acceptor

@

@ Pyrophospho- 2.7.6. Diphosphotransferaseskinase

transfer of pyrophosphatefrom ATP to an acceptorothe r than water

~

@ Nucleotidyl 2.7.7. Nucleotidyl Transferases tra ns fer of nucleotidylTransferase groups

@ Nucleotidyl 4.6.1. Phos phoru s oxygen lyases cyclization of a nucleosideCyclase triph osph ate by formation

of a pyrophosphate

@ Triphospho- 3.1.5.Triphosphoric mo noe ster transfer of triphosphate

hydrolase hydrolases from a nucleoside triphos-phate to water

@ Polynucleotide 6.5.1. Ligases forming phosphoric links two polynucleotidessynthase ester bonds to produce a polynucleotide

chain

@ Phospholipase 3.1.4. Phosphoric diester hydrolases hydrolytic cleavage of

phosphoglycerides

@ Nuclease 3.1.4. Phosphoric diester h ydrolases3.1. End o- and exonucleases nucleotide to water

transfer of a phosphopoly-

@ Phospho- 2.7.8. Transferases for oth er substituted transfer of a phosphom ono-

diesterase pho sph ate groups ester from a phosphodiester,other th an a polynucleotide,to water

3.1.4. Phosphoric diester hydrolases



2 Formation ofP-0 Bonds 225

range of commercially available enzymes issomewhat less than this. On occasions it maybe worthwhile to extract and purify a given en-zyme, however, this is not practical in general.

Consequently the following sections empha-size enzymes which are commercially availableor easily extractable. For each example a refer-ence to Methods in Enzymologyis given whichprovides recipes for purification and assay.

1.2.1.1 A cetate Kinase

Acetate kinase (EC 2.7.2.1) has been usedextensively for phosphorylation of adenosinediphosphate (ADP) (3) (Fig. 4) and will also

accept GDP, UDP, and CDP (HAYNIE ndWHITESIDES,990; CRANS t al., 1987). Themechanism of phosphoryl transfer has beenextensively studied (SPECTOR, 980) and pro-ceeds with net inversion at phosphorous. Theenzyme accepts acetyl phosphate (8) (Tab. 2) ,propionyl phosphate, and carbamoyl phos-phate, although the latter two cause reducedenzyme activity (30% and 18%, respectively)but phosphoenolpyruvate (4) is not accepted.Spontaneous hydrolysis of acetyl phosphate

(8) restricts the use of acetate kinase for ex-tended periods. The X-ray crystal structure ofacetate kinase shows an “actin fold” which isdiscussed in the section on glycerol kinase(Sect. 1.2.1.2) (Buss et al., 1997). Immobilizedacetate kinase from E. coli has high activity,but is less stable than the much more expen-sive enzyme from the thermophile, Bacillusstearothermophilus (TANG and JOHANSSON,1997).

1.2.1.2 Glycerol Kinase

Glycerol kinase (EC 2.7.1.30) is usually iso-lated from rat liver (THOMER nd PAULUS,1973) and commercial enzymes with high ac-tivity extracted from E . coli (KEE et al., 1988)and a Cellulomonussp. Rat liver glycerol kin-ase is most stable at pH 5 and readily mono-phosphorylates glycerol (15) Fig. 6), L-glycer-aldehyde (21) (Fig. 9), or dihydroxyacetone(LIN,1977; CRANS nd WHITESIDES,985a,b).ATP (1) s the best cofactor, although different

forms of the enzyme can accept uridine tri-

phosphate (UTP), guanosine triphosphate(GTP), and cytidine triphosphate (CTP). ADP(3) has been shown to inhibit glycerol kinasenoncompetitively at low concentrations, and

competitively at higher concentrations. Ace-tate kinase and glycerol kinase are members ofa structural super family which includes sugarkinases, hexokinase, heat shock protein 70, andactin.These are characterized by a two domainstructure (actin fold) with the topographyPPPaPaPa. The nucleotide binding site liesbetween the two domains which are joined bya putative hinge. Nucleotide hydrolysis elimi-nates some of the interdomain bridging inter-actions which enables formation of an openconformation. The large conformational

changes involved in these processes may be in-volved in regulation (HURLEY, 996; KABSCHand HOLMES,995).

1.2.1.3 Pyruvate Kinase

The most common source of pyruvate kin-ase is from rabbit muscle from which it can beisolated in crystalline form. It accepts guano-sine diphosphate (GDP) with approximately

the same tolerance as ATP (1) (KAYNE,973).However other nucleoside diphosphates(NDPs), such as cytidine diphosphate (CDP),are not so readily accepted, with reaction ratesapproximately 30% less than that of ADP (3).The enzyme shows a very high degree of speci-ficity towards the phosphate donor with onlyphosphoenolpyruvate (4) being accepted. Thereaction rate is highly dependent on the con-centration of monovalent cations, in particularpotassium and ammonium. One drawback isthat it is inhibited by ATP (l),hus neces-

sitating keeping ATP (1) levels low during re-action.

1.2.1.4 Creatine Kinaseand Arginine Kinase

These enzymes are both ATP :guanidinophosphotransferases which have high aminoacid homology (GROSS t al., 1995; MUHLE-BACH, 994) although they are found in two

different biological classes (WATTS, 1973;




MORRISON,973). Creatine kinase (E C 2.7.3.2)is isolated from vertebrate muscle (usuallyfrom rabbit muscle) and is stable below 50°Cbetween pH 6.5 and 9.5. It has narrow sub-

strat e specificity accepting only a small rangeof guanid ine based amino acids. Arg inine kin-ase ( E C 2.7.3.3) is found in most invertebratesand has similar stability and substrate prop er-ties to creatine kinase (DU MA S nd CAMONIS,1993; STRON Gnd ELLINGTON,995).

1.2.1.5 Adenylate Kinase

Adeny late kinase (myokinase, E C 2.7.4.3)

catalyzes the disproportionation of ATP (1)and adenosine monophosphate (2) (AMP) toA D P (3) (Fig. 4). The enzyme is widespreadthroughout organisms, with higher con centra-tions located in regions of high energy turn-over, such as muscle (N ODA , 973; KREJCO VAand HOR SKA , 997). Most other nucleotidesand inorganic triphospha te can be utilized, butat reduced reaction rates. A divalent cation(usually magnesium o r manganese) is neededfor the reaction to proceed (WILD t al., 1997).

1.2.1.6 Nucleoside Kinase

Nucleoside kinases catalyze the phosphoryltransfer from a nucleoside triphosphate (NTP)to a nucleoside, giving a nucleoside mono-phosphate (NMP). There are many differentclasses of these enzymes depending upon thenucleoside being phosphorylated (ANDERSON,1973).These enzymes are often specific for the

nucleoside.Thus,AM P (2) will only be form edfrom adenosine utilizing adenosine kinase(EC 2.7.1.20).Th ere is however, more flexibil-ity with the NTP don or in m ost cases; adeno-sine kinase accepts most o ther NTPs.

Adenylate

kinase

ATP + AMP 2 ADP

1 2 3

Fig. 4. Formation of ADP (3) from ATP (1) andAMP(2).

1.2.1.7 Hexokinase

Hexokinases (EC 2.7.1.1) have been usedextensively for the conversion of glucose to

glucose-6-phosphate (13) (CHENAULTt al.,1997). There are two m ain sources of hexo-kinase: from yeast and from m ammals (CO LO-WICK, 973).Both sources have certain aspectsin common, including strong inhibition byN-acetyl glucosamine derivatives (WILLSONet al., 1997; MAITY and JAR OR I, 997), broadsubstrate specificity towards the sugar accep-tor, and conversely narrow specificity towardsthe NTP used as the donor (ATP (1) is thebest).

1.2.1.8 Alkaline Phosphatase

The tw o mains sources of alkaline phosphat-ase ( E C 3.1.3.1) are bovine intestines (FE RN-LEY, 1971) and E. coli (REID and WILSON,1971). As its nam e suggests, the enzyme is

stable at basic pH (typically pH 9.8). Bovinealkaline pho spha tase has very high activity, ischeap, but has low stereospecificity (NATCHEV,

1987). Consequently it is used as a general pur -pose workhorse for the cleavage of mono phos-phates to alcohols. It will also catalyze trans-phosphorylation and some enzymes alsocleave pyrophosphates. The X-ray crystalstructure has been determined (MURPH Yt al.,1997; cf. KE IKINH EIM Ot al., 1996).

1.2.1.9 Phosphodiesterase

Phosphodiesterase enzym es are of ten calledvenom exonucleases since they are p resent inall snake venoms. One of the more commonenzymes used is snake venom phosphodies-terase (phosphodiesterase I, EC 3.1.4.1)(LASKOWSKI,971), which is a 5 -exonuclease,i.e., it specifically cleaves oligonucleotides togive 5 '-mononucleosides. This enzyme togeth -er with bovine intestinal phosphodiesterasehas high activity with non-nucleotide sub-strates. Although snake venom phosphodies-terases are known which act as 3'-exonucle-

ase, spleen phosphodiesterase is used morecommonly (phosphodiesterase 11, EC 3.1.16.1)(BERNARDInd BERNARDI,971).



1 Phosphoryl Transfer in General 227

Tab. 2. A Comparison of the Relative Phosphorylation Potentials of Phosphoryl Transfer Reagents

Cofactor Phospho rylation Potential[kJ mol-’1([kcal mol-’I)”

C O Fhosphoenolpyruvate 4 -61.9 ( -14.8)b

KM e 0 OP

Hz N1P

Methoxycarbonyl Phosphate 5 -51.9 ( - 12.4)’

Carbamyl Phosphate 6 -51.4 (-12.3)d

OH

WJ C 0 2 P,3-Diphosphoglycerate 7 -49.4 (-11.8)=

-43.1 (-10.3)bcetyl Phosphate 8

-43.1 (-10.3)”hosphocreatine 9

P Me

-30.5 ( - 7.3)’-41.8 ( - 10.0)’

ATP 1 -,A D P + PiATP 1 -+ A M P + PP,

-32.2 ( - 7.7)’rginine Phosphate 10

PPi

OP

- 9.2 (- 4.6)‘yrophosphate 11

Glucose-1-phosphate 12 -20.9 ( - 5.00)“

Glucose-6-phosphate 13 -13.8 (- 3.3)‘

Glycerol-1-phosphate 14 - 9.2 ( - 2.2)’

a Values are for pH 7.0.AT P hydrolysis is highly depend ent on magnesium ion co ncentration; BOLTE ndWHITESIDES1984); KAZLAUSKASnd WHITESIDES1985); SHIH nd WHITESIDES1977); LEHNINGER

(1975).




1.2.2 The Source of

Phosphate Group

Phosphoryl transfer necessarily means thata source of phosphate is needed. This is rarelyinorganic phosphate and almost invariably acomplex cofactor. A comparison of the freeenergies of hydrolysis of a number of phos-phorylating cofactors gives an indication of thephosphorylation potential of that species (Tab.2). The majority of these phosphorylatingagents have some biological role, although themost widely used in nature are the NTPs. ATP(1) acts predominantly in central pathways ofenergy metabolism, GTP is used to drive pro-

tein synthesis, CTP phospholipid biosynthesis,and UTP glycosidation. Those agents with alarge phosphorylation potential are theprimary sources of energy in biological pro-cesses. Phosphoenolpyruvate (4) and 1,3-di-phosphoglycerate (7) are both formed duringglycolysis and are used as a primary source ofphosphate. Phosphocreatine (9) and phospho-arginine (10) are found in muscle tissue in ver-tebrates and invertebrates, respectively. Thesecompounds act as an energy “store”, which isconverted back to ATF’ (1) for use when

primary energy sources (such as phosphoenol-pyruvate (4)) are not available. ATP (1) itselfthen serves as a means to transfer thephosphoryl groups to and from storage, andthus is recognized as a cofactor by many differ-ent enzymes. Although better phosphoryltransfer agents are available (cf. phosphoenol-pyruvate (4)), ATP (1)is essential for theseprocesses.

1.2.3 Cofactor Regeneration

The most commonly used NTP as a cofactorfor phosphoryl transfer is ATP (1).The activeform of ATP (1)has been found to be theMgATP2- species, which necessitates additionof a source of magnesium ions to the reactionmixture. However, the high cost of ATP (1)precludes the use of stoichiometric quantitiesin vitro.Since most enzymespeed ATP (1) as acofactor, regeneration of AMP (2), or morecommonly ADP (3) formed after the firstphosphoryl transfer is necessary. ATP (1) re-

generation can be achieved by chemical meansor by enzymatic methods. In general, the en-zymatic route is preferred. Most commonly aphosphorylating agent with a high phosphory-

lation potential is used to enable efficientcofactor regeneration. The differing combina-tions of enzyme and phosphate source are de-scribed later.

2 Formation of

P-0 Bonds

Reactions which involve the synthesis of a

P-0 bond all require a source of phosphate.This section is divided according to the co-factors used, concluding with a summary ofvarious P-O bond forming reactions (Tab. 3).

2.1 Inorganic Phosphate (Pi)

as the Phosphate Source

Glucose-1-phosphate (12) has been pre-pared by the action of phosphorylase-a (EC2.4.1.1) (Fig. 5) on soluble starch or dextrin in

the presence of inorganic phosphate and theglucose-1-phosphate (12) ormed transformedto glucose-6-phosphate (13) using phospho-glucomutase (EC 2.7.5.1) (WONGand WHITE-SIDES, 981).

An excellent study of the use of calf intes-tinal alkaline phosphatase with sodium pyro-phosphate as the phosphate source has beenmade. A variety of diols were used but most ofthe work was concentrated on the formationof glycerol-1-phosphate (14) from glycerol(15) (Fig. 6). The regioselectivity of this reac-

tion was found to be greater than 90% in favorof the primary hydroxyl group. However, thereaction was not enantioselective (PRADINESet al., 1988).

2.2 Nucleoside Triphosphates

(NTPs) as the Phosphate Source

The use of NTPs is one of the most effectiveand widespread enzymatic methods of phos-phoryl transfer. As noted earlier, regeneration




Starch

Phosphorylase-arOP

I Phosphoglucomu tase

G

Fig. 5. Synthesis of glucose-6-phosphate (13).

of NTPs is necessary to reduce the cost of thereaction. Several systems exist for this process(HEIDLASt al., 1992).

2.2.1 Phosphoenolpyruvate/

Pyruvate Kinase

Phosphoenolpyruvate (PEP) (4) is an at-tractive phosphorylating agent for ATP (1) re-generation since it has a high phosphorylation

potential (- 1.9 kJ mol-’). A drawback is thehigh commercial cost and problematic synthe-sis of the potassium salt (HIRSCHBEINt al.,1982), although WHITESIDESas developed anefficient enzymatic method (Fig.7) of generat-ing PEP (4) in situ from the relatively inexpen-sive D-(-)-3-phosphoglyceric acid (17)for thesynthesis of NTPs from NMPs (SIMON t al.,l989,1990).The use of this phosphoryl transfersystem is widespread and has broad applicabil-ity. For example, a whole range of ring fluor-inated hexose sugars have been transformed

to their corresponding 6-phosphates, as well as

OH

H O A O H 15

Alkaline

phosphataseIOH

PO&OH 14> 90%

OP

H o A O H 16

Fig. 6.

alkaline phosphatase.Selective phosphorylation of glycerol using

OH

“A~$7Phosphoglycerate

mutase

OP

Eholase

JOP

Ace 4

Fig. 7. Enzymatic synthesis of phosphoenolpyruv-ate (4).

those analogs containing sulfur or nitrogenwithin the pyranose ring (DRUECKHAMMER

and WONG, 985). The reaction is also highlystereospecific. One of the more important usesof enzymatic phosphate synthesis is the pre-paration of chiral phosphorothioates whichare used extensively for the elucidation of bio-logical mechanisms. The synthesis and uses ofphosphorothioates have been extensively re-viewed (LESNIKOWSKI,993; ECKSTEIN,983,1985). As an illustrative example, the 5 ‘ -

monothiophosphate (19) was transformed ex-clusively into one diastereoisomer of the“pseudo” triphosphate (20) (Fig. 8) (MORAN

and WHITESIDES,984).



230 4 Phosphorylution

Adenylate kinasePyruvate kinasePEP

Fig. 8. Stereospecific phosphorylation of a mono-thiophosphate(19).

Glycerol kinase and A TP (1)have been used

to perform kinetic resolutions of rac-glyceral-dehyde (21) Fig. 9) (WONG nd WHITESIDES,1983). ATP (1)was regenerated using PE P (4)and pyruvate kinase. Enantioselective phos-phorylation of glycerol (15) has also beenachieved with glycerol kinase (CRANS ndWHITESIDES,985a, b) and th e acetyl phos-phate @)/acetate kinase system (RIOS-MER-CADILLO and WHITESIDES,979). The structu-ral requirements of glycerol kinase were eva-luated in a survey of 66 glycerol analogs. This

indicated that one terminal hydroxyl groupand either a hydroxyl or methyl group at thesecondary position was necessary for transfor-mation (CRA NS nd WHITESIDES,985a).

The pyruvate kinasePEP (4) regenerationsystem has also been exploited in multienzymesequences a s illustrated by t he regeneration of

UTP from uridine diphosphate (UDP) for thein sifu synthesis of UDP-glucose (WON G t al.,1982), fo r use in Leloir glycosidation.

Hydrolysis of the anom eric linkage to ade-nine is a major pathway for the degradation ofadenosine derivatives in solution. Moreover,when bound to most enzymes the oxygen ofthe ribose ring of adenosine nucleotides does

OH

o& OH

21 rac-GlyceraldehydePyruvate

k;ixyruva t einase

PEP 4

OH

0 OH

2 1 DGlyceraldehyde

R

OH

22 L-Glyceraldehyde-3-phosphate

H A PR

14 sn-Glyceraldehyde-3-phosphate

Fig. 9. Specific phosphorylation of glyceraldehyderuc-(21)an d glycerol (15).

not participate in substantial hydrogen bond-ing. Consequently, the carbosugar analog ofATP (23) (Fig. 10) should be substantiallymore stable and should bind to enzymesequally well. Fortuitously the carbosugar ana-log of adenosine is a microbial natural prod-uct, aristeromycin (23a). This was converted

to the 6’-monophosphate (23b) sing a chem-ical method and then to the diphosphate (23c)

and triphosphate (23d) catalyzed by adenylate



2 Formation ofP -0 Bonds 231

N H 2

a R 1 = H H 6 6 H 2 4 R = F

b R 1 = PPyruvate

Pyruvate kinase

PEP 4

c R1 = PP

lPruvate

d R1 = PPP

Fig. 10. Carbosugar nucleoside triphosphates usedin phosphoryl transfer reactions.

kinase and pyruvate kinase, respectively (cf.SIMON t al., 1990).An identical sequence wasused to prepare the fluoroanalog (24d).Bothcarbosugar nucleotides (23d) nd (24d) func-tioned in place of ATP (1) in the synthesis ofglucose-6-phosphate (13) from glucose cata-lyzed by hexokinase and sn-glycerol-3-phos-phate (14) from glycerol (15) using glycerolkinase (LEGRANDnd ROBERTS,993).

2.2.2 Acetyl Phosphate/

Acetate Kinase

Acetyl phosphate (8) is a good phosphoryl-ating agent. It has a reasonable phosphoryla-tion potential and has been used extensivelywith acetate kinase in many enzyme reactionsdue to its relative inexpense. One drawback isthat acetate kinase is prone to slight inhibitionby acetate ions, which are formed during thereaction. Also, acetyl phosphate (8) is not as

stable in solution as most other cofactors.

Acetyl phosphate (8 ) can be prepared inlarge quantities as its diammonium salt fromeither phosphoric acid, ketene, and ammonia(WHITESIDESt al., 1975), or acetic anhydride,

phosphoric acid, and ammonia (LEWIS t al.,1979).The latter is easier to use, since it doesnot involve the preparation and handling ofketene. In both cases, the ammonium ionpresent reacts with magnesium ions during thephosphorylation reaction to give a precipitateof Mg(NH,)PO,, which causes essential mag-nesium ions to be removed from the reactionmixture. Reaction of acetyl phosphate (8) withminute quantities of ammonia present canalso cause problems. These can be circumvent-ed by using the potassium or sodium salts

which can be prepared either by ion exchangechromatography of the diammonium salt, orby direct preparation (CRANSand WHITE-SIDES, 1983).

2.2.3 Methoxycarbonyl Phosphate/

Acetate Kinase or Carbamate

Kinase

Both acetyl phosphate (8) and PEP (4) haveinherent advantages and disadvantages. Me-thoxycarbonyl phosphate ( 5 ) has been devel-oped (KAZLAUSKASnd WHITESIDES,985) asan easily synthesized alternative to acetylphosphate (8),with comparable phosphorylat-ing potential to PEP (4). Methoxycarbonylphosphate (5) also has the advantage in notcontaining ammonium groups, thus negatingthe problems associated with diammoniumacetyl phosphate. Following phosphoryl trans-

fer, the by-product of the reaction, methyl car-bonate, decomposes to give methanol and car-bon dioxide, thus making the workup easier. Itis readily prepared in an analogous way to ace-tyl phosphate with acetic anhydride and phos-phoric acid, but decomposes much more rapid-ly in aqueous solution. One of two enzymescan be used to catalyze phosphoryl transfer toATP (1):acetate kinase or carbamate kinase.The former has been used with methoxycarbo-nyl phosphate (9, reatine kinase, and ATP(1) to generate creatine phosphate (9) from

creatine.




2.2.4 Other Systems

Extensive work has been carried out on thechemoselective formation of ATP (1) fromA D P (3) using the macrocycle (25a) (Fig. 11)(HOSSEINI nd LEHN,1985, 1987, 1988, 1991;HOSSEINIt al., 1983).In the presence of acetylphosphate (8), magnesium ions, and a hexo-kinase, the complex (2%) is formed which isable to facilitate a turnover of ATP (1) suffi-cient to catalyze the formation of glucosed-phosphate (13) from glucose by hexokinase(FENNIRInd LEHN , 993).

The versatility of the use of enzymes can beillustrated in the use of the glycolytic pathwayto act as a means of ATP (1) regeneration(WEI and G ou x, 1992). With inorganic phos-phate and glucose as the “fuel” for the reac-tion, creatine was converted by creatine kinaseto creatine phosphate (9) in the presence ofATP (1)and the required enzymes for glucosemetabolism.

2.3 Coupling to Form

PhosphodiestersPhosphodiesters play an important role in

biological functions. DNA polymers containphosphodiester links as do coenzyme A andNA D(P )(H) .There is a wide range of enzymes

A DP

25a

that catalyze phosphodiester formation bycoupling of two components. Usually one ofthese is a NTP and thus use of a phosphatesource or cofactor is not necessary.

2.3.1 Coupling with ATP

v ro s in e residues in peptides have been spe-cifically phosphorylated at the phenolic posi-tion in a two step process. Initially a tyrosine5 ’-adenosine phosphodiester was formed us-ing glutamine synthetase adenylyltransferase(EC 2.7.7.42), which was selectively dephos-phorylated using micrococcal nuclease to givethe desired phospho-tyrosine (G IBS ON t al.,1990). In a similar manner (MARTINandDRUECKHAMMER,992), pantetheine 4 ’-phos-phate (26) was coupled with ATP (1)using de-phospho-CoA-pyrophosphorylaseEC2.7.7.3)and inorganic pyrophosphate (Fig. 12) to give3 ‘-dephospho-coenzyme A (27a). The 3 ‘-hy-droxyl was specifically phosphorylated by ATPusing dephospho-CoA-kinase (EC 2.7.1.24).

2.3.2 Coupling with CTP

N-Acetyl-neuraminic acid (28) (Fig. 13), is

a sialic acid which is a component of glyco-proteins and lipids. It was coupled with CTP

25b (nega t ive charges on oxygen not shown)

Fig. 11. Macrocyclic amino-crown ether used to catalyze formation of ATP (1) rom ADP (3).




0 26 Pantetheine phosphate

Dephospho-CoA-pyrophosphory lase

RO 011H

0 0 27

a R = Hb R = P Coenzyme A

phospho-Co A-kinase"xDP

Fig. 12. Formation of CoenzymeA (2%).

using cytidine-5 monophospho-N-acetylneu-

ramic acid synthase as a prelude to enzymaticglycosidation. CTP was generated by phos-

phorylation of CDP with PEP (4), catalyzedby pyruvate kinase. CDP was prepared in situby disproportionation of stoichiometric CMPwith CTP catalyzed by adenylate kinase(SIMONt al., 1988a).

2.3.3 Coupling with UTP

UDP-glucose has been prepared bycoupling of UTP with glucose-6-phosphate

(13) in the presence of UDP-glucose pyro-phosphorylase, inorganic phosphate, and phos-phoglucomutase. The unusual feature of thisreaction is the source of UTP. Yeast RNA wascleaved into oligonucleotides by nuclease PIwhich were further degraded to the nucleosidepyrophosphates using polynucleotide phos-phorylase. Pyruvate kinase mediated phos-phorylation by PEP (4) gave a mixture ofNTPs corresponding to ATP (24%), CTP(18%),UTP (28%), and GTP (30%).The mix-ture was used crude in the coupling with glu-

cose-6-phosphate (13) (WONGet al., 1983a).

This system has also been used with acetatekinase and acetyl phosphate (8) in place ofpyruvate kinase and PEP (4) (HAYNIE nd

WHITESIDES,990).

AcNHvoH COY

to"9OH

Fig. 13. Coupling of cytidine-5 -triphosphate and

N-acetylneuramic acid (28).




Tab. 3. Survey of P- 0 Bond Forming Reactions~

Substrate Product Enzym e C ofactor Phosphate Cofactor ReferenceSource Regeneration

D-Arabinose ~-a rab ino se-s - hexokinase ATP PE P pyruvate kinase BED NA RSK It al.

D-Glucose ~- gl uc os ed - hexokinase ATP PEP pyruvate kinase HIRSCHB EINt al.

phosphate (1988)

phosphate (1982)hexokinase ATP AcP acetate kinase CRA NSnd WHITE-

SIDES (1983)

D-Fructose ~-fr uct ose -6- hexokinase ATP AcP acetate kinase WO NG nd WH ITE-phosphate SIDES (1983)

D-Fructose ~-fruc tose-1 ,6- hexokinase ATP - - WONG t al.diphosphate and phos- (1983b)

phofructo-kinase

D-Ribose ~- rib os e-5 - ribokinase ATP PE P pyruvate kinase GROSS t al. (1983)phosphate5-p ho sp ho -~ - PR PP syn- ATP PE P pyruvate kinase GROSS t al. (1983)ribosyl-a-1- thase and Pi and adeny latepyrophosphate ribokinase kinase(PRPP)

D-Ribose-5- 5-p ho sp ho -~ - PR PP syn- ATP PE P pyruvate kinase GROSSt al. (1983)phosphate ribosyl-a-1- thase Pi and adenylate

pyrophosphate kinase

(PRPP)

D-Ribulose- ~-r ibu lose -1,5 - phospho- ATP AcP acetate kinase WO NG t al. (1980)5-phosphate bisphosphate ribulose

kinasephospho- ATP PE P pyruvate kinase GR OSS t al. (1983)ribulosekinase

2'-Deoxy 2'-deoxy adenylate ATP PE P pyruvate kinase LAD NERndAM P ATP kinase WH ITESIDES1985)

Adenine ATP adenylate ATP AcP acetate kinase BAU GHNt al.

kinase and (1978)adenosinekinase

UMP UTP" adenylate ATP PE P pyruvate kinase SIMONt al.kinase (1988b)

Deoxy- deoxycytidine deoxycyti- ATP - - KESSEL1968)cytidine triphosphate dine kinase

CMP CTP" adenylate CTP PE P pyruvate kinase SIMONt al.kinase (1988b)adenylate ATP PE P pyruvate kinase SIMONt al.kinase (1989)

nucleoside ATP PE P pyruvate kinase AUG E ndmonophos- GAUTHERONphokinase (1988)



3 Cleavage of P-0 Bonds 235

Tab. 3. Continued

Substrate Product Enzyme Cofactor Phosphate Cofactor ReferenceSource Regeneration

GM P GT P guanylate ATP PE P pyruvate kinase SIMON t al.kinase (1988b)

Dihydroxy- dihydroxy- glycerol AT P P E P pyruvate kinase WON G ndaceton e aceton e kinase WHITESIDES1983)

glycerol ATP AcP acetate kinase WO NG ndkinase WHITESIDES1983)

phosph ate kinase WHITESIDES1977)

phosphate

Creatin e creatine-6- creatine AT P Ac P acetate kinase SHIHan d

Arginine arginine arginine AT P PE P pyruvate kinase BOLTE ndphosph ate kinase WHITESIDES1984)

allo- 0-phospho- hydroxy- GTP - - HILES ndHydroxy-L- allo-hydroxy- lysine kinase HE ND ER SO N1972)lysine L-lysine

Hydroxy-L- 0-ph osph o- hydroxy- GTF’ - - HILESndlysine hydroxy-L- lysine kinase HEN DER SON1972)

L-Serine L-serine- L-serine Pi - - CAGEN nd

lysine

pho spha te transferase FRIEDM ANN1972)

L-Threonine L-threonine- L-serine Pi - - CAGEN ndpho spha te transferase FRIEDM ANN1972)

a Need ed t o initially gen erate the corresponding NDP.

2.3.4 Coupling with Other

Phosphate Sources

Phospholipase D has been used to couplevarious species with dimyristoyl-L-a-phos-phatidyl choline in order to synthesize possiblephospholipid inhibitors (WANGet al., 1993).These include (R)- nd (S)-proline and serinol.The reaction was found to be selective for pri-mary alcohols but nitrogen and sulfur nucle-ophiles were not accepted.

nique which has become a routine tool in theanalysis of nucleic acid oligomers, while re-maining relatively unused in synthetic work.This is largely due to the usual ease of cleavageof P-0 bonds by chemical methods, and is re-flected in the literature pertaining to enzymecatalyzed dephosphorylation (DAVIESet al.,1989;FABER, 995).

3.1 Hydrolysis of Nucleic Acid

Derivatives

The extensive use of phosphodiesterase andalkaline phosphatase enzymes for routineanalysis of nucleotides excludes a complete re-view of this area. Brief descriptions of themode of action of phosphodiesterase and alka-line phosphatase have already been made.

These enzymes are usually used together to

3 Cleavage of P-0 Bonds

In marked contrast to the formation of P-0bonds through the specific phosphorylation of

functional groups, dephosphorylation is a tech-




completely digest the oligomer of ribonucleicacid to e ither a 5 ’- r 3 ’-nucleoside dep endingupon the phosp hodiesterase chosen. The roleof alkaline phosphatase is usually to cleaveany phosphate mon oesters present, while thephosphod iesterase (or for that matter, any nu-clease) cleaves the phosphodiester link. Forexample, the 2 ’-phospho-trimer (30) (Fig. 14)was first dephosphorylated at the 2 ‘-positionby calf intestinal alkaline phosphatase andfurther digested with nuclease PI to give aden-osine, 5 ’-monophospho-adenosine and 5 ’-monophospho-uridine (SEKINE t al., 1993).Analysis of the digestion products is usuallycarried out by HPLC of the crude reaction

mixture. This method can prove to be a versa-tile tool, especially when modified phospho-diesters a re employed as “markers” for specif-ic groups. The thymidine oligomer (31) (Fig.15), modified with one methyl phosphonatelinkage, upon digestion with a 5 ’-snake venomphosphodiesterase gave thymidine, 5 ’-mono-phospho-thymidine, and the methylphospho-nate dimer (AGRAWALnd GOODCHILD,987).

Extensive mechanistic studies have beencarried out on both pho sphodiesterase and al-

kaline p hosphatase enzymes using substratessynthesized by specific enzyme phosphoryla-tions. Chiral thio-adenosine derivatives (such

A (2’-p) - PA (2l-p) IJ 30

Calf intestinalphosphatase

Nuclease P,

1IA PA PU

A + pA + pU

A = AdenosineU = Uridinep = phosphate

Fig. 14. Typical hydrolysis of phosphate esters

using nuclease and phosphatase enzymes.

TpTpTpTpTpTpT 31

Snake venom

phosphodiesterase

1+ pT + pTpT

T = Thymidinep = phosphatep = methyl phosphonate

Fig. 15. An example of th e use of phosphonates to

inhibit cleavage of phosphate dieste rs.

as (19)) have been used in determining themechanism of action of purple acid phosphat-ase. It was foun d that the enzymecatalyzes di-rect transfer of the phospho group to water(MU ELL ER t al., 1993). Isotopically labeleduridine (3 ‘ -5 ) adenosine phosphodiestershave also been used in this way with variousenzymes (SEELA t al., 1983).Spleen phospho -diesterase gave the 3 ’-uridine-monophos-

phate and adenosine, while nuclease PI gaveuridine and 5 ’-adenosine-monophosphate.

Enzymes that catalyze more specific de-phosphorylations are also known, althoughnot used as widely as those already men tioned .For example, ATPase catalyzes the hydrolysisof ATP (1) t o A D P (3) (WATTet al., 1984).

3.2 Hydrolysis of Substrates Othe r

than Nucleic Acid D erivativesThe hydrolysis of phosphate esters is chem-

ically a trivial step. The use of enzymes for thisprocess has therefore received little considera-tion other than for elucidation of enzymemechanism s, such as in the hydrolysis of p-ni-trophenol phosphates (NEUMANN,968; WIL-SON et al., 1964),or for mo re subtle processeswhere standard techniques are too harsh, aswith the hydrolysis of polyprenyl pyrophos-phates (FUJIIet al., 1982). The epoxy-farnesy l

pyrophosphate (32) Fig. 16), for example, wasdephosphorylated with alkaline phosphatase(KOYAMA t al., 1987,1990).



4 References 237

3 2

RO-,PLIZ; 3 6

H N

Fig. 16. Epoxy-farnesyl pyrophosphate (32).

Carbohydrates have also been used as sub-

strates, although the reaction is not of great

are usually the more valuable (WONG and

synthetic use since the phosphorylated species 2HO- P,

Alkaline

Phosphatase orPhosphcdiesterase

WHITESIDES,983).

NATCHEVas used the difference in reactiv-

ity of phosphodiesterase and alkaline phos-

phatase enzymes in the cleavage of various

1

Fig. 18. Hydrolysis of phosphoramidate esters.

0II

H0 ,P\% 3 4RO

Alkaline

Phosphatase1fi'

PhosphodiesteraseI3 5

HO =PLxHO

Fig. 17. Selectivities in the hydrolysis of phospho-

nate esters.

4 References

AGRAWAL,., GOODCHILD,. (1987), Oligodeoxy-

nucleoside methylphosphonates: Synthesis and

enzymic degradation, Tetrahedron Lett. 28,3539-3542.

ALLEN, . T., HEINTZELMAN,. R., TOONE,E. J.(1992), Pyruvate aldolases as reagents for stereo-

specific aldol condensation, J . Org. Chem. 57,426427.

ANDERSON,. I? (1973), Nucleoside and nucleotide

kinases, n: The Enzym es,3rd Edn., IX (BOYER,

D., Ed.), pp. 49-96. New York: Academic Press.

AuGE, C., GAUTHERON,. (1988), An efficient syn-

thesis of cytidine monophospho-sialic acids with

four immobilized enzymes, Tetrahedron Lef t. 29,

BAUGHN,. L., ADALSTEINSSON,.,WHITESIDES,.

M. (1978), Large-scale enzyme-catalyzed synthe-

sis of ATP from adenosine and acetyl phosphate.

Regeneration of ATP from AMP, J . Am . Chem.

SOC.100,304-306.BEDNARSKI,. D., CRANS,D. C., DICOSINIO, .,

SIMON, . S., STEIN,P. D., WHITESIDES,. M.,SCHNEIDER,. J. (1988), Synthesis of 3-deoxy-~-

manno-2-octulosonate-8-phosphateKDO-8-P)

from D-arabinose: Generation of ~-arabinose-5-

789-790.




BOLTE,.,WHITESIDES,. M. (1984), Enzy matic syn-thesis of a rginine phosphate with coupled ATPcofactor regeneration, Bioorg. Chem. 12,170-175.

BIJSS,K. A., INGRAM-SMITH,., FERRY,. G., SAN-

DERS, D. A., HA SSON , . S. (1997), Crystallizationof acetate kinase from Methanosarcina thermo-phila and prediction of its fold, Protein Sci. 6,2659-2662.

CAGEN, . M., FRIEDMA NN,. C. (1972), Enzym aticphosphorylation of serine, J . Biol. Chem. 247,3382-3392.

CHENAULT,. K., MAND ES, . F., HOR NBE RGE R,.R. (1997), Synthetic utility of yeast hexokinase.Sub strate specificity, cofactor regen eration, andproduct isolation, J. Org. Chem.62,331-336.

COLOWICK,. I? (1973),The hexokinases, n: The En-

zymes, 3rd Edn., IX (BOYE R,?

D., Ed.), pp. 1 4 8 .New York: Academ ic Press.CRA NS, . C., WHITESIDE S,.M. (1983), A conven-

ient synthesis of disodium acetyl phosphate foruse in in situ ATP cofactor regeneration, J. Org.

Chem. 48,3130-3132.CRANS,D. C., WH ITESIDES,. M. (1985a), Glycerol

kinase: substrate specificity, J . Am . Chem. SOC .

CRANS, . C., WHITESIDES,. M. (1985b), G lycerolkinase: synthesis of dihydroxyacetone phosphate,sn-glycerol-3-phosphate, and chiral analogs, J .

Am . Chem.SOC. 07,7019-7027.CRANS, . C., KAZLAUSKAS,. J., HIRSCHB EIN,. L.,

WONG,C.-H., ABRIL,O., WHITESIDES, . M.(1987), Enzy matic regene ration of adenosine5 '-triphosphate - acetyl phosphate, phosphoenolpyruvate, methoxycarbonyl phosphate, dihydrox-yacetone phosphate, 5-ph osph o-a-~ -ribosy lyro-phosphate, uridine-5'-diphosphateglucose, Me-thods Enzym ol. 136,263-280.

DAVIES, . G., GR EEN , . G., KELLY,D.R., ROBERTS,S. M. (1989), Biotransforrnations in PreparativeOrganic Chemistry: The Use of Isolated Enzym esand Whole Cell Systems in Synthesis, pp. 75-86.

San Diego, CA: Academ ic Press.DRAUZ, . , WALDMANN,. (1995), Enzyme Cataly-

sis in Organic Synthesis,A Comprehensive H and-book,pp, 505-545. W einheim: VC H.

DRUECKHAMMER,.G.,WONG, . H. (1985), Chem o-enzym atic synthes es of fluoro su gar phosphatesand analogs,J. Org. Chem. 50,5912-5913.

DUM AS, ., CAM ONIS,. (1993), Cloning and sequen-ce analysis of the cDNA for arginine kinase oflobster muscle, J . Biol. Chem. 268,21599-21 605.

ECKSTEIN,. (1983), Phosphorothioate analogs of

nucleotides - Tools for the inv estigation of bio-chemical processes, Angew. Chem. (Int. Edn.Engl.) 22,42343 9.

ECKSTEIN,. (1985), Nucleoside phosphorothioates,Annu. Rev. Biochem. 54,367402.

107,7008-7018.

FABER,K. (1995), Biotransformations in OrganicChemistry, pp. 105-107. Heidelberg, New York:Springer-Verlag.

FENNIRI,H., LEHN , .-M. (1993), Coupling of su-

pramolecular synthesis of ATP with ATP-con-suming enzyme systems, J . Chem. SOC., Chem.Commun., 1819-1821.

FERNLEY, . N. (1971), Mamm alian alkaline phos-phatase s, in: The Enzym es, 3rd Edn., IV (BO YER ,P. D., Ed.), pp. 417447. New York: AcademicPress.

FUJII ,H., KOYAMA,., OG UR A , . (1982), Efficientenzymatic hydrolysis of polyprenyl pyrophos-phates, Biochim. Biophys. Ac ta 712,716-718.

GIBSON, . W., HINES,W., Yu, Z., KENYON, . L.,

matic synthesis of phosphotyrosine-containing

peptides via adenylated intermediates, J . Am .Chem. SOC.112,8523-8528.

GROSS,A., ABRIL,O., LEWIS, . M., GE RES H, .,

WHITESIDES,. M. (1983), Practical synthesis of5-phospho-~-ribosyl-a-l-pyrophosphatePRPP):Enzymatic routes from ribose 5-phosphate or ri-bose,J.Am . Chem. SOC .105,7428-7435.

GROSS,M., LUSTIG, ,, WALLIMANN,., FURTER, .(1995), Multiple-state equilibrium unfolding of

guanidino kinases,Biochemistry 34,lO350-1 0 357.HAYNIE,. L., WHITESIDES,. M. (1990), Prepara tion

of a mixture of nucleoside triphosphates suitable

for use in synthesisof nucleotide phosphate sugarsfrom ribonucleic-acidusing nuclease -pl, a mixtureof nucleoside monophosphokmases and acetatekinase,Ap pl. Biochem. Biotechnol.23,205-220.

HEIDLAS,. E., WILLIAMS ,. W., WHIT ESIDE S,. M.(1992), Nucleoside phosphate sugars - synthesison practical scales for use as reagents in the enzy-matic preparation of oligosaccharide s and glyco-conjugates,Acc. Chem. Res. 25,307-314.

HILES,R. A., HENDERSON,. L. (1972), The partialpurification and properties of hydroxylysinekinase from rat liver,J. Biol. Chem. 247,64&651.

HIRSCHBEIN,. L., MAZENOD,. F WHITESIDES,.M. (1982), Syn thesis of phospho enolpy ruvate andits use in adenosine triphosphate cofactor regen-eration, J. Org. Chem. 47,3765-3766.

HOSSEINNI,. W., LEHN , .-M. (1985), Cocatalysis:Pyrophosphate synthesis from acetylphosphatecatalyzed by a macrocyclic polyamine, J. Chem.SOC., hem. Commun., 1155-1157.

HOSSEINNI, . W., LEHN , .-M. (1987), Bind ing ofAMP, ADP, and ATP nucleotides by polyam-monium macrocycles, Helv. Chim. Acta 70,

HOSSEINNI,. W., LEHN, .-M. (1988), Supramolec-ular catalysis: Substrate phosphorylations and

adenosine triphosphate synthesis with acetylphosphate catalyzed by a macrocyclic polyamine,J.Chem. SOC.,Chem.Commun.,397-399.

MCNEMAR,., VILLAFRANCA, J. J. (1990), Enzy-

1312-13 19.



4 References 239

HO SSEIN NI, . W., LEHN , .-M. (1991), Supramole-

cular catalysis of adenosine triphosphate synthe-

sis in aqueous solution mediated by a macrocyclic

polyamine and divalent metal cations, J . Chem.Soc., Chem.Commun.,451453.

HOSSEINNI,. W., LEHN , .-M., ME RTE S, . € (1983),

Efficient molecular catalysis of ATP-hydrolysis

by protonated macrocyclic polyamines, Helv.Chim. Acta 66,24542466.

HURLEY,. H. (1996), The sugar kinase heat-shock-

protein-70, actin super family - implications of

conserved structure for mechanism, Ann. Rev.Biophys. Biomol. Struct.25,137.

KABSCH,W., HOLMES, . C. (1995), Protein motifs.

2. The actin fold, FASEB J. 9,167-174.

KAYNE,F. J. (1973), Pyruvate kinase, in: The En-zymes, 3rd Edn., VIII (BOYER, P. D., Ed.),

pp. 353-382. New York: Academic Press.

KAZLAUSKAS,. J., WHITESIDES,. M . (1985), Syn-

thesis of methoxycarbonyl phosphate, a new re-

agent having high phosphoryl donor potential for

use in ATP cofactor regeneration, J. Org. Chem.

KEE, Y., LEE,Y. S., CHUNG,C. H., WAXM AN, .,GOLDBERG,. L. (1988), Improved methods for

purification and assay of glycerol kinase from

Escherichia coli, J. Chromatogr. Biomed. Appl.

KEIKINHEIMO,., LEHTONEN,., BAYKOV,., LAHTI,R., COOPERMAN,. S., GOLDMAN,. (1996), The

structural basis for pyrophosphatase catalysis,Structure 4,1491-1508.

KESSEL,D. (1968), Properties of deoxycytidine kin-

ase partially purified from L1210 cells, J. Biol.Chem. 243,47394744.

KNOWLES,. R. (1980), Enzyme-catalyzed phospho-

ryl transfer reactions, Annu. Rev. Biochem. 49,

KOYAMA,T., GUR A, . , BAK ER,. C., JAMIESON,.C., SCHOOLERY ,. A. (1987), Synthesis and abso-

lute configuration of 4-methyl juvenile hormone

1 (4-MeJH 1) by a biogenetic approach: a combi-

nation of enzymatic synthesis and biotransforma-

tion,J.Am . Chem.SOC.109,2853-2854.KOYAMA,., INOUE,H., O H N U M A , ., OGURA,K.(1990), Prenyltransferase reaction involving

enantiomeric discrimination. Enzymatic synthesis

of (S)-l0,ll-epoxyfarnesol from racemic 6,7-

epoxygeranyl diphosphate and isopentenyl

diphosphate, Tetrahedron Lett. 31,41894190.

KREBS, . G., BEAU, . A. (1979), Phosphorylation -

dephosphorylation of enzymes, Annu. Rev. Bio-chem. 48,923-959.

KREJCOVA,., HORSKA, . (1997), Adenylate kin-

ases,Chem . Listy 91,179-188.

LADNER, . E., WHITESIDES,. M. (1985), Enzymat-

ic synthesis of deoxy ATP using DNA as startingmateria1,J. Org. Chem. SO , 1076-1079.

50,1069-1076.

428,345-351.

877-919.

LASKOWSKI,., Sr. (1971), Venom exonuclease, in:

The Enzymes, 3rd Edn., IV (BOYER,P. D., Ed.),

pp. 313-328. New York Academic Press.

LEGRA ND, . M., ROBERTS,. M. (1993), Carbocyclic

adenosine phosphates as nucleotide mimics in

some enzyme-catalyzed reactions, J. Chem. SOC.,

Chem.Commun.,12841285.

LEHNINGER,. L. (1975), Biochemistry, 2nd Edn.,

p. 398. New York: Worth Publishers.

LESNIKOWSKI,.J.(1993), Stereocontrolled synthesis

of P-chiral analogs of oligonucleotides, Bioorg.Chem. 21,127-155.

LEWIS,. M., HAY NIE,.L., WHITESIDES,. M. (1979),

An improved synthesis of diammonium acetyl

phosphate, J. Org. Chem. 44,864865.

LIN,E. C. C. (1977), Glycerol utilization and its regu-

lation in mammals, Annu. Rev. Biochem. 46,

MAITY,H., JARORI,G. K. (1997), Yeast hexokinase

PII-bound nucleotide conformation at the active

site,Eur. J. Biochem. 250,539-548.

MARTIN, . P., DRUECKHAMMER,. G. (1992),Com-

bined chemical and enzymatic synthesis of co-

enzyme A analogs, J. Am. Chem. SOC. 114,

MORAN,. R.,WHITESIDES,. M. (1984),A practical

enzymatic synthesis of (S,)-adenosine 5 '-0-(1-

thiotriphosphate)((S,,)-ATP-a-S), . Org. Chem.49,704706.

MORRISON,. F. (1973),Arginine kinase and other in-

vertebrate guanidino kinases, in: The Enzymes,3rd Edn., VIII (BOYER, . D., Ed.), pp. 457486.

New York: Academic Press.

MUELLER, . G., CRO WD ER, . W., AVERILL , . A.,

KNOWLES,. R. (1993), Purple acid phosphatase:

A diiron enzyme that catalyzes a direct phospho

group transfer to water, J . Am . Chem. SOC.115,

2974-2975.

MUHLEBACH,. M., GROSS,M., WIRZ,T., WALLI-MANN, ., PERRIARD,. C., WYSS,M. (1994), Se-

quence homology and structure predictions of

the creatine kinase isoenzymes, Mol. Cell. Bio-chem. 133,245-262.

MURPHY,.E.,

STEC,B., MA, L., KANTROW ITZ,. R.

(1997), Trapping and visualization of a covalent

enzyme-phosphate intermediate, Nature Struct.Biol. 4,618-622.

NATCHEV,. A. (1987), Phosphono and phosphino

analogs and derivatives of the natural ami-

nocarboxylic acids and peptides. 1.Synthesis and

enzyme-substrate interactions of N-phosphono-,

and N-phosphinomethylated cyclic amides, Syn-thesis, 1079-1084.

NATCHEV,. A. (1988a), Organophosphorous analogs

and derivatives of the natural L-amino carboxylic

acids and peptides. 111. Synthesis and enzyme-

substrate interactions of D-, DL-, and L-5-di-

hydroxyphosphinyl-3,4-didehydrono~alinend

765-795.

7287-7288.




their cyclic analogs and derivatives, Bull. Chem.SOC.Jpn. 61,3711-3715.

NATCHEV,. A . (1988b),Total synthesis, enzyme-sub-strate interactions and herbicidal activity of

plumbem icin A and B (N-1409), Tetrahedron44,1511-1592.

NEUMANN,. (1968), Subs trate selectivity in the ac-tion of alkaline and acid phosphatases, J . Biol.Chem. 243,46714676.

NODA, . (1973),Adenylate kinase, in: The Enzym es,3rd Edn., VIII (BOYER,. D., Ed.), pp. 279-305.New Y ork: Academ ic Press.

Nomenclature Commitee of the InternationalUnion of Biochemistry (1984), Enzyme Nomen-clature.Orlando, F L Academ ic Press.

PRAD INES, ., KLAEBE,A. , P~RI E , ., PAUL,F.,MONSAN,. (1988), Enzyma tic synthesis of phos-phoric monoesters with alkaline phosphatase inreverse hydrolysis conditions, Tetrahedron 44,63734386.

REID,T. W., WILSON,. B. (1971), E. coli alkalinephosphatase, in: The Enzymes, 3rd Edn., IV(BOYER,. D., Ed .), pp. 373-415. New York: Aca -demic Press.

RIOS-MERCADILLO,V.., WHITESIDE S,. M. (1979),Enzym atic synthesis of sn-glycerol 3-phosphate,J.Am . Chem.SOC.101,5828-5829.

SEELA, ., O n , J., POTTER,B. V. L. (1983), Oxygenchiral phosphate in uridylyl(3’-5’) adenosine by

oxidatio of a phosphite intermediate: Synthesisand absolute configuration, J. Am. Chem. SOC .

SEKINE, ., I IMURA,., FURUSAWA,. (1993), Syn-thesis of a new class of 2‘-phosphorylated oligo-ribonucleotides capable of conversion to oligo-ribonucleotides,J. Org. Chem.58,3204-3208.

SHIH,Y. S., WHITESIDES,. M. (1977), Large -scaleATP-requiring enzymatic phosphorylation ofcreatine can be driven by enzymatic ATP regen-eration, J. Org. Chem. 42,41654166.

SIMON, . S., BEDNARSKI,. D., WH ITESIDES,.M.(1988a), Synthesis of CM P-NeuA c from N -ace-tylglucosamine- generation of CTP from CMPusing adenylate kinase, J. Am. Chem. SOC . 110,7159-7163.

SIMON, . S., BEDNA RSKI,. D., WH ITESIDES,. M.(1988b), Gen eration of cytidine 5 ’-triphosphateusing adenylate kinase, Tetrahedron Lett. 29,

SIMON, . S., GRABOWSKI,., WHITESIDES,. M.(1989), Preparation of phosphoenolpyruvatefrom D-( -)-3-phosphoglyceric acid for use inregeneration of ATP, J. Am. Chem. SOC. 111,8920-8921.

SIMON, . S. , GRABOWSKI,., WHITESIDES,. M.(1990), Covenient syn thesis of cytidine 5 ’-tri-phosphate, guanosine 5 ‘-triphosphate, and uri-dine 5’-triphosph ate and their use in the prepar-

105,5879-5886.

1123-1 126.

ation of UD P-glucose, UD P-glucuronic acid, andGDP-mannose,J. Org. Chem. 55,1834-1841.

SPECTOR,. B. (1980), Acetate kinase: a triple-dis-placement enzyme,Proc. Natl. Aca d. Sci. U SA 71,

2626-2630.STRONG,. J., ELLINGTON,. R. (1995), Isolation and

sequence analysis of the gene fo r arginine kinasefrom the Chelicerate arthropod, Limulus poly-phemus - insights into catalytically importantresidues,Biochim. Biophys. Acta 1246,197-200.

TANG,X. J., JOHANSSON,. (1997), A bioelectro-chemical metho d for the determination of aceta tewith immobilized acetate kinase, Analyt. Lett. 30,

THOMER,. W., PAULUS,. (1973), Glycerol and gly-cerate kinases, in: The Enzymes, 3rd Edn., VIII(BOER, P. D., Ed.), pp. 487-508. New York: Aca-demic P ress.

VOET,D., VOET,. (1995),Biochemistry, pp. 428-434;593-598. Ch ichester: John Wiley & Sons.

WANG,P., SCHUSTER,., WANG,Y. F., WONG, . H.

(1993), Syn thesisof phospholipid-inhibitor conju-gates by enzymatic transphosphatidylation w ithphospholipase D, J. Am. Chem. SOC. 115,

WAIT, D. R., FINDEIS, . A., RIOS-MERCADILLO,.

M., AuGB, J., WH ITESIDES,. M. (1984), An effi-cient chemical and enzymatic synthesis of nico-tinamide adenine dinucleotide (NAD+), J. Am.

Chem. SOC.106,234-239.WAITS,D. C. (1973), Creatine kinase (A denosine 5 ’-

triphosphate-creatine phosphotransferase), in:The Enzym es, 3rd Edn., VIII (BOY ER,. D., Ed.),pp. 383-455. New York: Academ ic Press.

WEI,L. L., Gou x, W. J. (1992),ATP cofactor regener-ation via th e g lycolytic pathway, Bioorg. Chem.20 ,6246 .

WHITESIDES,. M., SIEG EL , ., GARRETT,. (1975),Large-scale synthesis of diammon ium acetylphosphate, J. Org. Chem. 40,2516-2519.

WILD,K., GRAFMULLER,., WAGNER,., SCHULZ,.E. (1997), Structure, catalysis and supram olecularassem bly of adenylate kinase , Eur. J. Biochem.250,326-331.

WILLSO N, ., ALRIC,.,PERIE,., SANEJOUAND,. H.

(1997), Yeast hexokinase inhibitors designedfrom 3-D enzyme structure rebuilding,J. EnzymeZnhib. 12,101.

WILSON,. B., DAYA N,., CYR,K. (1964), Some prop-erties of alkaline phosphatase from Escherichiacoli. Transphosphorylation, J. Biol. Chem. 239,41824185.

WONG, .H.,WHITESIDES,. M. (1981),Enzyme-cat-alyzed organic synthesis: NA D(P)H cofactor re-generation by using glucose 6-phosphate and theglucose 6-phosphate deh ydrogenase from Leuco-nostoc mesenteroides, J. Am. Chem. SOC. 103,4890-4899.

2469-2483.

10487-10 491.



4 References 241

WONG,C. H.,WHITESIDES,. M. (1983), Synthesis ofsugars by aldolase-catalyzed condensation reac-tions, J. Org. Chem. 48,3199-3205.

WONG,C . H., MCCURRY,. D., WHITESIDES,. M.(1980), Practical enzy matic syn theses of ribulose1,s-bisphosphate and ribose 5-phosphate, J.Am.Chem. SOC.102,7938-7939.

WONG,C. H., HAYNIE,S. L., WHITESIDES,. M.(1982), Enzym e-catalyzed synthesis of N-acetyl-lactosamine with in situ regeneration of uridine5 ‘-diphosphate glucose and uridine 5 ‘-diphos-pha te galactose, J. Org. Chem. 47,5416-5418.

WONG,C. H., HAYNIE, . L., WHITESIDES,. M.(1983a), Prep aration of a mixture of nucleosidetriphosphates from yeast R N A : Use in enzymaticsynthesis requiring nucleoside triphosphate re-generation an d conversion t o nucleoside diphos-pha te sugars, J.Am. Chem. SOC.105,115-117.

WONG,C. H., MAZENOD,. P., WHITESIDES,. M.(1983b), Chemical and enzymatic syntheses of

6-deoxyhexoses. Conversion to 2,s dimethyl-4-hydroxy-2,3-dihydrofuran-3-onefuraneol) andanalogs, J. Org. Chem. 48,3493-3497.



5 Enzymes in CarbohydrateChemistry: Formation of GlycosidicLinkages

SABINE. FLITSCH

GREGORY. WATT

Edinburgh, UK

1 Introduction 245

2 Glycosidases 2452.1 Availability of Enzymes 246

2.2 Kinetic vs. Thermodynamic Glycosylation 246

2.3 Glycosyl Donors 247

2.4 Acceptor Substrates for Transglycosylations

2.4.1 Simple Alcohols as Acceptor Substrates 247

2.4.2 Diastereoselective Glycosylation 248

2.4.3 Glycosides as Acceptors 249

2.4.4 Synthesis of Glycoconjugates 251

2.5.1 Aqueous Systems 253

2.5.2 Use of Organic Solvents 254

247

2.5 Glycosylation by Reverse Hydrolysis 253

2.6 Transfer of Disaccharides Using Endoglycosidases 2552.7 Selective Hydrolysis for the Synthesisof Glycosides

3.1 Availability of Enzymes 256

3.2 Availability of Glycosyl Donors 256

3.3 Examples of Oligosaccharide Syntheses Using Glycosyltransferases 258

3.4 Synthesisof Glycoconjugates 261

3.4.1 Synthesis of Glycolipids 262

3.4.2 Glycopeptides and Glycoproteins 2633.4.3 Cell Surface Glycosylation 264

255

3 Glycosyltransferases 256

3.5 Solid Phase Synthesis 264

3.6 Acceptor-Donor Analogs 264





244 5 En zym es in Carbohydrate Chemistry: Formation of Glycosidic Linkages

3.7 Whole-Cell Glycosylation 2663.8 Glycosyltransferases Using Non-Nucleotide Donor Substrates 267

4 Combination of Glycosidases and Glycosyltransferases 2685 Practical Aspects of Synthesis and Analysis 268

5.1 Isolation and Purification of Enzymes for Biocatalysis 2685.2 Biocatalytic Conversions 2685.3 Analysis of Products 2695.4 Chromatographic Methods for the Isolation of Products 269

6 Conclusions 2707 References 270



2 Glycosidases 245

1 Introduction

The glycosidic linkage is used in Natu re as a

way of reversibly attaching carbohydrates toother biological materials and it forms thechains of oligo- and po lysaccharides. Next tothe peptide and phosphodiester linkages ofnucleic acids, it plays a pivotal role in manybiological events ranging from provision ofmechanical stability in cellulose and energystorage in amylose or glycogen to m ediation ofhighly specific inter- an d intracellular recogni-tion events. Although carbohyd rates are high-ly functionalized molecu les, conjugation near-ly always takes place through the anomericcenter as outlined in Fig. 1.The two partners ofthe glycosidic linkage a re generally called gly-cosy1 dono r (e.g., 1) and glycosyl acceptor (2),which is frequently, but not always an alcohol.

The intrinsic chemistry of monosaccharidesprovides for a sta rtling level of struc tural com-plexity of the ir polymers, which rende rs themideally suited as molecules for unique biologi-cal specificity.W hereas nucleic acids and poly-peptide chains are usually joined in a linear, di-valent fashion, the glycoside linkage alone can

result in two stereoisomers, thea-

nd p-ano-mers (e.g.,4 and 3). Furth er structural diversitycan be obtained if the glycosyl acceptor(R O H ) is multivalent, as in the case of saccha-rides. Thus, between only 2 different hexoses,20 different glycosidic linkages can be ob-tained.

The potential for forming chemically simi-lar, isomeric glycosidic linkages has p rovided amuch grea ter challenge to synthetic chemistrythan peptides a nd oligonucleotides. W hereasautoma ted synthesizers are now available for

the synthesis of the latter two, most oligosac-

Fig. 1

Ho6o +O H O H

1

charide synthesis requires lengthy an d labori-ou s synthetic routes and purification proto-cols. Rece nt progress in this area has b een re-viewed elsewhere (PAULSEN, 982; SCHM IDT,

1991; SCHMIDTnd KINZY,1994;BOONS, 996;FLITSCHnd WATT,1997).

A n alternative route to obtaining glycocon-jugates would be by using biological in vivomethods, which again have fou nd w ide appli-cations for proteins and larger oligonucleo-tides. However, in vivo synthesis has met withlimited success because glycoconjugates gen-erally occur in heterogeneous mixtures insmall quantities in biological sources a nd thusisolation from natural sources is not practicalin most cases. The m anipulation of the biosyn-

thetic pathways within the cell by genetic engi-neering techniques is very difficult, becauseeach glycosidic linkage is formed by an indi-vidual enzyme (glycosyltransferases) which inturn underlie complex control mechanisms.

O ne of the most successful method s for thepractical synthesis of defined oligosaccharidesand glycoconjugates has, therefore, bee n the in

vitro enzym atic synthesis, using isolated en-zymes as biocatalysts.The enzymes used in bio-synthesis ar e the glycosyltransferases (and gly-

cosidases), which have been shown to beamenable to use in v i m . Both types of en-zymes will be discussed in detail.

2 Glycosidases

It has been known for a long time that glyco-sidases,which normally catalyze the hydrolysisof glycosidic bonds (i.e., the backw ard re action

in Fig. l ) , can be used “in reverse” for the syn-

OR- H20 ., OH

Glycosyl acceptorGlycosyldonor

9 OR



246 5 Enzymes in Carbohydrate Chemistry: Formation ofGlycosidic Linkages

thesis of glycosidic inkages. System atic studiesusing glycosidases for the synthesis of definedsaccharides started in the early 1980s, andthere are now a large number of examples of

glycosidases that ar e useful in synthesis.

2.1 Availability of Enzymes

Glycosidases can be obtained from a largevariety of biological sources, ranging fromthermophiles, bacteria and fungi to mammal-ian tissue. They are generally stable and solu-ble enzym es. Since they ar e widely used as an-alytical tools for biochemical research, manyglycosidases are commercially available, al-

though not all glycosidases are suitable for syn-thesis. Glucuronidases, e.g., have so far failedto yield any glucuronides. Glycosidases are gen-erally highly specific for a particular linkage (a

vs.p) and for a p articular glycosyl donor. How -ever, they tend to accept a structurally diverserang e of glycosyl acceptors, which ma kes themuseful as general catalysts for glycosylation.A tthe same time, when the acceptor has severalhydroxyl groups, glycosidases might havepoor er regioselectivity, and can result in mix-

tures of isomers that need to b e separated.

2.2 Kinetic vs. Thermodynamic

Glycosylation

In the presence of glycosidases the thermo-dynamic equilibium of the reaction shown inFig. 1 ies generally strongly to the left in aque-ous medium, favoring hydrolyis of the glyco-sides (3) and (4). If glycosidases are used for theform ation of glycosidic linkages, reac tion con -ditions need to be chosen w hich take this intoaccount. In principle, there are two possibleapproaches: the equilibrium can b e shifted to-wards the glycosides in Fig. 1 by decreasing thewater concentration a nd/or increasing glycosyldonor and/or alcohol concentration. This can

be e ithe r achieved by using a large excess of al-cohol (2) or sugar (1)o r by working in organicsolvents.All these approaches have been used.

Alternatively, on e can take advantage of thefact that many glycosidase acceptor su bstrateshave a m uch higher activity com pared t o waterthan w ould be expected from the difference inconcentration.Thus, the enzymebound activat-ed intermediate (represented as an oxoniumion in Fig.2) reacts much faster with an alcohol( R ’ OH) to give (7) nd (8) than it does with

water. In ord er t o encoura ge this kinetic effectover thermodynamic equilibration the glyco-

R

B Fig. 2



2 Glycosidases 247

syl donor needs to b e a good substrate for theenzyme, which necessitates the use of activat-ed glycosyl donors, such as glycosyl fluoridesor aryl glycosides. This allows the reaction to

proceed rapidly before equilibration t o (5) cantake place. Kinetic glycosylations (or “trans-glycosylations”) can, therefore, be performedin aqueous medium with a moderate excess ofdonor or acceptor. A drawback, however, isthe requirement for multistep synthesis of theactivated g lycosyl donor.

2.3 Glycosyl Donors

Glycosidases that have been used for syn-

thesis generally catalyze the transglycosylationstep w ith very high stereospecificity, in mostcases with retention of th e anom eric configu-ration.A s a result, the activated glycosyl don or(6) needs to have the right anomeric configu-ration fo r the particular glycosidase used an dwill get selectively conv erted to only o ne of thetwo possible anom ers (7 or 8). For example, ap-galactosidase would require a p-galactosyldono r a nd w ould catalyze selectively the for-mation of th e p-galactosidic linkage.

Glycosidases tend also to be highly specificfor their donor sugar, although more catholicglycosidases are now being reported, whichhave not been systematically studied in enzy-matic synthesis.

The activated glycosyl don or (6) needs to bea good substrate fo r the enzym e, but needs al-so to be stable in aqueo us medium during the

Ho

POH

reaction. For these reasons, the most commonactivated glycosyl donors a re, therefore , glyco-syl fluorides, pheny l glycosides, nitro-, d initro-,and chloro-phenyl glycosides.A 1,2-oxazoline

derivative of N-acetylglucosamine has recent-ly been used as a donor for chitinase (Ko-BAYASHI et al., 1997). In addition, cheaplyavailable disaccharides such as lactose (for p-galactosidase) and cellobiose (for pglucosid-ase) have been used a s glycosyl donors. Mostof these donors are commercially available.The choice of glycosyl donor depends on th eenzyme system used and can markedly influ-ence yield and regioselectivity of the reaction ,an exam ple of which is given in Sect. 2.4.3.

2.4 Acceptor Substrates

for Transglycosylations

2.4.1 Simple Alcohols

as Acceptor Substrates

Alkyl glycosides have found wide applica-tions such as detergents in research and in in-dustry and are , therefore, good targets for en -

zym atic synthesis using glycosidases, providedthat inexpensive starting materials can be em-ployed. Allyl-, benzy l-, and trimethylsilylethylglycosides can be used as synthetic interm edi-ates carrying a temporary protecting group atthe glycosidic center. On e of th e e arlier appli-cations of transglycosylations for th e synthesisof sim ple glycosides was rep orte d by NILSSON,

a-galactosidase t q wP o

Fig. 3



248 5 En zym es in Carboh ydrate Chemistry: Formation ofGlycosidic Linkages

who found that a mixture of 0.1 M lactose (9)

and 0.2 M benzyl alcohol in phosphate bufferwith 5 mg P-galactosidase yielded 6% of ben-zyl P-glucoside (10) (NILSSON,988) (Fig. 3).

Similarly, allyl alcohol was glycosylated withraffinose (11) and a-galactosidase to give se-lectively the allyl a-galactoside (12).

The transglycosylation method is economicfor these galactosylations,because lactose andraffinose are inexpensive. Where synthetic ac-tivated glycosyl donors such as puru-nitro-phenyl glycosides need to be used, this methodbecomes too expensive for the synthesis of

simple glycosides. Here reverse glycosylationemploying organic solvents or the respectivealcohol as solvent is much more practical andrecent examples will be discussed later on.

a

2.4.2 Diastereoselective

Glycosylation

The diastereoselectivity of transglycosyla-tion with glycosidases has been investigatedwith more complex prochiral or racemic ac-ceptor substrates such as (U a, b),(15) nd (17)(Fig. 4). In the case of the cyclic meso-diols(13a, b) (n=l, 2); (GAIS t al., 1988) good dia-stereoselectivities (75%de and 96% de) wereobtained when the reactions were performedin 50% (v/v) aqueous acetone. Interestingly,nodiastereoselectivity was observed without ad-dition of the co-solvent.

In other cases reported so far, the selectiv-ities have been much lower. For example, for

p-galactosidase

E. coli

‘O H acetonehuffer *a--H

+ lactose

n = i : 2oD/o(96%de)

141r n-2: 189&(7!5%de)

pgalactosidase Ho

E. coli

HO+ lactose

OH

Ho

HoSOLH

p-galactosidaseE. coli

t 1 RS = 10:8.6

WJ, + lactose +

Fig. 4

H0 OH

nob o T O Hp OH

WS 10:7.7



2 Glycosidases 249

the galactosylation product of 2,3-epoxyprop-anol (15) €US values for the aglycon of (16)were only 7/3 (BJORK LINGnd GODTFREDSEN,1988). Galactosylation of 1,2-propanediol(l7)

resulted in a mixture of regio- and stereoisom-ers with poor diastereoselectivity (CROUT ndMACMANUS,990).

2.4.3 Glycosides as Acceptors

A much more challenging applicationof gly-cosidases than alkyl glycosides is the use ofsaccharides themselves as acceptor substrates,which results in the synthesis of oligosaccha-rides. This provides not only a high degree of

stereoselectivity of reaction but also the po-tential glycosylation of one of several hydrox-yl groups.Thus, 8 different P-galactosides canbe formed by galactosylation of methyl galac-

toside (21) with para-nitrophenyl galactoside(20), as a result of galactosylation of the 4 dif-ferent hydroxyl groups each of the acceptor(21) and the donor (20) itself. However, withthe right choice of reaction conditions, sub-strates and enzymes, reasonable regioselectiv-ities of reactions can be obtained, which has al-lowed the use of glycosidases for the synthesisof small oligosaccharides.

The competition of the donor substrate withacceptor can be overcome by using a large ex-cess of acceptor substrate, although this can

OMe

HO .OH Ho We

P O H

Ho

HOk.0o, p-galactosidase OH OH

majorHo P

p

Ho

OMe

k&-0- No, major OMe

Ho

OH

Ho OH

Ho

ap 24

Ho

bHog- 0

+ aa-galactosidase

Fig. 5



250 5 Enzym es in Carboh ydrate Chemistry: Formation of Glycosidic Linkages

present problems if the acceptor is valuable ordifficult to separate from product.

A bias of regioselectivity is generally ob-served in glycosidase-catalyzed reactions, of-

ten with preferential formation of the 1-6 link-age. However, there have been many reportsof good selectivities for other acceptor hydrox-yl groups. How the regioselectivity of disac-charide formation can be influenced by appar-ently subtle changes in glycosyl donor or ac-ceptor structure has been shown by NILSSON

(NILSSON,987) (Fig. 5 ) . For example, the 1-3P-glucosidic linkage was mainly formed (22),when the P-methylglycoside (21)was the ac-ceptor and para-nitrophenyl p-glucoside (20)

the donor, whereas the 1-6 linked disaccharide(24) was the main product using methyl a-gly-coside (23) s the acceptor substrate.

Even more pronounced selectivity was ob-

served with nitrophenyl glycosides (25)whichact both as acceptors and donors to give the(1-3) or (1-2) linked disaccharides (26) and(27) in different ratios, depending on whetherortho- or para-nitrophenyl glycosides are used.With the ortho-isomer of (25) the ratio of (26)to (27) was less than 1:6 whereas with thepara-isomer of (25) it was reversed (8:1).

Glycosidases can not only be used for disac-charide synthesis, but also for the assembly ofmore complex oligosaccharides. Using a p-N-

OH

OHo

NHAC HO29 NHAe

N4 8-Nacetylhexosaminidase * +HAe

I A. oryzae

OH

Ho

Ho HONHAC N W

3Q

B-N-acety lhexosaminidase

A. oryzae

pmannosidaseA. otyzae

m,Ho NHtc Ho ;II-1HAe Ho NHAC OH

Fig. 6

HO OH

IHo

Ho Ho

NuAc NHAC

;P 2 6 Y O



2 Glycosidases 251

acetylhexosaminidase from Aspergillus ory-zae, CROUT nd coworkers have synthesizedchitooligosaccharides starting from N-acetylglucosamine (Fig. 6). The first transglycosyla-

tion using para-nitropheny l G lcNAc (28) pro-ceeded with good selectivity for the 1-4isomer(30), although some 1-6 isomer (29) wasformed. The latter could be removed by selec-tive hydrolysis using the 1-6 specific P-N-ac-etylhexosaminidase from Jack bean. Furtherelongation to higher chito-oligosaccharidessuch as (31) (with (30) cting as the donor andacceptor went with rem arkable selectivity forth e 1-4 linkage, with no detection of the 1-6product (SING H t al., 1995). Equa lly, glycosy-lation with a p-mannosidase from A. oryzaeand para-nitrophenyl Man gave the trisaccha-ride (32) n 26% yield based on the donor asthe only regioisomer (SINGH t al., 1996).

NILSSONas reported recently on the two-step synthesis of a trisaccharide de rivative (35)which is related to antigens involved in hyper-acute rejection of xenotransplants (Fig. 7) .Both the thioethyl group and the 2-N-trichlo-roethoxycarbonyl group allow the furtherstraightforward chemical elaboration of the

trisaccharide either for analog synthesis orconjugation t o polymers. Both reactions wereperforme d w ith excess of glycosyl dono r an dgave 20% and 1 2% yields based on the accep-

tor substrates (33) and (M), espectively(NILSSON,997).

2.4.4 Synthesis of Glycoconjugates

Oth er than saccharides, glycosidases are al-so able to glycosylate many other classes ofnatural products such as show n in Fig. 8. 001

and colleagues have used galactosidase fromA . oryzae for the synthesis of cardiac glyco-sides such as (37) rom the precursor (36) n26% yield. These glycosides are chem ically un-stable and difficult to obtain by conventionalchemical synthesis which involves harsh gly-cosylation conditions and deprotection strate-gies (001 t al., 1984).

An important class of glycoconjugates areglycopeptides an d glycoproteins. Many ad-vances have been made in the chemical syn-thesis of glycopeptides, which requ ire largeamounts of glycosylated amino acid building

NI p-galactosidaseactose ~ HoHo&o& OH no NI SEt

20%ullera singularis

oA O A m , Ia ;ip

a-galactosidase

green coffee beans

Fig. 7



252 5 Enzy mes in Carbohyd rate Chemistry: Formation of Glycosidic Linkages

phenyl J3-Dgalactoside

A. oryzae

J3-galactosidase OH

26% OH

Ho BD0.K)

15% is

o-nitrophenylp-D-glucoside

Ho OH Y2o-nitrophenylJ3-D-galactoside

H o d w J3-galactosidase * HObodH C02H

!R A. oryzae 10%

Fig. 8

blocks, such as glycosyl serine derivatives. Suchbuilding blocks are accessible by enzymaticsynthesis of glycosyl serine derivatives usingtransglycosylation methodology. CANTACU-ZENE and colleagues have reported the synthe-sis of various protected galactosyl serine deriv-atives such as (39) from lactose and the semi-protected serine (38) n 15% yield based on

lactose (CANTACUZENEt al., 1991).

The need for adding excess of acceptor sub-strate in order to reduce competition by thedonor substrate can be a problem if the accep-tor is expensive. For example, the yield of (39)

based on (38) as a starting material is only 8%.This can be overcome (TURNER nd WEBBER-LEY, 991) by using excess donor instead, withslow addition of the glycosyl donor during the

reaction (with a syringe pump). Using this



2 Glycosidases 253

method, the glycosyl serine derivative (41)

could be prepare d from (40) in 25% yield. Al-ternatively, for the synthesis of glycosyl serinederivatives, the less expensive serine (42) itself

was used in large excess (30-fold) to g enerategalactosyl serine (43) directly (SAUERBREIndTHIEM,992; HEIDELBERGnd THIEM,997;NILSSONt al., 1997).

2.5 Glycosylation by Reverse

Hydrolysis

As discussed in Sect. 2.2, the alternative totransglycosylation is the use of glyco sidases inthe reverse hydrolysis, which has the advant-age that no expensive activated glycosyl do-nors are needed. Reverse hydrolysis isachieved by reducing the water/glycosyl donorratio such that the reaction is driven towardsformation of the glycosidic linkage, rathe r thanhydrolysis. This can e ither be achieved by us-ing a very large excess of acceptor in aqueoussystems, or by using the enzym e in organic sol-

vent mixtures, thus reducing the w ater activity.The use of organic solvents has the ad vantagetha t purification is easier, although no t all gly-cosidases can tolerate organic solvents.

2.5.1 Aqueous Systems

Di- and trisaccharides can be prepared byusing very high co ncentrations of acceptor anddonor. This makes use of the high solubility ofsugars in water and of the good stability of en -zymes under high sugar concentrations. Forexam ples incubating a solution of 30% N-ace-tyl glucosamine and 10% galactose with P-ga-lactosidase from Escherichia coli the 1-4 an d1-6 isomers (44) and (45) could be o btained inoverall yield of 7.6% (AJISAKAt al., 1988;AJI-

SAKAand FUJIMOTO,989) (Fig.9).Purificationof th e disaccharides was possible by activatedcarbon chromatography, since the affinity fordisaccharides is much higher than the affinityfor monosaccharides. In addition, the overallyield could be improved to 16% by using acon tinuou s process involving the circulation of

H&

&OH&

Hoo& OH Ho OH

Ho OH

tNHAC

0% OH

Hofbgalactosidase +

m o E. coli Ho

16% yield

Ho

OH Ncue

4 :44=4:1

Manal-GMan (2.2 g)

Manal-2Man (290 mg)- Manal3Man (33 mg)

a-mannosidase

Aspergillus niger

Ho&Ho

50°C Mana l-2ManalB Man (14 mg)H

10g

80%W h Manal-2Manal-6Man (3mg)

Fig. 9



254 5 Enzymes in Carbohydrate Chemistry: Formation of Glycosidic Linkages

the reaction mixture through columns of im-mobilized galactosidase and activated carbonin series.

Mannobioses a nd mann otrioses were isolat-

ed upon incubation of mannose (80% W/V)with an a-mannosidase from A. niger (AJISA-

KA et al., 1995) (Fig. 9).

2.5.2 Use of Organic Solvents

The use of organic co-solvents has been in-vestigated for both the kinetic and thermody-namic glycosylation with almond P-glucosid-ase using tert-butanol o r acetonitrile as co-sol-vents in orde r to improve yields. tert-Butanolwas particularly useful because it does not actas an acceptor sub strate, presumably becauseof steric hindrance, but the enzyme remainedstable in up t o 95%, with an optimum of activ-ity at abo ut 90% butanol in water giving up to20% yield. Total loss of activity was observedin 100% solvent (VIC nd THOMAS,992). Theyield could be improved by using the a cceptoralcohol itself as a solvent, in which case ally1and benzyl glucosides could b e synthesized in40% and 62% yields (VIC and CROU T, 995).

This methodology could also be used for diolssuch as 1,6-hexane diol using reverse hydroly-sis, such that, e.g., the glucoside (46)was ob-tained in good yield (61% ) from glucose (Fig.

10) (VICe t al., 1996).Reverse hydrolysis reactions normally pro-

ceed very slowly, but can be accelerated usinghigher temperatures if th e enzyme can toleratesuch incubation conditions. The P-glucosidasefrom almonds, e.g. (Fig. 10) gives higher yieldsat 50°C (VIC nd CROU T, 994).Recently, ther-mophile a nd thermostable glycosidases havebeen isolated which can be used a t even high-er temperatures or under focused microwaveirradiation (GELO -PUJIC t al., 1997) to giveproducts in good yields. Thus, glucosides (48)

and (49) were produced in up to 77% and20%, respectively, from (47) using crude ho-mogenate from Sulfolobus solfutaricus at110 "C in 2 h. C omm ercial recombinan t ther-mophilic enzyme libraries are now becomingavailable, which give a num ber of b iocatalystsfor screening. For example, the glycosidaseCLO NEZY MET M library developed by Re-combinant Biocatalysis Inc. has been used fo roptimizing the synthesis of N-acetyl lactos-amine (LI and WANG,1997).

O-OH

Ho 90%Vhl OH

* H O6o&Ho OH OH pglucosidase Ho OH

OPhHo

OH

a

(almond) 4

..G0Jo

OH

m 4t +

Fig. 10 4



2 Glycosidases 255

reaction conditions (KOBAYASHIt al., 1991)(Fig. 11). More recently, glucanohydrolaseshave been used to make tetra- and highersaccharides by using suitable acceptor sac-

charides. For example, a 1,3-1,4-P-~-glucan -glucanohydrolase from Bacillus licheniformiscan catalyze the formation of the tetrasac-charide Glc~l-3Glc~l-4Glc~l-3Glc~l-OMefrom G lcP1-3GlcP1-F and G lcP l-3G lcp l-OMe in 40%yield. By using a 7-fold excess ofacceptor substrate, autocond ensation could besuppressed (VILADOTt al., 1997).

2.6 Transfer of Disaccharides

Using Endoglycosidases

Most glycosidases that have been used asbiocatalysts are exoglycosidases,which hydro-lyze or synthesize terminal glycosidic linkages.However, a number of endoglycosidases ar eknown and even available, which would allowassembly of larger oligosaccharides from natu-ral o r synthetic building blocks.An interestingand useful example is the application of cellul-ase for such a “block synthesis” approach tosynthetic cellulose polysaccharides. KOBAYA-SH I and colleagues have reported that incuba-tion of fi-~-ce llobiosy l luoride (50) with cel-lulase from Trichoderma viride in a mixed sol-vent of acetonitrile/acetate buffer (5 :1) yield-ed predominantly water-soluble and insolublecello-oligosaccharides of DP larger than 22and sm aller than 8, respectively, depe nding on

2.7 Selective Hydrolysis for the

Synthesis of Glycosides

Regioselective glycosylation can also beachieved by selective hydrolysis of multigly-cosylated substrates. For example, the p otent

Ho

HOo

9p

Ho

Ho

Fig. 11

OH

..-oOH

h

B-alucoronidase 2OH

Fig. 12 a



256 5 Enzym es in Carboh ydrate Chemistry: Formation ofGlycosidic Linkages

analgesic morphine-6-glucoronide (52) wasobta ined by selective hydrolysis of the read ilyavailable diglucuronide (51) sing glucuronid-ase a s a catalyst (Fig. 12) (BROWN t al., 1995).

3 G1ycosyltransf er ases

Glycosyltransferases are the enzymes thatare naturally responsible for th e biosynthesisof glycosides. They ca talyze the transfe r of asaccharide from a glycosyl don or t o a glycosylacceptor with either inversion or retention ofthe anomeric center as outlined in Fig. 13.

Most of the transferases discussed here useglycosyl phosphates as donors, in particularthe sugar nucleo tides show n in Fig. 14. How ev-er, the glycosyl don or m ight also be a glycosideitself, such a s a di- o r polysaccharide. Since gly-cosyltransferases have evolved t o catalyze theform ation of a very specific glycosidic linkagein a biosynthetic pathway, they tend to b e high-ly selective both for acceptor and donor sub-stra tes as well as highly regio- and stereoselec-tive in the form ation of the glycosidic bond.

3.1 Availability of EnzymesAlthough a large number of am ino acid se-

quences for glycosyltransferases are nowknown (FIELDand WAINWRIGHT,995), thenum ber of enzymes comm ercially available isstill small compared to glycosidases, and theytend to b e relatively expensive enzymes.5 gly-cosyltransferases are currently sold throughcatalogs, namely p-l,4-galactosyltransferase,

a-2,3-sialyltransferase, -2,6-sialyltransferase,a-1,2-mannosyltransferase, and a-1,3-fucosyl-

transferase V.However, protocols fo r the isola-tion of transferases from natural sources are

well established (PALCIC, 994) and cloningand overexpression techniques have greatlyincreased the number of enzymes that areavailable for biotransformations in recent

years (GIJSEN t al., 1996). Although most ofthe glycosyltransferases come from higher or-ganisms, there are some examples of successfulheterologous overexpression in prokaryotes(WAIT e t al., 1997; WANG t al., 1993). PALC IChas recently compiled a useful list of trans-ferases that have been used for preparative-scale biotransform ations (PALC IC, 994).

3.2 Availability of Glycosyl Donors

Despite the variety of different glycosyl-transferases that are found in Nature, only avery small num ber of comm on glycosyl donorsar e used by the m ajority of transferases. Theirstructures are shown in Fig. 14. These are allcomm ercially available, but expensive , if reac-tions need to be done beyo nd a 1G 50 mg scale.In addition, glycosyltransferase reactions oftensuffer from produc t inhibition when stoichio-metric am ounts of glycosyl don or are used.

Both problems can be overcome by in situ

regeneration of the sugar nucleotide cofactorsuch that concentration s of the nucleotide sub-strates and products a re kep t low at all timesduring the biotransformation (ICHIKAW At al.,1994).Such sugar nucleotide regene ration usesthe biosynthetic enzymes in vitro and was firstdeveloped by WH ITESIDES nd colleagues(WO NG t al., 1982) in the early 1980s. Sincethen m any improvements have been reported,such that glycosyltransferases can now be usedto produce complex oligosaccharides on a kilo-gram scale (ICHIKAWAt al., 1994). The first

reported regeneration protocol was for UDP-Glc and UD P-G al as outlined in Fig. 15.Both

0




R-OH: UDP-GIC R-OH: UDP-GalR - N H k : U D P - O I C N A C H0 OH R - N H k : UDP-GelNAc H0

OH

GDP-Man

CMP-Ne u-5- Ac

0

Ho

GDP-FUCH O O H

Fig. 14

were generated from glucose-1-phosphate(53) and substoichiometric amounts of UTP.Synthesis of UDP-Glc is achieved by UDP-Glc pyrophosphorylase, with the by-product

inorganic phospha te (PPi) being hydrolyzed topho spha te using inorganic pyrophosphatase to

avoid enzyme inhibition. UDP-Glc can thenbe used directly in the reaction o r be convert-ed to UDP-Gal using UDP-Gal 4-epimerase.The latter equilibrium favors formation ofUD P-Glc, which can lead t o com plications ifthe subsequent galactosyltransferase can ac-cept UD P-Glc as a substrate. U D P generatedfrom th e glycosyltransfer reaction can be re-generated to UTP by using pyruvate kinaseand phosphoenolpyruvate as the phosphory-lating agent (ICHIKAWAt al., 1994).With such

cofactor regeneration, allyl-N-acetyl lactos-amine (55) was synthesized in 5 d o n a 1.7 gscale in 50% yield from (54) using equim olaramounts of (54), glucose-1-phosphate, and

phosphoenolpyruvate and 0.025 equivalents ofUDI?

When the m onosaccharide starting materialis expensive, as in the case of sialic acid, the re-generation system can include synthesis of themonosaccharide itself, as illustrated in Fig. 16for the regeneration of CM P-NeuAc, the com -mon donor substrate for sialyltransferases(ICHIKAWAt al., 1994). Synthesis of sialic acid(NeuAc) is achieved from N-acetyl mannos-amine (56) and pyruvate by using a sialyl al-dolase. The monosaccharide is then activatedto C MP-NeuAc by the action of CMP-NeuAc



258 5 Enzym es in C arbohydrate Ch emistry: Formation of Glycosidic Linkages

j3-1,4GalT b0&Ho Ho O\/\\dGL&I Nwc

L7-/apPPase ww%opQ*

OH 5s NHAC

UDP-Gal

phosphoenolpyruvate

pyruvate kinase

[=uD pyruvate

UDP-Glc

2Pi 53

Fig. 15

synthase using CTP. After transfer of the sialicacid to a suitable acceptor substrate by sialyl-transferase, the CMP is recycled to CTP n twosteps first using ATP and nucleoside mono-phosphate kinase to generate CDP and thenpyruvate kinase to generate CTPThe same py-

ruvate kinase can also be used to recycle ATPneeded for CDP formation.

3.3 Examples of

Oligosaccharide Syntheses

Using Glycosyltransferases

Glycosyltransferases are generally highlyselective for both their glycosyl donor and gly-cosy1 acceptor substrates and catalyze the for-mation of one specific glycosidic linkage with

very high stereo- and regioselectivity. Thismakes them particularly useful as biocatalystsfor the synthesis of much larger and complexoligosaccharides, where regioisomers thatwould be generated by glycosidases are diffi-cult to separate.

First examples of the use of glycosyltransfer-ases appeared in the early eighties (WONG tal., 1982) (AUGE t al., 1984) using P-lP-galac-tosyltransferase from bovine colostrum, whichhas become the best studied glycosyltransfer-ase and is now commercially available. Di- andtrisaccharides (Galpl4GlcNAc and Galpl-4-GlcNAcpl-6/3Gal) were prepared on a milli-mole scale using immobilized enzymes and re-generation systems for the sugar nucleotidecofactors. This work has been extended overthe past 15 years, perhaps driven by the discov-ery that oligosaccharide ligands such as the




a-sialyl transferaseacceptor-OH -m

CMPMP-NeuAc

ADP

co;

0

Fig. 16

blood group antigens sialyl Lewis x and sialyl

Lewis a are involved in cell adhesion between

activated endothelial cells and neutrophils and

have potential therapeutic applications in in-

flammation and cancer metastasis (FEIZI, 993;

LASKY,994; PAREKHnd EDGE,1994).Thus,the sialyl Lewisx structure (60) (Fig. 17)can be

assembled in only three steps from (57) using a

galactosyltransferase to form (S),ollowed by

sialylation to (59) and final fucosylation (ICHI-

KAWA et al., 1992). These reactions have been

performed on up to 0.5-1 kg scale and have

also been recently used for bivalent sialyl

Lewis x structure (DEFREES t al., 1995).

Using similar enzymes, the sialylated un -decasaccharide-asparagine conjugate (62),an

important side chain of glycoproteins,was pre-

pared from chemically synthesized (61) in

86% yield by enzymatic transfer of saccharide

units in a one-pot reaction (UNVERZAGT,996).

Besides focus on galactosyl-, sialyl-, and fu-

cosyltransferases,advances have been made in

studying glycosyltransferases involved in the

synthesis of the core structures of oligosaccha-



260 5 Enzym es in Carbohydrate Chemistry:Formation of Glycosidic Linkages

1) galactosyltransferaseE. Colialkaline phosphatase E. cob

GlcNAcpO-ally1

p1-4 galactosyltransferase

a 2 3 sialyltransferase

al-3 fucosyltransferase

Neu5Aca(2+3)Gal~(l4)GlcNAcp-O-allyl

1II

Galp(14)GlcNAcf3-0-allyl

Ia(l+S)Fuc

Neu5Aca(2+3)Galp(14)GlcNAcp-0-ally1 (u

0 NHzHo

%oOUDP

OH

2) a-2,6-sialyltransferase€.lkaline phosphatase E. m/iob Hox&ir-LCHV

Ho

fNeuSAca(l+6)Gal~(1+4)GlcNAt$(l +P)Manal

h6 Hn-Man8(1+)GlcNAcp( 1+4)GlcNA@l- N

0 NH2euSAca(1+6)Galp( 14)GlcNAcp( 1+P)Manal f

IpFig. 17

ride side chains of glycoproteins.A P-1,4-man-nosyltransferase from yeast involved in thebiosynthesis of the conserved trisaccharidecore of glycoproteins was overexpressed in E.

coli as a soluble HislO-fusion protein. Absorp-

tion of the enzyme on an immobilized metal(nickel) affinity column provided the manno-syltransferase directly from crude cell extractsas a functionally pure biocatalyst, which wasused in the synthesis of the glycolipid (64)




from synthetic (63a) (Fig. 18; W A ~t al.,1997). An a-l,2-mannosyltransferase fromyeast has also been overexpressed in E. coli(WA NG t al., 1993).

Most asparagine-linked oligosaccharides inproteins contain a common pentasaccharidecore but are very heterogeneous in furtherelaboration of the oligosaccharide structure(KORNFELDnd KORNFELD,985). Heteroge -neity is first introduced by the biosynthesis ofdifferent N-acetyl glucosaminidic linkages tothe pentasaccharide chore, each being catal-yzed by a different GlcNAc-transferase. Theseenzymes have been investigated by SCHACH-TE R and colleagues (BROCKHAUSENt al.,1988, 1992) and have been assigned asGlcNAc-transferases I-VI according to thelinkage they form , as shown in Fig. 19. All ofthese can be isolated in at least milliunitamounts, which is sufficient for milligramsynthesis. The GlcNAc-transferases transfe rGlcNAc not only to the natural glycoproteinor peptide side chain, but also to shorte r oligo-saccharide acceptors, which are syntheticallyavailable. Thus trisaccharide (65) is a substratefor G lcNAc-transferases I and I1 and (66) forGlcNAc-transferase V (Fig. 19) (KAURet al.,

1991; PALC IC,994).Glucuronosyltransferases, which transferglucuronic acid from UDP -glucuronic acid toacceptor substrates are very important en-zymes in the metabolism of endogenous andexogenous lipophilic compounds, since glucu-

Fig. 18

ronidation enhanc es water solubility and thusexcretability from the cellular milieu. Theseenzymes are, therefore, of great interest indrug metabolism, and a liver transferase

system has been used for th e synthesis of phe -nol and lipophilic steroid glucu ronates in goodyields (65% and 30%) on a milligram scale(GY GA X t al., 1991). When crude liver homo-genate was used all the necessary enzymes re-quired to convert glucose-1-phosphate toUDP-glucuronic acid were present in thepreparation, such that the m uch less expensiveglucose-1-phosphate could be employed as adonor substrate. On the other hand, the liverenzymes and UDP-glucuronic acid are nowcommercially available.

3.4 Synthesisof Glycoconjugates

One of the greatest advantages of usingglycosyltransferases is that comp lex glycocon-jugates such as lipids, peptides, proteins andeven cell surfaces can be glycosylated. Thus,oligosaccharides can be assembled directly onthe glycoconjugate in a biomimetic fashion,which is difficult with g lycosidases because of

the high acceptor concentrations required andpoo r regioselectivity. Equally, chemical syn-thetic methods of glycosidic bond formationoften fail on macromolecules, consequently,the oligosaccharide must be assembled beforeattachment to the macromolecule. Glycosyl-

p-Mannosyl-

transferase 14-19GDP-Man

HO 0 0

w.uo

\Ho

84 .o 0



262 5 En zym es in Carbohydrate Chemistry: Formation of Glycosidic Linkages

GlcNAc transferases

GlcNAcT V

GlcNAcT VI

GlcNAcT 1 1

GlcNAcT I l l

GlcNAcT IV

\

/

alcNAcbl-6

GlcNAcp14

GlcNAcpl-2Man a14

G l c N A c ~ l 4 M a n 1 4 R

GlcNAcb14Mana 1 4

GlcNAcfil-2

/GlcNAcT

I

Ho

noHO

OR Ho

+h

Ho O(W)FHsOH

OH

acceptor for GlcNAcT I and I &#acceptor for GlcNAcT V

Fig. 19

transferases are, therefore, important tools forthe biochemist for m odifying the g lycosylationof cell surfaces, proteins, an d lipids and for in-

vestigating structure-function relationships. Afew examples are given in the following sec-tions.

3.4.1 Synthesis of Glycolipids

Glycolipids are ubiquitous me mbran e com -ponents of cells and are involved in a widerange of biological processes, which makesthem important synthetic targets. Examples ofthe enzymatic synthesis of glycolipid interme -diates in glycoprotein biosynthesis were dis-

cussed earlier (Fig. 18; WATT et al., 1997;IMPERIALInd HENRICKSON,995).

Another important class of glycolipids ar e

glycosphingo lipids such as gangliosides, whichare commonly found in eukaryotic cells andparticipate in a number of cell recognitionevents (KARLSSON,989; HAKOMORI,990;ZELLER nd MARCHASE,992). These lipidscontain ceramide as their membrane-associat-ed hydrophobic anch or and com plex oligosac-charide headg roups, which contain sialic acidsin the case of gangliosides. The biosyntheticpathways of glycosphingolipids app ea r similarto th at of othe r glycoconjugates, in that theoligosaccharide headgroup is assembled byspecific glycosyltransferases from the reducing




end (BOUHOURSnd BOUHOURS,991). Al-though glycosidic linkages in glycolipids arequite sim ilar to those in glycoproteins, it ap-pears that distinctly separate glycolipid and

glycoprotein transferases are used in biosyn-thesis (MELKERSON-WATSONnd SWEELEY,1991; ITO et al., 1993). Indeed, glycoceramidesthemselves are often very poor substrates forthe glycoprotein glycosyltransferases discus-sed earlier. On the other hand, the glycolipid-specific transferases are often n ot available ona sufficient scale for application in millimoleglycolipid synthesis (PR EUSS t al., 1993).

Several practical solutions have be en foundto overcome these problems. ZEHAVInd col-leagues demonstrated that enzymatic glyco-sphingolipid synthesis was possible on solidsupport using bovine milk galactosyltransfer-ase (ZEHA VI t al., 1990). Mo re recently, thisapproach was extended to sialyltransferases,with transfer of the oligosaccharide productfrom solid support onto ceramide using cera-mide glycanase (NISHIMURAnd YAMADA,1997).Alternatively, more water-soluble chem-ical precursors of ceramide such as azido-sphingosine derivatives were good acceptorsubstra tes fo r g lycosyltransferases, which led

to efficient chemo-enzymatic routes of gangli-oside G M3 and analogs (GUILBERTt al., 1992;GUILBERTnd FLITSCH,994).

3.4.2 Glycopeptides

and Glycoproteins

Oligosaccharides are linked to peptides a ndproteins through two main types of linkage:the so-called “0-linkage’’ where the hydroxyl

groups of serine and threonine provide sitesfor glycosylation and the “N-linkage” whereglycosylation occurs through the amide ofasparagine (SCHACHTER,991; IMPER IALIndHENRICKSON,995).Thre e main types of enzy-matic method s have been reported for the syn-thesis of glycopeptides an d glycop roteins: first-ly, th e group of WONG as used sub tilisin-cata-lyzed glycopeptide condensation for the syn-thesis of ribonuclease glycoforms ( W m etal., 1997). Secondly, enzymes have been usedto catalyze the formation of the glycosidicbonds that attach the oligosaccharide to the

peptide side chain. Thirdly, glycosyltransfer-ases can exten d a n existing mon o- or oligosac-charide that is already attached to the proteinby fu rther glycosylation.

The most common monosaccharide linkedto the serine or threonine side chain in O -gly-can core structures is an a-GalNAc residue(SCHACHTER,991).Biosynthesis of the O-gly-can the n occu rs by successive add ition of indi-vidual monosaccharide units from the corre-sponding ac tivated sugar nucleotides. The firstglycosyltransferase in this pathway, peptideGalNAc-transferase, has been isolated and isavailable for the glycosylation of peptides(CLAUSENnd BENNEIT, 1996) (YA NG t al.,1992). Considerable effort has go ne into stud-ying the relationship between acceptor pep-tide sequence and enzyme activity. Althoughthe enzyme m ight glycosylate different threo-nine and serine residues within a given poly-peptide sequence at different rates, so far, nopeptide motif for 0-glycosylation has beenidentified (CLAUSENnd BENNEP, 1996), andhence specific glycosylation remains elusive.

Th e form ation of N-glycans occurs by a verydifferent biosyn thetic process. N-glycosylationis very selective for a tripeptide motif includ-

ing the asparagine that is to be glycosylated(Asn-Xaa-Thr/Ser, where Xaa cannot be pro-line). Furthermore, monosaccharide units arenot assembled sucessively on the polypeptide,but are first biosynthesized on a phospholipid(dolichol) and then transferred from th e lipidto the nascent polypeptide chain by an oligo-saccharide transferase (KORNFELDnd KORN-FELD, 985;IMPERIALInd HENRICKSON,995).The natural substrate for the oligosaccharyl-transferase is a tetradecasaccha ride lipid con-sisting of to GlcNAc, 9 mannose and 2 glucose

units, but the enzyme also accepts smallersubstrates down to the disaccharide-lipid(GlcNAc/31-4GlcNAc-lipid).A cru de enzyme-containing extract has bee n used to glycosylatepeptides containing the Asn-Xaa-Thr/Sermotif (LEE and COWAR D,993), but glycosyla-tion of full-length proteins ap pears t o be moredifficult (LIU et al., 1994; XU an d COW ARD,1997; IMPER IALI,997). Thus, the biomimeticapproach for the synthesis of N-glycosylatedproteins using the biosynthetic glycosyltrans-ferases, despite their availability, is still diffi-cult.



264 5 Enzym es in C arbohydrate Chem istry: Formation of Glycosidic Linkages

An interesting alternative non-biomimeticway of generating N-glycans has involved theuse of endo-P-N-acetyl-glucosaminidaseEn-do-A) from Arthrobacter protophormiae (FAN

et al., 1995), which transfers Man9GlcNAcfrom Man9GlcNAc2Asn onto GlcNAc-pep-tide to form Man9GlcNAc2-peptide. This en-zyme can also be used to make N-glycopeptideanalogs with C-glycosidic linkages to the pep-tide (WANG t al., 1997).

Galactosyltransferases, sialyltransferases,and fucosyltransferases previously describedcan also be used to extend oligosaccharides onpeptides and proteins (UNVERZAGTt al.,1994; UNVERZAGT,996; SEITZ and WONG,1997).The disadvantage of all these methods isthat they require initial attachment of a singlemonosaccharide specifically onto the polypep-tide chain, which can be achieved either chem-ically or by truncating the natural oligosaccha-ride back to a single GlcNAc residue, using ap-propriate glycosidases such as Endo-H. Fur-ther development of these methods should ul-timately lead to the synthesis of pure “glyco-forms” of proteins, which remains a majorchallenge to scientists (BILL and FLITSCH,1996; VERBERT,996; STANLEY,992).

3.4.3 Cell Surface Glycosylation

An interesting application carrying the useof glycosyltransferases even further into bio-logical systems is the glycosylation of cell sur-faces (PAULSENnd RODGERS,987). Thus de-sialylated erythrocytes were treated with sia-lyltransferases of different specificity resultingin samples of erythrocytes with differentsurface linkages (e.g., Neu-5-Aca2-6Galpl-

4GlcNAc-R,Neu-5-Aca2-3Galpl-(3)4GlcNAc-R, Neu-5-Aca2-3Galpl-3GalNAc-Thr/Ser,).These preparations were used to test the spec-ificity of sialo-oligosaccharide binding pro-teins such as influenza virus hemagglutinins bymeasuring its ability to agglutinate the resialy-lated cells.

3.5 Solid Phase Synthesis

Solid-phase synthesis is a method well es-tablished in peptide and nucleic acid chemis-

try, and there is a considerable interest in ap-plying it to carbohydrate chemistry. Advantag-es are the easeof purification and the potentialfor construction of combinatorial libraries.

However, development of solid-phase methodshas been slow in this area, because of the inher-ent difficulties of regio- and stereo-control inthe chemical formation of glycosidic linkages.

The usefulness of solid-phase synthesis inenzymatic glycolipid synthesis, where one ofthe major problems is the insolubility of theacceptor lipid in aqueous media, has alreadybeen discussed (ZEHAVI t al., 1990; NISHIMU-RA and YAMADA,997). Several examples ofoligosaccharide and glycopeptide synthesis us-ing galactosyl- and sialyltransferases havebeen reported since 1994,such as the synthesisof a sialyl Lewis x glycopeptide (SCHUSTERtal., 1994). One of the practical problems en-zyme-catalyzed reactions on solid support hasbeen the choice of enzyme-compatible poly-meric support, which has to render support-bound substrates available to the enzyme inaqueous media. The commonly used polysty-rene resins do not swell in water and are,therefore, generally unsuitable. Among theresins used successfully are controlled pore

glass (HALCOMBt al., 1994),polyethylene gly-col polyacrylamide co-polymers (MELDAL tal., 1994), polyacrylamide (KOEPPER, 994),and sepharose (BLIXT nd NORBERG,997).Soluble polyacrylamide supports can also beused in combination with glycosyltransferases(NISHIMURAt al., 1994; YAMADAnd NISHI-MUM, 1995; WIEMANNt al., 1994). Variouslinkers have been used for oligosaccharides onsolid support which can be cleaved with chy-motrypsin (YAMADA nd NISHIMURA,995;SCHUSTER t al., 1994), by hydrazinolysis

(HALCOMBt al., 1994),by hydrogenation (NI-SHIMURA et al., 1994), by photolysis (WIEMANNet al., 1994; KOEPPER,994), and by reductivecleavage of disulfide bonds (BLIXT nd NOR-BERG, 997).

3.6 Acceptor-Donor Analogs

Glycosyltransferases can show remarkablespecificity for both acceptor and donor sub-strates, often beyond even the terminal non-

reducing saccharide of the acceptor. For exam-




ple, an a-2,3-sialyltransferase isolated from

porcine submaxillary gland was highly specific

for acceptor substrates terminating with the

Galpl3GalNAc sequence found in 0-linked

oligosacchandes,but did not glycosylate thoseterminating in Galpl4Glc or Galpl4GlcNAc,

found in N-linked chains (REARICK t al.,

1979). Such findings have led to the general be-

lief that glycosyltransferases are highly specific

and cannot be used for the synthesis of analogs.

However, when unnatural, chemically synthe-

sized donor and acceptor analogs were further

investigated, a number of unnatural glycosidic

linkages could be formed on a milligram scale

although the relative rates of reaction were of-ten much reduced (WONG t al.,1995).

The best studied system is the p-1,4-galacto-

syltransferase from bovine colostrum (Fig.20).Instead of UDP-Gal as a donor substrate,

UDP-Glc, UDP-GalNAc, UDP-GlcNH,, and

OH OH

HoboOUDP + no* Muc RJ&oHo&o Ho

NHAC R

galactosyl-transferase

Ho

G a l - l p so R G a l - l p sO OH R

hBH

6z

Gal-1%

6Gal-1%o

*Ma0 OH R

LQwu

Gal-lp- HO R

1lOH

Gal-1%+&R

o

OH

u

Fig. 20



266 5 Enz yme s in Carbohydrate C hemistry: Formation of Glycosidic Linkages

all the 2,3,4,6-deoxy- and 6-deoxy-6-fluoro-UDP-Gal analogs were successfully trans-ferred to acceptor substrates (PALCIC ndHINDSGAUL,991; KAJIHARA t al., 1995). A n

even wider range of acceptor substrates hasbeen used to generate a number of non-natu-ral disaccharides such as (67-75) in Fig. 20.Substitutions in the 3- and 6-positions of theacceptor substrate ap pear to b e well tolerated,whereas any modifications in the 4-position,not surprisingly, eads to loss of activity. The 2-position appe ars also to be sensitive to change,such that a 2-hydroxyl group as in glucose canbe tolerated (more so in the presence of a co-enzyme, lactalbumin). Glucal is also a substra-te leading to (74), but mannose (the C-2-epi-me r) is not a sub strate.The enzym e is also verysensitive to negative charge, glucuronic acidand glucose-1-phosphates are not accepted(WO NG t al., 1995). When using acce ptor an-alogs with lower binding constants to the en-zyme the high regioselectivity of the enzymemight potentially be lost and it is, therefore,importan t to check authen tity of the glycosidiclinkage.This has mainly bee n d one using NMRmethods, in particular N O E effects, and look -ing at deshielding effects of ring protons

(WO NG t al., 1991; NISHID A t al., 1993a, b).A n interesting case of changed regioselectivityof the galactosyltransferase has been repo rtedby THIEMnd colleagues (NISHIDAt al., 1993a,b) for acce ptor substrates containing 3-N-ace-tyl substituents such as N-acetyl5-thio-gentos-amine which led to the formation of 1,l-O-linked disaccharide (75). This work has sug-gested that perhaps the N-acetyl group of th eacceptors N-acetyl glucosamine and N-acetyl5-thio-gentosamine is a key sub stituent in de-termining regioselectivity.

Various deoxy-U DP-GlcNA c analogs havebeen studied as substrate donors for N-acetylglucosaminyltransferasesI and I1 (GnT-I andGnT-11) from human milk. Whereas all wereaccepted by GnT-I, deoxygenation of HO-3completely abolished ability to act as a don or

Analog studies for sialyltransferases havebeen limited to 9-substituted CMP-Neu-5-Ac(ITO et al., 1993) and to various disaccharideacceptor analogs where the 2-NH Ac group hasbeen replaced by a number of substituentssuch as azido, 0-piv aloyl and other N -acyl sub-

(SRIVASTAVA et al., 1990).

stituents (BAISCH t al., 1996a, b). S ialyltrans-ferase have also been used for the synthesis ofLewis x analogs (ICHIKAWAt al., 1992) biva-lent (DEFR EES t al., 1995) and clustered sialo-

sides (SABESA Nt al., 1991).The substrate specificity of the Lewis x and

a fucosy ltransferases ( 4 3 nd a- l ,4 ) whichtransfers fucose to the internal GlcNAc resi-due in Ga lpl3/4G lcN Ac seems to be particu-larly broad (Fig. 21).Apa rt fr om deoxygenatedacceptor substrates (GOSSELIN nd PALCIC,1996), they can be either sialylated (WON G tal., 1995) or sulfated (AU GE t al., 1997).With-in the d ono r substrate, a large substituent canbe tolerated at the C-6 position, which hasbeen used to transfer carbon-backbone elon-gated G DP-fucose derivatives and e ven oligo-saccharides linked via a spacer to the C-6 posi-tion of GDP -fucose in one step to app ropriateacceptor sub strates (VO GEL t al., 1997).

3.7 Whole-Cell Glycosylation

As an alternative approach to in v i m syn-thesis of oligosaccharides and glycoconjugatesis the use of whole-cell systems. In particular in

the are a of glycoprotein synthesis, the a p-proach of “glycosylation engineering” by het-erologous expression of specific glycosyltrans-ferases such that the glycosylation pattern ofthe glycoproteins is changed, is an area of in-tense interest and has been reviewed else-where (STANLEY,992; VERBER T,996; BILLan d FLITSCH, 996).

In addition, recombinant whole cells, whichhad originally been developed as expressionsystems for glycosyltransferases, have beenused directly as catalysts for the synthesis of

oligosaccharides (HERRMANNt al., 1994;HERRMANNt al., 1995a). Th e advantag e of

such an approach is that the glycosyltransfer-ases do not need to be isolated, although pur-ification of the product from the crude cellmixtures is more difficult. Thus, E. coli XL1-Blue cells, harboring a plasmid expressing ana-1,2-mannosyltransferaseexpressed at a levelof about 1 Unit p er liter, were incubated w ithGDP-mannose and various mannose accep-tor sub strates (mannose, a-methyl-mannoside,Cbz-aManThr-OMe or Boc-Tyr-aManThr-Val-OMe) to give the corresponding a-1-2-




OR

NHAC

Y

OR

Fucosyltransf eraseoHO NHAC

HO

Ho

Alternative Acceptor Substrates: Alternative Donor Substrates:

" HO OH OR

Ho

ORHO

Ho OH

Y c F o ~ D pH

"& RY) Ho NHAc OR

R' = H, N w ~ A C U Fig. 21

mannosides in 42-75% isolated yields (HERR-MANN et al., 1994). Similarly, a hum an P-1,Cga-lactosyltransferase expressed in yeast was usedto synthesize N-acetyl lactosamine by incuba-tion with UDP-Gal and N-acetyl glucosamine

(HERRMANNt al., 1995a).

3.8 Glycosyltransferases Using

Non-N ucleotide Don or Substrates

A number of glycosyltransferases involvedin the biosynthetic formation of glycosidiclinkages do not use sugar nucleotides as shownin Fig. 14 as donor substrates, but more simpleglycosyl phospha tes or glycosides. In contrastto the previously discussed transferases, whichare sometimes called "Leloir glycosyltrans-

ferases" these "Non-Leloir glycosyltransferas-es" are involved both in oligosaccharide andpolysaccharide synthesis to catalyze mostly re-versible reactions.In v im hey have been usedas biocatalysts for the synthesis of sucrose

from glucose-1-phosphate and fructose usingsucrose phosphorylase, of trehalose from glu-cose an d glucose-1-phosphate, and for variouspolysaccharides (WONG t al., 1995).

Another important group of targets are cy-clodextrins, which have bee n synthesized enzy-matically using cyclodextrin glycosyltransfer-ases from bacteria (BORNAGHIt al., 1996).Al-

though polysaccharides such as starch and re-lated malto-oligosaccharides are the naturalglycosyl donor substrates, it has been shownthat a-maltosyl fluoride (Glc al-4 G lcal- F) canbe incorporated into cyclodextrins. The sub-



268 5 Enzym es in Carbohydrate Chemistry:Formation ofGlycosidic Linkages

strate specificity of this enzyme system hasbeen investigated, with the aim to synthesizenew deriva tives of cyclodextrins for supram o-lecular studies. It w as foun d that modifications

at C-6 of th e a-maltosyl fluoride were tolerat-ed by the system and that cyclothiomaltinscould be prepared by using 4-thio-a-maltosylfluoride as the activated disaccharide.

4 Combination of

Glycosidases and

GlycosyltransferasesGlycosyltransferases and glycosidases can

of cou rse be u sed in consecutive glycosylationreactions. This is particularly advantageouswhen a highly acceptor-specific transfe rase isused subsequently to a glycosidase-catalyzedstep, which produces regioisomers. Thus,NILSSONas repo rted the syn thesis of sialylat-ed trisaccharides by firstly using a p-galactos-idase from E. coli to obtain a mixture of pl -3 -and pl-4-linked Gal-GlcNAc-OMe disaccha-

rides. Since the p-D-galactoside 3-a-sialyl-transferase from porcine submaxillary glandshas a higher selectivity for Galpl3GlcNAcsubsequent sialylation of the product mixtureyielded selectively Neu-5-Aca2-3 Galpl-3-GlcNAc-OMe (NILSSON,989).

A further avantage of using a combinationof both enzymes is that the glycosidase-cata-lyzed reac tion is reversible, but the transferasereaction is irreversible, such that yields can beimproved by using one-pot synthesis (KRENand THIEM , 995;HERRMANNt al., 1993).

5 Practical Aspects

of Synthesis and Analysis

5.1 Isolation and Purification

of Enzymes for Biocatalysis

Many of the glycosidases discussed in thischapter are commercially available and inex-

pensive and particularly the thermophilic en-zyme libraries might provide a wide range ofuseful glycosidation biocatalysts in the future(LI and W ANG, 997).The number of glycosyl-

transferases available is still very limited andthese enzymes need to b e purified before use,but a numb er of reliable purification protocolsfrom mammalian tissue have been published(PALCIC, 994; SADLE R t al., 1982 ). Affinitychromatography using immobilized nucle-osides such as CDP-agarose for sialyltrans-ferases has been successful (PAULSONt al.,1977). The am oun t of transferases that can be

isolated from natural sources variies greatly. Aa-2,6-sialyltransferase from porcine liver con-tains 40 Un its per kg (AU GE et al., 1990)whereas 250 mL of human m ilk yielded about20 milliunits of an a-1,3/4-fucosyltransferase(PALCIC,994).

W here glycosyltransferases are usually onlyfound in very low abund ance in natural sourc-es, considerable effort has been spent on de-veloping efficient overexpression systems. So

far, a lot of the mammalian glycosyltransfer-ases have required mammalian expressionsystems, such as those reported for sialyltrans-ferases (GILLESPIEt al., 1992), fucosyltrans-

ferases (DE VR IES t al., 1995; DE VR IES t al.,

and GlcNAc-transferases (D’AGOSTAROt al.,1995). M ore efficient and h igher yieldingexpression was achieved with insect cells(BAISCHt al., 1996a,b) and yeast (MALISSARDet al., 1996;HERRMANNt a l., 1995b;BORSIG tal., 1995).The yeast system provided 200 Unitsof the p-1,4-galactosyltransferase rom a 290Lferm enta tion and 47 Un its of a-2,6-sialyltrans-ferase from a 150 L bioreactor.

E. coli expression at a level of about 1Unit

per liter in shake-flask cultures has beenachieved for two mannosyltransferases a-1,2-(WANG t al., 1993) and p-1,4- (WATT et al.,1997). The p-1,4-mannosyltransferase yieldedmuch better conversion, when an N-terminalhydrophobic peptide sequence was deleted,resulting in a mo re soluble enzyme.

1993;WONG et a]., 1992;BAISCH e t al., 1996a,b)

5.2 Biocatalytic Conversions

Biocatalytic conversions with glycosidasesand transferases have been reported for im-



270 5 Enz yme s in Carbohydrate Chemistry: Formation of Glycosidic Linkages

bon-celite chromatography (SINGHet al.,1995) has bee n useful as a purification step forsuch crude m ixtures.

Since glycosyltransferase-catalyzed reac-

tions tend t o be better yielding and highly se-lective, sepa ration of isomeric mixtures of olig-osaccharides is not required and the purifica-tion from enzymes, buffer, salts, and startingmaterial is often straightforward. HPLC separ-ation, gel filtration, ion exchange chromato-graphy or gel permeation chromatography(YAMASHITAt al., 1982) are the most widelyused techniques for oligosaccharide isolation,in particular on a m illigram scale.

6 Conclusions

Both natural and unnatural oligosaccha-rides and glycoconjugates are accessible bysynthetic enzymatic methods using either gly-cosidases o r glycosyltransferases.Both enzymesystems are highly stereospecific and oftensubstrate-specific, such that each particularglycosidic linkage req uires its own biocatalyst.

So far, there is no known general biocatalystfor fo rma tion of glycosidic linkages and , giventhe high degree of stereo- and regioselectivityrequired, such a catalyst would have limiteduse, given that oligosaccharide biosynthesisdoes not involve a tem plate as is available foroligonucleotide and polypeptide biosynthesis.A great deal of current research is, therefore,focused on finding glycosylation catalysts withnew stereo-, regio-, and substrate selectivities,and on then investigating their substrate spec-ificity, such tha t eventua lly perhaps a com pleteset of catalysts for each glycosidic linkage willbecome available. It should be kept in mindthat natura l sources might provide only limit-

7 References

AJISAKA,., FUJIMOTO,. (1989), Regioselective

synthesis of trisaccharides by use of a reversedhydrolysis activity of a- an d P-galactosidase, Car-bohydr. Res. 185,139-146.

AJISAKA, ., FUJIMOTO,. et al. (1988), Enzymicsynthesis of disaccharides by use of th e reversedhydrolysis activity of P-galactosidases, Carbo-hydr. Res. 180 ,3542 .

AJISAKA,., MATSUO,. et al. (1995), Enzym atic syn-thesis of mannobiosides and mannotriosides byreverse hydrolysis using a-m ann osid ase from As-pergillus niger, Carbohydr. Res. 270,123-130.

AUGE, ., DAVID,. et al. (1984), Synthesis with im-mobilized enzymes of two trisaccharides, one of

them active as the dete rminant of a stage antigen,Tetrahedron Lett. 25,1467-1470.AWE,C., MATHIEU,. et al. (1986), The use of an

immobilized cyclic multi-enzyme system to syn-thesize branched pen ta- und hexa-saccharides as-sociated with blood-group I epitopes, Carbohydr.Res. 151,147-156.

AWE,C., FERNANDEZ-FERNANDEZ,. et al. (1990),The use of immobilized glycosyltransferases inthe synthesis of sialo-oligosaccharides, Carbo-hydr. Res. 200,257-268.

AUGE, ., DAGRON,. et al. (1997), Syn thesis of sul-fated derivatives as sialyl Lewis a an d sialyl Lew-

is x analogs, in: Carbohydrate Mimics: Conceptsand Methods (CHAPLEUR,., Ed.), pp. 365-383.Weinheim: WILE Y-VC H Verlag.

BAISCH,.,OHRLEIN,. e t al. (1996a), Enzym atic fu-cosylation of non-natural trisaccharides withcloned fucosyltransferase VI. Bioorg. Med. Chem.

Lett. 6,759-762.BAISCH, ., OHRLEIN,. et al. (1996b), Enzymatic

a(2-3)-sialylation of no n-na tural oligosaccharideswith cloned sialyltransferase, Bioorg. Med. Chem.Lett. 6,755-758.

BILL, . M., FLITSCH,. L. (1996), Chem ical and bio-logical approaches to glycoprotein synthesis,

Chem. Biol. 3,145-149.BJORKLING,., GODTFREDSEN,. E. (1988), New en-zyme catalyzed synthesis of monoacyl galactogly-cerides. Tetrahedron 44.2957-2962.-

ed activities, since of th e thousands of possible BLIXT,O.,NORBERG,T.1997), Solid-phase enzy mat-oligosaccharide isomers (if o ne assumes every ic synthesis of a Lewis a trisaccharide using an ac-possible permutation of stereo- and regiose- ceptor reversibly bound to sepharose, J. Carbo-

lectivity) only a sm all subset actually occur in hYdr. Chem. 16, l4SM4.

natural systems.A very interesting area of re - BOONS,G-J. (1996), Strategies in oligosaccharidesynthesis, Tetrahedron 52,1095-1121.

catalysts using either rational site-specific synthesis of cyc~othiomaltins, them. sot.,(SETO et al., 1997) or random mutagenesis

techniques combined with directed evolution BORSIG,., IVANOV, S. x. t al. (1995), Scaled-up ex-(MOOREnd ARNOLD,996).

search therefore? he generation Of new bio-BORNAGHI,., UTILLE,,-p.et al. (1996), Enzymic

Chem. commun.,541-2542.

pression of human (~-2,6(iV)sialyltransferase n



7 References 271

Saccharomyces cerevisiae, Biochim. Biophys. Res.Commun. 210,14-20.

BOUHOU RS, ., BOUHOU RS,.-F. (1991), Geneticpolymorphism of rat liver gangliosides, J . Biol.Chem.

266 ,12 944-12 948.BROCKHAUSEN,., CARVER,. P. et al. (1988),Controlof glycoprotein biosynthesis. The use of o ligosac-charide substrates and HPLC to study the se-quential pathw ay for N -acetyl glucosaminyltrans-ferases I, 11,111,IV,V, and V1 in the biosynthesisof highly branched N-glycans by hen oviductmembranes, Biochem. Cell Biol. 66,1134-1151.

BROCKHAUSEN,., MOLLER, . et al. (1992), Con-trol of glycoprotein-synthesis- characterizationof (~-]4)-N-acetyl-P-~-glucosaminyltransferases

acting on the a-D-(1-]3) 0-linked and a-D-(1-]6)-linked arms of N-linked oligosaccharides,Carbo-

hydr. Res. 236,281-299.BROWN, . T., CART ER, . E. et al. (1995), Synthesisof morphine-5-glucuronide via a highly selectiveenzyme-catalyzed hydrolysis reaction, Tetrahe-dron Lett. 36,1117-1120.

CANTACUZENE,.,ATTAL, S. et al. (1991),Stereospe-cific chem oenzy matic synthesis of galactopyrano-syl-L-serine,Bioorg. Med. Chem. Lett. 1,197-200.

CLARK, . S. ,BANEWEE,. et al. (1990), Yeast oligo-saccharyltransferase:glycosylation of peptide sub-strates and chemical characterization of th e glyco-peptide product,J. Org. Chem.55,6275-6285.

CLAUSEN, ., B E N N E ~ ,. F? (1996), A family of

UDP-GalNAc :polypeptide N -acetylgalactosami-nyl-transferases control the initiation of mucin-type 0-linked glycosylation, Glycobiology 6,635-646.

CROUT, . H. G., MACMANUS,. A. (1990), Enzy-matic synthesis of glycosides using the P-galactos-idase of Escherichia coli: regio- and stereo-chem-ical studies, J . Chem. Soc., Perkin Trans. Z,

D’AGOSTARO,. A. F., ZIN GO NI,A .t al. (1995),Mo-lecular cloning and expression of cDN A enco dingthe rat UD P-N-acetyl glucosaminyltransferase11,J . Biol. Chem.270,15211-15221.

DEFREES,. A., KOSCH,W. et al. (1995),Ligand rec-ognition by E-selectin: synthesis, inhibitory activ-ity, and conform ational analysis of bivalent sialylLewis x analogs, . Am . Chem.SOC.117 ,6679 .

DEVRIES,T.,ORBERG,T.,O”, H.,VAN DE N EYNDEN,D. H. (1993),The use of human milk fucosyltrans-ferase in the synthesis of tum or-associated rimer-ic X determ inants Eur.J . Biochem. 216,769-777.

DEVRIES,T.,RNK A, . et al. (1995),Acceptor speci-ficity of different length constructs of human re-combinant a-l,3/4-fucosyltransferases eplace-men t of the stem region and the transmemb ranedomain of fucosyltransferase-V by p rotein-A re-

sults in an enzyme with G DP-fucosyl hydrolysingactivity,J. Biol. Chem. 270,8712-8722.

1865-1868.

FAN, J.-Q., TAKEGAWA,. et al. (1995), Enhancedtransglyco sylation activity of Arthrobacter p roto -phormiae endo-P-N-acetyl glucosaminidase inmedia containing organic solvents, . Biol. Chem.

FEIZ I, T. (1993), Oligosaccharides that mediatemamm alian cell-cell adhesion, Curr. Opin. Struct.Biol. 3,701-710.

FIELD,M. C., WAINWRIGHT,. J. (1995), Molecularcloning of eukaryo tic glycoprotein and glycolipidglycosyltransferases: a survey, Glycobiology 5 ,463-472.

FLITSCH,. L., WA n, G. M. (1997), Chemical Synthe-sis of Glycoprotein Glycans. Glycopeptides andRelated Compounds (LARGE,D. C., WA RRE N, .D., Eds.), pp.207-244. New Y ork: Marcel Dek ker.

GAIS,H.-J., ZEISSLER,. et al. (1988), Diastereose-

lective D-galactopyranosyl transfer to m eso-diolscatalyzed by P-galactosidases, Tetrahedron Lett.

GELO -PUJIC, ., GUIBE-JAMPEL,. et al. (1997),En-zymatic glycosidation in dry media under micro-wave irradiation, J . Chem. SOC.,Perkin Trans. I ,1001-1002.

GIJSEN , . J. M., QIAO,L. et al. (1996), Recent ad-vances in the chemo-enzym atic ynthesis of carbo-hydrates and carbohydrate mimetics,Chem. Rev.%,443-473.

GILLESPIE,., KEL M, . et al.(1992),Cloning and ex-pression of the Galbl-3G alNA c a-2,3-sialyltrans-

ferase,J. Biol. Chem.267,2100&21010.GOSSELIN,., PALCIC,M. M. (1996), Acceptor hy-droxyl group mapping for human milk a l - 3 andal-3 /4 fucosyltransferases,Bioorg. Med. Chem .4,2023-2028.

GUILBERT,., FLITSCH,. L. (1994), A short chemo-enzymic route to g lycosphingolipids using solubleglycosyltransferases, . Chem. Soc., Perkin Trans.I , 1181.

GUILBERT,., KHA N,T.H. et al. (1992),Chemo-enzy-matic synthesis of a glycosphingolipid,J. Chem.SOC.,Chem. Commun., 15261527.

GY GAX , ., SPIES,? et al. (1991),Enzym atic synthe-

sis of P-D-glucuron ides with in situ regenerationof uridine-5 ‘-diphos phog lucuron ic acid, Tetrahe-dron 47,5119-5123.

HAKOMORI,.-I. (1990), Bifunctional role of glycos-phingolipids,J . Biol. Chem.265,18713-18716.

HALCOMB,. L., HUANG, . et al. (1994), Solution-and solid-phase synthesisof inhibitors of H.pylo-ri attachm ent and E -selectin-mediated leukocyteadhesion,J. Am . Chem. SOC.116,11315-11 322.

HEIDELBERG,T.,HIEM,. (1997), Synthesis of novelamino acid glycoside conjugates,Carbohydr. Res.

HERRMANN,., ICHIKAWA,. et al. (1993), A new

multi-enzyme system for a one-po t synthesis ofsialyl oligosaccharides:combined use of P-galac-

270 ,17 723-17 729.

29,5743-5744.

301,145-153.



272 5 Enzym es in Carbohydrate Chem istry: Formation of Glycosidic Linkages

tosidase and a(2,6)sialyltransferase coupled withregeneration in sifu of CMP-sialyc acid, Tetrahe-dron Lett. 34,3091-3094.

HERRMANN,., WANG,!et al. (1994), Recom binant

whole cells as catalysts for the enzymatic synthe-sis of oligosaccharides and glycopeptides,Angew.Chem. (Int. E dn. E ngl.) 33,1241-1242.

HERRMANN,. F., ELL ING , . e t al. (1995a), Use oftransformed whole yeast cells expressing p-1,4-galactosyltransferase for the synthesis of N -acetyllactosamine,Bioorg. Med. C hem. Lett. 5,673476.

HERRMAN N, . F., KREZDORN,. et al. (1995b),Large-scale production of a soluble human p-1,4-galactosyltransferase using a Saccharomyces ce-revisiae expression system,Protein Expr. Purijic.

ICHIKAWA,., LIN ,Y.-C. et al. (1992), Chemical-enzy-

matic synthesis and conformational analysis ofsialyl Lewis-x and derivatives, . Am . Chem. SOC.

ICHIKAWA,., WANG, . et al. (1994), Regenerationof sugar nucleotide for enzym atic oligosaccharidesynthesis,Methods Enzym ol. 247,107-127.

IMPERIALI,. (1997), Protein glycosylation: he clashof the titans,Acc . Chem. Res.30,452459.

IMPERIALI,., HENRICKSON,T.. (1995),Asparagine-linked glycosylation: specificity and function of

oligosaccharyltransferase,Bioorg. Med. Chem.Lett. 3,1565-1578.

ITO, ., GAU DINO ,. et al. (1993), Syn thesis of bioac-

tive sialosides,Pure Appl. Chem. 65,753-762.KAJIHARA,., ENDO ,T. t al. (1995), Enzym ic trans-fer of 6 -mo dified D-galaCtOSyl residues : syn thes isof biantennary penta- and heptasaccharides hav-ing two 6-deoxy-~-galactose esidues at th e non-reducing end and evaluation of 6 -deo xy-~ -galac -tosyl transfer to glycoprotein using bovine p-(1,4)-galactosyltransferase and UDP-6-deoxy-~-galactose,Carbohydr. Res. 269,273-294.

KARLSSON,.-A. (1989), Anim al glycosphingolipidsas memb rane attachm ent sites for bacteria,Annu.Rev. Biochem. 58,309-350.

KAUR,K. J., ALTON , . et al. (1991), Use of N-acetyl

glucosaminyltransferasesI and I1 in the pre para-tive synthesis of oligosaccha rides,Carbohydr. Res.210,145-153.

KOBAYASHI,., KA SHIW A, . et al. (1991), Novelmethod for po lysaccharide synthesis using an en-zyme: the first in vitro synthesis of cellulose via anonbiosynthetic path utilizing cellulase as cata-lyst,J.Am. Chem.SOC. l3,3079-3084.

KOBAYASHI,. , KIYOSADA,. et al. (1997), A novelmethod for synthesis of chitob iose via enzym aticglycosylation using a sugar oxazoline as glycosyldonor, Tetrahedron Lett. 38,2111-2112.

KOEPPER,. (1994),Polymer-supported enzymic syn-

thesis on a preparative scale,Carbohydr. Rex 265,

6,72-78.

114,9283-9290.

161-166.

KORNFELD,., KORNFELD,. (1985), Assembly of

asparagine-linked oligosaccharides, Ann. Rev.Biochem. 54,631464.

KRAGL, ., GYM , D. et al. (1991), Enzymatic two-step syn thesis of N-acetyl neuram inic acid in theenzyme membrane reactor, Angew. Chem. (Int.Edn. E ngl.) 30,827-828.

KREN,V.,THIEM,. (1995),A m ultienzyme system forone-pot s ynthesis of sialyl T-antigen, Angew.

Chem. (Int. Edn. E ngl.) 34,893-895.KREN,V.,THIEM,. (1997), Simple and non-radioac-

tive method for determination of sialyltransfer-ase activity,Biotech nol. Tech. 11,323-326.

LASKY,. A . (1994), A sweet success,Nature Struct.Biol. 1,139-141.

LEE,J., COWARD,. K. (1993), Oligosaccharyltrans-ferase: synthesis and use of deuterium-labeled

peptide substrates as mimetic probes, Biochemis-try 32,6794-6801.LI, J., WANG,P.G. (1997), Chemical and enzymatic

synthes is of glycoconjugates. 2. High yielding re-gioselective synthesis of N-acetyl lactosam ine byuse of recom binant therm ophilic glycosidase li-brary, Tetrahedron Lett. 38,7967-7970.

LIU,Y.-L., HOOPS,G. C. et al. (1994), A com parisonof proteins and peptides as substrates for micro-soma1 and solubilized oligosaccharyltransferase,

Bioorg. Med. Chem .2,1133-1141.MALISSARD,., BORSIG, . et al. (1996), Recombi-

nant soluble ~-1,4-galactosyltransferases x-

pressed in Saccharomyces cerevisiae - purifica-tion, characterization and comparison with hu-man enzyme,Eur. J . Biochem. 239,340-348.

MELDAL,M., AUZANNEAU,-I. et al. (1994), A P E-G A resin fo r use in the solid-phase chemical-en-zym atic synthesis of glycopeptides, . Chem. SOC.,

Chem. Commun., 1849-1850.MELKERSON-WATSON,. J., SWEELEY,. C. (1991),

Purification to a pparen t homogen eity by immu-noaffinity chrom atography and partial character-ization of the GM 3 ganglioside-forming enzyme,CMP-sialic acid : actosylceramide a2,3-sialyl-transferase (SA T-l), from rat liver Golgi,J. Biol.

Chem. 266,4448-4457.MOORE,. C., ARNOLD,. H. (19%), Directed evolu-tion of a para-nitrobenzyl esterase for aqueous-or-ganic solvents, Nature Biotechnology 14,458-467.

NILSSON, . G. I. (198 7),A simp le strategy for chang -ing the regioselectivity of glycosidase-catalyzedformation of disaccharides,Carbohydr. Res. 167,

NILSSON, . G. I. (1988),A simple strategy for ch ang-ing the regioselectivity of glycosidase-catalyzedformation of disaccharides:Part 11. Enzymic syn-thesis in situ of various acceptor glycosides,Car-bohydr. Res. 180,53-59.

NILSSON, . G. I. (1989), Enzy mic synthesis of di- andtrisaccharide glycosides, using glycosidases and p-

95-103.



7 References 273

D-galactoside 3-a-sialyltransferase, Carbohydr.Res. 188,9-17.

NILSSON , . G. I. (1997), G lycosidase-catalyzessyn-thesis of di- and trisaccharide derivatives related

to antigens involved in the hyperacute rejectionof xenotransplants, Tetrahedron Lett. 38,133-136.NILSSON,. G . I., LJUN GER , . e t al. (1997), Glycosi-

dase-catalyzed synthesis of glycosylated aminoacids: Syn thesis of G alNA c a-Ser and GlcNA cp-

Ser derivatives,Biotechn. Lett. 19,8894392.NISHIDA,., WIEMAN N,. et al. (1993a),A new type

of galactosyltransferase reaction: transfer of ga-lactose to the anomeric position of N -acetyllac-tosamine,J . Am . Chem.SOC.115,2536-2537.

N I S H I D A , ~ . ,IEMANN,T.t al. (1993b), Extension ofthe p-Gall,l,-transfer to N-acetyl-5-thio-gentosa-mine by galactosyltransferase, Tetrahedron Lett.

NISHIMURA,.-I.,MATSUOKA,. et al. (1994), Chem o-enzymatic oligosaccharide synthesis on a solublepolymeric cam er, Tetrahedron Lett. 35,5657-5660.

NISHIMURA,. , YAMADA,. (1997), Trans fer of gan-glioside GM 3 oligosaccharide from a w ater solu-ble polymer to ceramide by ceramide glycanase.A novel approach for the chemical-enzymaticsynthesis of glycosphingo lipids, . Am . Chem.SOC.

OOI ,~ . ,ASHIMOTO,T.t al. (1984), Enzym ic synthe-sis of chemically unstable cardiac glycosides by p-galactosidase from Aspergillus oryzae, Tetrahe-

dron Lett. 25,2241-2244.PALCIC,M. M. (1994), Glycosyltransferases in gly-cobiology,Methods Enzym ol. 230,300-360.

PALCIC, . M., HEERZE, .D. et al. (198 8),The use ofhydrophobic synthetic glycosides as acceptors inglycosyltransferase assays, Glycoconjugate J . 5,4 9 4 3 .

PALCIC, . M., HINDSGAU L,. (1991), Flexibility inthe do nor s ubs trate specificity of pl,4-galactosyl-transferase: application in the synthesis of com-plex carbohydrates, Glycobiology 1,205-209.

PAREK H, . B., EDGE ,C. J. (1994), Selectins- glyco-protein targets for therapeutic intervention in in-

flammation, TZBTECH 12,339-345.PAULSEN,. (1982), Advances in selective chemical

syntheses of complex oligosaccharides,Angew.Chem. (Int. Ed n. Eng l.) 21,155-173.

PAULSEN,. C., ROD GE RS, . N. (1987), Resialylatederythrocytes fo r assessment of the specificity ofsialooligosaccharide binding proteins, MethodsEnzymol. 138,162-168.

PAULSON,.C.,BERANEK,. E. et al. (1977),Purifica-tion of a sialyltransferase from bovine colostrumby affinity chromatography on CDP-agarose, J .

Biol. Chem. 252,2356-2362.PREUSS,U., Gu, X. et al. (1993), Purification and

characterization of CMP-N-acetylneuraminicacid : actosylceramide (a2-3) sialyltransferase

34,2905-2906.

119,10555-10556.

(GM3-synthase) from rat brain, J . Biol. Chem.

REARICK,. I.,SADLER,. E. et al. (1979), Enzym aticcharacteriza tion of p-~ -g al ac to si de u2-3 sialyl-

transferase from porcine submaxillary gland, J.Biol. Chem.254,44444l51.SABESAN,., DUUS,. et al. (1991), Syn thesis of clus-

ter sialoside inhibitors for influenza virus, J . A m .Chem. SOC.113,5865-5866.

SADLER,. E., BEYER ,. A. et al. (1982), Purificationof mammalian glycosyltransferases,Methods En-zymol. 83,458-514.

SAUEREIREI,.,THIEM,. (1992), Galactosylation andglucos ylation by use of P-galac tosidase, Tetrahe-dron Lett. 33,201-204.

SCHACHTER,. (1991), Enzy mes assoc iated with gly-cosylation,C um Opin. Struct. Biol. 1,755-765.

SCHMIDT,. R. (1991), Synthesis of glycosides. Com -prehensive organic synthesis, 6,3 3 43 .SCHMIDT,. R., KINZY,W. (1994),Anomeric-oxygen

activation for glycoside synthesis. The trichloro-acetimidate method, Adv. Carbohydr. Chem.Bio-chem. 50,21-123.

SCHUSTER, ., WANG,P. et al. (1994), Solid-phasechemical-enzymatic synthesis of glycopeptidesand oligosaccharides, J . Am. Chem. SOC. 116,

SEITZ,O.,WO NG , .-H. (1997), Chemo enzymaticso-lution- and solid-phase synthes is of O-g lycopep -tides of the mucin domain of MAdCAM-1. A

general route to 0-LacNAc, 0-sialyl-LacNAcand 0-sialyl-Lewis-X peptides,J . Am . Chem. SOC.119,8766-8776.

SETO, . 0.L., PALCIC,M. M. et al. (1997), Sequentialinterchange of four amino acids from bloodgroup B to blood group A glycosyltransferaseboosts catalytic activity and progressively modi-fies substrate recognition in human recombinantenzymes,J . Biol. Chem. 272,14 133-14 138.

SINGH,S., PACKWOOD,. et al. (1995), Glycosidase-catalyzed oligosaccharide synthesis: preparationof N-acetyl chitooligosaccharides using the p-N-acetyl hexosam inidase from Aspergillus oryzae,

Carbohydr. Res. 279,293-305.SINGH, . , SCIGELOVA,. et al. (1996), Glycosidase-catalyzed synthesis of oligosaccharides: a two-step synthesis of the core trisaccharide of N -linked glycoproteins using the p-N-acetyl hexosa-minidase and the Pmannosidase from Aspergil-lus oryzae, Chem. Comm un.,993-994.

SRIVASTAVA,., ALTON,G. e t al. (1990), Com binedchemical-enzymic synthesis of deoxygenatedoligosaccharide analogs: transfer of deoxygenat-e d D - G ~ c ~ N A cesidues from their UDP-GlcpNA c derivatives using N-acetyl glucosamj-nyltransferase I, Carbohydr. Res. 207,259-276.

STANLEY,. (1992), Glycosylation engineering, Gly-cobiology 2,99-107.

268,26 273-26 278.

1135-1136.



274 5 Enzym es in Carbohydrate Chem istry: Formation of Glycosidic Linkages

TURNER,. J.,WEBBERLEY, . C. (1991), Stereospe-cific attachment of carbohydrates to amino acidderivatives using P-glucosidase and P-xylosidase.J. Chem. Soc., Chern. Comm un., 1349-1350.

UNVERZAGT,. (1996), Chemoenzymatic synthesisof a sialylated undecasaccharide-asparagine con-jugate, Angew. Chem. (Int. Edn. Engl.) 35,2350-2353.

JNVERZAGT,., KUNZ,H. et al. (1990), High-effi-ciency synthesis of sialyloligosaccharides andsialoglycopeptides, J . Am. Chem. SOC. 112,9308-9309.

JNVERZAGT,., KELM, . et al. (1994), Chemical andenzymatic synthesis of multivalent sialoglycopep-tides, Carbohydr. Res. 251,285-301.

VERBERT, . (1996), The fascinating challenge tomimic nature in producing recombinant glyco-proteins, Chimica Oggi, 9-14.

VIC, G., THOMAS,. (1992), Enzyme-catalyzed syn-thesis of alkyl p-D-glucosides in organic media,Tetrahedron Lett. 33,45674570.

VIC,G., CROUT,D. H. G. (1994), Synthesis of gluco-sidic derivatives with a spacer arm by reverse hy-drolysis using almond P-D-glucosidase, Tetrahe-dron:Asyrnrnetry 5,2513-2516.

VIC, G., CROUT,D. H. G. (1995), Synthesis of ally1and benzyl P-D-glucopyranosides and ally1 p-D-galactopyranoside from D-glucose or D-galactoseand the corresponding alcohol using almond p-D-glucosidase,Carbohydr. Res. 279,315-319.

VIC,G., HASTINGS,. J. et al. (1996), Glycosidase-cat-alyzed synthesis of glycosides by an improvedprocedure for reverse hydrolysis: Application tothe chemoenzymatic synthesis of galactopyrano-syl-(1-4)-0-a-galactopyranoside erivatives, Tet-rahedr0n:Asymmetry I, 973-1984.

VILADOT,.-L., MOREAU,. et al. (1997), Transglyco-sylation activity of Bacillus 1,3-1,4-p-~-glucan -glucanohydrolases. Enzymic synthesis of alter-nate 1,3-1,4-~-~-gluco-oligosaccharides,. Chem.SOC.,Perkin Trans. I 16,2383-2387.

VOGEL, ., BERGEMANN,. et al. (1997), Synthesis ofcarbon-backbone-elongated GDP-L-fucose de-rivatives as substrates for fucosyltransferase-cata-lyzed reactions,Liebigs An n. Rec. 3,601-612.

WANG, ., SHEN,G.-J. e t al. (1993), Enzymes in oligo-saccharide synthesis: active-domain overproduc-tion, specificity study, and synthetic use of an a-

1,2-mannosyltransferase with regeneration ofGDP-mannose,J. Org. Chem.58,3985-3990.

WANG,L. X., TANG,M. et al. (1997), Combinedchemical and enzymatic synthesis of a C-glyco-peptide and its inhibitory activity towards glycoa-midases,J.Arn. Chern.SOC.119,11137-11 146.

WATT,G. M., REVERS,. et al. (1997), Efficient enzy-matic synthesis of the core trisaccharide of N-gly-cans using a recombinant P-mannosyltransferase.Angew. Chem. (Int. Edn. Engl.) 36,2354-2356.

WIEMANN,T.,TAUBKEN,N.t al. (1994),Enzymicsyn-thesis of N-acetyl lactosamine on a soluble light-sensitive polymer, Carbohydr. Res.251, Cl-C6.

WILLENBROCK,.W., NEVILLE,. C.A. et al. (1991),The use of HPLC-pulsed amperometry for thecharacterization an assay of glycosidases and gly-cosyltransferases,Glycobiology 1,223-227.

WILSON, I. B. H., WEBBERLEY, . C., REVERS, L.,FLITSCH,. L. (1995), Dolichol is not a necessarymoiety in lipid-linked oligosaccharide ubstratesforthe mannosyltransferases nvolved in N-linked olig-osaccharide biosynthesis,Biochern.J.310, W 9 1 6 .

W m , K., SEARS,P. et al. (19!Z'), Enzymatic glycopro-tein synthesis: Preparation of ribonuclease glyco-forms via enzymatic glycopeptidecondensation andglycosylationJ Am. Chem.SOC.119,2114-2118.

WONG,C.-H., HAYNIE,. L. et al. (1982), Enzyme-catalyzed synthesis of N-acetyl lactosamine within situ regeneration of uridine 5 '-diphosphateglucose and uridine 5 -diphosphate galactose, J .Org. Chem. 47,5416-5418.

WONG,C.-H.,ICHIKAWA,Y.t al.(1991),Probing he ac-ceptor specificity of j3-1,4-galactosyltransferase orthe development of enzymatic synthesis of noveloligosaccharides,J. m. Chem.Soc.113,81374145.

WONG, ., DUMAS,. et al. (1992), Specificity inhibi-tion and synthetic utility of a recombinant humanc~-1,3-fucosyl-transferase.. Am . Chem. SOC.114,

WONG, .,HALCOMB,. L. et al. (1995), (Part 2) En-zymes in organic synthesis: Application to theproblems of carbohydrate recognition, Angew.Chern. (Int. Edn. Engl.) 34,521-546.

Wu, J.,TAKAYAMA,. et al. (1997), Quantitative elec-trospray mass spectrometry for the rapid assay ofenzyme inhibitors, Chem. Biol. 4,653-657.

Xu, T., COWARD,. K. (1997), C-13 and N-15-labeledpeptide substratesasmechanistic probes of oligosac-charyltransferase,Biochemistry 36,14683-14689.

YAMADA,., NISHIMURA,.-I. (1995), An efficientsynthesis of sialoglycoconjugateson a peptidase-sensitive polymer support, Tetrahedron Lett. 36,

YAMASHITA,., MIZUOCHI,. et al. (1982), Analysisof oligosaccharides by gel filtration, Methods En-

zymol. 83,105-124.YANG,W.,ABERNETHY,. L. et al. (1992), Purification

and characterization of a UDP-GalNAc-polypep-tide N-acetyl galactosaminyltransferase specificfor the glycosylation of threonine residues,J.Biol.Chem. 267,12709-12716.

ZEHAVI,U., HERCHMAN,. et al. (1990), Enzymicglycosphingolipid synthesis on polymer supports.11.Synthesis of lactosyl ceramide, Glycoconjugate

ZELLER,C. B., MARCHASE,. B. (1992), Ganglio-sides as modulators of cell function, A m . J . Phys-iol. 262, C1341-1355.

7321-7322.

9493-9496.

J. I, 29-234.



Applications



6 Industrial Biotransformations

UWE T. BORNSCHEUERGreifswald, Germ any

1 Introduction 278

2 Enantiomerically Pure Pharmaceutical Intermediates 2781.1 Basic Con siderations 278

2.1 Diltiazem 2792.2 Am ines 2792.3 6-Am inopenicillanic Acid 2802.4 7-Am inocepha losporanic Acid 2802.5 Nap roxen 2812.6 Pyrethroid Precursors 282

2.6.1 Via Oxynitrilases 2822.6.2 Via a Mo nooxy genase 2822.6.3 Via Es tera se 283

2.7 Produc tion of Chiral C, Building Blocks 283

2.8 O the r Pharmaceutical Intermediates 283

3.1 Natural Am ino Acids 2853.2 Non -Natural Am ino Acids 2853.3 Aspartam e 286

4.1 Acrylam ide/Nicotinam ide 2874.2 L-Carnitine 2884.3 Lipid Mo dification 289

3 Am inoAcids 285

4 O the r Applications 287

4.3.1 Coc oa Bu tter Equivalents 2894.3.2 O th er Structured Lipids 289

5 References 290





278 6 Industrial Biotransformations

1 Introduction

In the last two decades, the application of

biocatalysts in industry has considerably in-creased. This is due to a number of reasons, ofwhich the most important ones might be thebetter availability of enzymes in large quan-tities (mainly due to the progress in geneticengineering) and an increasing demand forenantiomerically pure compounds for phar-maceuticals/agrochemicals due to changes inlegislation. Furthermore, organic chemistsmore and more accept biocatalysts as a usefulalternative to chemical catalysts. The recentdevelopment in the field of directed evolution

(STEMMER,1994; ARNOLD, 998; BORN-

SCHEUER, 1998; REETZand JAEGER, 1999)might further increase the availability of suit-able biocatalysts.

1.1 Basic Considerations

The successful application of biotransfor-mations in an industrial environment dependson a number of factors, such as

0 availability (and price) of suitable en-

0 up- and downstream processing,0 competition with (in-house) chemical

methods/established processes,0 time requirements for process develop-

ment,0 wastewater treatment, solvent disposal,0 legislation/approval of the process or

0 public perception.

zymes or microorganisms,

product,

All these factors determine whether a biocata-lytic route is superior to a chemical one andthe final decision has to be made case-by-case.ROZZELL1999) pointed out five commonlyheld ideas (myths) about enzymes (too expen-sive, too unstable, productivity is low, redoxfactors cannot be recycled, they do not cata-lyze industrially interesting reactions). Indeed,the examples shown in his publication as wellas those shown here, clearly demonstrate thatthese myths are rather misconceptions.

In the following survey, some examples forthe industrial use of a wide range of biocata-lysts are given. They are organized by productnot by enzyme or microorganism used, be-

cause often several biocatalytic routes havebeen developed for the same target. Most ex-amples are taken from the recent literature.For further industrialized processes readersare referred to a number of excellent books(COLLINSt al., 1992,1997;SHELDON,993).

It should be pointed out that it is very diffi-cult to provide a complete overview, becauseinformation about commercialized processesis hard to get. Moreover, even if details are re-vealed in the (patent) literature, processesmight have changed in the meantime.

2 Enantiomerically Pure

Pharmaceutical

Intermediates

The worldwide sales of single-enantiomerdrugs are increasing dramatically, e.g., from1996 to 1997 by 21% to almost 90 billionUS$.Of the top 100 drugs,50 are marketed as singleenantiomers (STINSON, 998). As a conse-quence, the need for enantioselective technol-ogies is also increasing, and companies as wellas academia are making considerable effortsto develop efficient synthetic routes includingbiocatalysis and biotransformations.A considerable proportion of successful ex-

amples for the enzymatic synthesis of opticallypure intermediates is based on the action of li-pases. This is not surprising keeping in mind

that lipases probably represent the most easy-t o m e enzymes for organic synthesis. They donot require any cofactors, a wide range of en-zymes from various sources are commerciallyavailable, and they are active and stable in or-ganic solvents. In contrast to many other en-zymes, they accept a huge range of non-natu-ral substrates, which are often converted withhigh stereoselectivity. Structures of 12 lipaseshave been elucidated, and this might make thedesign of tailor-made biocatalysts feasible.Properties and applications of lipases are doc-

umented in several thousand publications, and



2 Enantiomerically Pure Pharmaceutical Intermediates 279

read ers are referred to a numb er of recent re-views (JAEG ER nd IZEETZ, 998; SCHMID ndVER GER , 998), book sections (KAZLAUSK ASand BORNSCHEUER,998; BORNSCH EUERnd

KAZLAUSKAS,999), or books (WOOLLE YndPETERSEN,994) for further reading.

2.1 Diltiazem

Both DSM-Andeno (Netherlands) and Ta-nabe P harma ceutical (Osaka, Japan) in collab-oration with Sepracor (Marlborough, MA)have commercialized lipase-catalyzed resolu-tions of (+ )-(2S,3R)-truns-3-(4-methoxy-phe-ny1)-glycidic acid methy lester (M PG M ), a key

precursor to diltiazem (HULSHOF nd Ros-

et al., 1996). The DSM -Ande no process useslipase from Rh izo m u c o r mie h ei (RM L), whilethe Tanabe process uses a lipase secreted bySerrutiu marcescens Sr41 8000. In both casesthe lipase catalyzed hydrolysis of the unwant-ed enantiomer with high enantioselectivity(E>100).The resulting acid spontaneously d e-composes to an alde hyde (Fig. 1).

In the Tanabe process, a m embrane reactor

and crystallizer combine hydrolysis, separa-tion, and crystallization of (+)-(2R,3S)-MPGM. Toluene dissolves the racemic sub-strate in the crystallizer and carries it to themem brane containing immobilized lipase. The

KAM, 1989;MATSUMAEt al., 1993,1994;FURUI

lipase catalyzes hydrolysis of the unwanted(-)-MPG M to the acid, which then passesthrough the mem brane into an aqueous phase.Spontaneous decarboxylation of the acid

yields an aldehyde which reacts with the bisul-fite in the aq ueous phase. In the absen ce of bi-sulfite, this aldehyde deactivates the lipase.The desired (+)-MPGM remains in the to-luene phase and circulates back to the crystal-lizer where it crystallizes. Lipase activity dropssignificantly after eight runs and the mem -brane must be recharged with additional lip-ase. Although the researchers detected n olipase-catalyzed hydrolysis of (- -MPGM,chemical hydrolysis lowered the apparentenantioselectivity to E= 135 und er typical re-

action conditions. The yield of cry stalline (+)-(2R,3S)-MPGM is >43% w ith 100% chemicaland e nantiome ric purity.

2.2Amines

BASF produces enantiomerically-pureamines using a Pseudomonas lipase-catalyzedacylation (BALKENHOHLt al., 1997). A keypart of the commercialization of this process

was the discovery that methoxyacetate estersreacted much faster than simple esters. Acti-vated esters are not suitable d ue to a compet-ing uncatalyzed acylation (Fig. 2). In the caseof ( R ) or (S)-1-phenylethylamine, classical

COOMelipase from + " " F ) H + MeOH

O "'"Me

n H \ A o Serratia marcescens

toluene-water/NaHSOs MeO

racemic membrane reactor (+)-(PR,3S)-M PGM

trans-isomer I 3steps

OM e

+ CO,

M e 0

Fig. 1. Commercial synthesis of diltiazem by Tanabe Pharmaceuticaluses a kinetic resolution

catalyzed by lipase from Serratia marcescens.




0 0L O M e NH2

c + xRPseudomonassp. lipase

I ,

racemateR = H, 4-Me,3-OMe

Fig. 2. Large-scale resolution of amines involves a Pseudomonas sp. lipase and m ethoxy acetic acid deriva-tives.

resolution via diastereomeric salts yielded atbest a n enantiomeric ratio of 98:2, whereas thebiocatalytic product contains less than 0.25%of the opposite enantiome r. In a similar man-ner a wide range of other amines is resolved.The process is run in a continuous manne r us-

ing immobilized lipase in a fixed-bed reactorand without additional solvents with an ann u-al production capacity of >2,000 metric tons.

2.3 6-AminopenicillanicAcid

Penicillin G acylase (PGA , penicillin amid-ase; for reviews, see BALDAROt a l., 1992;TUR-NER,1998) catalyzes hydrolysis of t he pheny l-acetyl group in penicillin G (benzylpenicillin) togive 6-aminop enicillanic acid (6 -APA ) (Fig. 3).

Penicillin G acylase also cleaves the sidechain in penicillin V, where the phenylacetylgroup is replaced by a phenoxyacetyl group.The commercially available enzymes are de-rived from E.coli strains. Both penicillin G andV are readily available from fermentation, sopenicillin manu facturers carry out a PGA -cat-alyzed hydrolysis to make 6-APA on a scale ofapproximately 5,000 metric tons per year(MATSUMOTO,992). They use 6-A PA to pre-pare semi-synthetic penicillins such as ampicil-

lin, where a D-phenylglycine is linked to thefree amino group of 6-APA, or amoxicillin,where a D-4-hydroxyphenylglycine is linked.

2.4 7-Aminocephalosporanic Acid

The enzymatic production of 7-aminoceph-alosporanic acid (7-ACA) represents an im-pressive example, where t he biocatalytic routewas highly superior to the well-established

chemical synthesis. W hereas an enzymaticroute to the closely related 6-aminopenicillan-ic acid (6-APA), a key precursor for the syn-thesis of semi-synthetic penicillins,was well es-tablished in the early 1970s and relies on theaction of penicillin G amidase (see Fig. 3), noenzymatic method could be found for 7-ACA.The biocatalytic route was developed byHoe chst researchers an d is currently conduct-ed at Biochemie GmbH (Austria). Two en-zymes are involved in the transformation ofcephalosporin C t o 7-AC A (Fig. 4) (CO NLO Ntal., 1995; BAYER nd WU LLBRAN DT,999). In

the first step, an immobilized D-amino acid ox-idase ( E C 1.4.3.3) converts cephalosporin C toa-keto-adipinyl-7-ACA. Hydrogen peroxidegenerated during this step is directly con-sumed in the decarboxylation to yield glutaryl7-AC A. Finally, glutaric acid is cleav ed by theaction of a n immobilized glutaryl amidase ( E C3.5.1.3) yielding 7-ACA. The commercial suc-cess was mainly based on ecological reasons. Incontrast to the chemical route, no toxic sub-

stances are produced and the wastewater isreadily biodegradable. Thus the amount ofwaste to be treated was reduced from 31 met-

penicillinG 6-APA

Fig. 3. Commercial process for the productionof 6-APA from penicillin G using penicillin G acylase.



2 Enantiomerically Pure P harmaceutical Intermediates 281

cephalosporin C

NH3 H202

o-amino acidoxidase

COOH 0

a- eto-adipinyl-7-ACA

I lutarylglutaricrnidasecid e & o y c H 3

0

COOH 0glutaryl-7-ACA

Fig. 4. Biocatalytic route to 7-amino cephalosporanic acid.

ric tons (chemical synthesis) to 0.3 metric ton s(enzymatic route ) in the p roduction of 1 met-ric ton of 7-ACA . The cost share for environ-mental protection on the total process costswas reduced from 21% to only 1% (BAYER

and WULLBRANDT,999).

2.5 Naproxen

Although a wide range of esterases (EC3.1.1.1) has been described in the literature(for reviews, see PHYTIAN,1998; B O R N -

SCHEUER and KAZLAUSKAS,999), heir indus-trial application is very limited. This is mostlydue to the restricted availability of esterasescomp ared to lipases (with the exception of pig

liver esterase). An other exception is carboxyl-estera se NP, which was o riginally isolated fro mBacillus sub tilis (strain Thail-8) and was clonedand expressed in B. subtilis (QUAX and BROEK-HUIZEN, 1994). The esterase shows very highactivity and stereoselectivity towards 2-aryl-propionic acids, which a re, e.g., used in th e syn-thesis of (S)-naproxen - (+)-(S)-2-(6-meth-oxy-Znaphthyl) propionic acid - a non-steroidal anti-inflammatory drug (Fig. 5). (S)-naproxe n is ca. 150-times mo re effective than(R)-naproxen, the latter also might promoteunwanted gastrointestinal disorders. D ue to its

selectivity toward Naproxen, this esterase isabbreviated in most references as carboxyl-esterase NP. It has a molecular weight of 32kDa, a pH optimum betw een p H 8.5-10.5, an da temperature optimum between 35-55 "C.

Carboxylesterase N P is produced as intracel-lular prote in; its structure is unknown .Although the process gave (S)-naproxen

with excellent optical purity (99% ee) at anoverall yield of 95%, it has not reached com-mercialization.

M e0qo/(R,S)-naproxen methylester

esterase NP

/ \e t i o npH 9.0,40°C

M e 0aoHM e0

(S)-naproxen (R)-naproxen methylester99% ee, 95%overall yield

Fig. 5. Synthesis of (S)-naproxen by kinetic resolu-tion of the (R,S)-m ethyl ester with carboxylesteraseNP followed by chemical racemization of the (R)-naproxen methylester (QUAX and BROEKHUIZEN,

1994).




In contrast, Chiroscience (UK) commercial-ized the resolution of the eth ylester of naprox-en at ton scale, in which anothe r recombinantesterase is used (BRO WNt al., 1999).

2.6 Pyrethroid Precursors

2.6.1 Via Oxynitrilases

Oxy nitrilases (also named hydroxy nitrile ly-ases, HNL) comprise a group of enzymeswhich are capable of creating a carb on xa rb onbond utilizing HC N as donor and an aldehyde(or keton e) as acce ptor (BU NCH , 998). Initial-ly, enzym es isolated fro m various plant sources

such as cassava, Sorghum bicolor, and Heveasp. had bee n employed. In recent years, espe-cially the groups of EFFENBERGERStuttgart,Germany) and GRIENGLGraz, Austria) suc-cessfully cloned the genes encoding e ither ( S ) -or (R)-specific HNLs into heterologous hoststhus making the enzym es available at com pet-itive prices and in sufficient am oun ts (WAJANT

et al., 1994; EFFENBER GER,994; WAJANT ndEFFENBERGER,996; HASSLA CHERt a l., 1997;JOHNSONnd GRIENGL,998).

A process based on a HNL-catalyzed reac-tion was developed by DSM in collaborationwith the university of Graz. A prerequisite fora successful commercialization was the usageof an already existing plant, which allowed asafe handling of HCN. Currently, DSM is pro-ducing (S)-m-phenoxybenzaldehyde cyano-

hydrine on a m ulti-ton scale. The prod uct is animportant precursor for the synthesis of pyr-ethroids such as (S, S)-fenvalerate and delta-methrin (Fig. 6). The (S)-oxynitrilase ex-

pressed recombinantly in the yeast Pichia pas-toris presumably originates from Hevea brasi-liensis and is produced by Roche Diagnostics(Penzberg, Germ any).

2.6.2 Via a M onooxygenase

BASF has commercialized a process inwhich (R)-2-phenoxypropionic acid is selec-tively hydroxylated to (R)-2-(4-hydroxyphen-0xy)propionic acid (H-POPS). H-POPS is an

interm ediate in the syn thesis of optically pureherbicides of the aryloxyphenoxypropionatetype. Out of 7,900 strains tested, the fungusBeauveria bassiana was identified (Fig. 7).Considerable effort was spent for strain im-provem ent by mutagenesis to increase produc-tivity as well as tolerance of the microorgan-ism towards high substrate and product con-centrations. Thus productivity was increasedfrom 0.2 g L- l d -l to 7 g L -l d- '.

Fig. 7. Synthesis of (R)-2-(4-hydroxyphenoxy)pro-

pionic acid by BASF uses Beauveria bassiana.

NC. A

(S,S)-fenvalerate

(8-HNL rom

+ HCN Hevea brasiieiensis-

Br

Fig. 6. Synthesis of pyrethroid precursors using a hydroxynitrilase(NHL) from Hevea brusiliensis.



2 Enantiomerically Pure pharmaceutical Intermediates 283

2.6.3 Via Esterase

(+ -trans-( 1R,3R ) chrysanthemic acid is animportant precursor of pyrethrin insecticides.An efficient kinetic resolution starting from

th e (f -cis-trans ethylester was achieved us-ing an esterase from Arthrobac ter globiformisresulting in the sole formation of the desiredenantiomer (>99% ee, at 77% conversion)(Fig. 8). The enzyme was purified and th e genewas cloned in E. coli (NISHIZAWAt al., 1993).In a 160 g scale process reported by Sum itomoresearchers, hydrolysis is performed at p H 9.5at 50 "C .Acid produced is separated throug h ahollow-fiber mem brane module and the e ster-ase was very stable over four cycles of 48 h(NISHIZAWAt al., 1995). It is u nclear, wheth erthe process is currently run on an industrialscale.

Arthrobacterglobiforrnis

esterase*CCOHOOEt,~~~

(+)-c is- rans- (+)-transchrysanthemicchrysanthemic acid acid (>99% ee)

Fig. 8. Synthesis of (+ ) - t ran schrysanthemic acid bykinetic resolution of the ethy lester using an esterasefrom Arthrobacter globiformis.

2.7 Production of Chiral C,Building Blocks

Optically active C , comp ounds are versatilebuilding blocks for organic synthesis, and themost important ones are epichlorohydrin, 3-chloro-1 2-propane diol, and glycidol. Stereo-selective synthesis can b e performed by chem-ical means (e.g., using th e Jacobsen-K atsukicatalyst, Sharpless epoxidation, or D-manni-

tol), as well as by a rang e of biocatalytic routes.KAS AI t al. (1998) provide an excellent over-view for both approaches.

For instance, DSM-Andeno (Netherlands)produce (R)-glycidol butyrate using porcinepancreatic lipase (Fig. 9) by kinetic resolution,but they did not reveal details of the process(LADNER nd WHITESIDES,984; KLOOST ER-M AN et al., 1988). Several groups have sincestudied this reaction and its scale-up in m oredetail (WALTS nd Fo x, 1990; W u e t al., 1993;VAN TOL t a]., 1995a,b).

"-prxokPL, E=13_ pprTolHO I

0 0 oi(R,S)-glycidyl (R)-glycidyl (3-gl ycido l

butyrate bulyrate

Fig. 9.Resolution of glycidol butyrate by D SM -An-den0 using porcine pancreatic lipase (PPL).

O th er groups described resolution of epox-ides using an epoxide hydrolase o r by m icrobi-al degradation. Researchers at Daiso, Japan,discovered an (R)-2,3-dichloro-l-propanols-similating bacterium, Pseudo mon as sp. OS-K-29 and a (S)-selective strain from Alcaligenessp. DS-K-S38 (Fig. 10). Bo th strains exhibitedexcellent selectivity and allowed for t he isola-

tion of 50% remaining substrate having either(S) or @)-configuration (KASAI t al., 1990,1992a,b). Although on e enantiomer is lost, it isclaimed that the fermentative process estab-lished in 1994 is cost-effective (KASAI t al.,1998). In a similar man ner, 3-chloro-1,2-pro-pane diol can be selectively degraded by m icro-organisms, and a fermentation process hasbeen developed on an industrial scale at D aiso(SUZU KI nd KAS AI, 991; SUZ UKI t al., 1992,1993).Furth er examples for industrial-scale bi-

ocatalysis by dehalogenases have recentlybeen reviewed (SWANSON ,999).

C l y O H Pseudomonassp ._ COHCI d

100% ee ,38%

C I T O H Alcaligenessp. CI-OH

OH OH

>99% ee . 40-45%

Fig. 10. Microbial reso lution of C,-building blocks

by Daiso , Japan.

2.8 Other Pharmaceutical

Intermediates

Glaxo developed a process to resolve (lS,2S)-trans-2-methoxy cyclohexanol, a secon-dary alcohol for the synthesis of a tricyclic p-

lactam a ntibiotic (ST EAD t al., 1996).The slowreacting enan tiomer neede d for synthesis is re -covered in 99% e e from an acetylation of th e




racemate with vinyl acetate in cyclohexane.Immobilized lipase B from Candida antarctica(CAL-B) and lipase from Pseudomonasfluorescens (PFL) both showed high enantio-

selectivity, but CA L-B was mo re stable overmultiple use cycles. Other workers had re-solved this alcohol by hydrolysis of its esterswith lipase from Pseudomonas cepacia (PCL),Candida rugosa (CRL), or pig liver acetonepow der (LAUM EN t al., 1989; HONIGandSEUFER-WASSERTHAL,990; BASAVAIAHndKRISHNA,994),but Glaxo chose resolution byacylation of the alcohol because it yields therequired slow-reacting alcohol directly (Fig.11).

CAL-B,37 s/Lvinyl acetate, 1.7 M

triethylamine, 0.16 M\,.OMe cyclohexane, 6-8 h.

racemic

Researchers reported a number of otherkilogram scale routes to pharmaceutical pre-cursors tha t invo lve lipases. Selected exam plesare sum marized in Fig. 12.

Other carboxylic acids are also importantintermediates in the synthesis of pharmaceuti-cals. Researchers have published kilogram-scale resolutions by using proteas es (Fig. 13).

Chiroscience (UK) also established a pro-cess for the large-scale production of a keyintermediate for the synthesis of the reversetranscriptase inhibitor Abacavir (Ziagen). Theprocess relies on a recombinant p-lactamaseyielding the required ( -)-lactam in >98% e e(E>400) and is oper ated by Glaxo-Wellcome(Fig. 14) (BRO WNt al., 1999).

>99% ee36% yield

antibiotic

Fig. 11.Glaxo resolves a b uilding block fo r antibiotic synthesis.

HO ,OBn

OAc *OMe

COOnPr

an anti-cancer drug hypercholestemic agents isopropenyl acetate

PATELt al. (1994)

A for cafbovir, an anti-HIV agent

MACKEITHt al. (1993, 1994)

for side chain of taxol, enantiomer used for anti- PCL, E = 32 - 68

SIH1996), HENEGARt al. (1997)

H H U

HOzCJy30 CI

elastase inhibitorexperimental treatment

for cystic fibrosisCVETOVICHt al. (1996)

LTD4 antagonist for asthma treatment(did not pass clinical trials)HUGHESt al. (1989, 1990)

4Bn

for a thromboxaneA2 CRL, vinyl acetate for antifungal agentantagonist configuration at '*'set by synthesis SAKSENAt al. (1995)

MORGANt al. (1997) MORGANt al. (1997)ATELt al. (1 992)

Fig. 12. Kilogram-scale rou tes to pharmaceu tical precursors involving lipases.



3 AminoAcids 285

subtilisinrenin inhbitors

MAZDIYASNIt al. (1993)

I0

P h H X p N > N

‘COOH

PGAfor renin inhibitors for a colhgenase inhibitor for p-lactam antibiotic

WIRZand SOUKUP1997) ZUIJEWSKIt al. (1991)

Fig. 13. Kilogram-scale routes to pharmaceutical precursors involving proteases oramidases.

L o X R

subtiliiin subtilisin

DOSWALDt al. (1994)

Fig. 14.Process for th e resolution of

(k )-2-azabicyclo[2.2.1]hept-5-en-3-

on e using enantiocomplementaryp-lactamases. The enzymefurnishing th e ( -)-lactam has bee ncloned an d is currently u sed in a

multi-ton scale process for the pro -duction of an Abacavir intermedia-te.

3 AminoAcids

0

(-)-lactarn>98% ee

H.

(*)-lactarn

H(+)-lactarn>98% ee

3.1 Natural Amino Acids

Awide range of naturally occu mn g (protein-ogenic) L-amino acids ar e currently produce d

by biotransform ation. Especially in Japan, sev-eral L-amino acids ar e produced by fermenta-tion (see C hapter 8, this volume). For instance,L-glutamic acid is produced annually at the300,000 metric ton scale by Ajinomoto, Japan,using Corynebacterium glutamicum. On th eother hand, L-amino acids can be producedfrom chemically synthesized racemic mixturesby kinetic resolution of, e.g., N-acetyl deriva-tives (see also Sect. 3.2). L-Aspartate is prod-uced by Tanabe Seiyaku and Kyowa HakkoKogyo (both Japan) from fumaric acid using

an aspa rtate am monia lyase as catalyst.

IHO Abacavir

3.2 Non-Natural Amino Acids

In principle, some of the ro utes de velopedfor the p roduction of L-amino acids can also beapplied for the synthesis of non-natural aminoacids which consist of eith er na tural a min o ac-ids, but in the D-configuration o r non -protein-

ogenic amino acids (usually also in the D-con-figuration). The m ost prominent examples areD-phenylglycine and D-4-hydroxyphenylglyc-ine, which are precursors of th e semi-syntheticpenicillins Ampicillin and Am oxicillin and arecurrently produced in a hydantoinasekarba-moylase process at a >1,000 tonnes per yearscale (see also Sect. 2.3).

Most non-natural am ino acids are producedcommercially via acylase-, hydantoinase- andamidase-catalyzed routes (for reviews, seeBOMMARIUSt al., 1992; KAMPHUIS t al.,

1992). A few examples a re shown in Fig. 15.




Acylase-catalyzed routes are best for theL-amino acids because acylases favor theL-enantiomer. Racemization of unreactedN-acyl derivative of the D-amino acid avoids

wasting half of the star ting material. Hy danto -inase-catalyzed routes are best for producingD-amino acids because the most comm on hyd-antoinases favor the D-enantiomer. Amidase-catalyzed resolutions are best for unusualamino acids whose derivatives are not sub-strates fo r acylases o r hydantoinases. For ex-ample, a-alkyl-a-amino acids are resolved byamidases as are the cyclic amino acids 2-piper-idine carboxylic acid (pipecolic acid) and pi-perazine-2-carboxylic acid. Amidases usuallyfavor the natural L -enantiomer.

NSC T echnologies has succeeded in produc-ing D-phenylalanine an d D-tyrosine on a multi-ton scale. The biocatalytic route is based on theshikimic acid metabolic p athway a nd relies ona D-transaminase acting on phenylpy ruvic acid(STINSON,998).

Kyowa Hakko Kogyo researchers also re-ported the syn thesis of D-alanine using a D,L-alaninamide hydrolase found in Arthrobactersp. NJ-26 (YAGASA KI nd OZA KI, 1998).D-Glutamate is produce d by the sam e compa-ny starting from che ap L-glutamate.A combi-nation of a glutamate racemase (from Lacto-bacillus brevis ATC C8287) and an L-glutamatedecarboxylase (from E. coli ATCC11246) al-lowed the synthesis of D-glutamate (99% ee)a t 50% yield by a two-step fermentation. In asimilar manner D-proline can be produced(YAGASAKInd OZA KI, 998).

Researchers at DSM (The Netherlands), de-scribed a new amidase activity discovered inOchrobactrum anthropi N CIMB 40321, whichaccepts a,a-disubstituted amino acids as sub-

strates. The amidase gene was successfullycloned and overexpressed in a host organism,allowing the application of the enzyme in acommercial process, probably dealing with the

synthesis of 4-methylthio- an d 4-methylsulfo-nyl substituted (2S,3R)-3-phenyIserines, (S)-(+ )-cericlamine or L-methylphenylalanine(VANDOOREN nd VA N DEN TW EEL, 992;VAN

DEN TW EEL t al., 1993; KAPTEIN t al., 1994,1995,1998;SCHOEMAKERt al., 1996).Chiroscience (UK) has scaled up the dy-

nam ic kinetic resolution of (S)-tert-leucine, anintermediate for the synthesis of conforma-tionally restricted peptides and chiral auxiliar-ies (TU RNE R t al., 1995; McC ague a nd TAY-LO R, 997). How ever, the only comm ercializedroute is a process by Degussa (H ana u, Germ a-ny) utilizing a leucine dehydrogenase. TheN AD H consumed during the reductive am ina-tion of the a-keto acid is recycled with a for-

mate dehydrogenase/ammonium formate sys-tem. Recycling of the cofactor allows for totalturnover num bers in the range of >lO,OOOmaking the process cost-effective (BOMMA-RIUS et al., 1992,1995).

3.3 Aspartame

The largest scale application (hundreds tothousands of tons) of protease-catalyzed pep-tide synthesis is the thermolysin ca talyzed syn-thesis of aspartame, a low calorie sweetener(Fig. 16) (ISOWAt al., 1979; OYA MA ,992; seeChapter 2, this volume). Precipitation of theproduct drives this thermodynamically con-trolled synthesis. The high regioselectivity ofthermolysin ensures that only the a-carboxylgroup in aspartate reacts. Thus, there is noneed to protect the P-carboxylate. The highenantioselectivity allows Tosoh t o use racemicamino acids; only the L-enantiomer reacts.

acylase hydantoinase arnidase arnidaseBOMMARIUSt al. (1992) EICHHORNt al. (1997)

Fig. 15.Examples of non-natural amino acids produced via acylase, hydantoinase, or amidase.

BOMMARIUSt al. (1992) KAMPHUISt al. (1992)



4 Other Applications 287

coo0

G O H + H3'+o~e thermolysin- HY bz ' 1 0 M eph > + OM e

PhhHY

Cbz 0

insolublexcess

Fig. 16. Commercial process for the production of aspartame (a-L-aspartyl-L-phenylalan-ine methyl ester).

4 Other Applications

4.1 Acrylamide/Nicotinamide

Although nitriles can also be hydrolyzed to

the corresponding carboxylic acid by strongacid or base at high temperatures, nitrile hy-drolyzing enzymes have the advantages thatthey require mild conditions and do not pro-duce large amounts of by-products. In addi-tion, during hy drolysis of dinitriles, they are of-ten regioselective so that only one nitrilegroup is hydrolyzed and they can be used forthe synthesis of optically active substances.

Th e enzym atic hydrolysis of nitriles followstwo different pathways (Fig. 17). Nitrilases

(E C 3.5.5.1) directly catalyze the conversion ofa nitrile into the corresponding acid plus am-monia. In the other pathway, a nitrile hydrat-ase (NH ase, E C 4.2.1.84; a lyase) catalyzes thehydration of a nitrile to the amide, which maybe converted to the carboxylic acid and amm o-nia by an am idase ( E C 3.5.1.4) (BUN CH, 998).

Both pathways occur in the biosynthesis ofthe phytohormone indole-3-acetic acid fromindole-3-aceton itrile fo r a nitrilase from Alcal-igenes faecalis JM3 (KOBAYASHIt al., 1993)and for the nitrile hydratase/amidase system in

Agrobacteriurn turnefaciens and Rhizobium sp.(KOBAYASHIt al., 1995).

Pure nitrilases and nitrile hydratases areusually unstable; so most researchers use them

nitrilaseRCN- COOH + NH3

rRCIOINH2ydratase 1midase

Fig. 17. Hydrolysis of nitriles follows two differentpathways.

in whole cell preparations. Furthermore, thenitrile hydrolyzing activity must be inducedfirst. .

Two applications based on nitrile hydratasein Rhodococcus rhodochrous originally isolat-ed by Yamada's group (NAGA SAW And YAMA-

DA, 995) have been com mercialized.The largeproduction of th e com modity chemical acryla-mide (Fig. 18) is perform ed by N itto Chem ical(Yokohama, Japan) on a >30,000 me tric ton sper year scale. Initially,strains from Rhodococ-cus sp. N-774 or Pseudornonas chlororaphisB23 were used, however the current processuses the 10-fold more productive strain R. rho-dochrous J1. The productivity is >7,000 gacrylamide per g cells at a conversion of acry-lonitrile of 99 .97%. Formation of acrylic acid is

barely detectable at the reaction temperatureof 2-4 "C. In laboratory-scale experim ents withresting cells up t o 656 g acrylamide per liter re-action mixtures were achieved (KOBA YASHItal., 1992).

O ne of the most important characteristics ofth e R. rhodochrous strain is its tolerance to-ward high concentrations of acrylamide (up to50%). Induction of NHase activity is per-formed using urea resulting in more than 50%of high molecular weight NHase of the totalsoluble protein. Cobalt ions are essential to get

active NHase. Besides acrylamide, a widerange of other amides can also be produced,e.g., acetamide (150 g L-'), isobutyram ide(100 g L-l), methacrylamide (200 g L-'),

propionamide (560 g L-'), and crotonam ide(200 g L- l) (KOBAYASHIt a l., 1992).

Furthermore,R. rhodochrous J1 also acceptsaromatic and arylaliphatic nitriles as sub-strates. For instance, also the conversion of3-cyanopyridine to nicotinamide (a vitamin inanimal feed supplem entation, Fig. 18) is cata-lyzed (NAGASAWAt al., 1988), which was in-

dustrialized by Switzerland's Lonza on a 3,000




metric tons per year scale in their plant inGuangzhou, C hina. In contrast to the chemicalprocess, no formation of the by-product nico-tinic acid is observed using the nitrile hydrat-

ase system.DuPont researchers described a route to

1,5-dimethyl-2-piperidone from 2-methylglu-taronitrile (MG N). MGN is first hydrolyzed to4-cyanopentanoic acid ammonium salt usingimmobilized cells of Acidovorax facilis 72W(ATCC55746) which contain a nitrilase. Theselectivity of this biocatalyst is >98% yieldingthe desired monoacid as major product incon cen trations between 150-200 g L-', whichafter chem ical hydrogenation furnishes the d e-sired product. Compared to the chemical

route, the biocatalytic approach yields a singlelactam pro duct, less waste, and higher overallyields (DICOSIMOt al., 1997; GAVA GANt al.,1998).

acrylonitrile

0

acrylamide

Scyano pyridine nicotinamide. a vitamin

Fig. 18. Com mercial production of acrylamide andnicotinam ide using resting cells of Rhodococcus rho-dochrous J1.

4.2 L-Carnitine

L-Carnitine [(R)-3-hydroxy-4-(trimethylam-monio)butanoate] belongs to a group of vita-

min-like nutrients. L-Carnitine plays an essen-tial role in the @ -oxidationof long-cha in fattyacids by mitochon dria an d is synthesized in thehuman liver from lysine and methionine. In -sufficient production causes muscle weakness,fatigue, and elevated levels of triglycerides inthe blood. Besides its use in infant, health,spor t and g eriatric nutrition, it might also leadto enhanced growth rates in the animal feedsector. Because D-carnitine comp etes with th eL-form for active transport into the e ukaryoticcell, the pure L-enantiomer is required for usein the medical, nutritional and feed applica-tions of carnitine. Severa l routes to L-carnitinehave been described in the litera ture includingresolution of racemic mixtures (see ZIMMER-M A N N et al., 1997, and references cited there -in). For instance, researchers at Daiso, Japan,reported a route starting from optically pureepichlorohydrin (KASAIand SAKAGUCHI,992;KASAI et al., 1998). How ever, it is unclearwhethe r the process has been commercialized.An other access by Italians Sigma-Tau might

be based on the reduction of y-chloroacetoac-etate ester to (R)- -chloro-P-hydroxy butan -oat e mediated by a P-ketoredu ctase from Suc-

charomyces cerevisiae.Large-scale production by Lonza (Switzer-

land) starts from easily available butyrobe-taine w hich is converted in a multi-enzyme re-action to L-carnitine (Fig. 19) (ZIMMERMANN

ATP, CoA AMP + PPi

Me& M e 3 6 y S c o Asynthetase0

butyrobetaine 4-butyrobetainyl-CoA

FADHp-H 0 hydrolase 0 ATP 0

dehydrogenase

H 2 0

Me&

(Rj-carnitine (100% ee) crotonobetainyl-CoA crotonobetaine

Fig. 19. Biocatalytic route to L-carnitine.



4 Other Applications 289

et al., 1997). The initially foun d most activestrain (HK 4) was related taxonomically to thesoil microorganisms Agrobacterium and Rhi-zobium. For large-scale production, a strain

improvement program was undertaken withthe aim to increase the tolerance of the straintowards high concen trations of sub strates andproducts. In addition, a de hydrogen ase activitywas irreversibly inactivated by fr am e shift mu-tation to avoid degradation via undesiredmetabolic pathways. The thus ob tained strain(HK1349) quantitatively converts 4-butyrobe-taine (and crotonobetaine) into L-carnitine(100% e e), which is excreted into the medium.Maximum productivity is in the rang e of 100 gL-’. Although a fed-batch process gives lowerproductivity, this route was chosen due to aneasier downstream processing and lower in-vestment costs. Resea rchers a t Lonza also de-veloped a recombinant production strain;however, this on e is not yet used in the process.

4.3 Lipid Modification

Alre ady in the early 1970s, he application oflipases for the modification of their natural

substrates-

fats and oils-

was exploited bymany researchers in industry and academia.Although strong efforts were undertaken touse lipases to obtain free fatty acids by hydrol-ysis or t o prod uce monoglycerides by glycerol-ysis, these enzym atic processes were nev er e f-fective enough to compete with well estab-lished chemical methods. Current processesrather exploit 1,3-regiospecificityof lipases toproduce so-called structured triglycerides(ST). ST are triglycerides modified in eitherthe type of fatty acid o r the position of the fat-

ty acids and the m ost promine nt exam ples arecocoa butter equivalents an d ABA lipids (tri-glycerides with fatty acids A at sn-1 and sn -3and fa tty acid B at sn-2-position). CHRISTOPH E(1998) provides an ex cellent overview of nutri-

tional properties and enzymatic synthesis ofstructured triglycerides.

4.3.1 Cocoa Butter Equivalents

Cocoa butter is predominantly 1,3-disatu-rated-2-oleyl-glyceride,where palmitic, stearicand oleic acids account for mo re tha n 95% ofthe total fa tty acids. Coco a butte r is crystallineand melts between 25 and 35 “C imparting thedesirable “mouth feel”. Unilever (COLEMANand MACR AE, 977) and Fuji O il (MATS UOtal., 1981) filed the first patents for th e lipase-catalyzed synthesis of cocoa butter equiva-lents. Both companies currently manufacture

cocoa butter equivalents on a multi-ton scaleusing 1,3-selec tive lipases to r eplac e palm iticacid with stearic acid a t the sn-1 and sn-3 posi-tions (for reviews, see MA CRA E,983;MACRAEand HAMMOND,985; QUIN LANnd MOORE,1993). The Unilever subsidiary Q uest-LodersCroklaan (The Netherlands) uses RML andpalm oil mid fraction o r high oleate sunfloweroil as starting materials (Fig. 20). Other suit-able starting oils are rape seed (ADLER-CREUT Z, 994) or olive oils (CHANGet al.,

1990).

4.3.2 Other Structured Lipids

An other structured triglyceride is Betapol, aformula additive for premature infants. Hu-man milk contains the more easily digestedOPO,but vegetable oils contain the PO 0 iso-mer (OPO: 1,3-dioley1-2-palmitoyl glyceride,POO: mixture of 1,2-dioley1-3-palmitoylglyc-eride and its enantiomer). Interesterification

of tripalmitin with oleic acid using RM L a t lowwater activity yields OPO. Evaporation re-moves excess fatty acids and crystallization re-moves remaining PPP. Betapol is currentlyproduced b y a U nilever subsidiary.

m16:O

1,3-selective ipase* E m j R l + EE::; + 3c16:o

oc16:o

OC160 ~ 1 8 : O m1eo

2 EocIB:~ 3 18:O

palm oil fraction co ma butter equivalent

Fig. 20. Lipase-catalyzed synthesis of cocoa-butter equivalent.




Other processes under development focuson the enrichment or isolation of polyunsatu-rated fatty acids such as eicosapentaenoic acid(EPA) or docosahexaenoic acid (D H A ) from

fish oil. Diets supplemented with EPA orD H A in the triglycerides are reported to showa range of nutritional benefits .

AcknowledgementsThe author thanks Dr. P. RASOR Roche Diag-nostics, Penzberg, Germany), Dr. J. PETERS(Bayer AG, Wuppertal, Germany), Dr. B.HAU ER BASF AG, Ludwigshafen, Germany),Dr.A. KIENERLonza, Visp, Sw itzerland), andProf. R.J. KAZLAUSKASMcGill University,

Montreal, Canada) for their support duringpreparation of the manuscript.

(1995), Synthesis and use of enantiomericallypure tert-leucine, Tetrahedron: Asym metry 6,2851- 2888.

BORNSCHEUER,. T. (1998), Directed evolution ofenzymes, Angew. Chem. (Int. Edn. Engl.) 37,3105-3108.

BORNSCHEUER,. T., KAZLAUSKAS,. J. (1999),H y-drolases in Organic Synthesis-Regio- and Stereo-selective Biotransformations. Weinheim: Wiley-VCH.

BROWN,. C.,TAYLOR, .J. C.,TAYLOR,. N., KEENE,P.A., BAUER,. (1999), Different cloning approa-ches for industrially useful biocatalysts,AbstractBoo k, Biotrans '99Conference, Taormina,Italy.

BUNCH,A. (1998), Nitriles, in: Biotechnology 2ndEdn., Vol. 8a (REHM,H.-J., REED,G., WHLER, .,STADLER,., Eds.), pp. 277-324. Weinheim: Wiley-VCH.

CHANG,M. K., ABRAHAM,., JOHN, .T. (1990), Pro-duction of cocoa butter-like fat from interesterifi-

5 References

ADLERCREUTZ.! (1994).

cation of vegetable oils,J . Am . Oil Chem. SOC.67,832-834.

CHRISTOPHE,. B. (Ed.) (1998),Structural ModifiedFood Fats: Synthesis, Biochemistry, and Use .Champaign, ILAOCS Press.

COLEMAN,. H., MACRAE,A. . (1977), Rearrange-ment of fatty acid esters in fat reaction reactants,

Enzvme-catalvzed b i d German Patent DE 2 705 608 (Unilever N.V.)modification, Bibtech nol. &net. E n; Rev. L12,

ARNOLD,. H. (1998), Design by directed evolution,Acc . Chem . Res.31,125-131.

BALDARO,., FUGANTI,., SERVI, ., TAGLIANI,.,TERRENI, . (1992), The use of immobilized pen-icillin G acylase in organic synthesis, in: MicrobialReagents in O rganic Synthesis (SERVI, ., Ed.), pp.175-188. Dordrecht: Kluwer Academic.

BALKENHOHL,., DITRICH, ., HAUER, ., LADNER,W. (1997), Optically active amines via lipase-cata-lyzed methoxyacetylation, J . Prakt. Chem. 339,

BASAVAIAH,., KRISHNA,! R. (1994), Pig liver acet-one powder (PLAP) as biocatalyst: enantioselec-

tive synthesis oftrans-2-alkoxycyclohexan-l-ols,

Tetrahedron50,10521-10530.BAYER,T., WULLBRANDT,. (1999), Enzymatische

Herstellung von 7-Aminocephalosporansaure,in: Industrielle Nutzung von Biokatalysatoren(HEIDEN,. , BOCK,A.-K.,ANTRANIKIAN,.,Eds.),pp. 187-194. Berlin: Erich-Schmidt-Verlag.

BOMMARIUS,. S.,DRAUZ, ., GROEGER,.,WAND-REY, C. (1992), Membrane bioreactors for theproduction of enantiomerically pure a-aminoacids, in: Chirality in Industry Vol. I (COLLINS,.N., SHELDRAKE,. N., CROSBY,., Eds.), pp. 371-397. Chichester: John Wiley& Sons.

BOMMARIUS,. S., SCHWARM,., STINGL,K., KOT-TENHAHN, M., HUTHMACHER,., DRAUZ, K.

231-254.

381-384.

(Chem.Abstr.87,166 366).COLLINS,A. ., SHELDRAKE,. N., CROSBY,. (Eds.)

(1992), Chirality in Industry I . New York JohnWiley & Sons.

COLLINS,A.., SHELDRAKE,. N., CROSBY,. (Eds.)(1997), Chirality in Industry IZ . Chichester: JohnWiley& Sons.

CONLON,H. D., BAQAI, ., BAKER, ., SHEN, . Q.,WONG,B. L. et al. (1995), Tko-step immobilizedenzyme conversion of cephalosporin C to 7-amin-ocephalosporanic acid, Biotechnol. Bioeng. 46,51&513.

CVETOVICH,. J., CHARTRAIN,., HARTNER,. W.,ROBERGE, ., AMATO, . S., GRABOWSKI,. J.(1996), An asymmetric synthesis of L-694,458, ahuman leukocyte elastase inhibitor, via novelenzyme resolution of p-lactam esters, J . Org.Chem. 61,65754580.

DICOSIMO,., FALLON,. D., GAVAGAN,. E. (1997),Preparation of 5- or 6-membered lactams fromaliphatic a,o-dinitriles, Int. Patent Appl.W09744318 (DuPont) (Chem. Abstr. 128,89231;Corr. of Us, 8589).

DOSWALD,.,ESTERMANN,H.,UPFER, ., STADLER,H., WALTHER, . et al. (1994), Large scale prepa-ration of chiral building blocks for the P3 site ofrenin inhibitors,Bioorg. Med. Chem. 2,403-410.

EFFENBERGER,. (1994), Synthesis and reactions of

optically-activecyanohydrins,Angew. Ch em. (Int.Edn. En gl.)33,1555-1564.



5 References 291

EICHHORN,., RODUIT,.-P., SHAW,N., HEINZMANN,K., KIENER,A.1997), Preparation of ( S ) - pipera-zine-2-carboxylic acid, (R)-piperazine-2-carboxy-lic acid, and (S)-piperidine-2-carboxylicacid bykinetic resolution of the corresponding racemiccarboxamides with stereoselective amidases inwhole bacterial cells, Tetrahedron:Asymmetry 8 ,

FURUI,M., FURUTANI,., SHIBATANI,., NAKAMOTO,Y., MORI,T. H. (1996), A membrane bioreactorcombined with crystallizer for production of opti-cally active (2R,3S)-3-(4-methoxyphenyl)glycidicacid methyl ester,J. Ferment. Bioeng. 81,21-25.

GAVAGAN,. E., FAGER,.K.,FALLON,. D.,FOLSOM,P. W., HERKES,. E. et al. (1998), Chemoenzymicproduction of lactams from aliphatic ap di ni tr i-les, . Org. Chem.63,47924801.

HASSLACHER,., SCHALL,M., HAYN,M., BONA,R.,

RUMBOLD,. et al. (1997), High-level intracellu-lar expression of hydroxynitrile lyase from thetropical rubber tree Hevea brasiliensisin microbi-al hosts,Protein Expr. Purif 11,61-71.

HENEGAR, . E., ASHFORD,.W., BAUGHMAN,. A.,SIH,J. C., Gu, R. L. (1997), Practical asymmetricsynthesis of (S)-4-ethyl-7,8-dihydro-4-hydroxy-lH-pyran0[3,4-flindolizine-3,6,10(4H)-trione,

key intermediate for the synthesis of irinotecanand other camptothecin analogs,J. Org. Chem.62,6588-6597.

HONIG,H., SEUFER-WASSERTHAL,. 1990), A gene-ral method for the separation of enantiomerictrans-2-substituted cyclohexanols, Synthesis,

HUGHES,. L.,BERGAN,. J.,AMATo, . S. ,REIDER,.J., GRABOWSKI,. J. J. (1989), Synthesis of chiraldithioacetals: a chemoenzymic synthesis of a no-vel LTD4 antagonist,J. Org. Chem.54,1787-1788.

HUGHES,. L., BERGAN,.J.,AMATo, . S. ,BHUPATHY,M., LEAZER,. L. et al. (1990), Lipase-catalyzedasymmetric hydrolysis of esters having remotechirallprochiral centers, J . Org. Chem. 55, 6252-6259.

HULSHOF,. A., ROSKAM,. H. (1989), Phenylglycid-ate stereoisomers, conversion products thereofwith e.g. 2-nitrophenol and preparation of diltia-zem, European Patent EP0343714 (Stamicarbon)(Chem.Abstr . 113,76603).

ISOWA,., OHMORI, ., ICHIKAWA,., MORI,K., No-NAKA, Y. et al. (1979), The thermolysin-catalyzedcondensation reactions of N-substituted asparticacid and glutamic acids with phenylalanine alkylesters, Tetrahedron Lett. 2611-2612.

JAEGER, .-E., REETZ,M.T. (1998), Microbial lipasesform versatile tools for biotechnology, TrendsBiotechnol. 16,396403.

JOHNSON,.V., GRIENGL, . (1998), Biocatalytic ap-plications of hydroxynitrile lyases, in:Advances inBiochemical Engineering/BiotechnologyVol. 63

2533-2536.

1137-1 140.

(SCHEPER,., Ed.), pp. 31-55. Berlin: Springer-Verlag.

KAMPHUIS,., BOESTEN,W. H. J., KAFTEN, B., HER-MES,H. F. M., SONKE,T. t al. (1992), The produc-tion and uses of optically pure natural and un-natural amino acids, in: Chirality in Indu stry Vol. I

(COLLINS, . N., SHELDRAKE,. N., CROSBY,.,Eds.), pp. 187-208. Chichester: John Wiley &Sons.

KAF~EIN, ., MOODY,H. M., BROXTERMAN,. B.,KAMPHUIS,. (1994), Chemo-enzymic synthesisof (S)-( )-cericlamine and related enantiomeri-cally pure 2,2-disubstituted-2-aminoethanols, .

Chem. SOC.,Perkin Trans. 1,1495-1498.KAETEIN,B., MONACO,V., BROXTERMAN,. B.,

SCHOEMAKER,.E., KAMPHUIS,. (1995), Synthe-sis of dipeptides containing a-substituted aminoacids; their use as chiral ligands in Lewis acid-ca-talyzed reactions,Recl. Trav. Chim. Pays-Ba s 114,

KAFTEIN, .,VANDOOREN,T.. G. M., BOESTEN,. H.J., SONKE, ., DUCHATEAU,. L. L. et al. (1998),Synthesis of 4-sulfur-substituted (2S,3R)-3-phenylserines by enzymic resolution. Enantiopu-re precursors for Thiamphenicol and Florfenicol,Org. Process Res. Dev. 2,lO-17.

KASAI,N., SAKAGUCHI,. (1992),An efficient syn-thesis of (R)-carnitine,Tetrahedron Lett. 33,1211-1212.

KASAI, ., TSUJIMURA,., UNOURA, ., SUZUKI,.(1990), Degradation of 2,3-dichloro-l-propanolby a Pseudomonas sp., Agric. Biol. Chem. 54,

KASAI,N., TSUJIMURA,., UNOURA, ., SUZUKI,.(1992a), Isolation of (S)-2,3-dichloro-l-propanol

assimilating bacterium, its characterization, andits use in preparation of (R)-2,3-dichloro-l-pro-panol and (S)-epichlorohydrin,J . Ind. Microbiol.10,3743.

KASAI,N., TSUJIMURA,., UNOURA, ., SUZUKI,.(1992b), Preparation of (S)-2,3-dichloro-l-pro-panol by Pseudomonas sp. and its use in the syn-thesis (R)-epichlorohydrin, J. Ind. Microbiol. 9,

KASAI,N., SUZUKI,., FURUKAWA,. (1998), ChiralC3 epoxides and halohydrins: Their preparationand synthetic application, J . Mol. C atal. B: Enzy -matic 4,237-252.

KAZLAUSKAS,. J., BORNSCHEUER,.T. (1998),Bio-transformations with lipases, in: BiotechnologyVol. 8a (REHM,H.., REED,G.,PUHLER,A., TAD-LER, . J.W., Eds.),pp. 37-191. Weinheim: John Wi-ley& Sons.

KLOOSTERMAN,., ELF ERIN K,~. M., IERSEL,. v.,ROSKAM,. H., MEIJER, .M. et al. (1988),Lipasesin the preparation of p blockers, Trends Biotech-

nol. 6,251-256.KOBAYASHI,., NAGASAWA,., YAMADA,. (1992),

231-238.

3185-3190.

97-101.




Enzymatic syntheses of acrylamide: a success

story not yet over, Trends Biotechnol. 10,402408.

KOBAYASHI,., IZUI, ., NAGASAWA,T.,YAMADA,.

(1993), Nitrilase in biosynthesis of the plant hor-

mone indole-3-acetic acid from indole-3-acetoni-

trile: Cloning of the Alcaligenes gene and site-di-

rected mutagenesis of cysteine residues, Proc.Natl. Aca d. Sci. US A 90,247-251.

KOBAYASHI,., SUZUKI,., FUJITA, ., MASUDA, .,

SHIMIZU,. (1995), Occurrence of enzymes in-

volved in biosynthesis of indole-3-acetic acid

from indole-3-acetonitrile in plant-associated

bacteria, Agrobacterium and Rhizobium, Proc.Natl. Aca d. Sci. US A 92,714-718.

LADNER, . E.,WHITESIDES,.M. (1984), Lipase-ca-

talyzed hydrolysis as a route to esters of chiral

epoxy alcohols,J.Am . Chem.SOC.106,7250-7251.

LAUMEN,., BREITGOFF, ., SEEMAYER,., SCHNEI-

DER, M. P. (1989), Enantiomerically pure cyclo-

hexanols and cyclohexane-1,2-diol derivatives,

chiral auxiliaries and substitutes for (-)-8-

phenylmenthol. A facile enzymic route, J . Chem.SOC.,Chem. Commun. 148-150.

MACKEITH, .A., MCCAGUE, ., OLIVO, . F., PAL-

MER,C. F., ROBERTS,. M. (1993), Conversion of

(- -4-hydroxy-2-oxabicyclo[3.3.0]oct-7-en-3-one

into the anti-HIV agent carbovir, J . Chem. SOC.,

Perkin Trans.I,313-314.

MACKEITH,R. A., MCCAGUE,R., OLIVO,H. F.,ROBERTS,. M.,TAYLOR,.J. C., XIONG, . (1994),

Enzyme-catalyzed kinetic resolution of 4-endo-

hydroxy-2- oxabicyclo[3.3.0]oct-7-en-3-one nd

employment of the pure enantiomers for the syn-

thesis of antiviral and hypocholestemic agents,

Bioorg. Med. Chem .2,387-394.

MACRAE, . R. (1983), Lipase-catalyzed interesteri-

fication of oils and fats, J . Am . Oil Chem. SOC.60 ,

MACRAE,A. ., HAMMOND,. C. (1985), Present and

future applications of lipases, Biotechnol. Genet.Eng. Rev. 3,193-217.

MATSUMAE,., FURUI,M., SHIBATANI,. (1993), Li-

pase-catalyzed asymmetric hydrolysis of 3-phenyl-

glycidic acid ester, the key intermediate in the

synthesis of diltiazem hydrochloride, J . Ferment.Bioeng. 75,93-98.

MATSUMAE,., FURUI,M., SHIBATANI,., TOSA,T.

(1994), Production of optically active 3-phenyl-

glycidic acid ester by the lipase from Serratia mar-cescens on a hollow-fiber membrane reactor, J .

Ferment. Bioeng. 78,5943.

MATSUMOTO, . (1992), Production of 6-APA, 7-

ACA, and 7-ADCA by immobilized penicillin

and cephalosporin amidases, in: Industrial Ap pli-cations of Imm obilized Biocatalysts (TANAKA,.,

TOSA, ., KOBAYASHI,., Eds.). New York: Marcel

Dekker.

MATSUO,., SAWAMURA,., HASHIMOTO,., HASHI-

291-294.

DA, w. (1981), The enzyme and method for

enzymatic transesterification of lipid, EuropeanPatent EP 0035883 (Fuji Oil Co.) (Chem . Abstr.96,4958).

MAZDIYASNI,., KONOPACKI,. B., DICKMAN,. A.,

ZYDOWSKY, . M. (1993), Enzyme-catalyzed syn-

thesis of optically-pure p-sulfonamidopropionic

acids. Useful starting materials for P-3 site modi-

fied renin inhibitors, Tetrahedron Lett. 34, 435-

438.

MCCAGUE,.,TAYLOR,.J. C. (1997), Four case stu-

dies in the development of biotransformation-

based processes, in: Chirality in Industry Vol. I1

(COLLINS, . N., SHELDRAKE, . N., CROSBY,.,

Eds) , pp. 183-206. Chichester:John Wiley& Sons.

MORGAN,., STOCKWELL,. R., DODDS, . R., AND-

RE W~,. R., SUDHAKAR,. R. et al. (1997),

Chemoenzymatic approaches to SCH 56592, a

new azole antifungal, J. Am . Oil Chem. SOC.74,

NAGASAWA,., YAMADA, . (1995), Microbial pro-

duction of commodity chemicals, Pure Appl.Chem.67,1241-1256.

NAGASAWA,., MATHEW,C. D., MAUGER,., YAMA-

DA,H. (1988), Nitrile hydratase-catalyzed produc-

tion of nicotinamide from 3-cyanopyridine in

Rhodococcus rhodochrous J1, Appl. Environ.Microbiol. 54,1761769.

NISHIZAWA,., GOMI,H., KISHIMOTO,. (1993), Pu-

rification and some properties of carboxylester-

ase from Arthobacter g lobijormis; stereoselective

hydrolysis of ethyl chrysanthemate, Biosci. Bio-tech. Biochem. 51,594-598.

NISHIZAWA, ., SHIMIZU, ., OHKAWA, ., KANAO-

KA, M. (1995), Stereoselective production of (+)-trans-chrysanthemic acid by a microbial esterase:

cloning, nucleotide sequence, and overexpression

of the esterase gene of Arthrobacter globiformis

in Escherichia coli, Ap pl. Environ. Micro biol. 61,3208-321 5.

OYAMA,K. (1992), Industrial production of asparta-

me, in: Chirality in Industry Vol. I (COLLINS,A..,SHELDRAKE,. N., CROSBY,.,Eds.), pp. 237-247.

Chichester: John Wiley & Sons.

PATEL,R. N., LIU,M., BANEWEE, ., SZARKA,. J.(1992), Stereoselective enzymatic hydrolysis of

(exo,exo)-7-oxabicyclo[2.2.1]heptane-2,3-dimeth-

anol diacetate ester in a biphasic system, Appl .Microbiol. Biotechnol. 37,180-183.

PATEL,R. N., BANERJEE,A.,O,R. Y., HOWELL,. M.,LI, W.3. et al. (1994), Enzymic preparation of

(3R-cis)-3-(acetyloxy)-4-phenyl-2-azetidinonea

taxol side-chain synthon, Biotechnol. Appl. Bio-chem. 20,23-33.

PHYTIAN, . J. (1998), Esterases, in: BiotechnologyVol. 8a (REHM,H. J., REED,G., PUHLER,A., TAD-

LER, . J. W., Eds.), pp. 193-241. Weinheim: Wiley-

VCH.

1361-1370.



5 References 293

QUAX, W. J., BROEKHUIZEN,. P. (1994), Develop-ment of a new Bacillus carboxyl esterase for usein the resolution of chiral drugs,Ap pl. Microbiol.Biotechnol. 41,425431.

QUINLAN,?, MOORE,S. (1993), Modification oftriglycerides by lipases: process technology andits application to the production of nutritionallyimproved fats,Inform 4,579-583.

REETZ,M. T., JAEGER, .-E. (1999), Superior bio-catalysts by directed evolution,Topic Curr. Chem.

ROZZELL,. D. (1999), Commercial scale biocataly-sis: myths and realities, Bioorg. Med. Chem. 7,

SAKSENA,A.., GIRIJAVALLABHAN,. M., LOVEY, .G., PIKE,R. E., WANG,H. et al. (1995), Highly ste-reoselective access to novel 2,2,4-trisubstitutedtetrahydrofurans by halocyclization: practicalchemoenzymic synthesis of SCH 51048, a broad-spectrum orally active antifungal agent, Tetrahe-dron Lett. 36,1787-1790.

SCHMID,. D., VERGER, . (1998), Lipases- nterfa-cial enzymes with attractive applications,Angew.Che m. (Int. Edn. E ngl.) 37,1608-1633.

SCHOEMAKER,. E., BOESTEN,W. H. J., KAPTEIN,.,Roos, E. C.,BROXTERMAN,. B. et al. (1996), En-zymic catalysis in organic synthesis. Synthesis ofenantiomerically pure Ca-substituted a-aminoand a-hydroxy acids,Acta Chem. Scand. 50,225-233.

SHELDON,. A. (1993), Chirotechnology - ndustrialSynthesis of Optically-Active Compounds. NewYork: Marcel Dekker.

SIH, . C. (1996),Application of immobilized lipase inproduction of Camptosar (CPT-ll), J . Am. Oi lChem. SOC.73,1377-1378.

STEAD,., MARLEY, ., MAHMOUDIAN,., WEBB,G.,NOBLE,D. et al. (1996), Efficient procedures forthe large-scale preparation of (lS,2S)-trans-2-methoxycyclohexanol, a key chiral intermediatein the synthesis of tricyclic p-lactam antibiotics,Tetrahedr0n:Asymmetry 7,2247-2250.

STEMMER,. l? C. (1994), DNA shuffling by random

fragmentation and reassembly:In vitro recombi-nation for molecular evolution, Proc. Natl. Acad .Sci. USA 91,10747-10751.

STINSON,. C. (1998), Counting on chiral drugs,Chem. Eng. News Sept. 21,83-104.

SUZUKI,T., ASAI, . (1991),A novel method for thegeneration of ( R ) - and (S)-3-chloro-1,2-propan-

ediol by stereospecific dehalogenating bacteriaand their use in the preparation of ( R ) -and (S)-glycidol,Bioorg. Med. C hem. Lett. 1,343-346.

SUZUKI,., KASAI,N., YAMAMOTO,., MINAMIURA,N. (1992), Isolation of a bacterium assimilating(R)-3-chloro-1,2-propanediolnd production of

(s)-3-chloro-1.2-~rooanediolsing microbial re-

ZOO, 31-57.

2253-2261.

SUZUKI, T., KASAI, ., YAMAMOTO, R., MINAMIURA,N. (1993), Production of highly optically active(R)-3-chloro-1,2-propanediol sing a bacteriumassimilating the (S)-isomer,Ap pl. Microbiol. Bio-technol.40,273-278.

SWANSON,! E. (1999), Dehalogenases applied to in-dustrial-scale biocatalysis,Curr. Opin . Biotechnol.

TAYLOR,.J. C., MCCAGUE,., WISDOM, ., LEE,C.,DICKSON, . et al. (1993), Development of thebiocatalytic resolution of 2-azabicyclo[2.2.1]-hept-5-en-3-one as an entry to single-enantiomercarbocyclic nucleosides, Tetrahedron:Asymmetry

TURNER,M. (1998), Perspectives in Biotransforma-tions, in: Biotechnology 2nd Edn., Vol. 8a (REHM,H.-J., REED,G., PUHLER, ., STADLER,., Eds.),pp. 15-17. Weinheim: Wiley-VCH.

TURNER,N.., WINTERMAN,. R., MCCAGUE,., PAR-RAIT, . ~.,TAYLOR,.J. C. (1995), Synthesis of ho-mochiral L-(S)-tert-leucine via a lipase catalyzeddynamic resolution process, Tetrahedron Lett. 36 ,

10,365-369.

4,1117-1128.

1113-1116.VA N DEN TWEEL,W. J. J., VAN DOOREN,T.. G. M., DE

JONGE,. H., KAPTEIN, ., DUCHATEAU,. L. L.,KAMPHUIS,. (1993), Ochrobactrum anthropiNCIMB 40321: a new biocatalyst with broad-spectrum L-specific amidase activity,App l. Micro-biol. Biotechnol. 39,296-300.

VA N DOOREN,. J. G. M., VANDEN TWEEL,W. J. J.(1992), Enzymic preparation of optically activecarboxylic acids, Eur. Patent Appl. EP494716(DSM) (Chem.Abstr . 117,210758).

VAN TOL, J. B. A., JONGEJAN,. A., DUINE,. A.(1995a), Description of hydrolase-enantioselec-tivity must be based on the actual kinetic mecha-nism: analysis of the kinetic resolution of glycidyl(2,3-epoxy-l-propyl) butyrate by pig pancreaslipase, Biocatal. Biotransform. U,9-117.

VAN TOL, . B. A., KRAAYVELD,. E., JONGEJAN,. A.,DUINE,.A. (1995b),The catalytic performance ofpig pancreas lipase in enantioselective transeste-rification in organic solvents, Biocatal. Biotrans-

form. 12,119-136.WAJANT, ., EFFENBERGER,. (1996), Hydroxynitrilelyases of higher plants, Biol. Chem. 377, 611-617.

WAJANT, H., MUNDRY,K. W., PFIZENMAIER,.(1994), Molecular cloning of hydroxynitrile lyasefrom Sorghum bicolor (L.).Homologies to serinecarboxypeptidases,Plant. Mol. B iol. 26,735-746.

WALTS,A. ., FOX, . M. (1990),A lipase fraction forresolution of glycidyl esters to high enantiomericexcess, US Patent 4923810 (Genzyme) (Chem.Abstr. 113,113879).

WIRZ,B., SOUKUP, M. (1997), Enzymatic preparationof homochiral2-isobutvl succinic acid derivatives.

solution,J. Ferment. Bioeng. 73,44?-448. Tetrahedron:Asymmetry 8,187-189.




WOOLLEY,., PETERSEN,. B. (1994), Lipases: TheirStructure, Biochemistry, and Application. Cam-bridge: Cambridge U niversity Press.

Wu, D. R., CRA M E R,. M., BELFORT,G. (1993), Kine-tic resolution of racem ic glycidyl bu tyra te using amultiphase membrane enzyme reactor: experi-ments a nd mod el verification, Biotechn ol. Bioeng.

YAGASAKI,., OZA KI, . (1998), Industrial biotrans-formations for the production of D-aminO acids,J . Mol. Catal. B: Enzymatic 4,1-11.

41,979-990.

ZIMMERMANN,. P., ROBINS,K. T., WERLEN, .,HOEKS, . W. J. M. M. (1997), Bio-transform ationin th e produ ction of L-carnitine, in: Chirality in In-dustry Vol. I1 (COLLINS, . N., SHELDRAKE,. N.,CROSBY,., Eds.), pp. 287-305. New Y ork: Jo hnWiley & Sons.

ZMIJEWSKI,. J., JR., BRIGGS,B. S.,THOMPSON,. R.,WRIGHT,. G. (1991), Enantioselective acylationof a p-lactam in termediate in the synthesis of lor-acarbef using penicillin G amidase, TetrahedronLett. 32,1621-1622.



7 Regioselective Oxidation of

Aminosorbitol with Gluconobacter

oxydans, Key Reaction in theIndustrial 1-Deoxynoj rimy cin

Synthesis

MICHAELCHEDELWuppertal, Germany

1 Introduction 2962 Nojirimycin, l-Deoxy nojirimycin, an d Miglitol 2973 Three Ro utes to Produc e l-Deoxynojirimycin 298

3.1 Extraction from Plants 2983.2 Ferm entation 2983.3 Chem ical Synthesis 299

4 Concept for a Combined Biotechnological-ChemicalSynthesis of l-Deoxyno jirimycin5 Oxidation of l-Amino-l-Deoxy-D-Sorbitol by Gluconobacter oxydans 3006 Biotransformation of N-Formyl-l-Amino-l-Deoxy-D-Sorbitoln the Industrial Scale 3037 Fermentative Production of Gluconobacter oxydans Cells on the Industrial Scale 3048 Conclusions and Outlook 306

9 References 307

299





296 7 Regioselective Oxidation ofAminosorbitol with Gluconobacter oxydans

1 Introduction

1-Deoxynojirimycin is the pre cursor t o the

poten t a-glucosidase inhibitor M iglitol, its N-substituted analog (GERARD et al., 1987;Scorr an d TATTERSALL, 1988). M iglitol waslaunched as a new therapeutic drug for thetreatment of non-insulin-dependent diabetesmellitus in 1998 (BERG MA N,999) in E urope,North America, and some other countries. Thestructures of 1-deoxynojirimycin an d of Migli-to1 are shown in Fig. 1.

The industrial production of l-deoxynoji-rimycin is a new example of a combined bio-technological-chemical synthesis carried outat a large scale. This synthesis integrates the

selectivity offered by an enzymatic reactioninto a sequence of chemical steps and resultsin an elegant, short and economical process(KINAST nd SCHEDEL,981). The biotechno-logical key reaction is the reg ioselective oxi-dation of 1-amino-1-deoxy-D-sorbitolo 6-amino-6-deoxy-~-sorbose.Whole cells of Glu-

conobacter oxydans having a high specific ac-tivity of the mem brane bound polyol dehy-

"OH D-glucose

OH

nojirimycin

'OH

H 1 deoxynojirimycinOH

HOH Miglitol

OH

Fig. 1.The structures of ~-glucose, ojirimycin, 1-deoxynojirimycin,and Miglitol.



2 Nojirimycin, 1-Deoxynojirimycin, and Miglitol 297

drogenase ar e used as catalyst. This central bi-otransformation step is flanked by four chemi-cal reactions.

The fruitful combination of biotechnology

and chem istry opens up new an d creative syn-thetic pathways. On e of the best know n exam -ples is the sy nthesis of vitam in C (REICHSTEINand GRUSSNER,934; KULHAN EK,970; BOU D-RANT,990). In this efficient process the oxida-tion of D-sorb itol to L -sorbose with Glucono-bacter oxydans (Acetobacter suboxydans) ispart of a five-step synthesis starting from D-

glucose. Afte r its publication by REICH STEINtwas for more than 60 years the method ofchoice - due to the short reaction sequence,the cheap availability of D-glucose as startingsubstrate, and the chemical stability of theintermediates. Various improvements havebeen introduc ed ove r the years.

The ind ustrial synthesis of l-deoxynojirimy-cin is similar to the vitamin C process. In bothcases D-glucose is the starting material a nd theability of G . oxydans to regioselectively oxi-dize straight chain polyhydric alcohols havinga well defined steric configuration is em-ployed. However, in th e vitamin C synthesis D-

sorbitol, a naturally occurring hexitol, is the

substrate of the microbial oxidation stepwhereas in the l-deoxynojirimycin case achemically modified polyol not found in na-tu re is used. Th e oxidation of such a terminallymodified polyol was possible because G. oxy-duns dem ands a specific steric configuration atone side of the substrate to be oxidized where-as structure and size of the other part of themolecu le are of little impo rtance.

The present chapter describes the develop-ment of the combined biotechnological-chem-ical l-deoxyno jirimycin synthesis, comp ares it

to other production alternatives, and sum-marizes relevant aspects of the oxidation ofl-amino-l-deoxy-D-sorbitol ith G. oxydansat the production scale.

The name Gluconobacter oxydans is usedaccording to the taxonomical classification de-scribed in Bergey’s Manual of Systematic Bac-teriology (DE LEYet al., 1984). n some earlierliterature this bacterium was named Gluco-nobacter suboxydans or Acetobacter suboxy-duns. If such literature is cited the name usedby the authors is given.

2 Nojirimycin,

1 Deoxynojirimycin, and

Miglitoll-Deoxynojirimycin is a natural compound

which occurs in plants an d in m oths feeding onsuch plants; it can also be found in fermenta-tion broths of various bacteria. Its isolationfrom natural sources was first described in apatent applied for by the Japanese companyNippon Shinyaku in 1976 (YAGIet al., 1976,1977;MU RA I t al., 1977); h e inventors namedthis strong a-glucosidase inhibitor M oranolineand claimed its extraction from Morus plants

and its use as an tidiabetic agent.Fig. 1 shows the structure of l-deoxynojiri-

mycin. It can be described as 1,5-dideoxy-1,5-imino-D-glucitol or as 5-hydroxymethyl-2,3,4-trihydroxypiperidine. This piperidinose showsstructural similarities to D-glucose; it has thesame steric configuration at four carbon at-oms, however, nitrogen replaces oxygen in thering. Ten years before l-deoxynojirimycin wasfound as a natural compound PAULSEN1966)already had published its synthesis from 6-

amino-6-deoxy-~-sorbose. he early work onl-deoxynojirimycin was closely related tostudies with its pare nt compound nojirimycin(Fig. 1) first isolated in 1965 as po ten t antibac-terial substance from the fermentation brothof Streptomyces roseochromogenes R-468 andlater also detected in the growth medium ofother Streptomyces and of Bacillus strains(NISHIKAWAnd ISHIDA, 965; INO UY E t al.,1966,1968; SHIDAt al., 1967a,b; ARGO UDELISand REUSSER,976; ARGOU DELISt al., 1976;SCHMIDTt al., 1979). l-Deoxy nojirimycin can

be prep ared from nojirimycin by catalytic hy-drogenation o r reduction with sodium borohy-dride (INOUYEt al., 1966).

Both nojirimycin and l-deoxynojirimycinstrongly inhibit a-glucosidases isolated fromthe intestinal tract of animals and m an (N IWAe t al., 1970; RE ESE t al., 1971; MU RA I t al.,

et al., 1979; KODA MAt al., 1985) and immedi-ately attracted the interest of various researchgroups. A s lead structures they offered a signif-icant potential for the development of new

therapeutic drugs to treat diabetes. Many se-

1977;FROMMER et a l., 1978a,b, 1979a;SCHMIDT



298 7 Regioselective Oxidation of Aminosorbitol with Gluconobacter oxydans

mi-synthetic N-substituted derivatives of l-de-oxynojirimycin have bee n prepared (JUN GE tal., 1979; MA TSU MU RAt al., 1979a) and Migli-tol, the N-hydroxyethyl analog, was identified

as one of the most interesting candidatesshowing a desired enzyme inhibitory profile(G ER ARD t al., 1987; SCOTT nd TATTERSALL,1988;JUNGEt a l., 1996).

Miglitol was develope d by Bayer as drug forthe treatment of type I1 (non-insulin-depen-den t) diabetes m ellitus. It was autho rized in1996 for marke ting in Europ e (trade name: Di-astabol@ ) n d the U nited S tates (Glysetm) andlaunched in 1998 in coopera tion with licensingpartners first in Germany and later in manyother countries including the United States,

Cana da, an d Australia. Miglitol is the third a-

glucosidase inhibitor to reach th e m arket - af -ter Aca rbose (GlucobayB, Precosea ), a secon-dary metabolite of pseudotetrasaccharidestructure produced by fermentation of Acti-noplanes sp. (SCHMIDTt al., 1977; CLISSO LDand EDWARDS,988;BISCHOFF, 994) and Vog-libose (Basen @ ), n N-substituted valiolaminederivative which is on the market in Japan andsome other Far Eastern countries (H ORII t al.,1982;NAKAMURAt al., 1993; OD AK A nd IKE-

DA, 1995).

3 Three Routes to Produce

1 Deoxynojirimycin

For the industrial production of Miglitol 1-deoxynojirimycin is needed as starting ma teri-al. There are th ree principal routes to prepa re

l-deoxynojirimycin (HUG HES and RUDG E,1994; STCJTZ,1999):

(1) Extraction from plants like the mul-

(2) Fermentation of various Bacillus or

(3) Chem ical synthesis following different

berry tree.

Streptomyces strains.

synthetic strategies.

3.1 Extraction from Plants

l-Deoxynojirimycin can be extracted withwater or polar solvents from the bark, the

roots, and the leaves of the mulberry tree(M oru s alba, M. bombycis, M . nigra) and alsofrom o ther plants including Jacobinia suberec-ta and J. tinctoria, the Chinese herbs Vespaenidus and Bombyx batryticatus, and two Eu-phorbiaceae species, h e rain forest tree Endo-spermum medullosum and the liana Omphaleaqueenslandiae (YAGI t al., 1976,1977;MATSU-MURA et al., 1980;EVANSt al., 1985; DA IGO tal., 1986;KITE t al., 1991;YAMADAt a l., 1993;SIMMONDSt al., 1999).The extraction of l -d e-oxynojirimycin from plant material as a ro ute

of production on the industrial scale has, how-ever, never seriously bee n regard ed as feasible.The content of l-deoxynojirimycin in plants islow a nd the purification process would not besimple as structurally very similar substancesare also present in these plants like nojirimy-cin and l-deoxymannojirimycin (D AIG O t al.,1986; AS AN O t al., 1994a,b).

Moreover, the sufficient and steady supplywould be difficult to establish, pose an im-mense logistic problem, and need the avail-

ability of sufficient land.

3.2 Fermentation

The ferm entative route offers an interestinglarge scale production alternative. A numberof Bacillus and Streptomyces strains werefound to synthesize l-deoxynojirimycin inappreciable amounts and secrete it into the m e-dium. SCHM IDTt al. (1979) and FROMMERtal. (1978a, 1979b ) isolated l-deoxyno jirimycin

from the culture broth of Bacillus amylolique-faciens, B. polym yxa , an d B. subtilis. FROMMERand SCHMIDT1980) claimed a fermentationprocess using Bacillus strains; they w ere ableto isolate 360 g of pur e l-deoxyno jirimycinbase after a multistep purification processstarting from 360 L of fermentation broth. 1-Deoxynojirimycin production using Strepto-myces lavendulae was reported by MATSUMU -RA et al. (1979b), MU RAO nd MIYATA 1980),in a patent of Amano Pharmaceuticals (SUMI-TA and MUR AO, 980), and by EZURE t al.

(1985). HA RD ICKt al. (1991,1992) and HAR-



4 Concept for a Combined Biotechnological-Chemical Synthesis of 1 -Deoxynojirimycin 299

DICK and HUTCHINSON1993) investigated thebiosynthesis of 1-deoxynojirimycin in Strepto-myces subrutilis and Bacillus sub tilis var. niger.

To produce 1-deoxynojirimycineconomical-

ly by fermentation a yield approaching 50gper liter would be necessary to reach a price ofless than 100US$ per kg. It does not seem un-realistic that this goal can be achieved as someof the wild type strains reported in the literatu-re already produced 1-deoxynojirimycin ingram quantities. However, an intensive geneticstrain development program and the optimiza-tion of the fermentation process over severalyears would be necessary.

3.3 Chemical Synthesis

There are many publications describing dif-ferent synthetic routes to 1-deoxynojirimycin;they were reviewed by NISHIMURA1991),HUGHES nd RUDGE(1994), JUNGE et al.(1996), and LA FERLA nd NICOTRA1999).The synthesis of 1-deoxynojirimycinpresents astereochemical challenge due to the presenceof four contiguous chiral centers and the samefunctional group occurring several times. The

various synthetic routes described in the litera-ture may be classified into three groups:

Starting material is D-glucose, the nitro-gen is introduced at C-1,the hydroxylgroup at C-5 is oxidized to the ketogroup, and 1-deoxynojirimycin sobtained by reductive ring closure. Anexample is the original synthesis pub-lished by PAULSENPAULSEN,966,1999;PAULSENt al., 1967).Starting material is again D-glucose, the

nitrogen is introduced at C-5 whilemaintaining the configuration presentin D-glucose followed by reductive ringclosure. This synthetic strategy was firstdescribed for nojirimycin by INOUEetal. (1968). The synthetic routes in thisand the first group take advantage ofthe fact that already three correct ste-reocenters needed in 1-deoxynojirimy-cin are contributed by the homochiralstarting material D-glucose.A number of other syntheses start from

building blocks which do or do not

have chiral centers like glucosylamine(BERNOTASnd GANEM,985),manno-se or idose derivatives (FLEET,989;FLEET t al., 1990), artaric acid (IIDA t

al., 1987), pyroglutamic acid (IKOTA,1989),or phenylalanine (RUDGE t al.,1994).An example of a chemo-enzy-matic process has been reported byWONGet al. (1995). In their synthesisdihydroxyacetone phosphate is stereo-selectively connected to 3-azido-2-hydroxypropan-a1 catalyzed by rabbitmuscle aldolase.

All synthetic routes are complex; most of themhave a relatively large number of individual

reaction steps and need a substantial amountof protection group chemistry. In some of thesyntheses proposed the lack of selectivity incrucial steps led to significant amounts of oneor more of the 15other possible stereoisomers.1-Deoxynojirimycin production by chemicalsyntheses on an industrial scale following theabove mentioned strategies was not economi-cal in those cases considered.

4 Concept for a Combined

Biotechnological-Chemical

Synthesis of

1 DeoxynojirimycinAn interesting concept for a combined bio-

technological-chemical synthesis was pro-posed by KINASTnd SCHEDEL1981) which is

shown in Fig. 2. It took advantage of the abilityof Gluconobacter oxydans to regio- and ster-eoselectively oxidize polyol substrates withhigh productivity according to the rule of BER-TRAND and HUDSONBERTRAND,904;HANNet al., 1938).The key reaction in the proposedthree-step synthesis was the regioselective mi-crobial oxidation of l-amino-l-deoxy-D-sorbi-

to1 to 6-amino-6-deoxy-~-sorbose. his crucialstep was flanked by well established and com-paratively simple chemical reactions: the re-ductive amination of D-glucose to give 1-ami-

no-1-deoxy-D-sorbitol and the stereoselective



300 7 Regioselective Oxidation ofArninosorbitol with Gluconobacter oxydans

NH2 ringclosure,CHO

OH reducpe NH2 GluconobacterHo{oH aminabon f: oxydans Hot:eduction)

HOO

OH

D-glucose 1 amino-D+ orbitd B-amino-~-so rbose 1deoxynojirimycin

Fig. 2. Concept for the combined biotechnological-chemical ynthesis of 1-deoxynojirimycin.

reductive ring closure of 6-amino-6-deoxy-~-sorbose to lead to 1-deoxynojirimycin.

The strategy of this synthesis was compar-able to the original synthesis by PAULSEN(PAULSEN,966; PAULSENt al., 1967): Use of

D-glucose as starting material which containsat carbon atoms 2, 3, and 4 three correctstereocenters, introduction of nitrogen at car-bon atom C-1, oxidation of the hydroxyl groupat C-5 followed by reductive ring closure. Useof th e selectivity of the polyo l dehydrogenaseenzyme of G . oxydans to oxidize l-am ino-l-deoxy-D-sorbitol made a sho rt reaction se -quence possible with no need for protectiongroup chemistry. Regarding the final stereo-selective ring closure and reduction step it was

known from the studies of PAULSEN t al.(1967) that the microbial oxidation product 6-amino-6-d eoxy-~-so rbose existed in solutionat alkaline pH -values in the piperidinose formand could readily be reduced with appropriatereducing agents like sodium boroh ydride (KO-BERNICK, 1982) to give 1-deoxynojirimycin.

5 Oxidation of 1-Amino-l-

Deoxy-D-Sorbitol by

Gluconobacter oxydans

1-Amino-1-deoxy-D-sorbitol,the substrateof the microbially catalyzed key reaction, is aterminally modified polyol. For the success ofthe proposed 1-deoxynojirimycin synthesis itwas essential that the introduction of the ami-no group at carbo n atom C-1 did not influencethe capability of Gluconobacter oxydans tooxidize the hydroxyl group a t carbon a tom C-5according to the Bertrand-Hudson rule (BER-

TRAND , 904; HA" et al., 19 38). This ru le

states that G . oxydans regioselectively oxidizesthe m iddle of thre e termina l hydroxyl groupsif they have the D-erythro configuration. Ac-cording to the literature th e size and structureof the remaining part of the molecule had in

the m ajority of cases little o r no effect on th eoxidation capability of Gluconobacter (KUL-HANEK,989).There are mo re than 30 publica-tions demonstrating the successful oxidationof the penu ltimate hydroxy l grou p of polyolscarrying the favorable Bertrand-Hudson con-figuration a t one end of the m olecule but a rechemically modified o r differ in c arbon chainlength or configuration at the o ther en d; somerelevant publications are summarized in Tab.1. BERTRA ND 1898, 1900), FULMER and

UNDERKOFLER1947), HANN t al. (1938),HANN n d HUDSON1939), ONISHI nd Su zu-KI (1969), MOSES nd FERRIER1962), BER -

MACLAY t al. (1942), ETTELand LIEBSTER(1949), PRATT and RICH TMY ER1955), andothers oxidized polyols with different c arbonchain length ranging from glycerol to heptitolsand o ctitols. Particularly th e oxida tion of sev-eral heptitols showed that the configurationoutside the Bertrand-Hudson group was of lit-tle effect on the oxidation reaction (PRAT T nd

RICHTMYER,955). It is of interest tha t even acyclitol (inositol) could be oxidized (D U N N IN Get al., 1938;CHARGAFF nd M AGA SAN IK,946).JONES and MITCHELL 1959), JONES t al.(1962), and HO UG H t al. (1959) used Aceto-bacter suboxydans for the oxidation of modi-fied pentitols and B OSSHAR Dnd REICHSTEIN(1935), HO UGH et al. (1959), JONESt al.(1961a, b), KU LHANE Kt al. (1977), BU DE S~N -s a t al. (1984), and TIWA RI t al. (1986) re-ported the oxidation of several chemicallymodified hexitols. In these studies various de-

oxy-, deoxyamino-, deoxyhalogen- an d deoxy-

TRAND and NITZBERG1928), TILDEN 1939),



5 Oxidation of I-Amino-1-Deoxy-D-sorbitolby Gluconobacter oxydans 301

Tab. 1.Oxydation of Various Polyols According to the Rule of BERTRANDnd HUDSON

~______

Polyol(s) Oxidized References

Glycerol

meso-Erythritol

D- Arabitol

Ribitol

D-Sorbitol

D- Allitol

D-Mannitol

D-Fructose

meso-, -, d-, epi-Inositol

alpha-Glucoheptit

D-alpha-Mannoheptitol (perseitol)

D-alpha, beta-Glucooctitol

Other polyols

D-Manno-D-gala-heptitol (perseitol)

L-Gluco-L-gulo-heptitol (alpha-glucoheptitol)

Volemitol (alpha-sedoheptitol)

D-Gluco-D-ido-heptitol

beta-Sedoheptitol

(D-altro-D-gluco-heptitolor L-gulo-D-talo-heptitol)

meso-Glycero-allo-heptitolD-Glycero-D-altro-heptitol

D-Erythro-D-galacto-octitol

2-Deoxy-D-erythro-pentitol

1-Deoxy-1-S-ethyl-D-arabitol

~-Sorbitol-3-methyl-etherI-Deoxy-D-glucitol

6-Deoxy-~-allitol

2-Deoxy-~-sorbitol

1-Deoxy-1-S-ethyl-D-glucitol1-Deoxy-1-S-ethyl-D-arabitol

1-Deoxy-1-S-ethyl-D-galactitol

2-Amino-2-deoxy-~-glucitol

2-Acetamido-2-deoxy-~-glucitol5-Deoxy-~-ribitol

6-Deoxy-~-galactitol

5-O-Methyl-~-ribitol

6-O-Acetyl-~~-galactitol

6-Deoxy -6-S-ethyl-L-galactitol

1-0-Acetyl-DL-galactitol2-Acetamido-2-deoxy-~-glucitol1-Deoxy-1-N-methylacetamido-D-glucitol

2-Acetamido-l,2-dideoxy-~-glucito1

2-Deoxy-2-fluoro-~-glucitol3-Deoxy-3-fluoro-~-mannitol2-Deoxy-~-arabino-hexitol

L-Fucitol (6-deoxy-~-galactito1)D-Rhamnitol

omega-Deoxy sugar alcohols

BERTRAND1898)

BERTRAND1900)

WHISTLERnd UNDERKOKER1938)

FULMERnd UNDERKOFLER1947)

Hu e t al. (1965)

HANN t al. (1938)

ONISHInd SUZUKI1969)

MOSESnd FERRIER1962)

BERTRAND1904)

CARR t al. (1968)

CUSHINGnd DAVIS1958)

AVIGARDnd ENGLARD1965)

DUNNINGt al. (1938)

CHARGAAFnd MAGASANIK1946)BERTRANDnd NITZBERG1928)

HA" et al. (1938)

HANN nd HUDSON1939)

TILDEN1939)

MACLAYt al. (1942)

E ~ Lnd LIEBSTER1949)

STEWARTt al. (1949)

PRAI-I et al. (1952)

STEWARTt al. (1952)

PUTT and RICHTMYER1955)

CHARLSONnd RICHTMYER1960)

LINEK t al. (1979)

JONES nd MITCHELL1959)

BOSSHARDnd REICHSTEIN1935)

KAUF M ANN and REICHSTEIN1967)

FULMERnd UNDERKOFLER1947)

HOUGHt al. (1959)

JONES t al. (1961a)

JONES t al. (1961b)

KULHANEKt al. (1977)

TIWARIt al. (1986)

RICHTMYERt al. (1950)ANDERSONnd LARDY1948)

BOLLENBACKnd UNDERKOFLER1950)

BUDESfNS& et al. (1984)




sulfuralditols were employed as substrates.ANDERSONnd LARDY 1948), RICHTMY ERtal. (1950), and BOLLENBACKnd UNDER-KOFLER (1950) oxidized w-deoxy sugars and

HOUGH t al. (1959) in addition w-0-methyl-and w-deoxy-acetyl sugars. In these polyols thehydroxyl group a t the third to last carbon a tomwas oxidized. To stay in accord with the Ber-trand-Hudson rule the authors regarded theterminal methyl group as equal to hydrogen,i.e., the terminal two-carbon groupCH,-CH OH was considered as a slightly ex-tended CH,OH group.

All the ab ove mentioned and in Tab. 1 sum-marized polyols could be oxidized accordingto the rule of BETRANDnd HUDSONith G.

oxyduns. In some studies the yield reportedwas, however, low an d the fermen tation timeneeded long (up to two or three weeks). Earli-er reviews summarizing the oxidation of poly-hydric alcoh ols by G. oxyduns were written byFULMER nd UNDERKOFLER1947), JANKE(1960), SPENCE Rnd GOR IN 1968), and KUL-HANEK (1989).

Exam ples of polyhydroxy compounds w hichwere not oxidized by G. oxyduns although theyhave the favorable Bertrand-Hudson configu-

ration are D-glucose dimethylacetal (ISELIN,1948), D-mannose diethylmercaptal (HA”et al., 1938), the 1,l-diethyldithioacetals ofD-arabinose, D-glucose, and D-galactose andthe 1 l-dibenzyldithioacetal of D-arabinose(HO UG H t al., 1959). These results indicatethat th ere are ad ditional factors which are rel-evant for the oxidation of ope n chain polyols.

1-Amino-1-deoxy-D-sorbitol contains theBertrand-Hudson-configuration and couldreadily be oxidized to 5-amino-5-deoxy-~-sor-bose employing various G. oxyduns strains

(KINAST nd SCHEDEL ,980a). The oxidationproduct 6-amino-6-deoxy-~-sorbose as, how-

ever, not stable in water a t near neu tral pHunder the conditions of the biotransformationreaction: spontaneous ring closure occurredfollowed by removal of water leading to

3-hydroxy-2-hydroxymethyl-pyridineFig. 3 ) .PAULSENt al. (1967) had rep orted that und ertheir conditions the pyridine formation from6-amino-6-deoxy-~-sorbosewas quantitativeaf ter five days. D ue to th e loss of 6-amino-6-deoxy-L-sorbose the necessary high l-deoxy-nojirimycin yield could not be obtained. To

prevent the spontaneous intramolecular ringclosure reaction the amino group had to beprotected. Various protection groups couldsuccessfully be emp loyed. In a series of p atentsdifferent protection groups were claimed

which could be removed either by hydrogeno-lytic cleavage (KINA ST nd SCH EDEL ,980b),under alkaline o r acidic conditions (K INAST tal., 1982; SCHRO DERnd STLJBBE,987), or en-zymatically (SCHU IT, 1993). Aston ishingly,many of these N -protected polyols were read i-ly oxidized by G. oxyduns even if the protec-tion group used - ike the benzyloxycarbonylgroup -was large and significantly owered thewater solubility of the substrate to be em-ployed.

In the reaction sequence of the combinedbiotechnological-chemical synthesis of 1 de-oxyno jirimycin two add itional steps were thusincluded (Fig. 4): Introduction of a favorableprotection grou p before the oxidation step andits removal afterwards prior to reduction.Small and hydrophilic groups like formyl-,dichloroacetyl-, or trichloroacetyl- whichcould be cleaved off under acidic or alkalineconditions were prefe rred. All further w ork tobe described in the following sections wasdone w ith N-formyl-l-amino-l-deoxy-o-sorbi-

to1 as exam ple.

OH

I-amino-o-sorbitd 6-amino-~-sorbose3-hydroxy-2-hydroxymethyl-

pyridine

Fig. 3. Formationof 3-hydroxy-2-hydroxymethyl-pyridinero m 6-amino-6-deoxy-sorbose.



6 Biotransformation of N-Formyl-I-Amino-1-Deoxy-D-Sorbitolt Industrial Scale 303

OH

OH OH OH OH OH

ldeoxynoj ir imycinamino-~~orbose 6-smino-L+orbosel+mlno-D+orbitol ’*minO-DBorbitd

N-protected N-protectedglUCOSS

Fig. 4. The combined biotechnological-chemical synthesis of 1 deoxynojirimycin.

6 Biotransformation of

N-Formyl-1-Amino-1-

Deoxy-D-Sorbitol on the

Industrial ScaleThe large scale industrial production of

L-sorbose as part of the vitamin C process iscarried out with Gluconobacter oxyduns cellsgrowing on D-sorbitol under fermentationconditions (ROSENBERGt al., 1993).N-Formyl1-amino-1-deoxy-D-sorbitolould, however,not be oxidized in the same way: This modifiedpolyol did not promote growth of G. oxydanswhen added in high concentrations instead of

D-sorbitol as sole polyol substrate to a yeastextracthalt medium. The same result was ob-tained when other protecting groups were em-ployed. N-Formyl-amino-1-deoxy-D-sorbitoleven inhibited growth in a concentration de-pendent manner when added together with D-

sorbitol. The biotransformation reaction had,therefore, to be carried out as a two-step pro-cess (SCHEDEL,997): In the first step G. oxy-duns biomass was produced by fermentationon a “good” growth substrate, preferably D-

sorbitol, separated from the broth and stored

under cooling as slurry in a stirred tank. In thesecond step N-formyl-1-amino-1-deoxy-D-sor-bitol which had been dissolved in water wasoxidized with G. oxydans biomass as catalystunder “resting cell” conditions with intensiveaeration and at a slightly acidic pH value. Theproductivity of the microbial oxidation reac-tion using G. oxydans strain DSM 2003 wasvery high: 250 kg of substrate per m3 couldquantitatively be oxidized in less than 16 hwith a biomass concentration of about 2.5 kgcell dry weight per m3. The specific oxidation

activity was around 50 mmol of substrate oxi-

dized per hour and g cell dry weight.A typicaltime course is shown in Fig. 5.

It is of interest to note that under these con-ditions resting G. oxydans cells oxidized D-sor-bitol, N-formyl-1-amino-1-deoxy-D-sorbitol,or other polyols completely uncoupled: The

electrons removed from the substrate weretransferred to oxygen without conservation ofmetabolically utilizable energy; only heat wasproduced. The non-existence of a respiratoryregulation was the prerequisite for this indus-trial biotransformation process employingresting cells suspended simply in water. Theuncoupled dehydrogenation of monosaccha-rides by G. oxydans had been pointed out byKULHANEK(1989) and USPENSKAYAndLOITSYANSKAYA1979).

Four main aspects of the N-formyl-l-amino-1-deoxy-D-sorbitolbiotransformation reactionwere important on the production scale underreaction conditions as described above, i.e., aGluconobacter biomass concentration ofabout 2.5 kg m P3 and a substrate concentra-tion of 250 kg m-3:

(1) High oxygen demand (about 6 kg0, mP3h-’).

(2) Removal of ihe process heat (about

(3) Labor intensive handling of large quan-

(4) Stability of the G . oxydans biocatalyst.

10KW m-3).

tities of solid substrate.

Regarding point (1) and (2) an adequatelyequipped stirred fermenter was necessary tomeet the high oxygen demand and to effec-tively remove the heat which was produced bythe biological reaction and introduced by thestirrer. To cope with the solid substrate hand-ling - about 10 t of material had to be filledinto the fermenter within a reasonably short

period of time - an automated bigbag delivery



304 7 Regioselective OxidationofAminosorbitol with Gluconobacter oxydans

300

55 250P

100

O i

0 1 2 3 4 5 6 7 8 9 1 0 1 1 1 2 1 3

Time [h]

100

80

60

40

20

0

0 1 2 3 4 5 6 7 8 9 1 0 1 1 1 2 1 3

Time m]

system to charge the fermenter was installed.Due to instability problems after prolongedstorage it was not possible to supply N-formyl-1-amino-1-deoxy-D-sorbitols a concentratedsolution which could be pumped from a stor-age tank to the medium preparation tank. Theactivity loss of G . oxydans cells under “restingcell conditions” limited their reuse and alsothe possibilities to develop an immobilized

biocatalyst. After one biotransformation cycleGluconobacter cells had lost about half of theirspecific activity (SCHEDEL, 997). Successfulimmobilization was also hindered by the veryhigh oxygen demand and the enormous sub-strate turnover rate which caused a severetransport limitation within immobilizationpearls. The oxidation of D-sorbitol toL-sorbose with immobilized Gluconobactercells had repeatedly been reported in the liter-ature (SCHNARRt al., 1977;STEFANOVAt al.,1987; TRIFONOVt al., 1991; KOSSEVA t al.,

1991).

Fig. 5. Time course of the oxidation ofN-formyl-1-amino-1 deoxy-D-sorbitolonthe production scale with Gluconobacteroxydarts DS M 2003.

7 Fermentative Production

o f Gluconobacter oxydans

Cells on the Industrial Scale

The fermentative production of Glucono-bacter oxydan s cells was carried out in a yeastextracthalt medium containing D-sorbitol;

other oxidizable substrates like glycerol couldalso be used. Fig.6 gives a typical time courseof a fermentation on the 30 m3 scale with G.oxydans strain DSM 2003 on D-sorbitol.The fi-nal biomass yield was low: about 5 g cell dryweight per liter. Only a very small portion ofthe D-sorbitol was used as carbon source by G.

oxydans DSM 2003and flowed into the metab-olism, about 95% of the substrate was oxidizedto L-sorbose and deposited in the medium.Correspondingly, the respiratory coefficientfor such a “tight” Gluconobacter strain was

low throughout the fermentation and ranged



7 Fermentative Production of Gluconobacter oxydans Cells at Industrial Scale 305

Fig. 6. Time course of the Gluconobacter oxy-duns fermentation on the production scale(strain D SM 2003).

around a value of 0 . 1 5 . A ~eported in the liter-atu re th e specific oxidation activity was high-est at the entry to the stationary phase (BAT-ZING and CLAUS,1973; WHITE and CLAUS,1982).Therefore, when the D-sorbitol substratewas used up and the biomass yield no longerincreased, the cells were separated from themedium and stored as a cooled slurry.

The Gluconobacter biomass productioncould successfully be run at the 30 L laborato-

ry scale as a continuous process under chem o-stat conditions w ith D-sorbitol limitation; mo rethan 25 volumes of medium were passedthrough the ferm enter with no loss in specificoxidative activity of the Gluconobacter cellsobtained (SCHEDEL,997). There are severalreports in the literature that the L-sorbose fer-mentation using Gluconobacter was carriedout continuously (see, for instance: ELSWORTHet al., 1959; MULLER, 966;NICOLSKAYAt al.,1984). In the pharmaceutical industry, how-ever, continuous fermentations to be run at

large scale under completely contamination

0 5 10 15 20 25 30

Time [h]

200

150

100

50

0

0 5 10 15 20 25 30

Time [h]

free conditions are very rarely the system ofchoice. The re are several reasons for this: In-creased risk of non-sterility, possible geneticinstability of the production organism duringlong-term fermentation necessitating exten-sive analytical characterization work to assureconstant p rodu ct quality, high investment intostorage tanks upstream and downstream fromthe fermen tation itself. The p roduction of G l u -

conobacter biomass has, however, a goo d chance

to become one of the very few large scale con-tinuous fermentations in the pharmaceuticalarea. This is m ainly due to the fact that no sig-nificant additional investment will be neededneither for the continuous preparation of me-dium o r its storage under sterile conditions norfor the harvest and storage of large volumes of

product. The medium can be continuouslymixed together from three liquid streams (wa-ter, D -sorbitol syrup, concentrated basal m edi-um containing salts and yeast ex tract; see Fig.7) and passed via a continuous sterilizer into

the ferm enter.The microbial biomass is a small



306 7 Regioselective OxidationofAm inosorbitol with Gluconobacteroxydans

sobitol sv r uD1>El--++

medium'

continuous Iater Isterilizer

supernatant(waste water)

+ iomass s lu r ry(product)

*concentratedyeast extractlsalt

production fermenter continuousseparator

Fig. 7. Concept for the continuous fermentation of Glucon obacter oxydans on the production scale underchem ostat conditions.

volume product which is harvested continu-ously with an existing separator; the culturesupernatant, the large volume side product isdirectly passed to the wastewater plant with-out intermediate storage. The existing batchprocess plant needs, therefore, only minoramendments to set up a fully continuous Glu-conobucter fermentation.

The enzymatic activity for the oxidation ofN-protected 1-amino-1-deoxy-D-sorbitols lo-cated in the membrane fraction. Kinetic stud-ies indicated close similarity between D-sorbi-

to1 and N-formyl-1-amino-1-deoxy-D-sorbitoloxidation. It is likely - but not yet proved -that both substrates are oxidized by the sameenzyme which belongs to either of the twotypes of membrane bound D-sorbitol dehydro-genases described for G . oxydans: The cyto-chrome containing three-subunit enzyme witha total molecular weight of about 140,000Dadescribed by SHINAGAWAt al. (1982) andCHOIet al. (1995) or the one-subunit enzymewith a molecular weight of around 80,OOO Dareported by HOSHINOt al. (1996). The mem-

brane bound polyol dehydrogenases do not

need cofactors like NAD or NADP as report-ed for the sorbitol dehydrogenases which hadbeen isolated from the soluble fraction byCUMMINSt al. (1957) or HAHN t al. (1989).

8 Conclusions and

Outlook

The combined biotechnological-chemicalsynthesis of 1-deoxynojirimycin is meanwhilea well established process on the industrialscale which enables the economic productionof this precursor to Miglitol. Combination ofbiotechnology and chemistry has led to a shortand elegant synthesis with little protectiongroup chemistry.

Regarding potential safety risks and envi-ronmental aspects the central biotransforma-tion step has significant advantages: The mi-crobially catalyzed polyol oxidation is run

under physiological conditions and - apart



9 References 307

ration of enantiomerically pure cyclic aminoaldi-

tols, total synthesis of 1-deoxynojirimycinand 1-

deoxymannojirimycin, Tetrahedron Lett. 26,

BERTRAND, . G. (1898), Action de la bactCrie du

sorbose sur les alcools plurivalents, C. R. Acad.Sci. 126,762-765.

BERTRAND,. G. (1900), Sur l’oxidation de 1’Crythri-

te par la bactCrie du sorbose; production d’un

nouveau sucre: YCrythrulose,C. R. Acad. Sci. 130,

1330-1333.

BERTR AND, . G. (1904), Etude biochimique de la

bactCrie du sorbose,Ann. Ch im. Phy s. 3,181-288.

BERTRAND, ., NITZBERG, . (1928), La fonction

cktonique de l’a-glucoheptulose,C. R. Acad. Sci.

BISCHOFF,. (1994), Pharmacology of a-glucosidase

inhibition, Eur.J. Clin. Invest. 24,3-10.

BOLLENBACK,. N., UNDERKOFLER,.A. (1950),The

action of Acetobacter suboxydans upon wdesoxy

sugar alcohols,J.Am . Chem. SOC.72,741-745.

BOSSHARD,W., REICHST EIN,. (1935), l-Sorbose-4-

methylather, Helv. Chim .Acta 18,959-961.

BOUDRANT,. (1990), Microbial processes for ascor-

bic acid biosynthesis: a review,Enzym e M icrob.Technol.U,22-329.

BUDESfNSK?, M., C E R ~ ,., DOLEZALOVA,., KUL-HANEK,M.,PAcAK,.,TADRA,M.1984),4-Deoxy-

4-fluoro-~-fructose: Preparation and structure,

Coll. Czech. Chem. Commun.49,267-274.

CARR,. G., COGGINS,.A., HOUGH,.,STACY, . E. ,WHITING,. C. (1968), Microbiological oxidation

of allitol to L-ribo-hexulose by Acetomonas oxy-duns, Phytochemistry 7 , 1 4

CHARGAFF,., MAGASANIK,. (1946), Oxidation of

stereoisomers of the inositol group by Acetobac-ter suboxydans,J. Biol. Chem .165,379-380.

CHARLSON,. J., RICHTMYER,.K . (1960), The iso-

lation of an octulose and an octitol from natural

sources: D-glycero-D-manno-octulose and D-

erythro-D-galacto-octitol from the avocado and

D-glycero-D-manno-octulose from Sedum spe-

cies,J.Am . Chem. SOC.82,3428-3434.

CHOI, .-S., LEE,E.-H.,RHEE, .-K. (1995), Purifica-

tion of a membrane-bound sorbitol dehydrogen-

ase from Gluconobacter suboxydans, FEMSMicrobiol. Letters U5,45-50.

CLISSOLD,. P., EDWARDS,. (1988), Acarbose, a

preliminary review of its pharmacodynamic and

pharmacokinetik properties, and therapeutic po-

tential, Drugs 35,214-243.

CUMMINS,. T., CHELDELIN,. H., KING,T. S. (1957),

Sorbitol dehydrogenases in Acetobacter suboxy-dam, J . Biol. Chem.226,301-306.

CUSHING,. B., DAVIS,R . V. (1958), Purification of

fructose on ion exchange resins,J.Am. Pharm. As -

DAIGO,., INAMORI,Y.,TAKEMOTO,T.1986), Studies

1123-1126.

134,1172-1175.

SOC. 47,765-768.

from acids and bases needed for pH-control -involves no toxic, easly inflammable, or other-wise hazardous substances. The microorgan-ism used is safe. All organic components are

readily degradable.Compared to a fully chemical synthesis,

however, a significant investment is necessaryinto a large fermenter with state-of-the-arttechnology.A working volume of 30,50,or 100m3 is generally not standard in chemical pro-duction. The Gluconobacter fermentation stephas to be run under completely contaminationfree conditions, a challenge to the equipmentand the qualification of the operators. TheGluconobacter fermentation produces a largevolume wastewater stream with a high biolog-

ical oxygen demand. If the fermentation is tobe run continuously the capacity of thewastewater plant will soon set the limit to theoverall process.

9 References

ANDERSON,., LARDY, . A. (1948), The synthesis of

D-fructomethylose by biochemical oxidation, J.

Am . Chem. SOC.70,596597.

ARGOUDELIS,. D., REUSSER,. (1976), Process for

producing antibiotic U-51640, US Patent 3998698.

ARGOUDELIS,. D., REUSSER,., MIZSAK, . A . ,BACZYNSKYJ,. (1976), Antibiotics produced by

Streptomyces ficellus 11. Feldamycin and noji-

rimycin, J.Antibiot. 29,1007-1014.

ASANO,N., SEKI, ., TOMIOKA,., KIZU,H., MAT-S U I , K. (1994a), N-Containing sugars from M o w

alba and their glycosidase inhibitory activities,

Carbohydr. Res.259,243-255.

ASANO,N., OMIOKA, ., KIZU,H., MATSUI,K.(1994b), Sugars with nitrogen in the ring isolated

from the leaves of Morus bombycis, Carbohydr.Res.253,235-245.

AVIGARD,., ENGLARD,. (1965),5-Keto-D-fructose.

I. Chemical characterization and analytical deter-

mination of the dicarbonyl hexose produced by

Gluconobacter cerinus, J. Biol. Chem. 240,229C

2296.

BATZING,. L., CLA US, . W. (1973), Fine structural

changes of Acetobacter suboxydans during

growth in a defined medium, J. Bacteriol. 113,

BERGMAN, . D. (1999), Major new drugs. Part 2,

BERNOTAS,. C., GANEM, . (1985), Efficient prepa-

1455-1461.

Cornmun. Pharm. 90,3539.




on the constituents of the water extract of theroot of mulberry tree (Morus bombycis KOIDZ),Chem. Pharm. Bull. 34,2243-2246.

DE LEY, ., GILLIS,M., SWINGS,. (1984), Family VI.Acetobacteraceae, in:Bergey s Manual ofSystem-atic Bacteriology, Vol. 1 (KRIEG, . R., HOLT, . G.,Eds.), pp. 267-278. Baltimore: William and Wil-kins.

DUNNING,. W., FULMER, . I., GUYMON,. F., UN-DERKOFLER, L. A. (1938), The growth and chemi-cal action of Acetobacter suboxydans upon i-ino-sitol,Science 87,72.

ELSWORTH,., TELLING, . C., EAST,D. N. (1959),The investment value of continuous culture, J.

Ap pl. Bacteriol. 22,138-152.EITEL, V., LIEBSTER,. (1949), Oxydation biochi-

mique de la volCmite, Coll. Czech. Chem. Com -mun.40,80-90.

EVANS,.V.,FELLOWS,. E., SHING,. K. M., FLEET,G. W. J. (1985), Glycosidase inhibition by plant al-kaloids which are structural analogues of mono-saccharides,Phytochemistry 24,1953-1955.

Ezu RE, ~., ARUO,.,MIYAZAKI,., KAWAMATA,.(1985), Moranoline (1-deoxynojirimycin) fer-mentation and its improvement, Agric. Biol.Chem.49,1119-1125.

FLEET,G.W. J. (1989), Homochiral compounds fromsugars,Chem . in B ritain 25,287-292.

FLEET,G. W. J., CARPENTER, . M., PETURSSON,.,RAMSDEN,. G. (1990), Synthesis of deoxynoji-rimycin and ofnojirimycin Slactam, TetrahedronLett .31,409-412.

FROMMER,., SCHMIDT,. D. (1980), 1-Desoxynoji-rimycin-Herstellung,Germ an Patent Offenlegung2907 190.

FROMMER,., MULLER, ., SCHMIDT,. D., PULS,W.,KRAUSE, .-P. (1978a), Inhibitoren fur a-Gluc-osidasen,Germ an Patent Offenlegung2658561.

FROMMER,., MULLER, ., SCHMIDT,. D., PULS,W.,KRAUSE, .-P., HEBER,U. (1978b), Verfahren zurHerstellung von Saccharase-Inhibitoren, die imVerdauungstrakt wirksam werden, aus Bacillen,D E 2658563.

FROMMER,., MULLER, ., SCHMIDT,. D., PULS,W.,KRAUSE, . 2 (1979a), Inhibitoren fur a-Gluc-osidasen Germ an Patent Offenlegung 2726898.

FROMMER,., MULLER, ., SCHMIDT,. D., PULS,W.,KRAUSE, .-P. (1979b), Inhibitoren fur Glykosid-Hydrolasen aus Bacillen,Germ an Patent Offenle-gung 2726899.

FULMER,. I., UNDERKOFLER,. A. (1947), Oxida-tion of polyhydric alcohols by Acetobacter sub-oxyda ns, Zowa State Coll.J.Sci.21,251-270.

GERARD,., HILLEBRAND,., LEF&BVRE,. J. (1987),Assessment of the clinical efficacy and toleranceof two new a-glucosidase inhibitors in insulin-

treated diabetics, Int. J. Clin. Pharmacol. 25,483-488.

HAHN,H., HEETHOFF,M., LIST,H.-J., GASSEN, . G.,JANY,K. D. (1989), Polyol-dehydrogenase fromGluconobacter oxydans: Identification, cloningand expression of the NADH-dependent sorbitoldehydrogenase 11, DECHEMA BiotechnologyConferences 3, pp. 345-349. Weinheim: VCH.

HA", R. M., HUDSON, . S. (1939), Proof of thestructure and configuration of perseulose (L-gala-heptulose), J.Am . Chem. SOC.61,336-340.

HA", R. M.,TILDEN,.B., HUDSON,.S. (1938),Theoxidation of sugar alcohols by Acetobacter sub-oxydans,J.Am . Chem.SOC.60,1201-1203.

HARDICK, . J., HUTCHINSON,. W. (1993), The bio-synthesis of 1-deoxynojirimycin n Bacillus subti-lis var. niger, Tetrahedron49,6707-6716.

HARDICK,. J., HUTCHINSON,. W.,TREW,.J., WEL-LINGTON,. M. H. (1991), The biosynthesis ofdeoxynojirimycinand deoxymannonojirimycin nStreptomyces subrutilis, J . Chem. SOC., Chem.Commun. 729-730.

HARDICK,. J.,HUTCHINSON, . W.,TREW, . J.,WEL-LINGTON, E. M. H. (1992), Glucose is a precursorof 1-deoxynojirimycin and l-deoxymannonoji-rimycin in Streptomyces subrutilis, Tetrahedron48,6285-6296.

HORII, . , KAMEDA,., FUKASE, . (1982), N-substi-tuted pseudo-aminosugars, their production anduse, European Patent 56194.

HOSHINO,T.,JIMA, .,SUGISAWA,T.1996), D-Sorbi-to1 dehydrogenase, European Patent 728840.

HOUGH, ., JONES,. K. N., MITCHELL,. L. (1959),The oxidation of some terminal-substituted poly-hydric alcohols by Acetobacter suboxydans, Can.J. Chem.37,725-730.

Hu, C. L., MCCOMB,.A., REN DIG ,~.. (1965), Iden-tification of altro-heptulose and L-threitol as pro-ducts of meso-erythritol oxidation by Acetobactersuboxydans, Arch. Biochem. Biophys. 110, 35CL353.

HUGHES, . B., RUDGE, . J. (1994), Deoxynojirimy-cin: Synthesis and biological activity, Nut. Prod.Rep. 11,135-162.

IIDA, ., YAMAZAKI,., KIBAYASHI,. (1987), Total

synthesis of (+)-nojirimycin and (+)-1-deoxyno-jirimycin,J. Org. Chem .52,3337-3342.IKOTA, . (1989), Synthesis of (+)-1-deoxynojirimy-

cin from (S)-pyroglutamic acid, Heterocycles 29,

INOUYE,.,TSURUOKA,T.,IIDA,T.1966),The struc-ture of nojirimycin, a piperidinose sugar antibio-tic,J.Antibiot. 19,288-292.

INOUYE,., TSURUOKA,., ITO,T., NIIDA,T. (1968),Structure and synthesis of nojirimycin, Tetra-hedron 24,2125-2144.

ISELIN, . (1948), Oxidations by Acetobacter suboxy-duns,J. Biol. Che m.175,997-998.

ISHIDA,., KUMAGAI,., NIIDA, .,HAMAMOTO,.,SHOMURA,. (1967a), Nojirimycin, a new antibio-

1469-1472.



9 References 309

von 1,5-Didesoxy-1,5-imino-D-glucitolnd des-sen N -Derivaten, European Patent 55431.

KOD AM A, ., TSU RUOKA ,., NIWA,T., INOUYE, .(1985), Molecular structure and glycosidase-inhi-bitory activity of nojirimycin bisulfite adduct, J .

Antibiot. 38,116-118.KOSSEVA,M., BESCHKOV,., POPOV, . (1991), Bio-

transformation of D-sorbitol t o L-sorbose by im-mobilized cells of Gluconobacter suboxydans in abubble column, J . Biotechnol. 19,301-308.

KULHAN EK, . (1970), Ferm entation processes em -ployed in vitamin C synthesis, Adv. App l. Micro-biol. l2 , l l -30 .

KULHANEK, . (1989), Microbial dehydrogenationof monosaccharides, Adv. Appl. Microbiol. 34,

KULHANEK, ., TADR A,M., PACAK , ., TREJBALOVA,H., CER&, M. (1977), 5-Deoxy-5-fluoro-~-sorbo-se originating from 2-deoxy-2-fluoro-~-glucitolby fermentation with Acetom onas oxydans, FoliaMicrobiol. 22,295-297.

LA FERLA,B., NICOTRA,. (1999), Synthetic m ethod sfor the preparation of iminosugars, in: Iminosu-gars as Glycosidase Inhibitors. Nojirimycin andBeyond (STiiTz,A. E., Ed.), pp. 69-92. Weinheim:Wiley-VCH.

LINEK,K., SANDTNEROVA,., STICZAY,., KOVAC IK,V. (1979), Th e synthesis of 4-doxy-~ -glycero-pen-tulose by biochemical dehydrogenation of 2-deoxy-D-erythro-pentitol, Carbohydr. Res. 76,290-294.

MACLAY, . D., HA", R. M., HUD SO N, . S. (1942),Some studies on L-glucoheptulose, J . Am . Chem.S O C . 64,1606-1609.

MATSUMURA,., ENOMOTO,.,AoYAGI,Y.,YOSHIKI-NI,Y.,KURA,K.,YAGI,M. (1979a), Neue N-substi-tuierte Moranolinderivate, German Patent Offen-legung 2915037.

MATSUMURA,., ENOM OTO, ., AOYAGI, ., EZURE,Y.,YOSHIKUNI,Y.,YAGI,. (1979b),Verfahren zurGewinnung von Moranolin, German Patent Of-fenlegung 2850467.

MATSUMURA,., ENOMOTO,.,AOYAGI,Y.,YOSHIKU-NI,Y., YAGI,M. (1980), Mo ranoline, Chem. A bstr.93,66581s.

MOSES,V., FERRIER,R. J. (1962), Th e biochemicalpreparation of D-XylOSe an d L -ribu lose . Detail s ofthe action of Acetobacter suboxydans on D-arabi-tol, ribitol and other polyhydroxy compounds,Biochem. J . 83,8-14.

MULLER,. (1966), Untersuch ungen iiber die Sorbo-segarung in stationarer u nd kontinuierlicher Kul-tur, Zbl. Bakt. 1u), 49-378.

MURAI,H., OHATA, ., ENOMO TO,., YOSHIKUNI,.,KONO,T., YAGI, M. (1977), 2-H ydroxy methy l-3,4,5-trihydroxypiperidin,Extraktionsverfahrenzu seiner Herstellung und seine Verwendung alsArzneimittel, German Patent Offenlegung

141-182.

tic. I Taxonomy and fermentation, J . Antibiot. 20,62-65.

ISHIDA, ., KUM AGA I, ., NIIDA,T., HAMAMOTO,.,SHOMURA,T.1967b), Nojirimycin, a new antibio-tic. I1 Isolation, characterization and biologicalactivity,J . Antibiot. 2 0 , 6 6 7 1 .

JANKE, . (1960), Die Essigsauregarung, Handbuchder Pflanzenphysiologie Vol. 12, pp. 670-746. Be r-lin: Springer-Verlag.

JONES,. K. N., MITCHELL, . L. (1959),The synthesisof 5-deoxy-5-S-ethyl-~-threo-pentulose,an. J .

Chem. 37,1561-1566.JONES,. K. N., PERRY,M. B., TURNER, J. C. (1961a),

The synthesis of acetamido-deoxy ketoses byAcetobacter suboxydans. Part I, Can.J . Chem. 39,

JONES,. K. N., PERRY,M. B.,TURNER,. C. (1961b),The synthesis of acetamido-deoxy ketoses byAcetobacter subo xydans. Part 11,Can.J . Chem.39,2400-2410.

JON ES, . K. N., PERR Y,M. B.,TURNER,. C. (1962),The synthesis of acetamido-deoxy ketoses byAcetobacter suboxydans. Part 111, Can. J . Chem.

JUNGE,B., KRAUSE,H.-P., MOLLER,L., PULS, W.(1979), Neu e Deriv ate von 3,4,5-Trihydroxypipe-ridin, Verfahren zu ihrer Herstellung und ihreVerwendung, European Patent 947.

JUNGE, ., MATZKE,M., STOLTEFUSS,. (1996), Che-mistry and structure-activity relationships of glu-cosidase inhibitors, in:

Handbook of Experimen-tal Pharmacology: Oral Antidiabetics (KUHL-MANN,., PULS,W., Eds.), pp. 411482. Berlin:Springer-Verlag.

KAUFMANN,., REICHSTEIN, . (1967), 6- D es ox y- ~-idose, 6-Desoxy-~-sorbose nd 6-Desoxy-~-pisco-se, Helv. Chim. Acta 50,2280-2287.

KINAST,G.,SCHE DEL, . (1980a), Verfahren zur H er-stellung v on 6-Amino-6-desoxy-~-sorbose,uro-pean Patent 8031.

KINAST, .,SCHEDEL,M. (1980b),Verfahren zur He r-stellung von N-substituierten Derivaten des 1-Desoxynojirimycins, European Patent 12278.

KINAST,G., SCHE DEL, . (1981), Vierstufige 1-Des-oxynojirimycin-Synthese mit einer Biotransfor-mation als zentralem Reaktionsschritt, Angew.Chem. 93,799-800.

KINAST,G., SCHEDEL,M . , K ~ BE RN I CK ,. (1982),Verfahren zur Herstellung von N-substituiertenDerivaten von 1-Desoxynojirimycin, EuropeanPatent 49858.

KITE,G. C.,FELLOWS,. E.,LEES,D. C., KIT CH EN, .,MONTEITH, . B. (1991), Alkaloidal glycosidaseinhibitors in nocturnal and diurnal uraniinemoths and their respective foodplant genera, En-dospermum an d Omphalea, Biochem. System.

Ecol. 19,441445.K ~ B E R N I C K ,. (1982), Verfahren zur H erstellung

965-972.

40,503-510.




2656602.MURAO,., MIYATA,. (1980), Isolatio n and charac-

terizatio n of a new treh alase inhib itor, S-G I,Agric. Biol. Chem .44,219-221.

NAKAMURA,T.,TAKEBE,., KUD OH, .,TERADA ,.,TANDOH,. et al. (1993), Effect of an a-glucosid -ase inhibitor on intestinal fermentation an d faec-al lipids in diabetic patients, J. Int. Med. Res. 21,257-267.

NICOLSKAYA,., BELAYKOVA,., STATKEVITSCH,.,POMORZEVA,., ABRAMOVA,. (1984), Plant forcon tinuo us microbiological oxid ation of D -sorbi-to1 to L-sorbose, Khim. Farm. Zh . 18,879-892.

NISHIKAWA,., ISHID A, . (1965), A new antib ioticR-468 active against drug-resistantShigellaJ An-tibiot. 18,132-133.

NISHIMURA,. (1991), The synthesis and biologicalactivity of glucosidase inhibitors, J . Synth. Org.Chem.49,846-857.

NIW A, ., INOUYE, ., TSURUO KA,., KOAZE, ., NI-IDA, T. (1970), “Nojirimycin”as a potent inhibitorof glucosidase,Agr. Biol. Chem .34,966-968.

ODAKA,H.,KEDA,H.1995),Voghbose (AO-128):Ahypoglycemic agent,J. Takeda Res. Lab .54,21-33.

ONISHI, ., SUZUKI,. (1969), Microbial productionof xylitol from glucose, App l. M icroblol. 18,1031-1035.

PAULSEN,. (1966), Kohlenhydrate mit Stickstoffoder Schwefel im “Halbaceta1”-Ring, Angew.

Chem.78,495-510.PAULSEN,. (1999), The early days of m onosacchari-

des c ontaining nitrogen in the ring, in: Iminosu-gars as Glycosidase Inhibitors. Nojirimycin andBeyond (STUTZ,A. E., Ed.), pp. 1-7. Weinheim,Wiley-VCH.

PAULSEN, ., SANG STER ,., HEYNS, . (1967), Mono-saccharide mit stickstoffhaltigem Ring, XIII. Syn-these und Reaktionen von Keto-piperidinosen,Chem. Ber. 100,802-815.

PRATT,. W., RICHTMYER,. K. (1955), D-Glycero-D-allo-heptose, L-allo-heptulose, D-talo-heptuloseand related su bstances derived from the additionof cyanide to D-allOSe,J.Am . Chem.SOC.77,6326-6328.

PRATT, . W., RICHTMYER,. K., HUDSON, . C.(1952), D-Idoheptulose and 2 ,7-an hyd ro-p-~ -ido-heptulopyranose, J . Am. Chem. SOC. 74, 2210-

2214.REESE,E. T., PAR RISH ,. W., ETTLINGER,. (1971),

Nojirimycin and D-glUCOnO-1,5-laCtOne as inhibi-tors of carbohydrases, Carbohydr. Res. 18 , 381-388.

REICHSTEIN,.,GRUSSNER,. (1934), Eine ergiebigeSynthese der 1-A scorbinsaure C-Vitamin),Helv.Chim. Acta 17,311-328.

RICHTMYER,. K., STEWART,. C., HUD SON, . S.

(1950), ~-Fu co-4 -keto se, new sug ar produced bythe action of Acetobacter suboxydans on L-fuci-

to1,J. Am . Chem. Soc. 72,49344937.ROSENBERG,.,SVITEL,., ROSENBERGOVA ,.,STUR-

DfK, E. (1993), Optimization of so rbose produc-tion from sorbitol by Gluconobacter oxydans, A c-

ta Biotechnol. 35,269-274.RU DG E, . J., COLLINS,., HOLMES,. B., BAKER,R.

(1994),An enantioselectivesynthesisof deoxyno-jirimycin, Angew. Chem. (Int. Edn. Engl.) 33,2320-2322.

SCHEDEL, . (1997), Oxidation von Aminosorbitdurch Gluconobacter oxydans: Ein wichtigerSchritt d er technischen M iglitolsynthese, Ab-stracts of 15th Annual Meeting of Biotechnolo-gists, p. 178. Fran kfur t: DE CH EM A.

SCHMIDT, . D., FRO MM ER, ., J U N G E, ., M ULLER,L., WINGENDER,. et al. (1977), a-Glucosidaseinhibitors. New complex oligosaccharides of

microbial origin, Naturwissenschajien 64, 535-536.

SCHMIDT,. D., FROMMER,W., MULLER,L., TRU-SCHEIT, E. (1979), Glucosidase-Inhibitoren ausBazillen,Natunvissenschajien 66,584-585.

SCH NA RR, . W., SZA REK ,W. A., JONES,. K. N.(1977), Preparation and activity of immobilizedAcetobacter suboxydans cells, Appl. Environ.Microbiol. 33,732-734.

SCHRODER,., STUBBE,M. (1987), Verfahren zurHerstellung von 1-Desoxyn ojirimycinund dessenN-Derivaten,European Patent 240868.

S c m , H. (1993), Enzy matic deacy lation of acyl-aminosorboses,US Patent 5177004.

SCOTT, . R., TATTERSALL,. B. (1988), a-Gluco sid-ase inhibition in the treatm ent of non-insulin-de-pend ent d iabetes mellitus,Diabet. Me d. 5 , 4246 .

SHINAGAWA,., MATSUSHITA, ., AD AC HI,O.,

AMEYAMA,. (1982), Purification and character-ization of D-sorbitol dehy drogen ase from mem-brane of Gluconobacter suboxydans var. a,Agric.Biol. Chem.46,135-141.

SIMMONDS,. S. J., KITE,G. C., PORTER, .A . (1999),Taxonomic distribution of iminosugars in plantsand their biological activities, in: Iminosugars asGlycosidase Inhibitors. Nojirimycin and Beyond(STOTZ,A. E., Ed.), pp. 8-30. Weinheim: Wiley-VCH.

SPENCER,. F.T., GOR IN,.A. J. (1968), Microbiologi-cal transformations of sugars and related com-pounds, Progr. Ind. Microbiol. 7,177-220.

STEFANOVA,., KOSEVA,M., TEPAVICHAROVA,.,BESCHKOV,. (1987), L-Sorbose productio n bycells of the strain Gluconobacter suboxvdans en -trapped in a polyacrylamide gel,Biotechnol. Lett.

9.475-477.STEWART, . C., RICHTMYER,. K., HUDSON,. S.

(1949),The oxidation of volem itol by Acetobactersuboxydans and by Acetobacter xylinum, J . Am .Chem.SOC. 1,3532-3534.

STEWART,. C., RICHTM YER,. K., HUDSON,. S.



9 References 311

nobacter oxydans, Mikrob iologiya 48,400405.WHISTLER, . L., UNDERKOFLER,. A. (1938), The

production of 1-erythrulose by t he action of Ace-tobacter suboxydans upon eryihritol, J.Am . Sac.

WHITE,S. A., CLAUS,G. W. (1982), Effect of in-tracytoplasmic membrane development on oxi-dation of sorbitol and othe r po lyols by Glucono-bacter oxydans,J. Bacteriol. 150,934-943.

WONG,C.-H., HALC OMB ,. L., ICHIKAWA,., KAJI-MOTO, T. (1995), Enzyme in de r organischen Syn-these: Das Problem d er molekularen Erkennungvon Koh lenhydraten (Teil 1),Angew. Chem. 107,453-474.

YAGI,M., KO UNO , ., AOYAGI, ., M UR AI,H. (1976),The structu re of Moranoline, a piperidine alkalo-id from Morus species, Nippon NogeikagakuKaishi 50,571-572.

YAGI,M., KUONO ,., AO YA GI,., MURAI,H. (1977),The structure of Moranoline, a piperidine alkal-oid from Morus species,Chem.Abstr. 86,167851r.

YAMADA, H., &A, I., NAGAI, ., MATSUMOTO,.,KIYOHARA,., OMU RA,. (1993), Screen ing of cr-

glucosidase inhibitor from Chinese herbs and itsapplication on the quality control of mulberrybark, Chem. Abstr. 119,146433r.

60,2507-2508.

(1952), L-Guloheptulose and 2,7-an hydro-P -~-gu-loheptulopyranose, J. Am . Chem. SOC.74,2206-2210.

STUTZ,A. E. (Ed.) (1999),Zminosugars as Glycosid-

use Inhibitors. Nojirimycin and Beyond, Wein-heim: Wiley-VCH .SU MIT A, ., MURAO, . (1980),Fermentative produc-

tion of 1-deoxyno jirimycin by cultivating Strep--tomyces microorganisms Japanese Patent55120792.

TILDEN, . B. (1939), The p repa ration of perseulos eby oxidation of perseitol with Acetobacter sub-oxydans,J. Bacteriol. 31,629-637.

TIWAR I, . N.,DHAWALE, . R.,SZA REK , . A., HAY,G.W., KROPINSKI,. M . B. (1986),A synthesis of

2-deoxy-~-arabino-hexitolnd its oxidation to 5-

deoxy-D-threo-hexulose (“~-deoxy-~-fructose”)

using immobilized cells of Gluconobacter oxy-duns, Carb ohydr. Res. 156,19-24.TRIFONOV,A.,TEFANOVA,., KONSTANTINOV,.,TE-

PAVICHAROvA, I. (1991), Biochemical oxidation of

D-sorbitol to L -sorbose by imm obilized Glucono-bacter oxydans cells, Ap pl. B iochem. Biotechnol.28129,397405.

USPENSKAYA,. N., LOITSYANSKAYA,. S. (1979), Ef -fectivenes s of the utilization of g lucose by Gluco-



8 Engineering Microbial Pathwaysfor Amino Acid Production

IANG. FOTHERINGHAMMt. Prospect, IL, USA

1 Introduction 3142 Bacterial Production of L-Phenylalanine 314

2.1 The Comm on Arom atic and L-Phenylalanine Biosynthetic Pathways 3142.2 Classical Mu tagenesis/Selection 3152.3 Deregu lation of D A H P Synthase Activity 3172.4 Dereg ulation of Chorismate Mu tase and Prep hen ate Deh ydratase Activity 3172.5 Precursor Supply 3202.6 Secretion of Pheny lalanine from the Cell 320

3 Bacterial Production of L-Lysine and L-Threonine 3213.1 B iochemical Pathway Engineering in Coryn ebacteria 3213.2 Enginee ring the Pathways of L-Lysine and L-Threonine Biosynthesis

4.1 A Novel Biosynthetic Pathway to L-2-Aminobutyrate 3244.2 A Novel Biosynthetic Pathway to D-Phenylalanine 325

3214 Enginee ring Novel Biochemical Pathways for Am ino Acid Produc tion 324

5 Conclusions 3276 Referen ces 328





314 8 Engineering Microbial Pathways for A mino Acid Production

1 Introduction

Many thousands of tons of different amino

acids are produced annually for a variety of in-dustrial applications. In particular, monosodi-um glutamate, phenylalanine, aspartic acid,and glycine are produced as human food addi-tives and lysine, threonine, methionine, andtryptophan are manufactured as animal feedadditives (ABEand TAKAYAMA,972; BATT e tal., 1985; GA RN ER t al., 1983; SHIIO, 986).These and other amino acids are also used aspharmaceutical interm ediates and in a varietyof nutritional and health care industries. Inmany cases bacteria are used to produce theseamino acids through large scale fermentation,particularly when a high degree of enantio-meric purity is required. In some cases chemo-enzymatic methods have also bee n successful-ly employed, particularly for L-phenylalanineand non-proteinogenic am ino acids.

The development of strains for fermenta-tion app roaches initially consisted of classicalrandom mutagenesis methods and screeningwith toxic amino acid analogs. In this way com -mercially viable amino acid overproducing

strains w ere developed. However, in the 1980sthis approach began t o be complemented bythe availability of molecular biological meth-ods and a rapid increase in the understandingof the molecular genetics of amino acid pro-duction in particular bacterial species. Theavailability of plasm id vector systems, trans-formation protocols, and general recom binantDN A methodology enabled biochemical path-ways to b e altered in much more precise waysto further enhance amino acid titer and pro-duction efficiency in Escherichia coli and in

the coryneform bacteria Corynebacterium andBrevibacterium, the most prominent aminoacid producing bacteria. This chapter willcenter upon the most significant aspects of

biochemical pathway engineering such as th eidentification of rate limiting enzymatic stepsan d the elimination of allosteric feedback inhi-bition of key enzymes. Con siderable emp hasiswill be placed up on th e shikim ate and phenyl-alanine pathways of E. coli which are amongthe best characterized biochemically and ge-netically and have been the subject of some ofthe most widespread efforts in am ino acid pro-

duction. It will also address aspects of meta-bolic engineering which have been applied inthe coryneform organisms due to the rapidlyincreasing understanding of the mo lecular ge-

netics of those organisms, and their impor-tance in amino acid produ ction. Additionally,it will examine the increasing potential t o con-struct comp letely nov el biochemical pathwaysin organisms such as E. coli in order to pro-duce non-proteinogenic or unnatural aminoacids.

2 Bacterial Production of

L-Phenylalanine

2.1 The Common Aromatic and

L-Phenylalanine Biosynthetic

Pathways

Efforts t o develop L-phenylalanine overpro-ducing organisms have been vigorously pur-sued by Nutrasweet Company, Ajinomoto,

Kyowa Hakko Kogyo, and others. The focushas centered upon bacterial strains which havepreviously demonstrated the ability to over-produce othe r amino acids. Such organisms in-clude principally the coryneform bacteria,Brevibacterium j7avum (SHIIO t a]., 1988) an dCorynebacterium glutamicum (HAGINOandNAKAYAMA,974b; IKED A t al., 1993) usedin L-glutamic acid production. In additionEscherichia coli (TRIBE, 987) has been exten-sively studied in L-phenylalanine manufacturedue to th e detailed characterization of the m o-

lecular genetics and biochemistry of its aro-matic amino acid pathways an d its amenabilityto recom binant D NA methodology. The bio-chemical pathway w hich results in th e synthe-sis of L-ph enylalanine from cho rism ate is iden-tical in each of these organisms and shown inFig. 1.The comm on aromatic pathway to chor-isma te in E. coli is shown in F ig.2. In each casethe L-phenylalanine biosynthetic pathwaycomprises three enzymatic steps from choris-mic acid, the product of the common arom aticpathway. The precursors of th e com mon a ro-matic pathway, phosphoenolpyruvate (PEP)



2 Bacterial Production of i>-Phenylalanine 315

Prephenatehorismate

&A%

Fig. 1.The biosynthetic pathwayfrom chorismate to L-phenyl-alanine in E. coli K12. L-Phenylalanine

___,Chor sma emutase

I P h W

Aromatic

transaminase

(W r B )Aspartatetransaminase

( a sPC)

“TCI

OH

Prephenatedehydratase

(PheNI- 0

Phenylpyruvate

and erythrose-4-phosphate (E4P) derive re-spectively from the glycolytic and pentosephosphate pathways of sugar metabolism. Al-

though the intermediate compounds in bothpathways are identical in each of the organ-isms, there are differences in the organizationand regulation of the genes and enzymes in-volved (HUDSONnd DAVIDSON,984; OZAKIe t al., 1985; SH IIOet al., 1988). Nevertheless,the principal points of pathway regulation arevery similar in each of the three bacteria(HUD SON nd DAVIDSON,984; IKEDA t al.,1993; SH IIO n d S UGIMOT O,981b). Converse-ly, tyrosine biosynthesis which is also carriedout in three biosynthetic steps from choris-

mate, proceeds through different intermedi-ates in E. coli than in the coryneform organ-isms (HUDSO Nnd DAVIDSON,984; SHIIO ndSUGIMOTO,981b).

In E. coli, regulation of phenylalanine bio-synthesis occurs both in the comm on aromat-ic pathway, and in th e term inal phenylalaninepathway and th e vast majority of efforts to de-regulate phenylalanine biosynthesis have fo-

cused upon two specific ra te limiting enzymat-ic steps. These a re the steps of the aroma ticand the phenylalanine pathways carried outrespectively by the enzymes 3-deoxy-~-arab-

ino-heptulosonate-7-phosphateDAHP) syn-thase and prephenate dehydratase. Classicalmutagenesis approaches using toxic amino

acid analogs and the molecular cloning of thegenes encoding these enzymes have led tovery significant increases in the capability of

host strains to overproduce L-phenylalanine.This has resulted from the elimination of th eregulatory mechanisms controlling enzymesynthesis and specific activity. The cellularmechanisms which govern th e activity of the separticular enzymes ar e complex and illustratemany of the sophisticated m eans by which bac-teria control gene expression and enzyme ac-tivity. Chorismate mutase, the first step in the

phenylalanine specific pathway a nd the shiki-ma te kinase activity of the common arom aticpathway are subject to a lesser degree of regu-lation and have also been characterized in de-tail (M ILLAR t al., 1986;NELMS t al., 1992).

2.2 Classical Mutagenesis/Selection

Classical methods of strain improvementhave been widely applied in the developmentof phenylalanine overproducing organisms

(SHIIO t al., 1973; TOK OROt al., 1970; TRIB E,



316 8 Engineering Microbial Pathways fo r Amin o Aci d Production

Erythrose 4-phosphate3-Deoxy-~-arabino-

heptulosonate 7-phosphate

DAHP synthase (aroF-Tyr)Isoenzymes (aroG-Phe)

(aroH-Trp)

OH P H2

OH

PhosphoendpyruvateDehydroquinatesynthase(aroB)I

OH

Shikimate

Shikimate Dehydroquinate

dehydrogenase dehydratase(aroEl (aroD)

e---

OHH

3-Dehydroshikimate 3-Dehydroquinate

Shikimate kinases( I -aroK, 11-aroL)I ate synthase 'f"

Shikimate 3-phosphate 5-Enolpyruvylshikimate- Chorismate3-phosphate

Fig. 2. The common aromatic pathway to chorismate in E. coli K12.The mnemonic of the genes involved are

shown in parentheses below the enzymes responsible for each step.

1987; TSUCHIDAt al., 1987). q ro si n e auxo-trophs have frequ ently been used in efforts toincrease phenylalanine production throughmutagenesis. These strains o ften already ov er-produce phenylalanine due to the overlappingnature of tyrosine and phenylalanine biosyn-thetic regulation (HAGINO nd NAKAYAMA,1974b; HWANG t al., 1985). Limiting tyrosine

availability leads to partial genetic and allo-

steric deregulation of common biosyntheticsteps (WALLACE nd PITTARD,969). Suchstrains have been subjected t o a variety of mu-tagenesis procedures to further increase theoverall titer and the efficiency of phenylala-nine production. In general this has involvedthe identification of mu tants which display re-sistance to toxic analogs of phenylalanine or

tyrosine such as @-2-thienyl-~,~-alaninend



2 Bacterial Production of L-Phenylalanine 317

whereas the aroH gene product is subject tomaximally 40% feedback inhibition by trypto-phan (GARNERt al., 1983).In Brevibacteriumf lavum, DAHP synthase forms a bifunctional

enzyme complex with chorismate m utase a ndis feedback inhibited by tyrosine and phenylal-anine synergistically but not by tryptophan(SHIIOt al., 1974;SUGIMOTOnd SHIIO,980).Similarly in Corynebacterium glutamicumDAHP synthase is inhibited most significantlyby phenylalanine and tyrosine acting inconcert (HAGINOnd NAKAYAMA,975b) but,unlike Brevibacterium flavum, reportedly itdoes not show tyrosine mediated repression oftranscription (HAGINOnd NAKAYAMA,974c;SHIIO t al., 1974).

Many analog resistant mutants of these o r-ganisms display reduced sensitivity of DAHPsynthase to feedback inhibition (HAGINOndNAKAYAMA,974b; OZAKI t al., 1985; SHIIOand SUGIMOTO,979; SHIIO t al., 1974; SUGI-M O T 0 et al., 1973;TRIBE,987). In E. coli th egenes encoding the DAHP synthase isoen-zymes have been characterized and se quenced(DAVIES nd DAVIDSON,982; HUDSON ndDAVIDSON,984;RAY et al., 1988). The mecha-nism of feedb ack inh ibition of th e aroF, aroG,

an d aroH isoenzymes has been studied in con-siderable detail (RAY e t al., 1988) and variantsof the aroF gene on plasmid vectors have beenused to increase phenylalanine overproduc-tion. Simple replacement of transcriptionalcontrol sequences with powerful constitutiveor inducible promoter regions and the use ofhigh copy number plasmids has readily en-abled overproduction of the enzyme (ED-WARDS e t al., 1987;FOERBERGt al., 1988),andreduction of tyrosine m ediated feedbac k inhi-bition has been described using resistance to

the aroma tic amino acid analogs P-2-thi-enyl-D,L-alanine and p-fluoro-D,L-phenylalanine(EDWARDSt a l., 1987;TRIBE,987).

p-fluoro-D,L-phenylalanine (CHOI nd TRIBE,1982; HAGINOnd NAKAYAMA,974b; TRIBE,1987). Such m utants can be readily identifiedon selective plates in which the analog is

present in the growth medium. The m utationsresponsible for the phenylalanine overproduc-tion have frequently bee n located in the genesencoding the enzymatic activities DAHP syn-thase, chorismate mutase, and prephenate de-hydratase. In turn this has promp ted m oleculargenetic approaches to fu rther increase phenyl-alanine production through the isolation andin vitro manipulation of these genes as de-scribed below.

2.3 Deregulation of D A H PSynthase Activity

The activity of DAHP synthase commitscarbon from intermediary metabolism to thecommon aromatic pathway converting equi-molar amounts of PEP an d E4P to DAHP(HASLAM,974). In E. coli there are three iso-enzymes of DAHP synthase of comparablecatalytic activity encoded by the genes aroF,

aroG, and aroH (PITTARDnd GIBSON,970).Enzyme activity is regulated respectively bythe aroma tic am ino acids tyrosine, phenylala-nine, and tryptophan (BROWN,968; BROWNand SOMERVILLE,971; CAMAKARISnd PIT-

TARD, 1969).In each case regulation is mediat-ed both by repression of gene transcriptionand by allosteric feed back inhibition of the en-zyme, though t o different degrees.

The aroF gene lies in an operon with tyrAwhich encodes the bifunctional protein choris-

ma te mutase/prephenate dehydrogenase. Bothgenes are regulated by the TyrR repressor pro-tein complexed with tyrosine. The aroG geneproduct accounts for 80%of the total DAHPsynthase activity in wild type E. coli cells. Th earoG gene is repressed by the TyrR repressorprotein complexed with phenylalanine andtryptophan. Repression of aroH is mediatedby tryptophan and the TrpR repressor protein(BROWN,968).The gene p roducts of aroF andaroG ar e almost completely feedback inhibit-ed respectively by low concentrations of tyro-

sine or phenylalanine (GARNERt al., 1983)

TARD, 1973; I M et al., 1971;WALLACE an d PIT-

2.4 Deregulation of Chorismate

Mutase and Prephenate

Dehydratase Activity

The three enzymatic steps by which choris-mate is converted to phenylalanine appear to

be identical between C. glutamicum , B. flavu m,



318 8 Engineering Microbial Pathways forAm ino Acid Production

and E. coli although only in E. coli have de-tailed reports appeared upon the charac-terization of the genes involved. In each casethe principal regulatory step is that catalyzed

by prephenate dehydratase. In E. coli, choris-mate is converted firstly to prephenate andthen to phenyl pyruvate by the action of a bi-functional enzyme, chorismate mutase/pre-phenate dehydratase (CMPD) encoded by thepheA gene (HUDSON nd DAVIDSON,984).The final step, in which phenyl pyruvate is con-verted to L-phenylalanine, is camed out pre-dominantly by the aromatic aminotransferaseencoded by fy rB (GARNERt al., 1983). How-ever both the aspartate aminotransferase, en-coded by aspC and the branched chain amino-transferase encoded by ilvE can catalyze thisreaction (FOTHERINGHAMt al., 1986). Phenyl-alanine biosynthesis is regulated by control ofCMPD through phenylalanine mediated at-tenuation of pheA transcription (HUDSONndDAVIDSON,984) and by feedback inhibition of

the prephenate dehydratase (PD) and choris-mate mutase (CM) activitiesof the enzyme. In-hibition is most pronounced upon the prephen-ate dehydratase activity, with almost total inhi-bition observed at micromolar phenylalanine

concentrations (DOPHEIDE et al., 1972;GETHING nd DAVIDSON,978). Chorismatemutase activity in contrast is maximally inhib-ited to only 40% (DOPHEIDEt al., 1972). In B.j lavum, prephenate dehydratase and choris-mate mutase are encoded by distinct genes.Prephenate dehydratase is again the principalpoint of regulation with the enzyme subject tofeedback inhibition, but not transcriptional re-pression, by phenylalanine (SHIIO and SUGI-MOTO,976; SUGIMOTOnd SHIIO, 974). Chor-ismate mutase which in this organism forms a

bifunctional complex with DAHP synthase ismaximally inhibited by phenylalanine and ty-rosine to 65%, but this is significantly dimin-ished by the presence of very low levels oftryptophan (SHIIO and SUGIMOTO, 981a).Expression of chorismate mutase is repressedby tyrosine (SHIIO nd SUGIMOTO,979; SUGI-MOTO and SHIIO, 985).The final step is carriedout by at least one transaminase (SHIIO t al.,1982). Similarly in C.glutumicum, the activitiesare encoded in distinct genes with prephenatedehydratase again being the more stronglyfeedback inhibited by phenylalanine (DE Bo-

ER and DIJKHUIZEN,990; HAGINO nd NA-KAYAMA,974a, 1975a).The only transcription-al repression reported is that of chorismatemutase by phenylalanine. The C. glutamicum

genes encoding prephenate dehydratase andchorismate mutase have been isolated andcloned from analog resistant mutants of C.glu-tamicum (IKEDA nd KATSUMATA,992; OZAKIet al., 1985) and used along with the clonedD A W synthase gene to augment L-phenylala-nine biosynthesis in overproducing strains ofC. glutumicum (IKEDA nd KATSUMATA,992).

Many publications and patents have de-scribed successful efforts to reduce and elimi-nate regulation of prephenate dehydratase ac-tivity in phenylalanine overproducing organ-isms (FOERBERGt al., 1988; NELMS t al., 1992;TRIBE, 987).As with DAHP synthase the ma-jority of the reported efforts have focusedupon the E. coli enzyme encoded by the p h eAgene which is transcribed convergently withthe tyrosine operon on the E. coli chromosome(HUDSON nd DAVIDSON,984). The detailedcharacterization of thepheA regulatory regionhas facilitated expression of the gene in a va-riety of transcriptional configurations leadingto elevated expression of CMPD, and a num-

ber of mutations in pheA which affect phenyl-alanine mediated feedback inhibition havebeen described (BACKMANnd BALAKRISH-NAN, 988; EDWARDSt al., 1987; NELMSt al.,1992). Increased expression of the gene isreadily achieved by cloning pheA onto multi-copy plasmid vectors and deletion of the nu-cleotide sequences comprising the transcrip-tion attenuator illustrated in Fig. 3.Most muta-tions which affect the allosteric regulation ofthe enzyme by phenylalanine have been iden-tified through resistance to phenylalanine ana-

logs such as P-Zthienylalanine, but there areexamples of feedback resistant mutations aris-ing through insertional mutagenesis and genetruncation (BACKMANnd BALAKRISHNAN,1988;EDWARDSt al., 1987).Two regions of theenzyme in particular have been shown to re-duce feedback inhibition to different degrees.Mutations at position Trp338 in the peptide se-quence desensitize the enzyme to levels ofphenylalanine in the 2-5 mM range but are in-sufficient to confer resistance to higher con-centrations of L-phenylalanine (BACKMANnd

BALAKRISHNAN,988;EDWARDSt al., 1987).



2 Bacterial Production of L-Phenylalanine 319

-1 0-35-G T A T C G C C A ? C G C G C C T T C G G G C G C G T T T T T T ( Y M Y U C M T

A AAAaTCACTTAAGGAAACAAACATGAAACACATACCGTTTTTCTTCGCATTCTTTTTTACCTTCCCC

AllENUATOR REGIONTGAATGGGAGGCGTTTCGTCGTGTGAAACAGAATGCGAAGAATGCG~GACGAACAAT~GGCC~CCAAATCGGG

M T S E N P1 0-35-

G A A T T C T P T T T T G ? T G A C A G C C T G A A A A C A G T A C ~ ~ A ~ T A C T ~ ~ C A C ~ ~ C ~ ~ C T A T G A ~ T C ~ ~ C C C-ig. 3. A. Seq uenc e of the wild type prom oter region of the E.coli K1 2ph eA gene. The -35 and -10hexamers

are indicated. Bases retained in the deregulated promoter are in bold type. Sequence involved in the at-tenuato r region is shown u nderlined. B. Sequence of the deregulated pheA promo ter region used to p rodu-ce chorismate m utase/pre phen ate dehydratase. Altered bases in the -10 region are show n underlined.

COOINGSEQUENCE

Mutations in the region of residues 304-310 (JN305-JN308) are shown in Fig.4, in compar-confer almost total resistance to feedback inhi- ison to the profile of wild-type enzymebition at L-phenylalanine concen trations of at (JN302). It is not clear, if the m echanism of re-least 200 mM (NELMS t al., 1992). Feedb ack sistance is similar in either case, but the differ-inhibition profiles

PI

c

c

c)

n

en

of four such variants ence is significant to com mercial application as

120

100

80

60

40 IN302- N305

20 IN306- N307- N308

0

0 50 100 150 200

L-phrnylalanlno concentration (mM)

Fig. 4. L-Phenylalanine m ediated feedback inhibition of wild type E. coli K12 prephenatedehydratase (JN302) and four feedback inhibition resistant enzyme variants(JN305-JN308).Activity is expressed as a percen tag e of norm al wild-type enzyme activity.



320 8 Engineering Microbial Pathways fo r Am ino Aci d P roduction

overproducing organisms readily achieve ex-tracellular concentrations of L-phenylalanineover 200 mM.

2.5 Precursor Supply

R ate limiting steps in the comm on aroma ticand phenylalanine biosynthetic pathways areobvious targets in the deve lopmen t of phenyl-alanine overproducing organisms. However,the detailed biochemical and genetic charac-terization of E. coli has enabled additional ar-eas of its metabolic function t o be specificallymanipulated to determine their effect on aro-matic pathway throughput. Besides efforts to

eliminate add itional lesser points of aromaticpathway regulation, attempts have been ma deto enhance phenylalanine production by in-creasing the supply of aro matic pathway pre-cursors an d by facilitating exodus of L-phenyl-alanine from the cell. The precursors of thecommon aromatic pathway D-erythrose 4-phosphate and phosphoenolpyruvate are therespective products of the pentose phosphateand glycolytic pathways. Precursor supply inaroma tic amino acid biosynthesis has be en re-

viewed recently (BERRY,1996; FROST ndDRATH S, 995). Theoretical analyses of thepathway and the cellular roles of these metab -olites suggest that the production of arom aticcompounds is likely to be limited by PEPavailability (PATNAIKnd LIAO,1994; PATNAIKet al., 1995), since phosphoen olpyruvate is in-volved in a number of cellular processes in-cluding the generation of metabolic energythrough the citric acid cycle (MILLER t al.,1987) and the transpo rt of glucose into the cellby the phosphotransferase system (POSTMA,

1987).Strategies to reduce the drain of PE P bythese processes have included mutation of sug-ar transport systems to reduce P EP dependentglucose transport (FLOR ES t al., 1996) andmodulation of the activities of pyruvate kin-ase, P E P synthase, and P EP carboxylase whichregulate PEP flux to pyruvate, oxalo-acetate, and the citric acid cycle (BERR Y, 996;BLEDIG, 994; MILLER t al., 1987). Similarly,the availability of E4P has been increased byaltering the levels of transketolase, the en zym eresponsible for E4P biosynthesis (DRATHSt

al., 1992). In general these e fforts have been

successful in directing additional flux of P E Por E4P into the aromatic pathway, althoughtheir effect has not proved to be as predictableas the deregulation of rate limiting pathway

steps and the ir overall impact on L-phenylala-nine overpro duction has not been well charac-terized.

2.6 Secretion of Phenylalanine

from the Cell

Exodus of phenylalanine is of manifestimportance in the large scale production of

phenylalanine as the am ino acid is typically re-covered only from the extracellular medium

an d w ashed cells. Lysis of cells to recover addi-tional phenylalanine is not generally practicaland so L-phenylalanine remaining w ithin thecells following washing is usually lost. Sincecell biomass is extremely high in large scalefermentations this can represent up to 5% oftotal phenylalanine produced. Additionally,since L-phenylalanine is known to regulate itsown biosynthesis at multiple points throughfeedback inhibition, attenuation, and TyrRmediated repression, methods to reduce intra-

cellular concentrations of phenylalaninethrough reduced uptake or increased exodusthroughout the fermentation have beensought. Studies have addressed the means bywhich L-phenylalanine is both taken up andexcreted by bacterial cells, and whether thiscan be altered to increase efflux. In overpro-ducing strains it is likely that most phenyl-alanin e leaves the cell by passive diffusion butspecific uptake and exodus pathways also ex-ist. In E. coli L-phen ylalanine is taken up by a tleast two perme ases (WHIPP and PITTARD,

1977), on e specific for ph enylalanine encodedbyph eP, and the other encoded by aroP a gen-eral aromatic amino acid permease respon-sible for the transport of tyrosine and trypto-phan in addition to phenylalanine. The genesencoding the permeases have bee n cloned andsequenced and E. coli mutants deficient in ei-ther system have been characterized (H ON OR Eand COLE, 99QP1et al., 1991).The effect of anaroP mutation has been evaluated in L-phenyl-alanine overproduction (TR IBE, 987). n addi-tion specific export systems have been identi-

fied which though not completely character-



3 Bacterial Production of L-Lysine and L-Threonine 321

ized have been successfully used to increaseexodus of L-phenylalanine from strains ofE. coli (FOTHERINGHAMt al., 1994;MULCAHY,1988). In one such system the Cin invertase of

bacteriophage P1 has been shown to induce ametastable phenylalanine hyper-secretingphenotype upon a wild type strain of E. coli.The phenotype can be stabilized by transientintroduction of the invertase on a temperaturesensitive plasmid replicon and is sufficient toestablish L-phenylalanine overproduction inthe absence of any alterations to the normalbiosynthetic regulation (FOTHERINGH AMt al.,1994).

The incremental gains made from the vari-ous levels at which phenylalanine b iosynthesis

has been addressed have led to the high effi-ciency production strains currently in use. Al-most all large scale L-phenylalanine manufac-turing processes in present operation are fer-men tations employing bacterial strains such asthose d escribed abo ve in which classical straindevelopment and/or molecular genetics havebeen extensively applied to bring aboutphenylalanine overproduction. The low sub-strate costs and the economics of scale associ-ated w ith this approach have resulted in signif-

icant economic advantages.

3 Bacterial Production of

L-Lysine and L-Threonine

3.1 Biochemical Pathway

Engineering in Corynebacteria

The ferm entative production of L-glutamateand amino acids of th e a sparta te family such aslysine and threonine, has been dominated bythe use of the coryneform organisms such as C.glutamicum. Several strains of th e corynebac-teria have been used for decades to produce L-

glutamic acid (AB E and TAKAYAMA,972;GARNER t al., 1983; YAMADA t al., 1973).These organisms have been found to secretethe amino acid at very high concentrationsthrough changes in the cell mem brane inducedeither by biotin limitation or by the add ition of

various surfactants (CLEMENTnd LANEELLE,

1986;DUPERRAYt al., 1992).The enzyme glu-tamate dehydrogenase encoded by the gdhgene is principally responsible for the biosyn-thesis of L-glutamate in the corynebacteria

from the keto acid precursor 2-ketoglutarate(TESCH t al., 1998). Although relatively fewgenetic studies have been published on this en-zyme (BOERMA NNt al., 1992),many advancesin both the genetic organization and th e tech-nology to manipulate DN A in the corynebac-teria have been made in the last two decades(ARCHER nd SINSKEY,993; MARTIN t al.,1987; SANTAMARIA t al., 1987). This has ena-bled molecular genetic approaches to comple-ment conv entional strain developm ent for theproduction of amino acids such as L-threoninean d L-lysine in Corynebacterium (ISHIDA t al.,1993,1989;JEITEN and SINSKEY,995). Theseadvances have been thoroughly reviewedquite recently (JETEN e t al., 1993;JEITEN ndSINSKEY,1995; NAMPOOTHIRInd PANDAY,1998) and will be summarized here with onlykey aspects relevant to microbial pathway en-gineering emphasized.

The use of E . coli has been a valuable tool tothe researchers in the corynebacteria as manyof t he ge nes encoding the key enzymes in the

biosynthesis of these amino acids we re initial-ly isolated through complementation of thecorresponding m utants in E. coli (JEITENandSINSKEY,995). Similarly, transconjugation ofplasmids from E. coli to various Corynebacte-rium strains, facilitating gene disruption andreplacem ent, has helped characterize the rolesof specific genes, particularly in lysine andthreonine biosynthesis (PUHLER t al., 1990;SCHWARZERnd PUHLER, 991). The recentdevelopm ent of Corynebacterium cloning vec-tors has fu rther facilitated th e studies of gene

regulation and expression in these strains(JEITENet al., 1994; MAR TIN t al., 1987;MIWAet al., 1985).

3.2 Engineering the Pathways of

L-Lysine and L-Threonine

Biosynthesis

The biosynthesis of the aspartate derivedamino acids is significantly impacted by the

regulation of aspartokinase, the enzyme gov-



322 8 Engineering Microbial Pa thways forAm ino Acid Production

erning the flow of carbon entering the initialcommon pathway, shown in Fig. 5, and encod-ed by the ask gene in C. gfutamicum.In thecase of C. glutamicum and B. flavum asparto-

kinase is tightly regulated by concerted feed-back inhibition, mediated by threo nine and ly-sine (SHIIOand MIYAJIMA,969). Throughclassical mutagenesishelection using the toxiclysine analog S-(a-aminoethy1)-D,L-cysteine,

lysine overproducing strains of C. glutamicum

were obtained which frequently carried vari-ants of aspartokinase resistant to feedback in-hibition by lysine and/or threo nine (KALINOW-SKI et al., 1991;TOSAKAnd TAKINAMI,986).Seque nce comparisons between the wild typeand mutant ask gene sequences assigned

mutations conferring feedback resistance tothe @-subunitof the enzyme (FOLLEITIEt al.,

1993; KALINOWSKIt al., 1991), although themechanism of feedback inhibition and of resis-tance remains unknown. Similar mutationshave been identified in the @-subunitof C. ac-tofermentum and in B. flavum (SHIIO t al.,1993,1990).

By analogy to the feedback resistant muta-tions in prephenate dehydratase and theDAHP synthases of the E. cofiphenylalanine

and common aroma tic pathways, the identifi-cation of fee dback resistant variants of aspar-tokinase has contributed significantly to ef-forts to engineer lysine and threonine overpro-ducing strains of Corynebacterium. Introduc-tion of feedback resistant Ask variants to ly-sine overproducing strains of C. gfutamicum

and C. lactofermentumenhance d extracellularlysine levels, the latter in a gene dosage depen-dent manner (CREMER t al., 1991; JETTEN etal., 1995). Overexpression of the wild type di-hydropicolinate synthase, on a multi-copy

plasmid also leads to elevated lysine produc-tion (CREMERt al., 1991) as m ight be expec-ted, as this is the branch point of lysine andthreonine biosynthesis in Corynebacterium.No

increase in extracellular lysine was observedwith the overexp ression of each of the remain-ing enzymes of the p rimary lysine biosyntheticpathway. A secondary pathway of lysine bio-synthesis exists in C.gfutamicum,which allowscom plemen tation of mutations in the ddh geneencoding diaminopimelate dehydrogenase,but was unable a t least in the wild type s tate to

provide high throughput of precursors to D,L

diaminopimelate (CREMER et al., 1991;SCHRUMPFt al., 1991).

As a converse approach t o the funneling ofcarbon to th e lysine specific pathways by o ver-

production of dihydropicolinate synthase, theelimination of this activity by m utation of th edapA gene led t o significant overprodu ction ofL-threonine (SHIIO t al., 1989) to over 13 gL -l in a stra in of B. flavum which also carrieda feedback inhibition resistant aspartokinase.This was determined to be comparable inproductivity to the more typical overproduc-ing strains of B. flavum carrying mutations inhomoserine dehydrogenase which eliminatethe normal threonine mediated feedback in-hibition. This is typically selected by the now

familiar mutagenesis/selection methods in-volving resistance to the toxic threonine ana-log D,L-a-amino-@-hydroxyvalericcid. Sim-ilar levels of threonine overproduction havebeen repo rted in recombinant E. coli strains inwhich the threonine ope ron from a classicallyderived overproducing mutant was cloned ona m ulti-copy plasmid and re-introduced t o thesame host (MIWAet al., 1983). Significantlygreater titers of threonine production havebee n reported by introducing the same cloned

threonine operon into mu tant strains of B. fla-vum grown with rigorous plasmid mainte-nance (ISHIDA t al., 1989).

The ability of the corynebacteria to over-produce these amino acids sufficiently forcomm ercial use is partly d ue to th e extraord i-nary efficiency with which this type of organ-ism can secrete amino acids. For this reason ithas been useful in the prod uction of isoleucine(COLON t al., 1995) and other amino acids(NAMPOOTHIRInd PANDEY,998) without ne-cessitating the elimination of feedback inhibi-

tion of key enzymes in the biochemical path-ways involved. Nevertheless, in the examplesdescribed above the av ailability of detailed in-formation on th e molecular genetic control of

metabo lic and biosynthetic pathways is of fun-dame ntal importanc e in achieving optimal lev-els of amino acid production. Similarly the ap-plication of recombinant DNA methodologyto a wider array of microbes is enhancing thepotential to effectively screen for mutationswhich relieve feedback inhibition, frequentlythe m ost crucial aspect of deregulating biosyn-

thesis of primary metabolites such as amino



3 Bacterial Production of L-Lysine and ~- Th reo nin e 323

OHI

L-Aspartyl

phosphate

L-Aspartate

semialdehyde

L-Ly sine

0

Aspartokinase

Aspartate semialdehyde

deh y drogenase

( a s 41

Homoserinedehy drogenase1 (horn)

OHIL-Homoserine

H2N)K-

v

H2N

0

L-Isoleucine L-Threonine L-Methionine

Fig. 5. The common pathway of the aspartate derived amino acids in corynebacteria.

The mnemonicof the genes involved are shown in parentheses below the enzymes re-

sponsible for each step. Dotted lines indicate multiple enzymatic steps.



324 8 Engineering Microbial Pathways fo r Amin o Acid Production

acids.As a results th e potential to engineer m i-crobial pathways is increasing sign ificantly andincludes opportunities to construct novel bio-chemical routes for the biosynthesis of non-

proteinogenic am ino acids.

4 Engineering NovelBiochemical Pathways forAmino Acid Production

The biosynthesis of most proteinogenic am i-

no acids including those described above di-rectly o r indirectly involves a transam inationstep (GAR NERt al., 1983). Biochemical stud-ies on a number of the bacterial L-amino acidtransaminases (aminotransferases) respon-sible for this reaction have revealed that theseenzymes frequently display broad substratespecificity. The arom atic transam inase of E .coli, e.g., is capable of synthesizing aspa rtateand leucine in addition to the arom atic aminoacids phenylalanine and tyrosine. Similarly, heE. coli branched chain transaminase can syn-thesize phenylalanine and m ethionine in addi-tion to the branched chain am ino acids, isoleu-cine, leucine, and valine (CH IHAR A,985). Therelatively relaxed substrate specificity of mi-crobial transaminases has been useful in thedevelopm ent of biotransformation approache sfor the synthesis of non-proteinoge nic L-ami-no acids, which are now in increasing dem andas intermediates for the synthesis of peptidom i-metic pharmaceuticals. Additionally the avail-ability of a highly efficient screen has enab led

the cloning of a number of genes encoding D-amino acid transaminase from a variety of mi-crobial species. O ne of these genes, from Bacil-lus sphaericus, has been en gineered into a syn-thetic pathway in E. coli along with an L-aminoacid deaminase from Proteus myxofaciens andan amino acid racemase from Salmonella ty-

0 R transarninase ~

+ I i 3 J c o * - -RKCOz-

phimurium to enable fermentative productionof a number of D-amino acids from L-aminoacid starting m aterials. Using these approach -es, com pounds such a s D-ph enylalanine, L-ter-

tiary leucine, and both isomers of 2-aminobu-tyrate have been produced (AG ER t al., 1997;FOTHERINGHAMt al., 1997 ). Som e asp ects ofthese processes are described below.

4.1 A Novel Biosynthetic Pathway

to L-2-Aminobutyrate

In addition to the identification of the ap-prop riate biosynthetic enzym e, the feasibility

of all biotransformation processes dependsheavily upon other criteria such as the avail-ability of inexp ensive starting ma terials, th e re-action yield and the complexity of product re-covery. In the case of transaminase processes,the reversible natu re of the reaction as shownin Fig. 6 and the presence of a keto acid by-product is a concern which limits the overallyield and purity of the amino acid product, andhas led to efforts to increase the conversionbeyond the typical 50% yield of product(CRUM P nd ROZZELL, 992; TAYLOR t al.,1998). A dditionally, there are cost considera-tions in the large scale preparation of ketoacid substrates such as 2-ketobutyrate, whichar e not commodity chemicals.

These issues are very readily addressed inwhole cell systems as additional microbial en-zymes can be used to generate substrates frominexpensive precursors a nd convert unwantedby-products to more easily handled com-pounds. For the biosynthesis of 2-aminobuty-rate using the E . coli aromatic transaminase

(FOTHERINGHAMt al., 1999), the engineeredE. coli incorporates the cloned E. coli K12 ilvAgene encoding threonine deaminase to gener-ate 2-ketobutyrate from the commodity am inoacid L-threonine, and the cloned alsS gene ofBacillus sub tilis 168 encoding acetolactate syn-thase which eliminates pyruvate, the keto acid

Fig. 6.Transaminase reaction scheme.Transaminases occur as L-amino acid or D-amino acid specific.



4 Engineering Novel Biochemical Pathways fo r Am ino A cid Production 325

by-product of the reaction, through the form a-tion of non-reactive acetolactate. The B.subri-lis enzyme is preferable to the acetolactatesynthases of E. coli,which also use 2-ketobuty-

rate as a substrate, due to their overlappingroles in branched chain amino acid biosynthe-sis. The process is operated as a whole cellbiotransformation in a single strain with thereaction scheme as shown in Fig. 7 producing27 g L- ' of L-Z aminob utyrate. The effect ofthe concerted action of these three enzymeson the process of L-2-aminobutyrate biosyn-thesis is very significant. Process econom icsbenefit from the use of an inexpensive aminoacid such as L-threonine as the source of 2-ke-tobutyrate and L-aspartic acid as the aminodonor. Secondly with the additional acetolac-tate synthase activity present the ratio of L-2-aminobutyrate to L-alanine, the major amino

acid impurity increases to 22.5 :1 from 2.4: 1reducing the com plexity of the product recov-ery. Incremental improvements remain to bemade in the prevention of undesired catabo-

lism of substrates by metabolically activewhole cells as the overall produ ct yield is only54% although the substrates are almost entire-ly consumed.The detailed understanding of E.coli metabolism and genetics increases thelikelihood of eliminating such undesirable ca t-abolic side reactions.

4.2 A Novel Biosynthetic Pathwayto D-Phenylalanine

Unlike the L-amino acid transaminases(LAT), very few D-amino acid transaminases(DAT) (E.C. 2.6.1.21) have been extensively

* Y O 2 -

6 0 ,

aminotransferase

acetolactatesynthase7

cozHO&acatdactate

HOa-acetoin

Fig. 7. Transaminase/acetolactatesynthase coupled biotransformation reaction for biosynthesis of L- andD-2-arninobutyrate.



326 8 Engineering Microbial Pathways fo r Am ino A cid Production

studied. One of the best understood is thatfrom Bacillus sp. YM 1, which has been cloned,sequenzed (TANIZAWAt al., 1989), an d th ecrystal structure solved (SUGIO t al., 1995).

Recently, the da t genes from Staphylococcushuemolyticus (PUCC I t al., 1995),),Bacillus li-cheniformis (TAYLOR and FOTHERINGHAM ,1997), and Bacillus sphaericus (FOTHERING-HA M et al., 1998a) have been cloned, se-quenced, and overexpressed in E. coli.Thesefactors, plus their broad substrate specificitymak e them ideal candidates for use in the pro-duction of D -amino acids.

As with the LATs, the DATs have an equi-librium constant nea r unity. Fortunately, manyof the procedu res devised to optimize product

formation with the LATs a re also applicable tothe DATs. Product insolubility, high substrateconcentrations, unstable keto acid products(e.g., oxaloacetate), and the removal of the ke-to acid produc t enzym atically (e.g., conversionof pyruvate to acetolactate) can be used t o al-ter th e equilibrium of the DA T reaction in fa-vor of th e D-amino acid prod uct. The supp ly ofD-amino acid donor can be problematical be-cause most D-amino acids are either unavail-able as comm odity chemicals or are expensive.

This problem is generally solved by initially us-ing an L-amino acid such as L-aspartate, L-glu-tam ate, or L-alanine, which is conver ted in situto a racemate using an appropriate racemase.The DN A sequences of a number of racemas-es are available in the GenBank DNA data-base (NCBI, Bethesda, MD, USA) and thegenes are easily obtainable by PC R m ethodol-

ogy.The whole cell bioconversion reaction de-

scribed fo r L-amino acids is equally applicableto th e biosynthesis of D-am ino acids using the

D-amino acid transaminase as shown in Fig. 7.This has been successfully applied to the pro-duc tion of a nu mb er of D-amino acids. Theseinclude both D-2-aminobutyrate (FOTHERING-HAM et al., 1997) and D-glutamate.Theoretical-ly, this rou te is applicable to the p rodu ction ofother D-amino acids, as long as the keto acidcannot serve as a substrate for acetolactatesynthase, and the p roduct is not enzymaticallyracemized by the rec ombina nt cell.

SODA nd colleagues (GALK IN t al., 1997a,b; NAKAJIMAt al., 1988) have described a pro-

cedure which uses a thermostable DAT from

Bacillus sp. YM 1 in a coupled reaction withL-alanine dehydrogenase and alanine racem-ase (KUR ODA t al., 1990; TAN IZAW At al.,1988) such tha t p yruvate is recycled to D-ala-

nine via L-alanine. The process requ ires anN A D H rege neration system, which is suppliedby the action of a cloned, thermostable for-mate dehydrogenase (GA LKINt al., 1995 ).Thegenes encoding each enzyme were cloned andexpressed in a single E. coli strain. The overallprocedure works efficiently for D-glutamateand D-leucine while other keto acids testedshowed poor yields a n d o r poor enantiomericexcess. These problems were attributed to anumber of causes including reaction of th eL-alanine dehydro genase w ith a numb er of ke-

to acids tested, racemization of some of theamino acid products by the thermostable ala-nine racemase, and transam ination of keto ac-ids by the L-amino acid transaminases of thehost. Other drawbacks include the need forrelatively low sub strate conce ntrations (by in-dustrial standards) an d the high cost of the ke-to acid supply.

A versatile, modular approach for the pro-duc tion of D-amino acids is currently being de -veloped which addresses both D-amino acid

donor and the a-k eto acid acceptor supply, butdoes not require exogenous regeneration of

co-factors. A generic depiction of this ap -proa ch is shown in Fig. 8. The key step is the in-clusion of th e lad gene of Proteus myxofaciens,encoding th e L-amino acid deam inase which isused to produce the a-keto acid in situ fromthe L-amino acid which is provided as sub-strate or overproduced by the host. In eithercase, the deamination occurs intracellularly.The D -amino acid required as am ino donor issimilarly gene rated in situ from the L-isomer

by an appropriate cloned racemase. The twosubstrates are then converted by a clonedDAT to produce the corresponding D-aminoacid. Because the reaction is carried o ut und erfermen tative conditions, the reaction is shiftedtowards 100% yield by catabolism of the a -k e-to product and by regeneration of the aminodon or by cellular enzymes. This system can beused to produce D-amino acids with highenantiomeric excess even in the presence ofthe endogenous L-transaminases of the hostorganism. Using a combination of the ap-

proaches described, D-phenylalanine can be



5 Conclusions 327

L-amino add

D-amino acid lransaminase D-Pmino( p m a d n s s r r t l a )

D-Pmino aad donor u-keto add( tMI4DdUWD-P) ( p p v . t e , a-KG or O M )

L-amino acid

L-aminoadd as amino donor( e x k n u l l y fed;L&, ~ g l ur LIMP)

Fig. 8. Eng inee red biochemical pathway for whole cell biosynthesis of D-phenylalanine.

produced in high yields with 99% enantio-meric excess (FOTHERINGHAMt al., 1998b).D-Tyrosine can similarly b e syn thesized using afed batch fermentation process but with im-proved overall yields due to th e insolubility ofthe final product (R. F. SENKPIEL,npublisheddata). The promiscuous substrate range ofL-amino acid deaminase (L-AAD)and DAT

facilitates the use of this type of approach togenerate a wide range of D-amino acids fromtheir L -amino acid counterparts.

5 Conclusions

A variety of bacterial species have beenused for almost 40 years in the large scale fer-mentative production of amino acids for in -dustrial uses. The coryneb acteria have bee n

most widely used in this application due to

their propensity to overproduce and excretevery high concentrations of amino acids underspecific process conditions an d their early de-velopment for monosodium glutamate pro-duction. However, other organisms includingE. coli have also been successfully emp loyed inthe p roduction of am ino acids such as phenyl-alanine.

The engineering of bacterial metabolicpathways to improve comm ercial fermentativeprod uction of am ino acids has traditionally in-volved the u se of relatively crude, no n specificmutagenesis methods coupled to repeatedrounds of arduous screening for resistance totoxic amino acid analogs. Although inelegantin execution, these methods nevertheless ledto highly efficient and economically importantorganisms for large scale production of manyamino acids, even though the approach be-com es increasingly self-limiting as pro duction

organisms accum ulate non-specific mutations.



328 8 Engineering Microbial Pathways fo r Am ino A cid Production

With the increasing understanding of thebiochemistry and molecular genetics of aminoacid production in bacteria and the availabilityof recombinant DNA methodology in the

1980s the rate limiting steps of biosyntheticpathways and precursor supply were ad-dressed through the cloning and characteri-zation of the genes encoding the key biosyn-thetic enzymes. This led to a greater under-standing of the molecular mechanisms in-volved in biosynthetic pathway regulation andthe characterization of mutations introducedpreviously throu gh classical methods. Deregu-lation of gene expression, re-introduction ofspecific genes on multi-copy plasmids, and therelief of e nd produc t mediated feedback inhi-

bition has led to even greater titers of productfrom amino acid overproducing bacterialstrains.

Con sequently, many successes of amino acidoverproduction in bacteria have come fromthe powerful hybrid approach of rational de-sign and more traditional mutagenesis/selec-tion methodology. As microbial DNA se-quence information becomes increasinglyavailable through genomic sequencing thepossibility to combine genes from diverse or-

ganisms in a single host is providing new op-portunities to further engineer specific hostbacterial strains for amino acid overproduc-tion. Accordingly, there are increasing exam -ples of E. coli strains containing genes frommultiple organisms to enable efficient wholecell biosynthesis of D-amino acids and non-proteinogenic L-amino acids.

A s our unde rstanding of m icrobial molecu-lar biology increases and with the availabilityof newer mutagenesis methods such as errorprone PC R and gene shuffling the capability

to engineer m icrobial biosynthetic pathways toproduce natural and unnatural amino acidswill con tinue to increase an d diversify.

AcknowledgementI would like to thank D r. PAUL AYLORnd Dr.DAVID ANTALEONEor their assistance in thepreparation of this chapter.

6 References

ABE,S., TAKAYAMA,. (1972), in: Amino A cid-Pro-ducing M icroorganisms. Variety and Classijicati-on. pp. 3-38. Tokyo: Kodansha.

AGER,D. J., FOTHERINGHAM,. G., LANEMAN,.A.,PANTALEONE,. P., TAYLOR,. P. (1997), Thelargescale synthesis of unnatural amino acids,Chim. Oggi 15,ll-14.

ARCHER,. A. C., SINSKEY,A.. (1993),The DNA se-quenc e and minimal replicon of th e Corynebacte-rium glutamicum plasmid pSR1: evidence of acommon ancestry with plasmids from C. diphthe-riae,J . Gen. M icrobiol. 139,1753-1759.

BACKMAN,. C.,BALAKRISHNAN,. (1988),Plasmidsencoding ph eA gene products resistant tophenylalanine inhibition for use in microbial m a-nufacture of phenylalanine, US . Patent 4753883.

BAIT, C. A., FOLLETTIE,. T., SHIN,H. K., YEH,P.,SINSKEY, . J. (1985), Genetic engineering of

coryneform bacteria, Trends Biotechnol. 3, 305-

310.

BERRY,. (1996), Improving production of aromaticcompounds in Escherichia coli by metabolic en-gineering, Trends Biotechnol. 14,250-256.

BLEDIG,. A . (1994), Investigation of the genetic re-gulation affecting carbon flux through phospho-enol pyruvate in Escherichia coli K12, Thesis,University of Warwick, England.

BOERMANN,. R., EIKMANNS,. J., SAHM, . (1992),

Molecular analysis of the Corynebacterium gluta-micum gdh gene encoding glutamate dehydro-genase, Mol. M icrobiol. 6,317-326.

BROWN,. D. (1968),Regulation of aroma tic aminoacid biosynthesis in Escherichia coli K12, Genet-ics 60 Pt . l), 1-48.

BROWN,. D.,SOMERVILLE,. L. (1971),Repressionof aromatic amino acid biosynthesis in Escheri-chia coli K12,J. Bacteriol. 108,386-399.

CAMAKARIS,. ,P ~ A R D ,. (1973),Regulation of ty-rosine and phenylalanine biosynthesis in Escheri-chia coli K12. Properties of the tyrR gene pro-duct,J. Bacteriol. 115,1135-1144.

CHOI, . J., TRIBE, . E. (1982),Contin uous produc-tion of phenylalanine using an Escherichia coli re -gulatory mutan t, Biotechnol. Lett. 4,223-228.

CLEMENT,^., LANEELLE,. (1986),Glutam ate excre-tion mechanism in Corynebacterium glutamicum:triggering by biotin starvation or by surfactantaddition,J . Gen. Microbiol. U2,925-929.

COLON,. E., NGUYEN,.T., JETTEN, M. S. M., SINS-KEY, A. J., STEPHANOPOULOS,. (1995), Produc-tion if isoleucine by overexpression of i lvA in aCorynebacterium lactofermentum threonine pro-ducer, Ap pl. Microbiol. Biotechnol. 43,482488.

CREMER,., EGGELING,., SAHM , . (1991), Control



6 References 329

of the lysine biosyn thesis sequen ce in Corynebac-terium glutamicum as analyzed by overexpressionof the individual corresponding genes,Appl . En-viron. Microbiol. 57,1746-1752.

CRUMP,. P., ROZZELL,. D. (1992), in: BiocatalyticProduction of Am ino Acid s and Derivatives (RO-ZZELL, D. J., WAGNER,., Eds.), pp. 43-58. NewY ork Oxford Un iversity Press.

DAVIES,W. D., DAVIDSON ,. E . (1982),The nucleoti-de sequence of aroG, the gene for 3-deox y-~-ara-binoheptulosonate-7-phosphateynthetase (phe)in Escherichia coli K12, Nucleic Acids Res. 10 ,40454058.

DE BOER,L., DUKHUIZEN,. (1990), Microbial andenzymic processes for L-phenylalanine produc-tion, Adv. Biochem. Eng./Biotechnol. 41 (Microb.Bioprod. ) ,1-27.

DOPHEIDE,.A. A., CREWTHER ,

?,DAVIDSON,. E .(1972), Chorismate mutase-prephenate dehy-

dratase from Escherichia coli K12.11. Kinetic pro-perties, J . Biol. Chem. 247 ,4447452 .

DRATHS, . M., POMPLIANO,. L., CON LEY, . L.,FROST,. W., BERR Y, . e t al. (1992), Biocatalyticsynthesis of a roma tics from D-glucose: the role oftransketolase,J . Am . Chem. SOC. 14,3956-3962.

DUPERRAY,., JEZEQUEL,., GH AZI, ., LETELLIER,L., SHECHTER,. (1992), Excretion of glutamatefrom Corynebacterium glutamicum triggered byamine surfactants, Biochim. Biophys. Acta 1103,25C258.

EDWARDS, . R., TAYLOR,. P., HUNTER,M. G., Fo-THERINGHAM,. G. (1987), Europ. Patent Appl.229,161.

F. (1996), Pathw ay engineering for the productionof aromatic compounds in Escherichia coli, Nat-ure Biotechnology 14,620-623.

FOERBERG,., ELIAESON,., HAEGGSTROEM,.(1988), Correlation of theoretical and experim en-tal yields of phenylalanine from non-growingcells of a rec Escherichia coli strain, J . Biotechnol.

FOLLETTIE, . T., PEOPL ES, .P.,AGOROPOULOU,.,

SINSKEY,. J. (1993), Ge ne struc ture and expres-sion of the Corynebacterium flavum N13 ask-asdoperon,J. Bacteriol. 175,4096-4103.

FOTHERINGHAM,. G., DACEY, . A., TAYLOR,. P.,

SMITH,T.., HUNTER,M. G. e t al. (1986),The clon-ing and sequence analysis of the aspC and tyrBgenes from Escherichia coli K12. Com parison ofthe primary structures of the aspartate amino-transferase and aromatic aminotransferase of E.coli with those of the pig aspartate aminotrans-ferase isoenzymes,Biochem. J . 234,593-604.

FOTHERINGHAM,. G., TON, ., HIG GINS , . (1994),Amino acid hypersecreting Escherichia coli and

their preparation, US Patent 5,354,672.

FLOR ES, ., XIAO, ., BERRY, A., BOL IVAR ,.,VALLE,

7,319-332.

FOTHERINGHAM,. G., PANTALEONE, . P., TAYLOR,P. P. (1997), Bioca talytic produ ction of unna turalamino acids, mono esters, and N-protected deri-vatives, Chim. Oggi 15,33-37.

FOTHERINGHAM,. G., BLEDIG, . A., TAYLOR,. P.

(1998a), Cha racterization of the genes encodingD-amino acid transaminase and glutamate racem -ase, two D-glutamatebiosynthetic enzymes of Ba-cillus sphaericus ATCC 10208, J . Bacteriol. 180,43194323.

FOTHERINGHAM,. G., TAYLOR, . P., TON,J. L.(1998b), Preparation of D-amino acids by directfermentative means, US Patent 5,728,555.

FOTHERINGHAM,. G., GRIN TER , ., PAN TALEO NE,.P., SENKP IEL,. F.,TAYLOR,. P. (1999), Engineer-ing of a nov el biochemical pathway for the bio-synthesis of L-2-aminobutyricacid in Escherichiacoli K12, Bioorg. Med. Chem . 7,2209-2213.

FROST,. W., DRATHS, . M. (1995), Biocata lytic syn-theses of arom atics from D-glucose: renew ablemicrobial sources of aromatic compo unds,Annu.Rev. M icrobiol. 49,557-579.

GALK IN, ., KULAKOVA ,., TISHKOV,., ESAKI,N.,SODA , . (1995), Cloning of formate dehydrogen-ase gene from a methanol-utilizing bacteriumMycobacterium vaccae N10,Appl.Microbiol. Bio-technol.44,479483.

GALKIN,A., ULAKOVA,.,YAMAMOTO,.,TANIZA-WA,K.,TANAKA,. (1997a), Con version of 0-ke-to acids to D-aminO acids by coup ling of fou r en-zyme reactions,J . Ferment. Bioeng. 83,299-300.

GALKIN, ., KULAKOVA,., YOSHIMURA,., SODA,

K., ESAK I, . (1997b), Syn thesisof optically activeamino acids from a-keto acids with Escherichiacoli cells expressing heterologous genes, Appl .Environ. M icrobiol. 63,46514656.

GAR NER, . C., HERRMA NN,. M., YOSHINAGA ,.,NAKAMORI,. 1983), n:Am ino A cids: Biosynthe-sis and Genetic Regu lation (HERRMANN,. M.,SOMMERVILLE,. L., Ed s), Reading, MA: Addi-son-Wesley.

GET HING , . J., DAV IDSON,. E. (1978), Chorismatemutase/prephenate dehydratase from Escheri-

chia coli K12. Binding studies with the allostericeffector phenylalanine, Eur. J. Biochem. 86,

HAGINO, ., NAKAYAMA,. (1974a), Production ofarom atic am ino acids by m icroorganisms. VI. R e-gulatory properties of preph enate dehydrogenaseand prephenate dehydratase from Corynebacteri-um glutamicum, Agric. Biol. Chem . 38,2367-2376.

HAGINO, ., NAKAYA MA,. (1974b), Production ofaromatic amino acids by microorganisms. IV.

L-Phenylalanine production by analog-resistantmutants of Corynebacterium glutamicum,Agr. Bi-01. Chem. 38,157-161.

HAGINO, ., NAKAYAMA ,. (1974c), Produ ction of

165-174.




aromatic amino acids by microorganisms. V.

D A H P [3-deoxy-~-arabinoheptulosonate-phos-phate] synthetase an d its control in Corynebacte-rium glutamicum, Agric. Biol. Chem. 38, 2125-2134.

HAGINO, ., NAKAYAM A,. (1975a), Production ofaromatic amino acids by microorganisms. VIII.Regulatory properties of ch orismate mutase fromCorynebacterium glutamicum, Agric. Biol. C hem.

HAGINO, ., NAKAYAM A,. (1975b), Production ofaromatic amino acids by microorganisms.X. Bio-synthetic control in aromatic amino acid produc-ing mutants of Corynebacterium glutamicum,Agric. Biol. Chem. 39,351-361.

HASLAM, . (1974), The Shikimate Pathway. NewYork: Halsted .

HON ORE, ., CO LE, .T. (1990), Nucleotide sequenceof the aroP gene encoding the general aromaticamino acid transport protein of Escherichia coliK-12: homology with yeast transport proteins,Nucleic Acids Res. 18,653.

HUDSON, . S., DAVIDSON,. E. (1984), Nucleotidesequence and transcription of the phenylalanineand tyrosine operons of Escherichia coli K12, J .

Mol. Biol. 180,1023-1051.HWANG,. O., GIL, G. H ., CHO ,Y. J., KANG, K. R.,

LEE, . H., BA E, . C. (1985),The fermentation pro-cess for L-phenylalanine production using anauxotrophic regulatory mutant of Escherichiacoli, A pp l. Microbiol. Biotechnol. 22,108-113.

ICHIH AR A, . (1985), in: Transaminuses (CHRISTEN,P.,METZLER,. D., Eds.), p. 430. New York JohnWiley & Sons.

IKEDA,M., KATSUMATA,. (1992), Metabolic en-gineering to produce tyrosine or phenylalanine ina tryptophan-producing Corynebacterium gluta-micum strain, Appl. Environ. Microbiol. 58,781-785.

IKEDA,M., OZA KI, A., KATSUMA TA,. (1993),Phenylalanine production by metabolically en-gineered Corynebacterium glutamicum with thepheA gene of Escherichia coli, Ap pl. Microbiol.

Biotechnol. 39,318-323.IM , S . W. K., DAVIDSON,., PITTARD, . (1971),Phenylalanine and tyrosine biosynthesis in Es-cherichia coli K-12. Mutants derepressed for 3-deoxy-D-arabinoheptulosonic cid 7-phosphatesynthetase (phe), 3-deoxy-~-arabinoheptulosonicacid 7-phosphate synthetase (tyr), chorismatemutase T-prephe nate dehydrogenase, and transa-minase A, J. Bacteriol. 108,400-409.

ISHIDA, ., YOSHINO,., MAKIHARA,., SATO,K.,ENEI,H., NAKAMORI,. (1989), Improvement ofan L -threonine producer derived from Brevibac-terium flavum using threon ine operon of Escheri-

chia coliK-12,

Agric. Biol. Chem.53,2269-2271.

39,331-342.

ISHIDA,., SATO,K., HASHIGU CHI,., ITO,H., EN EI,H., NAKAMORI,. (1993), High fermentative pro-duction of L-threonine from ac etate by a Brevi-bacterium flavum stabilized strain transformedwith a recombinant plasmid carrying the Escheri-chia coli thr operon, Biosci. Biotechnol. Biochem.

JETTEN,M. S. M., SINSKEY,. J. (1995), Recent ad-vances in the physiology and genetics of aminoacid-producing bacteria, Crit. Rev. Biotech nol. 15 ,

JETTEN,M. S. M., GUBLER,M. E., MCCORMICK,.M., COLON, . E., FOLLE TTIE, . T., SINSKEY,. J.(1993), Molecular organization and regulation ofthe biosynthetic pathway for aspartate-derivedamino acids in Corynebacterium glutamicum,Proc. Ind. Microorg. pp. 97-104.

JETTEN,M. S. M., FOL LETTIE , . T., SINSKEY,. J.(1994), Metabolic engineering of Corynebacteri-umglutamicum,Ann. h!Y Aca d. Sci. 721 (Recom-binant DN A Technology 11), 12-29.

JETTEN,M. S. M., FOLLETTIE, . T., SINSKEY, . J.(1995), Effect of differen t levels of aspartokinaseon the lysine production by Corynebacterium lac-tofermentum, Appl. Microbiol. Biotechnol. 43,76-82.

KALINOWSKI,., CREMER, ., BACHMANN,., EGGE-LING, ., SAHM , ., PRU EHL ER,. (1991), Gen eticand biochemical analysis of the aspartokinasefrom Corynebacterium glutamicum, Mol. Micro-biol. 5,1197-1204.

KURODA,. , TANIZAWA,., TANA KA , ., SODA,K.,SAKAMOTO,. (19W), Alan ine dehydrogen asesfrom two Bacillusspecies with d istinct thermosta-bilities: molecular cloning, DN A and protein se-quence determination, and structural comparisonwith other NAD(P) +-dependent dehydrogen-ases,Biochemistry 29,1009-1015.

MARTIN,. F., SANTAMA RIA,., SANDOVA L,., DELREAL,G., MATEOS, . M. et a]. (1987), Cloningsystems in amino acid-producing corynebacteria,Bio/Technology 5,137-146.

MILLAR,G., LEWENDON,., HUNTER,M. G., COG-

GINS,. R . (1986), The cloning and expression ofthe aroL gene from Escherichia coli K12. Purifi-cation and com plete amino acid sequence of shik-imate k inase 11, the aroL -gene product, Biochem.J . 237,427-437.

MILLER,. E., BACKMAN,. C., O'CONNOR,M. J.,HATCH, .T. (1987), Produ ction of phenylalanineand organic acids by phosphoenolpyruvate car-boxylase-deficient mutants of Escherichia coli, J .

Ind. Microbiol. 2,143-149.MIWA, .,TSUCHIDA,.,KURAHASHI,.,NAKAMORI,

S., SAN O, ., MOM OSE , . (1983), Construction ofL-threonine overproducing strains of Escherichiacoli K-12 using recombinant DNA techniques,

57,1755-1756.

73-103.



6 References 331

Agric. Biol. C hem. 47,2329-2334.MIWA,K., M AT SUI, ., TERABE, M., ITO, K., ISHIDA,

M. et al. (1985), Construction of novel shu ttle vec-tors and a cosmid vector for the glutamic acid-producing bacteria Brevibacterium lactofermen-turn and Corynebacterium glutamicum, Gene 39,

MULCAHY,. (1988), Thesis,University of Dund ee,Scotland.

NAKAJIMA,., TAN IZAWA ,.,TANAKA, ., SODA, .(1988), Enantioselective synthesis of variousD-amino acids by a multi-enzyme system, J . Bio-technol.8,243-248.

NAMPOOTHIRI,., PAND EY, . (1998), Genetic tun-ing of coryneform bacteria for the overproduc-tion of amino acids, Process Biochem. 33,

NELMS,., EDWARDS,. M., WARWICK,., FOTHE-RINGHAM,. (1992), Novel m utations in the ph eAgene of Escherichia coli K-12 which resu lt in hig-hly feedback inhibition-resistant variantsof chorismate mutase/prephenate dehydratase,Ap pl. Environ. Microbiol. 58,2592-2598.

OZAKI,A., KATSUMATA,., OKA,T., FURUYA, .(1985), Cloning of the genes concerned inphenylalanine biosynthesis in Corynebacteriumglutamicum and its application to breeding of aphenylalanine producing strain, Agric. Biol.Chem.49,2925-2930.

PATN AIK, ., LIAO, J. C. (1994), Engineering of

Escherichia coli central metabolism for aromaticmetabolite production with near theoreticalyield, Ap pl. Environ. Microbiol. 60,3903-3908.

PATNAIK,., SPITZE R, . G., LIAO, . C. (1995), Path-way engineering for production of aromatics inEscherichia coli: confirmation of stoichiometricanalysis by independent modulation of AroG,TktA, and Pps activities, Biotechnol. Bioeng. 46,

PI, J., WOOKEY,. J., PIITARD,A. J. (1991), Cloningand sequencing of the pheP gene, which enco desthe ph enylalan ine-specific transport system ofEscherichia coli, J . Bacteriol. 173,3622-3629.

PITTARD,., GIBSON,. (1970), Regulation of biosyn -thesis of arom atic amino acids and vitamins,Curr.Top. Cell Regul. 2 , 2 9 4 3 .

POSTMA,. W. (1987), in: Escherichia coli and Salmo-nella Cellular and Molecular Biology (NEID-HARDT, F. C., Ed.), pp. 127-141. Washington, D CASM P ress.

PUCCI,M. J., THANASSI,. A., Ho, H.-T., FALK, . J.,DOUGHERTY,. J. (1995),Staphylococcus hemo ly-ticus contains two D-glutamic acid biosyntheticactivities, a glu tamate racemase and a D-aminoacid transaminase,J . Bacteriol. 177,336-342.

PUHLER,A., KASSING,., WINTERFELDT,., KALI-

NOWSKI,.,SCHAEFER,A.t al. (1990),A novel sy-

281-286.

147-161.

361-370.

stem for genetic engineering of amino acid pro-ducing corynebacteria and the analysis of AEC -resistant mutants, Proc. ECB.5 (CHRISTIANSEN,.,MUNCK,.,VILLADSEN,.,Eds.), pp. 975-978. Co-

penhagen: Munksgaard.RAY, . M.,YANOFSKY,., BAUER LE, . (1988), Muta-

tional analysis of the catalytic and feedback sitesof the tryptophan-sensitive 3-deoxy-~-arabino-heptulosonate-7-phosphate synthase of Escheri-chia coli, J . Bacteriol. 170,550&5506.

Identification of a prom oter sequence in the plas-mid pUL340 of Brevibacterium lactofermentumand construction of new cloning vectors forcorynebacteria containing two selectable mar-kers, Gene 56,199-208.

SCHRUMPF, ., SCHW ARZER ,., KALINOWSKI, J.,PUHLER,A,, EGGELING,., SAHM,H. (1991), Afunctionally split pathway for lysine synthesis inCorynebacterium glutamicum, J . Bacteriol. 173,451M516.

SCHWARZER,A.,UHLER, . (1991), Manipulation ofCorynebacterium glutamicum by gene disruptionand replacement, Bio/Technology 9,84-87.

SHIIO,. (1986),Tryptophan,phenylalanine, and tyro-sine,Prog. Ind. Microbiol. 24,(Biotechnol. Am inoAcid Prod.), 188-206.

SHIIO, ., MIYA JIMA,. (1969), Concerted inhibitionand its reversal by end p roducts as aspa rtate kin-ase in Brevibacterium flavum, J. Biochem. 65,849-859.

SHIIO, ., SUGIMOTO,. (1976), Altered prephe natedehydratase in phenylalanine-excreting mutantsof Brevibacterium flavum, J. Biochem. 79, 173-183.

SHIIO, ., SUGIMOTO,. (1979), ' h o components ofchorismate mutase in Brevibacterium flavum, J .

Biochem. 86,17-25.SHIIO, ., SUGIMOTO,. (1981a), Effect of enzyme

concentration on regulation of dissociable choris-mate mutase in Brevibacterium flavum, J . Bio-chem. 89,1483-1492.

SHIIO,., SUGIMOTO,. (1981b), Regulation at meta-

bolic branch p oints of aromatic amino acid bio-synthesis in Brevibacterium flavum, Agric. Biol.Chem. 45,2197-2207.

SHIIO, ., ISHII,K., YOK OZEK I,. (1973), Produ ctionof L-tryptophan by 5-fluorotryptophan resistantmutants of Bacillus subtilis,Agr. Biol. Chem. 37,

SHIIO,.,SUGIMOTO,., MIYAJIMA,. (1974), Regula-tion of 3-deoxy-~-arabino-heptulosonate-phos-phate synthetase in Brevibacterium flavurn, J .

Biochem. 75,987ff.SHIIO, ., M ORI,M., OZAKI,H. (1982), Am ino acid

aminotransferases in an amino acid-producingbacterium. Brevibacterium flavum, Agric. Biol.

SANTAMARIA,. ., MARTfN, J. F., GIL, J. A. (1987),

1991-1200.




Chem. 46,2967-2977.SHIIO, I., SUGIMOTO, ., KAWAMURA , . (1988),

Breeding of phenylalanine-producing Brevibacte-rium flavum strains by removing feedback regu-lation of both the two key enzymes in its biosyn-thesis, Agric. Biol. Chem. 52,2247-2253.

SHIIO, ., TOR IDE, ., YOKO TA, ., SUGIMOTO,.,K A-WAMURA,. (1989), Threonine manufacture bydihydropicolinate synthase-deficient Brevibac-terium, Can. Patent App l. CA N 113:38946.

SHIIO, .,YOSHINO , . ,SUGIMOTO,. (1990),Isolationand properties of lysine-producing m utan ts withfeedback-resistant aspartokinase derived from aBrevibacterium flavum strain with citrate syntha-se- und pyruvate kinase-defects ans feedback-re-sistant phosphoen olpyruvate carboxylase, Agric.Biol. Chem. 54,3275-3282.

SHIIO, ., SUG IMOTO,., KAWAMURA,. (1993), Isola-tion and properties of a-ketobutyrate-resistantlysine-producing mutants from Brevibacteriumflavum, Biosci. Biotechnol. Biochem. 57,51-55.

SUGIMOTO,. , SHIIO, . (1974), Regulation of pre-phenate dehydratase in Brevibacterium flavum , J .

Biochem. 76,1103-1111.SUGIMOTO,., SHIIO,I. (1980), Purification a nd pr o-

perties of bifunctional 3-deoxy-~-arabino-heptu-losonate 7-phosphate synthetase-chorismatemutase component A from Brevibacterium flu-vum, J . Biochem. 87,881-890.

SUGIMOTO,. , SHIIO, . (1985), Enzymes of commonpathway for arom atic amino acid biosynthesis inBrevibacterium flavum and its tryptophan-prod-ucing mutants, Agric. Biol. Chem. 4 9 , 3 9 4 3 .

SUGIMOTO,. , MAKAGAWA,.,TsucHIDA,T.,SHIIO,.(1973), Regulation of aromatic amino acid bio-synthesis and production of tyrosine andphenylalanine in Brevibacterium flavum, Agr.Biol. Chem. 37,2327-2336.

SUGIO, . , PETSKO,G. A., MANN ING,. M., SODA,K.,RINGE,D. (1995), Crystal structure of a D-aminoacid aminotransferase: how the protein controlsstereoselectivity, Biochemistry 34,9661-9669.

TANIZAWA,., OHSHIM A,A., CHEIDE GGER ,A.,NA-

GAK I, . ,TANAKA , . , SODA,K. (1988),Thermost-able alanine racemase from Bacillus stearother-mophilus: D N A and pro tein sequence determina-tion an d secondary stru cture prediction, Bioche-

mistry 27,1311-1316.TANIZAWA,., ASAN O, ., MASU,Y.,KURAMITSU,.,

KAGAMIYAMA,. et al. (1989),Theprimary struc-ture of thermostable D-aminO acid aminotrans-ferase from a thermophilic Bacillus species andits correlation w ith L-amino acid aminotrans fera-ses, J . Biol. Chem. 264,2450-2454.

TAYLOR,. P., FOTHERINGHAM,. G. (1997),Nucleoti-de sequence of the Bacillus licheniformis ATCC10716 da t gene and comparison of the p redictedamino acid sequence with those of othe r bacterialspecies,Biochim. Bioph ys. Acta 1350,38-40.

TAYLOR, P., PANTALEONE,. P., SENKPIEL,. F.,Fo-THERINGHAM,. G. 1998),Novel biosynthetic ap-proaches to the production of unnatural aminoacids using transaminases, Trends Biotechnol. 16,

412-418.TESCH,M., EIKMAN NS,. J., DE GR AAF,A.A. , AHM,

H. (1998),Am mon ia assimilation in Corynebucte-rium glutamicum and a glutamate dehydrogen-ase-deficient mutant, Biotechnol. Lett. 20, 953-957.

TOK ORO , ., OSH IMA, ., OK II,M., YAMAGUCHI,.,TAN AKA , ., KINOSHITA,. (1970),Microbial pro-duction of L-phenylalanine from n-alkanes, Agr.Biol. Chem. 34,1516-1521.

TOSAKA,O., TAKINAMI,. (1986), Lysine, Prog. Znd.Microbiol. 24 (Biotechnol. Amino Acid Prod.),152-172.

TRIBE,D. E. (1987),Prep aration and use of novel Es-cherichia coli strains to manufacture phenylalani-ne , US Patent 4,681,852.

TSUCHIDA,., KUBOTA, ., MORINAGA,., MATSUI,H., ENE I,H., YOSHINAGA,. (1987),Production ofL-phenylalanine by a mutant of Brevibacteriumlactofermentum 2256,A gric. Biol. Chem .51,2095-2101.

WALLACE,. J., P I ~ A R D ,. (1969), Regulator genecontrolling enzymes concerned in tyrosine bio-synthesis in Escherichia coli, J . Bacteriol. 97,1234-1241.

WHIPP,M. J., PITTARD,A.. (1977),Regulation of aro-matic amin o acid transpo rt system s in Escherichiacoli K-12,J.Bacteriol. 132,453-461.

YAMA DA, . , KINOSHITA,., TSUNODA,., AIDA ,K.(1973), The Microbial Production of AminoAcids. New York: Halsted.



9 Biotechnological Production of

Natural Aroma Chemicals byFermentation Processes

JURGEN ABENHORSTHolzminden, Germ any

1 Introduction 3342 Ferm entation Processes 334

2.1 Ferm entative Productio n of Vanillin 3342.1.1 Ou tline of Diffe rent Processes Ap plied to the Formation of Vanillin 3342.1.2 Produc tion of Ferulic Acid as an Interm ediate for Vanillin Produc tion 3362.1.3 Production of Vanillin from Ferulic Acid 338

2.2 Ferm entative Production of Flavor-Active Lactones 3402.3 Fermentative Productio n of Carboxylic Acids 344

2.4 Fermentation of Oth er Natural Aro ma Com pounds 3452.4.1 Production of Flavor-Active Aldehy des and Alcohols 3452.4.2 Pheno lic Co mpounds 3462.4.3 Methyl Ke tones 347

3 Conclusions 3474 Referen ces 347





334 9 Biotechnological Production ofNatural Aro ma Chemicals by Fermentation Processes

1 Introduction

The demand for natural flavors by the cus-

tomers of th e Western world increased duringthe last decade. Naturalness is correlated bymany people with health and other positivemeanings. In E urope and the U SA natural fla-vors a re clearly defined legally.

In the E U Directive 88/388/EEC the term“natural” is allowed for a flavoring substancewhich is obtained “by appropriate physicalprocesses (including distillation and solventextraction) or enzymatic or microbiologicalprocesses from material of vegetable or ani-mal origin either in the raw state or after pro-

cessing for hum an consumption by traditionalfood prepa ration processes (including drying,torrefaction, and ferm entation)”.

In the US A th e following definition of natu -ral flavors app lies: “Th e term natural flavo r o rnatural flavorin g mean s the essential oil oleo-resin, essence or extractive, protein hydroly-sate, distillate, or any product of roasting, hea-ting, or enzymolysis, which con tains the flavor-ing constituents derived from a spice, fruitjuice, vegetable, or vegetable juice, edible

yeast, herb, bark, bud, root, leaf, or similarplant material, meat, seafood, poultry, eggs,dairy products, or fermentation productsthereof, whose significant function in food isflavoring rather than nutritional . .” (CFR,1993).

Since the 1970s several reports appeared onthe microbial and enzymatic synthesis of fra-grance an d a roma chemicals (for reviews, seeSCHINDLERnd SCHMID, 982; and SCHAR PF,JR. et al., 1986).

2 Fermentation Processes

2.1 Fermentative Production

of Vanillin

Vanillin is the m ost important aroma chem-ical with an annual consumption of about12,000 tons. It is broadly used for the flavo ringof many different foods like chocolate, ice-

cream, baked goods or liqueurs, and also inperfum ery. Vanillin has an intensely sweet andvery tenacious creamy-vanilla-like odo r and atypical sweet vanillin-like taste.

Synthetic vanillin is currently producedfrom guaiacol and lignin (CL AR K, 990).Na tural vanilla flavor is classically obtained

from the alcoholic extracts of cured vanillapods (RANADIVE,994). D ue to the limitedavailability of vanilla pods and the fluctuationsof delivery based on vary ing harvest yields, de-pending on the weather, efforts were under-taken to find alternative sources.

If th e price of natu ral vanillin was calculatedon the yield in cure d vanilla pods, which com-prise less than 2% vanillin, it would be in the

range of 2,000 U.S. $ kg -l, which is extremelyhigh compa red to this na ture identical vanillinobtain ed by chem ical synthesis that only costsabout 12 U.S.$ kg-l. This is one of the incen-tives of the flavor industry for the search ofnatural vanillin from o ther sources.

2.1.1 Outline of Different

Processes Applied to the Formation

of VanillinSince the beginning of the 1990s a num ber

of patents for the biotechno logical productionof vanillin have been disclosed. In som e earlierstudies vanillin was mentioned only as an in-termediate in the microbial catabolism of dif-ferent chemicals of plant origin. For example,TADASAescribed the degradation of eugenolwith a Corynebacterium sp. (TAD ASA , 977)and a Pseudomonas sp. (TADASAnd KAYAHA-RA, 1983) and found vanillin as one of the in-

termediates. Many studies have been perfor-med on the d egradation of lignin. Ferulic acidhas frequently been used as a simple modelcompound in studies for the d egradation of li-gnin po1ymers.A num ber of w orkers found va-nillin as an intermediate in the ferulic acid ca-tabolism of different microorganisms (TOMSand WOOD,1970; SUTHERLANDt al., 1983;

The first patent for the alternative produc-tion of na tural vanillin was disclosed by DO L-FINI et al. (1990). This was not a re al biotech-

nological process because there turmeric oleo-

Om K, 1985).



2 Fermentation Processes 335

316 "C, 1680 psi

resin, in which curcumin is the main ingre-dient, was subjected to th e action of heat andpressure in the presence of water by a conti-nuous or batch process to produce vanillin

(Fig. 1). This process yields only 20% of vanil-lin.

The first enzymatic process for the produc-tion of vanillin was disclosed by YOSHIM OTOtal. (1990a). They used a dioxygenase , isolatedfrom a new Pseudomonas sp. TMY1009 (YOS-HIMOTO et al., 1990b).A s substrates they usedisoeugenol (Fig. 2c) and 1,2-bis(4-hydroxy-3-methoxypheny1)ethylene (Fig. 2a) or the cor-responding glucoside (Fig.2b).

The first microbial fermentation process forthe production of vanillin was disclosed byRABENHORSTnd HOPP 1991) with Haar-mann & Reim er. They used bacteria of thegenera Klebsiella, Proteus, and Serratia andeugenol or isoeugenol as substrate (Fig. 3).Only w ith isoeugenol reasonable v anillin con-centrations of 3.8 g L-' were obtained after9 d.

MARKUSt al. (1993) with Q uest disclosed adifferent enzymatic approach for the forma-tion of van illin. They used lipoxygenase, alsoknown as lipoxidase (EC.1.13.11.12), an en-

zyme com mon in different plants like soy, po-tato, cucumber, grape, and tea. As substratesthey used be nzoe siam resin, in which the main

\H,CO

component is coniferyl benzoate, or isoeuge-no1 and eugenol. Only with isoeugenol theyreached reasonable vanillin concen trations ofabout 10 g L-' within 4 d. With benzo e siam

resin and e ugenol only concentrations of 2 to4 g L-' and 0.3-0.5 g L-' were obtained, re-spectively. D ue to the high price of lipoxygen-ase this process seems to be commercially un-attractive. This is also true for all processesbased on natu ral isoeugenol, because it is only

available from some essential oils (ylang-ylang, nutmeg) as a m inor comp onent. There-fore, the p rice of this subs trate is unacceptablyhigh.

An interesting approach was followed by aUS biotech company, ESCAgenetics Corp.,who tried to prod uce vanillin by plant cell cul-ture (KNUTH nd SAHAI, 991) in ferm enters.With a cell line with an extremely high dou-bling time of 106 h they obtaine d con centr a-tions of 11,900 mg L - ' vanillin under high celldensity conditions (27 g cell dry weight per L;

SAHAI, 994) but they see med not to be able toscale up the process to a com mercial scale, be-cause the product never reached the marketand the company does not exist anymore.A very ambitious approach was followed by

FROST nd coworkers, who w ere trying t o ob-tain vanillin from glucose by metabolic engi-neering in Escherichia coli via the shikim ic ac-

CHO OH

w

Fig. 1.Hydrolysis of curcumin according to DOLFINIt al. (1990).



336

a)

9 Biotechnological Production of Natural Ar om a Chemicals by Fermentation Processes

/U0

CH CH= H

d:Hioxygenase * OHC+

--

Fig. 2. Formation of vanillin with dioxygenase acco rding to YOSHIMOTOt al. (1990a, b).

SerratiaCH, -CH = H4:~+ marcescens OHC

Fig. 3. Vanillin form ation from isoeugenol by Serratia marcescens according to RABENHORSTnd HOPP

(1990).

id pathway (LI and FROST, 998). The main 2.1.2 Production of Ferulic Acid

as an Intermediate for Vanillinimiting steps are a lack of a sufficient cate-

chol-0-methvltransferase activitv for the for-mation of vkil li c acid from p;otocatechuicacid, and a lack of selectivity for the aryl alde-hyde dehydrogenase, so that about 10% of iso-vanillin were synthesized. Until now, obvious-ly, only parts of the pathway have been cloned

within one strain.

Production

Ferulic acid is an ubiquitously found phenol-ic cinnamic acid derivative in plants. It islinked to various carbohydrates as glycosidic

conjugates, occurs as an ester or amide in var-




ious natural produc ts, and is a part of the com-mon plant polymer lignin (ROSA ZZA t al.,1995).Free fe rulic acid is not easily available innature (STRACK,990).

In a p atent ANT RIM nd HAR RIS isclosedthe isolation of ferulic acid from corn hulls(ANTRIMnd HARR IS, 977). Du e to the chem-ical alkali hydrolysis of the corn hulls to c leavethe ester bonds, the resulting material cannotbe considered as natural.

Another possible source is eugenol, themain component of the essential oil of theclove tree Syzigium aromaticum (syn. Eugeniacariophyllus). It is a cheap natural aromachemical and available in sufficient amountson the world market. It is used in many per-

fume and arom a compositions due to its orien-tal an d spicy clove odor (BAU ER t al., 1990).When eugeno l is metabolized by different bac-teria, ferulic acid is formed as an intermediate(TADASA,1977; TADASAand KAYAHARA,1983).

In 1993, HOP P nd RABENHORSTisclosed apatent application with Haarmann & Reimerfor a new Pseudomonas sp. HR 199, isolatedfrom soil, and a process for the production ofmethoxyphenols from eugenol. With this fer-

mentation process ferulic acid con centrationsof 5.8 g L-' within 75 h (Fig. 4) were reached.

10000

s 7000

6000

2 5000

E

m

a 4000

' 000

a 2000

:::rE

1000

0

D ue to the toxicity of eugen ol a repe ated addi-tion of the substrate was necessary. They ob-tained a high molar conversion of eugenol to

ferulic acid of mo re tha n 50%.

By variation of the process parametersoth er interesting chem icals like coniferyl alco-hol o r vanillic acid could also be obtained ingood yields (RABENHORST,996).

In the meantime the process has been fur-ther optimized for higher volumetric and m o-lar yield and is run now successfully on a 40 m3scale.

Because the degradation of eugenol is ex-

tremely rare within the genus Pseudomonas(STEINB~CHEL,ersonal com munication), andthe process of ferulic acid production is com-

mercially important, the genes and enzymes ofthe m etabolic pathway were isolated and char-acterized (STJZINBUCHELt al., 1998). Thispathway with the genes and enzymes, there -fore, is outlined in Fig.5.

During these studies it was found that vanil-lin is also an interme diate of the degrada tionof eugenol, but it was detected only occasion-ally in the culture broth in trace am ounts. Thereaction of ferulic acid to vanillic acid is ob-viously very effective, probably du e to the

higher reactivity and toxicity to th e cells of th efree aldehyde.

s

8

s

4000

3000a0

2000-:im

1000

0

0 8 16 24 32 40 48 56 64 72

hours

Fig. 4. Production of ferulic acid by Pseudomonas sp. HR 199. A Eugenol addition; x eugenol con-centration;* ferulic acid concentration;+ coniferyl alcohol concentration; vanillic acid concentration.



338 9 Biotechnological Production ofNatural Ar om a Chemicals by Fermentation Processes

Eugenol Coniferyl alcohol Coniferyl aldehyde

CH,- CH= H,

Eugenol

hydmxyla=*

+(H OCH,

CH =CH-CHzOH CH =CH-CHO

Coniferyl alcohol

dehydrqenase

* b O C H 3

OH

@OCH,Q-

OH

Vanillin

CHO

Enoyl-CoA-

OCH,I OH be ty l -CoAI I

Van illic acid

COOH

Feruloyl-CoA I Ferulic acid

//O

S-COA\CH=CH-CI

//O

I ‘OH

vCH-CH-C

Fendoyl-CoA-

P O C H ,H P O C H 3H

Protocatechuic acid 13-Carboxy-cis-cis-muconoate

COOH COOH

OH OH

Fig. 5. Eug enol degradation pathway in Pseudomonas sp . HR 199 (STEINBUCHELt al., 1998).

2.1.3 Production of Vanillin from

Ferulic Acid

There were several attempts to use ferulicacid as a substrate for the production of natu-ral vanillin. The chemical reaction looks veryeasy: a hydrolytic cleavage with the release of

an acetate residue should give vanillin (Fig.6).

In 1992, a process from LABUDA t al. wasdisclosed for Kraft General Foods, who pro-duced vanillin from ferulic acid with differentmicroorganisms such as Corynebacterium glu-tamicum, Aspergillus niger, Pseudomonas p u-tida, and Rhodotorula glutinis in the presenceof sulfhydryl compounds such as dithiothreitol

(LABUDA t al., 1993).The vanillin concentra-




H,CO*CH=CH-COOH

HOMFig. 6. Vanillin forma tionfrom ferulic acid. Ferulicacid

tions were quite low and did not exceed210 mg L- l after 1,296 h. D ue t o the use of th eexpensive sulfhyd ryl com pounds, the low yield,and long process times this process seems notto be commercially feasible. 'Ikro groups atIN R A (Institut National de la Recherche Agro-nomique) together with Pernod-Ricard havebeen working for several years on a fungal bio-conversion process for the p roduction of vanil-lin from ferulic acid (GRO SS t al., 1991; FAL-CONNIER et al., 1994;MICA RD t al., 1995; LES-AGE-MEESSENt al., 1996a, b, 1997).They useda strain of Pycnoporus cinnabarinus, but hadproblems to channel the metabolic flux fromferulic acid to vanillin. They found at leastthree pathways for the metabolism of ferulicacid. A reductive pathway leads to coniferylaldehyde, another on e by the propenoic sidechain deg radation to vanillic acid, and a thirdon e via laccase to unsoluble polymers. The

formed vanillic acid is either decarboxylatedto m ethoxyhydroquinone or reduced to vanil-lin and vanillyl alcohol. So the concentrationof vanillin was only 64 mg L-' after 6 d (FAL-CONNIER et al., 1994).To avoid the decarboxy-lation of vanillic acid they found that it wasadvantageous to use cellobiose as carbonsource and fed additional cellobiose prior tothe addition of ferulic acid. This resulted in anincrease of van illin up to 560 mg L-' after 7 d.They postulated that cellobiose acts either asan easily metabolizable carbon source, re-

quired for the reductive pathway to occur, oras an inducer of th e cellobiose quinone oxido-reductase, a known inhibitor of vanillic aciddecarboxylation (LESAGE-MEESSENt al., 1997).An other approach w as a two-step process withthe combination of two fungal strains, i.e., anAspergillus niger fermentation together withth e Pycnoporus cinnabarinus or Phanero-chaete chrysosporium , where one strain con-verted ferulic acid to vanillic acid, while thesecond strain reduced vanillic acid to vanillin.But here vanillin concentrations were also<1gL-1 .

\

CHXOOH Vanillin

Th e process with the highest yields of vanil-lin so far is that disclosed by Haarmann &Reimer (RABENHORSTnd HOPP,1997).

In this process a filamentous growing acti-nomycete H R 167 was used. This strain wasidentified by the DSMZ, Braunschweig, Ger-many, as a new Amycolatopsis species.This re-sult was based on the typical fatty acid pa tternof isolanteiso- plus 10-methyl-branched satu-rated and unsaturated fatty acids plus 2-hy-droxy fatty acids, which are diagnostic for re-presentatives of the genus Amycolatopsis.

Since H R 167 also displayed the typical ap-pearance of representatives of the genus A m y -colatopsis - beige to yellow sub strate myceli-um and, on some m edia, also a fine white ae ri-al mycelium - he assignment of H R 167 to thegenus Amycolatopsis was additionally sup-ported by the morphology.

The com parison of the fatty acid patterns of

the strain with the e ntries of th e D SM fatty ac-id databases by principal component analysisresulted in an assignment to Amycolatopsismediterranei for the new strain. However, thesimilarity index is too low (0.037 and 0.006) topermit definitive species assignm ent.

Additionally performed analyses of the 1 6srDNA and comparison of the diagnostic par-tial sequences with the 16s rDNA databaseentries of Amycolatopsis type strains support-ed the results of the fatty acid analyses.In th ecase of the 16 s rDN A sequences, high homol-

ogy could also not be demonstrated with theknown Amycolatopsis type strains. On the ba-sis of the grea t differences in the 16 s rDN A se-quence s (similarity 99.6%), it has to be statedthat H R 167 is a new strain of a previously un-known Amycolatopsis species.

With this new strain a fed-batch fermenta-tion w as established. Within 32 h vanillin con-cen trations of 11.5 g L -' vanillin with a mo laryield of nearly 80%were obtained on th e 10L

scale (Fig. 7).This process has b een successfully scaled up

to the 40 m3 production scale.



340 9 Biotechnological Production ofNatural Aro ma Chemicals by Fermentation Processes

100

90

80

70

60

40

30

20

10

0

0 5 10 15 20 25 30 35

Time [h]

Fig. 7.Time course of vanillin production with Amycolatopsis sp. HR 167 in a 10L fermenter.+Vanillinconcentration;+ erulic acid concentration ;+erulic acid addition;- Oz.

The vanillin is then isolated from the fer-mentation broth by well established physicalmethods like extraction, distillation, and crys-tallization.

The genes and the enzymes of this reactionwere also characterized in Pseudomonas sp.(Fig.5 ) .Ferulic acid is first activated by a ferul-ic acid CoA synthase to ferulic acid CoA ester.This ester is then cleaved by an enoyl CoA hy-dratase to vanillin (NARBAD t al., 1997; GAS-

SO N et al., 1998; NARBA D nd GASSON,998;STEINBUCHELt al., 1998).Vanillin is then fur-ther metabolized via vanillic acid, protocate-chuic acid, and 3-carboxymuconolactone and3-oxoadipate to succinyl-CoA, which is part ofthe tricarboxylic acid cycle (PRIEFERTt al.,1997; OVERHAGEt al., 1999).

2.2 Fermentative Production of

Flavor-Active Lactones

A number of lactones are important flavorchemicals. Most important is y-decalactone. Ithas a typical peachy flavor (Tab. 1).Many pat-ents and articles on the formation of y-deca-lactone have been published over the years.

Most of the processes use ricinoleic acid, themain fatty acid in castor oil, or esters thereoffor the formation of y-decalactone.The forma-tion of y-decalactone in the catabolism of ri-cinoleic acid by yeasts of the genus Candidawas first observed by O K U I t al. (1963). Ricin-oleic acid is degraded by four successive cycles

of P-oxidation into 4-hydroxydecanoic acid,




Tab. 1.Sensoric Descriptions of Flavor Active Lactones

Nam e Formula Sensoric Description

y-Decalactone oily-peachy, extraordinarily tenacious odo r; verypowerful, creamy-fruity, peach-like ta ste in concen-

4 0 rations <5 ppm

y-Dodecalactone fatty-peachy, somew hat musky od or

g-Octalactone sweet-herbaceous coconut-like od or

very powerful a nd tenacious sweet creamy, nut-likeodo r with a heavy fruity un dertone; taste is creamy,sweet coconut-peach-milk-like <2 ppm

SDecalactone

powerful fresh-fruity, oily odo r, pear-peach-plum-like taste a t 2-5 ppm

SDodecalactone

soft, lactonic, milk-like, cream -like, little fruity,peach-like, coconut m ilk note

SOctalactone

very ten acious , soft, lactonic, sweet, fatty, creamy,

butter-like, fruity flavor of peach or a pricot

6-Pent yl-2-pyrone

which lactonizes at lower pH values to y-de-calactone (GATFIELD t al., 1993; Fig. 8). It isworth mentioning that the microbial lactoneformation results in the same enantiomericconfiguration of the lactone as it is found inpeaches and o ther fruits.

A number of different yeasts have been

used for the production of y-decalactone.In the first patent FARBOODnd WILLIS

(1983) with IFF claimed the production ofy-decalactone with Yarrowia lipolytica andvarious oth er yeasts from ricinoleic acid, butthe yields were less than 1 g L-'. In a patentWINKet al. (1988) with Hoechst disclosed aprocess for the produ ction of peach flavor withMonilia fructicola. Afte r extraction of the fer-mentation b roth and distillation they obtaineda m ixture of y-octalactone (Tab. l ) ,y-decalac-tone, and phenyl ethanol in a conce ntration of0.2 g L-1-1 g L-1.

CHEETH AM t al. (1988) used Sporobo-lomyces odorus and Rhodotorula glutinis.

GATFIELD t al. (1993) reported the use ofcastor oil and Candida lipolytica for the pro-duction of y-decalactone. They obtain ed y-de-calactone con centrations of 5 g L-' within 2 d.In the fermentation broth they identified an

additional lactone, 3-hydroxy-y-decalactone(Fig. 9), as a by-product of y-decalac tone pro -duction. It is probably produ ced by th e lacton-ization of 3,4-dihydroxy decanoic acid, whichis form ed as a result of the P-ox idation of 4-hy-droxy decanoic acid.

The group of SPINNLERtudied the y-deca-lactone production with Sporidiobolus spp. likeSporidiobolus salmonicolor, Sporidiobolusruinenii, Spo ridiobo lus johnso nii, and Sporidi-obolus pararoseus. These strains are very sen-sitive to y-decalactone, while the open form,i.e., 4-hydroxydecanoic acid, is less toxic (Du-



342 9 Biotechnological Production ofNatural Arom a Chemicals by Fermentation Processes

(CH,),-COOH 12-hydroxyoctadeo9enoic acid

OH

CH,COSCoAR-oxidation

(CH,),-COOH 10-hydroxy hexadec-7eno ic acid

CH3-(CH2)57OH

3 CH,COSCoA3 R-oxidation

C H 3 - ( C H , ) , y COOH 6-hydroxy dodec3enoic acid

OH

6-hydroxy dodecanoic acid

Ieduction

C H 3 - ( C H 2 ) 5 v C O O H

OH I

R-oxidation

4-hydroxy decanoic acid

OH

cyclization

y-decalactone

Fig. 8. P-Oxidation pathway of ricinoleic acid for the formation of ydecalactone in Yurrowia lipolytica (after

GATFIELDt al., 1993).

FOSSB et al., 1998).So the yields of y-decalac-tone in fermentations with these yeasts werealways very low (< 1 g L-').

In another patent application CARDILLOtal. (1989) used Aspergillus niger, Pichia etch-

ellsii,and Cladosporium suaveolens to produce

y-decalactone from castor oil. Additionallythey produced 1 g L-' of y-octalactone from10 g L-l coconut oil.NICAUDt al. (1996) used a multiple auxo-

trophic mutant, Yarrowia lipolytica POlD, to

obtain y-decalactone in high yields. At the end




4-hydroxy decanoic acid

OH

partial &oxidation

OOH

O H

OH

3,4-dihydroxy decanoic acid

3-hydroxy-y -decalactone

Ido

cyclization

Fig. 9. Formation of 3-hydroxy- ydecalactone.

of the growth phase they transferred the con-

centrated biomass into an uracil limited medi-

um where the biotransformation took place.

After 75 h the culture broth yields 9.5 gL-' of

the product (PAGOT t al., 1997). They usedmethyl ricinoleate as substrate.A different approach has been followed by

KUMIN and M U N C H (1997). They used Mucorcircillenoides for the biotransformation of

ethyl decanoate. After 60 h they reached

10.5g L-' y-decalactone.

A synopsis of the different ways to y-deca-

lactone has been presented recently by

KRINGS nd BERGER1998).

A Japanese group at T. Hasegawa Co. (Go-CH O et al., 1995) studied the biotransformation

of oleic acid to optically active y-dodecalac-tone. They established a two-step process.

First, oleic acid is oxidized to lo-hydroxystear-

ic acid with a newly isolated gram-positive

bacterial strain, bakers' yeast was then used

for the biotransformation of the hydroxy acid

to ( R ) -y-dodecalactone (Fig. 10).The biotrans-

formation yield from oleic acid to y-dodeca-

lactone was 22.5% on the flask scale.

A process for the production of Sdecalac-

tone and Sdodecalactone was established by

VAN DER SCHAFT t al., 1992). They used the

PFw (VAN DER SCHAFT and DE LAAT,1991;

massoi lactones 2-decen-1,5-olideand 2-dode-

cen-1J-olide, which are the main components

of massoi bark oil with 80% and 7%, respec-

tively, as substrate.These unsaturated lactones

were hydrogenated with baker's yeast (Fig.11). By stepwise addition of 2-decen-5-olide

they obtained about 1.2 g L-' Sdecalactone

H,C - CHJ7 - H = H- CH,), - OOH

bacterial hydroxylation

OHI

H3C- CH,), - H - H- CH,), - OOH

yeast &-oxidationand cyclization

Fig. 10. Formation of ydodecalactone from oleic

acid.



344 9 Biotechnological Production ofNatural Aroma Chemicals by Fermentation Processes

A 0 decend-olide

yeast reduction

Fig. 11.Formation of Gdecalactone from massoi lac-tone.

within 16 h. By addition of P-cyclodextrin thisprocess could be accelerated significantly.They also found that a number of molds per-formed the same reaction.

The same reaction with the use of differentbacteria has also been patented by Hasegawa(GOCHOt al., 1998).The best results were ob-

Tab. 2. Carboxylic Acids U sed in Flavor Compo sitions

tained with Pseudomonas putida ATCC 33015.

With this strain they were able to isolate near-ly 6 g L-' Gdecalactone from the fermenta-tion broth after 48 h of incubation.

2.3 Fermentative Production of

Carboxylic Acids

A number of short-chain carboxylic acidsare also important as natural flavor ingre-dients, either for useper se or as substrates forthe production of a broad number of flavor es-ters, which can be obtained by enzymatic syn-thesis with lipases and esterases (for a review,see GATFIELD,992). Acetic acid, which isprobably the largest by volume and commonlyused as vinegar is not mentioned here.

In Tab. 2 examples of several important car-boxylic acids and their sensory descriptionsare presented.

Nam e Structure Sensoric Description"

n-Butyric acid CH,-CH+H,.jOOH powerful, penetrating, diffusive sour o dor, reminis-cent of rancid bu tter; used in butte r, cheese, nut,fruit, and othe r flavors

powerful, diffusive sou r odo r, slightly less repulsive,and also less buttery tha n n-butyric acid; in extrem edilution alm ost pleasant, fruity; th e taste is, in pro -per dilution and with adequ ate sweetening, pleasantcreamy-fruity, while buttery-cheesy no tes dom inatein the absence of swe eteners

H&

/

Isobutyric acid

CH-COOH

H3C

very diffusive, acid-acrid, in mo dera te dilution chee -sy, unpleasant od or of poo r tenacity; in extre medilution th e odor becomes m ore agreeable, herb-aceous and dry; used in nut and coffee flavors dueto its peculiar warm-herbaceous taste in concentra-tions <20ppm

Isovaleric acid H,C\

CH-CH,-COOH/

H3C

2-Methyl butyric acid H3C- CH,- C H-C OO H pungent, acrid odo r reminiscent of Roquefort chee-se; acrid taste, which becomes quit e pleasant a ndfruity-sour at dilutions <10ppm; used in cheese,butter, cream , and chocolate flavors

I

CH3

Propionic acid HC- CH,- COOH pungen t sour odor reminiscent of sou r milk, cheese,or s our bu tter; used in raspberry, strawberry,cognac, butt er flavors

a Sensoric descriptions according to ARCTANDER,969




For the production of butyric acid two dif-ferent w ays are feasible. O ne rou te is the clas-sical anaerobic ferm entation w ith C lostridiumbutyricum (VANDAK t al., 1996). The other

one is the m icrobial oxidation of butano l withGluconobacter roseus (GATFIELDnd SAND,1988).With this process abou t 1 3 g L-' butyr-ic acid was obtained in high molar yield. Theoxidation of butano l was also studied by DRU-AUX et al. (1997) with Acefobacfer aceti.After abioconversion phase of 60 h with continuousfeeding of the substrate they reached butyricacid concentrations abo ut 39 g L-', with 86%molar yield. They also tested several othe r al-cohols like isoamyl alcohol, 3-methylbutanol,and 2-phenylethanol, but there the molaryields were significantly lower a nd no productconcen trations were presented .

Propionic acid can be obtained by de novosynthesis with Propionibacterium shermanii orPropionibacterium acidipropionici on cheapsubstrates like whey (CARRONDOt al., 1988;CHAMPAGNEt a l., 1989). Propionic acid canalso be produced by oxidation of propanolwith Gluconobacter oxydans (SVITELandSTURDIK, 995). This p rocess gives nearly44 g L-' after 70 h.

The alcohol oxidation with Gluconobacterroseus was also applied by GATFIELDndSANDor the production of several other acids.So after addition of 20 g L-' isobutanol to a30 L fermentor 21 g L-' isobutyric acid wasobtained after 1 5 h.

The addition of 10 g L-' 2-methyl butanolto the Gluconobacter roseus culture resulted in7 g L-l 2-methyl butyric acid after 24 h;10 g L-' isoamyl alcohol were oxidized to11g L-' isovaleric acid.

The alcohols used as substrates for the mi-crobial oxidations are cheaply available fromfuse1 oil residues from ethanol fermen tationsof yeast.

2.4 Fermentation of Other Natural

Aroma Compounds

2.4.1 Production of Flavor-Active

Aldehydes and Alcohols

Aldehydes are another important group offlavor chemicals. Most im portant are after va-nillin, which has been mentioned above, the

so-called "green notes". These a re trans-2-hex-enal, hexanal, and the alcohols cis-3-hexen-l-01, trans-2-hexen-1-01,and hexanol.

These compounds are formed from linolen-ic and linoleic acid through th e action of lipox-ygenase and hydroperoxide lyase. This reac-tion typically occurs in plant tissues such asleaves after dam age (HATANAKA,993).

As early as 1965 COLLINSnd KALNINSfound 2-hexenal together with acetaldehydeand 2-methylbutanal in the culture broth of

the o ak wilt fungus Ceratocystisfagacearum. Acommercial process for the production ofthese green notes does not use microorgan-isms but enzym es. So soybean lipoxygenase is

used to produc e hydroperoxy acids which arethen cleaved by lyases from plant homogen-ate s of guava (MULLERt al., 1993), or the fol-iage of other plants (HOLTZ t al., 1995). Theobtained aldehydes can be reduced by bakers'yeast to th e corresponding alcohols. The hy-droperoxide lyase of banana leaves has been

Tab. 3. Alcohol Conversion toAldehydes by Pichia pastoris (after MURRAYtal., 1988;MURRAY nd DUFF, 990)

Product Concentration Conversion Period

[g L-'I [hl

Acetaldehyde 30.8Propionaldehyde 28Butyraldehyde 14Isobutyraldehyde 10.8Isovaleraldehyde 11.3

Hexanal 13.8

1224

48

12

24



4 References 347

The only process with which higher yields of

vinyl guaiacol have been obtained was donewith Bacillus pu milus . In an aqueous-organictwo-phase system LEE et al. (1998) obtained

9.6 g L - l vinyl guaiacol from ferulic acid with-in 10h.

In an analogous reaction with vanillic acidguaiacol is obtained as product (Fig. 12b).Guaiacol has a powerful, smoke-like, some-what medicinal odor and a warm, medicinaland somewhat sweet taste, but accompaniedby a burning sensation. Guaiacol is used in fla-vor compositions fo r co ffee, vanilla, whisky, to-bacco, and in several fruit and spice complexes(ARCTANDER,969).

2.4.3 Methyl Ketones

Methyl ketones, especially 2-pentanone,2-heptanone, 2-nonanone, and 2-undecanone,are the flavor impact compounds of blueveined cheese, which is classically producedwith the mold Penicillium Toqueforti. Thesetypical flavors are used for flavoring of prod-ucts such as salad dressings, soups, crackers,and cakes.

The methyl ketones are formed due to theaction of several enzymes of the mold duringcheese ripening, like lipase for the release offree fatty acids from milk triglycerides, whichact as precursors for the forma tion of m ethylketones. The C-6, C-8, C-10,and C-12 fatty ac-ids act as substrates for the P-ketoacyl decar-boxylase, which generates the correspondingCn-l methyl ketones. The biosynthesis of themethyl ketones has bee n reviewed by KINSEL-LA and HWANG 1976). Based on this basicknowledge several processes for the biotech-

nological production of methyl ketones havebeen developed (LUKSAS, 973; WINKet al.,1988;LARROCHEt al., 1989).

3 Conclusions

For the prod uction of natural flavor chemi-cals microbial fermentation is a competitivetechnology to obtain chemicals which cannot

be isolated from plant extracts, essential oils,

or oth er sources. It is important to realize thatchemical synthesis, which can produce thesesimple chemicals much mo re cheaply, cannotcom pete du e to the legal status of na tural fla-

vors. The increasing knowledge of the enzy-matic pathways for the biosynthesis and me-tabolism of arom a chem icals gives new oppo r-tunities t o c reate tailor-made genetically engi-neered p roduction strains.This is actually don efor the production of vanillin.

A problem remains in the low demand forextremely active flavor specialities like somepyrazines or sulfur containing chemicals,where the annual consumption is sometimesonly in the area of hundreds of grams or a fewkilograms. This prevents the establishment of

extensive research programs for these chemi-cals.

The formation of flavors by the use of en-zymes has not been mentioned within thischapter, but this is an equally important bio-technological method for the production ofnatural flavors (for reviews, see GATFIELD,1992;BERGER,995; SCHREIER,997).

4 References

AGO, S., KIKUCHI,Y. (1998), European Patent857 789.

ANTRIM,. L., HARRIS, . W. (1977), US Patent4,038,481.

A R ~ A N D E R ,. (1969), Perfume and Flavor Chemi-cals. Montclair, N J Published by the au thor.

BAUER, ., GARBE, ., SURF JURG,. (1990), Com-mon Fragrance and Flavor Materials. 2nd Edn.,Weinheim:VCH.

BERGER,.G. (1995),Aroma Biotechnology.Berlin:Springer-Verlag.

CARDILLO,., UGANTI,., SACERDOTE,., BARBE-N I, M., CAB EL LA,^, SQUARCIA,F.1989), Europe-an Patent 356 291.

CARRONDO,. J.T.,CRE S~ O,. P. S. G.,MOURA,M. J.

(1988), Production of propionic acid using xyloseutilizing Propionibacterium, App l. Biochem. Bio-

technol. 17,295-312.CWAGNE, C . F’., BAILLAGEON-COTE,., Gou-

LET, J. (1989),Whey fermentation by immobilizedcells of Propionibacterium shermanii, J. Appl.Bacterial.66,175-184.

CHEETHAM, P. S. J., MAUME, . A ., DE ROOJI, . F.

(1988),European Patent 0258 993.



348 9 Biotechnological Production of Natural Ar om a Chemicals by Fermentation Processes

CLARK,G. S. (1990), Vanillin, perfum, Flavor 15,45-54.

Code of Federal Regulations (1993), 21 Food andDrugs, Parts 100-169, revised April 1,1993. Wash-ington, DC: National Archives and Records Ad-ministration.

COLLINS, . P., KALNINS, . (1965), Carbonyl com-pounds produced by Ceratocystis fagacearum,A m . J. Bot. 52,751-754.

DEGRASSI, ., DE LAURETO,. P.,BRUSCHI, . V.(19959, Purification and characterization of feru-late and p-coumarate decarboxylase from Bacil-lu s pumilus, Appl. Environ. Microbiol. 61,326-332.

DOLFINI,. E., GLINKA,., BOSCH,A. C. (1990), USPatent 4,927,805.

DRUAUX, ., MANGEOT, ., ENDRIZZI, ., BELIN,J.-M. (1997), Bacterial bioconversion of primaryaliphatic and aromatic alcohols into acids: effectsof molecular structure and physico-chemical con-ditions, . Chem . Tech. Biotechnol. 68,214-218.

DUFF, . J. B., MURRAY, . D. (1988), Comparison offree and immobilizedPichiapastoris cells for con-version of ethanol to acetaldehyde, Biotechnol.Bioeng. 31,790-795.

DUFF,S . J. B., MURRAY, . D. (1989), Oxidation ofbenzyl alcohol by whole cells of Pichia pastorisand by alcohol oxidase in aqueous and non-aque-ous reaction media, Biotechnol. Bioeng. 34, 153-159.

DUFF,S. 3. B., MURRAY,W. D. (1991), US Patent5,010,005.DUFOSSE,L.,FERON,G.,MAUVAIS,G.,BONNARME,P.,

DURAND, ., SPINNLER,.-E. (1998), Productionof y-decalactone and 4-hydroxy-decanoic acid inthe genus Sporidiobolus, J . Ferment. Bioeng. 86,

FALCONNIER,., LAPIERRE,., LESAGE-MEESSEN,.,YONNET,., BRUNERIE,. et al. (1994),Vanillii asa product of ferulic acid biotransforrnation by thewhite-rot fungus Pycnoporus cinnabarinus1-937:Identification of metabolic pathways,J . Biotech-nol. 37,123-132.

FARBOOD, . I., WILLIS,B. J. (1983), WO Patent83/01072.FUGANTI,. ,MAZZOLARI,., ZUCCHI, ., BARBENI,

M., CISERO,M.,VILLA,M. (1995), On the mode ofbiodegradation of cinnamic acids by Colletotri-chum gloeosporoides: a direct acces to 4-vinylphenol and 2-methoxy-4-vinylphen01,iotechnol.Lett. 17,707-710.

GASSON, . J., KITAMURA,., MCLAUCHLAN,. R.,NARBAD,., PARR,A. J. et al. (1998), Metabolismof ferulic acid to vanillin. A bacterial gene of theenoyl-SCoA hydrataselisomerase superfamilyencodes an enzyme for the hydration and cleav-

age of a hydroxy cinnamic acid SCoA thioester,

169-173.

J . Biol. Chem. 273,4163-4170.GATFIELD,. L. (1992), Bioreactors for industrial

production of flavors: use of enzymes, in: Biofor-mation o f Flavors (PATTERSON,. L. S., CHARL-WOOD,.V., MACLEOD, ., WILLIAMS,A.A.,ds.),pp. 171-185. Cambridge: The Royal Society ofChemistry.

GATFIELD,. L., SAND,T. (1988), European Patent289 822.

GATFIELD,.L., GUNTERT,M., SOMMER, ., WERK-HOFF, P. 1993), Some aspects of microbiologicalmanufacture of flavor-active actones with partic-ular reference to y-decalactone, Chem. Mikrobi-01. Technol. Lebensm. 15,165-170.

GOCHO,. , RUMI,K., TSUYOSHI,. (1998), US Patent5,763,233.

GOCHO,., TABOGAMI,., INAGAKI,., KAWABATA,C., KOMAI, . (1999, Biotransformation of oleicacid to optically active y-dodecalactone, Biosci.Biotech. Biochem. 59,1571-1572.

GROSS, .,ASTHER,M., BRUNERIE,. (1991),Europe-an Patent 453 368.

HATANAKA,. (1993), The biogeneration of greenodor by green leaves, Phytochemistry 34 , 1201-1218.

HAUSLER, ., MUNCH, . (1997), Microbial produc-tion of natural flavors,ASM News 63,551-559.

HAUSLER, ., LERCH,K., MUHEIM,A., SILKE,N.

(1997),European Patent 801 133.HOLTZ,R. B., MCCULLOCH, . J., GARGER, . J.

(1995),WO Patent 95123413.HOPP,R., RABENHORST,. (1993), European Patent583 687.

HUANG,.,DOSTAL,., ROSAZZA,. I? N. (1994), Pur-ification and characterization of a ferulic acid de-carboxylase from Pseudomonas fluorescens, J.

Bacteriol. 176,5912-5918.KINSELLA,. E., HWANG, . H. (1976), Enzymes of

Penicillium roqueforti involved in the biosynthe-sis of cheese flavor, Crit. Rev. Food Sci. Nutr. 8,191-228.

KNUTH,M., SAHAI, .(1991),US Patent 5,057,424.KRINGS, ., BERGER, . G. (1998), Biotechnological

production of flavors and fragrances, Appl. Mi-crobiol. Biotechnol. 49,1-8.KUMIN, B., MUNCH.,T. (1997), European Patent

LABUDA,. M., GOERS,. K., KEON,K. A. (1992), USPatent 5,128,253.

LABUDA,.M., GOERS,. K., KEON,K. A. (1993), Mi-crobial bioconversion process for the productionof vanillin, in: Progress in Flavor Precursor Stud-ies, Proc. Int. Conf. Wurzburg, Germany, Septem-ber 30-October 2, 1992 (SCHREIER,., WINTER-

Allured Publishing.

795-607.

HALTER, P.,Eds.), pp. 477482. Car01 Stream, IL

LARROCHE, .,ARPAH,M., GROS,.-B. (1989), Meth-



4 References 349

fragrances, in: Frontiers of Flavors, Proc. 5th Int.Flavor Conf., Porto Karras, Chalkidiki, Greece,1-3 July, 1987 (CHA RA LAM BOU S,., Ed.), pp.1-18. Am sterdam : Elsevier Science Publisher.

MURRAY,. D., DUFF,S.J. B., LANTHIER,. H. (1989),US Patent 4,871,669.

NARBAD,., GAS SON , . J. (1998), Me tabolism of fe-rulic acid via vanillin using a novel CoA-depen-den t pathway in a newly isolated strain of Pseud-omonas jluorescens, Microbiology 144, 1397-1405.

NARBAD, ., RHODES, . J. C., GASSO N, . J., WAL-TON, N. J. (1997),WOPatent 97/35999.

NAZARETH,. , MAVINKURVE,. (1986), Degradationof ferulic acid via 4-vinyl guaiacol by Fusariumsolani (Mart.) Sacc., Can. J . Microbiol. 32, 494-497.

NICAUD,. M., BELIN,. M., PAGOT, ., END RIZ ZI, .(1996),French Patent 2734-843.

NOZAKI,M., WASHIZU,., SUZUK I, ., K ANISAWA,.(1995), Microbial oxidation of alcoho ls by Candi-da boidinii, in: Bioflavor 95 (ETIEVANT,P.,SCHREIER,?,E ds ), pp. 255-260. Paris: INR A.

OKUI,S., UCHIYAM A,., M IZUG AK I, . (1963), Me-tabolism of hy droxy fatty acids. 11. Intermediatesof the oxidative breakdown of ricinoleic acid bygenus Candida, J . Biochem. 54,536-540.

OTOK,G. (1985), Degradation of ferulic acid by Es-cherichia coli, J . Ferment. Technol.63,501-506.

OVERHAGE,., KR ESS E,A . ., PRIEFERT,H., SOMMER,H. , KRAM MER,. et al. (1999), Molecular charac-terization of the genespcaG and pcaH, encodingprotocatechuate 3,4-dioxygenase, which are e s-sential for vanillin catabolism in Pseudomonas sp.strain HR 199, Appl. Environ. Microbiol. 65,

PAGOT, ., ENDRIZZI,., NICAUD,. M., BEL IN,. M .(1997), Utilization of an auxo trophic strain of theyeast Yarrowia Iipolytica to improve gamma-de-calactone production yields, Lett. Ap pl. Microbi-01.25,113-116.

PRIEFERT, ., RABENHORST,., STEINBUCHEL,.(1997), Molecular characterization of genes ofPseudomonas

sp. Strain H P 199 involved in bio-conversion of vanillin to pro tocatechu ate, J . Bac-teriol. 179,2595-2607.

RABENHORST,. (1996), Produ ction of methox yphe -nol-type natural aroma chemicals by biotransfor-mation of eugenol with a new Pseudomonas sp.,Ap pl. Microbiol. Biotechnol. 46,47&474.

RABENHORST,., HOPP,R. (1991), European Patent405 197.

RABENHORST,., HOPP,R. (1997), German Patent19532 317.

RANADIVE,. S. (1994),Vanilla- ultivation, curing,chemistry, technology and comm ercial products,

Dev. Food Sci. 34,517-577.

951-960.

yl-ketone production by Ca-alginateEudragitRL entrapped spores of Penicillium roqueforti,Enzy me M icrob. Technol.11,106-112.

LEE, .-Y.,VOLM, . G., ROSAZZ A,. P. N. (1998), De -carboxylation of ferulic acid to 4-vinyl guaiacolby Bacillus pumilus in aqueous-organic solventtwo-phase system, Enz yme Microb. Technol. 23,

LESAGE-MEESSEN,., DELAITRE,M., HAO N,M., As-THER, M. (1996a), WO Patent 96/08576.

LESAGE-MEESSEN,., DELAITRE,M., HAON ,M.,THI-BAULT, J. F., CECCALDI,. C. et al. (1996b), A two-step bioconversion process for vanillin produc-tion from ferulic acid combiningAspergillus nigerand Pycnoporus cinnabarinus, J . Biotechnol. 50,

LESAGE-MEESSEN,.,DELAITRE,M., HAO N,M.,THI-BAULT,. F., CECCALDI,. c.et a]. (1997),An at -tempt to channel the transformation of vanillicacid into vanillin by controlling methoxy hydro-quinone formation in Pycnoporus cinnabarinuswith cellobiose, Appl. Microbiol. Biotechnol. 47,393-397.

LI, K., FROST,. W. (1998), Synthesis of van illin fromglucose,J.Am . Chem. SOC.120,10545-10546.

LUKSAS,. J. (1973), US Patent 3,720,520.MALARCZYK,., KOCHM ANSKA-RDEST,., APALO VIC,

R., LEONOWICZ,. (1990), Production of vinylguaiacol, isoeugenol and other aromas from lig-nin-related materials by Nocardia, Prog. Biotech-nol. 6,403-409.

MARKUS,. H., PETERS,A. L. J., Roos, R. (1993),European Patent 0542 348.

MICA RD, ., R ENA RD , ., THIBAULT,. F., LESAGE-MEESSEN,., HAON,M. et al. (1995), Est-il pos-sible d’obtenir de la vanilline naturelle i partir depulpes d e betterave?, Colloq. -Inst. Natl. Rech. Ag -ron. 71,149-153.

MOLINARI,.,VILLA,R., MANZONI, ., ARAG OZZI-NI , F. (1995),Aldehyde production by alcohol ox-idation with Gluconobacter oxydans, Appl. Mi-crobiol. Biotechnol. 43,989-994.

MOLINARI,. ,ARAGOZZINI,., CABRAL ,. M. S., PR A-ZERES,

D.M.

F. (1997), Con tinuous p roduc tionof

isovaleraldehyde through extractive bioconver-sion in a hollow-fiber mem brane b ioreactor, En-zym e Microb. Technol.20,604-611.

MULLER, ., GAUTIER, ., DEAN,C., K UH N, .-C.(1993), WO Patent 93/24644.

MURRAY,. D.,DUFF,S.J.B. (1990),Bio-ox idation ofaliphatic and aromatic h igh-molecular weight al-cohols by Pichia pastoris alcohol oxidase, Appl .Microbiol. Biotechnol. 33,202-205.

MURRAY,. D., DUFF, S. J. B., LANTH IER,?H.,ARM-STRO NG , . W., W ELSH,D. W., WILLIAMS, . E .(1988), Development of biotechnological pro-cesses for the production of natural flavors and

261-266.

107-1 13.



350 9 Biotechnological Production ofNatural Ar om a Chemicals by Fermentation Processes

ROSAZZA,. P. N., HUANG, ., DOSTAL,., ROSEAU,

B. (1995), Review: Biocatalytic transformation of

ferulic acid: an abundant natural product, J . Ind.Microbiol. 15 ,457471 .

SAHAI, . 1994), Plant tissue culture, in: BioprocessProduction of Flavor, Fragrance and Colo r Ingre-dients (GABELMAN,., Ed.), pp. 239-275. New

York: John Willey& Sons.

SCHARPF,R.,L. G., SEITZ, . W., MORRIS,. A., FAR-

BOOD, M. I. (1986),Generation of flavor and odor

compounds through fermentation, in: Biogenera-tion of Aromas, ACS Symp. Ser.317 (PARLAMENT,

T. H., CROTEAU,., Eds), pp. 323-346. Washing-

ton, D C American Chemical Society.

SCHINDLER,., SCHMID, . D. (1982), Fragrance or

aroma chemicals-microbial synthesis and enzy-

matic transformation,Proc. Biochem. 5,243.SCHREIER, (1997), Enzymes and flavor biotechnol-

ogy, in: Adva nces in Biochemical EngineeringlBi-otechnology Vol. 55 (SCHEPER,., Ed.), pp. 51-72.Berlin: Springer-Verlag.

STEINBUCHEL,., PRIEFERT, H., RABENHORST,.

(1998), European Patent 0 845 532.STRACK,. (1990), Metabolism of hydroxycinnamic

acid conjugates, Bull. Liaison - Groupe Poly-

phenols 1 5 ,5 5 4 4 .SUTHERLm, J. B., CRAWFORD, . L.,POEm O 111,

A. L. (1983),Metabolism of cinnamic,p-coumar-

ic, and ferulic acids by Streptomyces setonii, Can.J. Microbiol. 29,1253-1257.

SUTHERLAND,. B., TANNER, . A., MOORE, . D.,FREEMAN,. P., DECK,.,WILLIAMS, . J. (1995),Conversion of ferulic acid to 4-vinyl guaiacol by

yeasts isolated from frozen concentrated orange

juice,J. Food Prot. 58,1260-1262.SVITEL,., STURDIK,. (1995), n-Propanol conver-

sion to propionic acid by Gluconobacter oxydans,Enz yme Microb. Technol.17,549-550.

TADASA, . (1977), Degradation of eugenol by a

microorganism,Agric. Biol. Chem. 41,925-929.TADASA, ., KAYAHARA,. (1983), Initial steps of

eugenol degradation pathway of a microorgan-

ism,Agric. Biol. Chem. 47,2639-2640.TOMS, ., WOOD, J. M. (1970), The degradation of

trans-ferulic acid by Pseudomonas acidovorans,Biochemistry 9,337-343.

VANDAK,., TELGARSKY,., STURDIK,. (1996),In-

fluence of growth factor supplementation on bu-

tyric acid production from sucrose by Clostridiumbutyricum , Folia Microbiol. 40,669-672.

European Patent 425 001.

BOSCH, ., COHEN.,A. M. (1992), Microbial pro-

duction of natural Sdecalactone and Gdode-

calactone from the corresponding cY,/3-unsaturat-

ed lactones in Massoi bark oil, Appl. Microbiol.Biotechnol. 36,712-716.

WINK, ., FRICKE,U., DEGER,H.-M., MIXICH, .,

BAUER, ., JUSTINSKI,. (1988), German Patent3701 836.

WINK,.,VOELSKOW,., GRABLEY,., DEGER,H. M.(1988),European Patent 286 950.

YOSHIMOTO,T.,AMEJIMA,., HANYU,., KOMAI,.

(1990a),Japanese Patent 02195871.YOSHIMOTO,., SAMEJIMA, ., HANYU, ., KOMAI,T.

(1990b ),Japanes e Patent 02200192.ZAGO,A.,DEGRASSI,., BRUSCHI,.V. (1995),Clon-

ing, sequenzing, and expression in Escherichiacoli of the Bacillus pumilus gene for ferulic acid

decarboxylase, Appl. Environ. Microbiol. 61,4484-4486.

VAN DER SCHAFT,F!H., DE LA AT,^. T. A. M. (1991),

VAN DE R SCHAFT, P.H., TE R BURG,N., VAN DEN



10 Synthetic Applications of

Enzyme-Catalyzed Reactions

JANE MCGREGOR-JONESTempe, AZ, USA

1 Introduction 353

2 Aliphatic Biotransformation Products 353

2.1 Hydrolysis/Condensation Reactions 353

2.1.1 Pig Liver Esterase 353

2.1.2 a-Chymotrypsin 354

2.1.3 Thermolysin Peptide Synthesis 355

2.1.4 Hog Kidney Acylase 356

2.1.5 Microbial Esterase 357

2.1.6 Porcine Pancreatic Lipase 358

2.2 Enzyme Catalyzed Reduction Reactions 3592.2.1 Yeast Alcohol Dehydrogenase 359

2.2.1.1 Reduction of Ketones and Aldehydes 359

2.2.1.2 Reduction of P-Keto Esters 359

2.3 Oxidation Reactions 362

2.3.1 Horse Liver Alcohol Dehydrogenase 362

2.3.2 Pseudomonas putida 363

2.3.3 Lipoxygenases 363

2.4.1 Farnesyl Pyrophosphate Synthetase 365

2.4.2 Aldolases 365

2.4.3 Acyloin Condensation 366

2.4.4 Cyanohydrin Formation 366

2.5 Multiple Enzyme Reactions 366

3.1 Hydrolysis/Condensation Reactions 367

2.4 Carbon-Carbon Bond Formation Reactions 365

3 Alicyclic Biotransformation Products 367

3.1.1 Pig Liver Esterase 367

3.1.2 Lipases 368

3.1.2.1 Arthrobacter Lipase 368

3.1.2.2 Enzymic Ester Resolutions 3683.1.2.3 Porcine Pancreatic Lipase 368

3.1.3 Yeast 370

Biotechnology Second, Com pletely Revised EditionH.-J. Rehm and G.Reed

copyright OWILEY-VCH Verlag Gm bH,2000



352 10 Synthetic Applications ofEnzyme-Catalyzed Reactions

3.2 Enzym e Catalyzed Redu ction Reac tions 3703.2.1 Yeast 3703.2.2 Rhizop us arrhizus 371

3.3.1 Horse Liver Alcohol Deh ydrogena se 3723.3.2 Micro bial Enzym es 373

3.3 Oxidation Reac tions 372

3.3.2.1 Pseudornonas putid a 3744 Polycyclic Biotransform ation Produc ts 376

4.1 Hydrolysis/Condensation Reac tions 3764.1.1 Pig Liver Este rase 3764.1.2 a-Ch ymo trypsin 376

4.2.1 Yeast 3774.2.2 Microorganism Redu ctions 3784.2.3 Ho rse Liver Alcoh ol Dehyd roge nase 379

4.3.1 Microorgan ism Ox idation 3804.3.2 Hydroxylation of Steroids 381

4.4.1 Yeast Cyclases 3824.4.2 R eactions Involving Acetyl Co A 382

4.5.1 Com binations of Dehy drogenases 382

4.2 Enzym e Catalyzed Redu ction Reac tions 377

4.3 Oxidation Reac tions 380

4.4 Carbon-Carbon B ond Formation Reac tions 382

4.5 Exploiting Com binations of Enzym e Specificity 382

4.6 Multiple Enzyme Reac tions 3825 Heterocyclic Biotransformation Produc ts 384

5.1 Hydrolysis/Condensation Reac tions 3845.1.1 Pig Liver Ester ase 384

5.1.2 8-(L- -Aminoadipyl) -L-cysteinyl-D-valine (ACV) Synthetase 3855.1.3 Lipases 3855.1.3.1 Pseudomonas Lipases 3855.1.3.2 Mucor Species Lipases 385

5.1.4.1 Transacylations Using Penicillin Acy lase 3865.1.4 Penicillin Acylase 386

5.2 Oxidation Reac tions 3865.2.1 Horse Liver Alcohol Dehy drogena se 3865.2.2 Dihydro folate Redu ctase 3875.2.3 Yeast 387

5.3 Carbon-Carbon B ond Formation Rea ctions 3885.3.1 Rabbit Muscle Ald olase 388

5.4 Nucleo tide Chem istry 3895.4.1 Phosp horylation 3895.4.2 G en e Synthesis 3895.4.3 Base Exchange 389

6 Aroma tic Biotransform ation Products 3906.1 Hydrolysis/Condensation Rea ctions 390

6.1.1 a-Chy mo trypsin 3906.2 Oxidation Reac tions 390

6.2.1 Hyd roxylation 3906.2.2 Oxidative Deg radation 391

6.3.1 Acyloin Con densation 3936.3 Carbon-Carbon B ond Formation Reac tions 393

7 Referen ces 393



2 Aliphatic Biotransformation Products 35 3

1 Introduction

The ability of microorganisms and enzym es

to act as selective and chiral catalysts has bee nacknowledged for many years (see Ch apter 1,this volume), but it is only in m ore rece nt timesthat biotransformations have become accept-ed as routine pro cedures in organic synthesis.Synthetic chemists have been hesitant t o learnthe unfamiliar techniques required to use mi-croorganisms and enzymes, especially sincenew synthetic reagents an d catalysts now easethe synthesis of complex molecules that onceseemed unthinkable. W hile nonbiological cata-lysts will continue to be developed and im-proved, the en ormo us range of biological cata-lysts provides a com pleme ntary set of rea gentswhich have unique and competitive advan-tages. The relative merits of the chiral pool,asymm etric synthesis, and biotransformationsas means for preparing chiral intermediatescan be judged from an extensive review of thesynthesis of chiral pheromones (MORI,1989).

Enzyme catalyzed biotransformations suchas the fermentation of sugar to ethanol byyeast featured in some of the earliest scrip-

tures. Modification and optimization continuestoday. Countless examples of biotransforma-tions can be foun d in the w orld around us: cit-ric acid, obtained through fermentation, isused in the food and drink industry as a fla-voring agent; the penicillins and cephalospo-rins are both secondary metabolites producedby fungi and are used clinically as antibacterialagents; anti-inflammatory steroids are pre-pared from more common naturally occurringsteroids and bacterial proteases a re used as ad-ditives in biolog ical wash ing powders.

The aim of this chapter is to demonstratethe wide utility and applicability of biotrans-formation products in synthesis today, withparticular reference to those that are flexiblesynthons and which can be used to reach com -mon targets of general interest. Such examplesof utility abound in the pharmaceutical indus-try where enantiomerically pure interm ediatesare necessary, and are frequently easily ob-tained through biotransformations.

The chap ter has been subdivided accordingto the structure of the biop rodu ct, i.e., aliphat-ics, alicyclics, polycyclics, heterocy cles, and aro -

matics. Each division is further divided ac-cording to the type of biochemical transform a-tion used to produce the bioproduct. Only afew examples in each category are given -this

chapter serves only as an overview of theusefulness of biotransformations. More ex-haustive texts should be consulted for furth erinformation (DAV IES t al., 1989; DR AU Z ndWALDMANN,995; FABER,1995; WONG andWHITESIDES,994; JONES , 986).

2 Aliphatic Biotransforma-

tion Products

2.1 Hydrolysis/Condensation

Reactions

2.1.1 Pig Liver Esterase

Hydrolytic enzymes require no cofactors,but p H control is frequently essential. Pig Liv-er Esterase (PLE) (GREENZAIDnd JENCKS,1971) catalyzes the stereoselective hydrolysis

of a wide varie ty of esters.The enantioselectivehydrolysis of substituted glutarate diesters (1)to half esters (2) (Fig. 1)can be rationalized byJONES ’ odel (JON ES, 993; see C hapter 4, hisvolume).

If the undesired ester is formed the twoenantiom eric series may be intercon verted byselective manipulation of the ester and acidgroups (e.g., ( S ) - and (R)-(3b)) or by selectivehydrolysis of a methyl ester (2c) n the pres-ence of a t-butyl ester. The la tter me thod wasexploited in a synthesis (WAN G t al., 1982) of

the antibiotic negamycin (6) (Fig.2) (HAMADAet al., 1970; KON DO t al., 1971).

The half ester (2c)was also clev erly exploit-ed in the synthesis of trans-carbapenems (KA-RADY et al., 1981; SALT ZM AN Nt al., 1980).PL Ehydrolysis of the diester (lc) gave half ester(2c)with high enantiom eric purity w hich wasfurther enhance d by crystallization. Cleavageof the carbobenzyloxy (Z) group by hydrogen-ation and cyclization using Mukaiyama’s re-agent gave the p-lactam (9) in exce llent yield(Fig. 3) (KOBAYA SHIt al., 1981). Diastereo se-

lective alkylation adjacent to the chiral center



354 10 Synthetic Applications of Enzym e-Catalyzed Reactions

--

PIE iLiBH4 0 0Z (S)-3b

62-93% yie ldM e 0 2 C C OzM e M a 2 C C 02H i B H ~

1 2 ii H+

a R = OH, R' = Meb R = H, R' = Me

c R = NHZ, R' = H

a 99% ee

b 9 0 % ee

c 93% ee

Fig. 1. Pig liver esterase hydrolysis of glutarate diesters.

(R)-3b

NHZ

HOzC&C02'Bu1 sobutylene, H2S04

HZ

M e o z c ~ c o 2 H 2 NaOH, H@ R2 c

~~

S t e p s

QH NH2Steps

H NH2 0 Me

H2N&cT!JVC02H - NCH

I

6 N e g a m y c i n fI 5

Fig. 2. The use of pig liver esterase in negam ycin synthesis.

C02M e+ H R ~

===3

1 H2, Pd-C2 Ph3P, F'ySSF'y

8O2H

7

a R = Et, R1= Ac; (+)-PS-5

c R = (R) -MeCH(0H)- ,R1= H ; (+) -Thienamycin COzMe 2c

b R = 'Pr, R1 = Ac; (+)-PS-6

NH ZFig. 3. Carbapenern antibiotic synthesis.

(8) and elaboration of the pyrroline ring com-pleted the synthesis (OKANOt al., 1983).

matic or hydrophobic groups, however it alsohas strong esterase activity. It can be used inplace of PLE in the hydrolyses of C-3 substi-tuted glutarate diesters and generally exhibitsopposite enantiospecificities. The stereoselec-tivity is easily rationalized by overlaying thestructure of the ester with that of the natural L-

amino acid substrates.

2.1.2 a-Chymotrypsin

a-Chymotrypsin is a protease with a prefer-ence for L-amino acid residues that bear aro-




2.1.3 Thermolysin PeptideSynthesis

Despite the e me rgence of several impo rtantmethods and coupling reagents for peptidebond formation over the last 50 years, almostall involve the possibility of side reactions a ndracemization. Conditions in chemical synthe-ses must be carefully controlled to minimize oreliminate these reactions. Condensation ofamino acids using proteolytic enzymes (the re -verse of the usual reaction) takes place effi-ciently under mild conditions, opening up auseful synthesis of peptides (ISOW A t al.,1977b). It is the thermodynamic products thateventually accumulate in enzyme catalyzedprocesses, as with all reactions. Unfortun atelythe equilibrium constant for peptide forma-tion usually lies heavily on th e side of the co n-stituent am ino acids, but it can be displaced bythe fo rmation of insoluble peptides, by contin-uous product removal (e.g., two-phase reac-tions), the use of low water systems, or reac-tions under kinetic control.

Th e exploitation of the differen t specificitiesof various proteolytic enzymes provides con-

vergent, multi-peptide bond syntheses in goodyields from the component amino acids(GLAS S, 981). In some cases these reactionshave been run o n a multi-ton scale, for exam-ple in the formation of the aspartam e precur-sor (12) Fig.4).Aspartam e is a dipeptide syn-thetic sweetener, more commonly knownunder the brand name NutrasweetB.

The co upling of L -phenylalanine methy l es-te r (11)with N-carbobenzyloxy-L-asparticcid( lob) to give the dipeptide (12) Fig.4) s cata-

lyzed by immo bilized thermolysin (OYA MA tal., 1981), with a yield of 82%. The use of animmobilized enzyme (e.g., on a solid supportof silica or porous ceramics) helps to drive the

reaction in the required direction with highyields, as do es the insolubility of the dipep tide(12). hermolysin is highly specific in its cata-lytic activity, so that protection of the sidechain carboxyl group of the aspartic acid de-rivative is not necessary. A commercial pro-duction route of the synthesis of aspartame(13) involves the u se of racemic phenylalaninemethyl ester ruc-(ll), in which the unreactiveD-isomer is recovered an d recycled.

Thermolysin and other proteolytic enzymessuch as papain, pepsin, and nargase are com-monly used in the formation of small peptidessuch as di-, tri-, an d tetrapeptides. Th e synthe -sis of peptides, even by conv ergent techn iquesrequires m any steps, consequently the yields ofindividual steps need to be extremely high ifthe route is to be viable. A n outstanding exam-ple is provided by the octapeptide angiotensinI1 (Fig. 5) (ISOWA t al., 1977a) which was pre-pared in an overall yield of 85%.The generalprinciples of peptide coupling by enzym es areillustrated by synthese of leucinyl- and m ethio-

nyl-enkephalins (15) and (16) (KULLMAN,1981) and dynorphin (17) (KULLMAN,982)(Fig. 6). In these reactions selectivity is pre-dominantly determined by th e residues at thecarboxylic acid terminus of the peptide chain.Thus chym otrypsin preferentially couples thecarboxylic group of terminal a roma tic aminoacids with m ost othe r amino acids. In contrast,for papain catalyzed couplings the pe nultimateam ino acid at th e carboxylic terminus is pref-erably phenylalanine, but other hydrophobic

*

1 2 R = Z H2

T h e r m o l y s i n

. OH 82%yield

1 3 R = H J F ’ d / C

-

H2N C02Me

H A s p a r t a m e

H02C’HN?J- PL-10 L-11

a R = H

b R = Z

Fig. 4. Enzymatic aspartame synthesis.



356 10 Synthetic Applica tions of Enzyme-Catalyzed Reactions

H-Asn-Arg-Val-Tyr-Val-His-Pro-Phe-OH

14 Angiotensin I1

Fig.5. Angiotensin 11.

Papain80% 80%

t t

t tTyrGly-Gly-Phe-R

70 % 7 0 %

Chymouypsin

15 R =

Leu52 %

yield16 R = Met 52% yieldEnkephalins

Papain Trypsin71 % 80% 65% 64%

I t 1 .Tyr-Gly-Gly-Phe-Leu-Arg- Arg-Ile

t t t72 % 52% 70%

Chymonypsin

17 Dynorphin 50% yield

Fig.6. Synthesis of leucyl and methionyl enkephal-ins.

amino acids such as leucine and valine a re alsoaccepted. The na ture of the terminal am ino ac-id plays only a minor role. In the examplesshown (Fig. 6), the influence of the aromaticam ino acid on papain catalyzed couplings en-

ables the coupling of glycine to Tyr-Gly, andextends a further residue in the coupling of

pheny lalanine to Tyr-Gly-Gly. Finally, trypsincatalyzed couplings require arginine or lysineto b e present at the carboxylic terminus, but d onot discriminate between amino acid coupling

partners. Carboxypeptidases have comple-mentary selectivity, i.e., they discriminate t heamino terminus of the peptide chain.

2.1.4 Hog Kidney Acylase

Th e first ma jor application of the resolutionof enantiome rs by enzymes was the use of hogkidney acylase (HKA) to hydrolyze N-acylamino acids (GREENSTEIN,954). HKA ac-com mod ates a broad ran ge of structural typesand is highly stereoselective. A s with virtuallyall mam malian enzym es that act on amino ac-ids it is selective for the L-enan tiomer (Fig. 7).The remaining N-acyl D-enantiomer ~ - ( 2 1 )srecycled via chemically induced racemization.This reaction is still of industrial importancetoday (CHIBATAt al., 1976).

Ho g acylase I was used by BALD WINBALD-WIN et al., 1976) in the first stereocontrolledsynthesis of a penicillin from a peptide. Incu-

bation of the chloroacetyl amide (23) withHog kidney acylase I and separation of the un-reacted amide, followed by acid hydrolysisgave the D-isodehydrovaline (24) in 60% yield(Fig. @ .T he material ob tained was identical toa sample obtained by degradation of penicil-lin.

In another example, MORI and IWASAWA(1980) prepared both enantiomers of thre0-2-amino-3-methylhexanoic acid (27) by enzy-matic resolution. These are precursors torhreo-4-methylheptan-3-01(28), a pheromone

component of the smaller european elm barkbeetle (Fig.9).

Racernizerac-21 D- 2 1 L-22

Fig.7. L-Stereospecificityof acylases.



2 Aliphatic Biotransforrnation Products 357

2.1.5 Microbial Esterase

HK.4 60%yield1tJ

BzHN

0)Xx25

C 0 2 M e%

C 0 2 M e

Fig. 8. Stereocontrolled synthesis of a penicillinsystem.

"Pr

NHAc

rac-261HKA

(3S,4S)-threo28

Fig. 9. Enzymatic resolution to pheromone pre-cursors.

Methyl trans-(2R,4R)-2,4-dimethylglutarate(30) is a useful chiral synthon for macrolide

and polyether antibiotics (Fig. 10).O ne of th emost convenient routes t o this intermediate isby microbial esterase catalyzed hydrolysis ofthe racemic diester (29). The unreactive trans-diester (2S,4S)-(29) cannot be easily recycledto give further amounts of (2R,4R)-(30) be-cause its base mediated epimerization alsoproduces the unwanted meso-cis-diester(2S,4R)-(29) (CHEN t al., 1981). The bifunc-tional chiral synthon (2R,4R)-(30) is a poten talprecursor to the polyether antibiotic calcimy-

cin (31) a metab olite of Streptomyces microor-ganisms (Fig. 11) (EVANSt al., 1979). Similar-ly, trea tm en t of dimethy l cis-2,4-dimethyl glu-tarate (29) (Fig. 12)with microbial esterase re-sults in stereospecific hydrolysis of the pro-Rester group in high yield (75%, 98%ee). Pigliver esterase preferentially cleaves the p r o 4ester grouping of the meso-diester (29), butwith a poor ee of 64% .

The bifunctional chiral synthon (2R,4S)-(30) represents a partial structural unit com-monly encountered in macrolide antibiotics,

M a 2 C 0 2 M e

r a p 29

Gliodadium

roseumI(2 S,4S) 29

( 2 R , 4 R ) - 30 : 50% yield

up to 98%ee

Fig. 10. A useful chiral synthon for antibiotics



358 I0 Synthetic App lications ofEnzyme-Catalyzed Reactions

Me02 C02H

31 Calcimycin

Fig. 11. Synthesis of calcimycin.

MeOzCxx 0 2 M e

Gliodadium

roseurnJMeOz

(2S,4R)-30

Fig.12. Stereospecific microbial esterase hydroly-sis.

and was used in COREY’Sotal synthesis oferythronolide B (32) (Fig. 13) (COREY t al.,1978a, b). The erythromycins, produced by thefungus Streptomyces erythreus, constitute oneof the most important, widespread and effec-tive of all known families of antibiotics com-

monly used against penicillin resistant Stuphy-

32 Erythronol ide B

Fig.13. ErythronolideB.

lococcus strains, and are the drugs of choice totreat Legionnaires’ disease. Another macro-lide antibiotic obtainable through the chiralsynthon (2R,4R)-(30) is methmycin (HIRAMAet al., 1979).

Numerous polyether antibiotics are also ac-cessible, including monensin (35) (Fig. 14)(COLLUM t al., 1980a), which has a stereo-chemically complex array of 17 asymmetric

centers. The approach to monensin is based onthe synthesis and coupling of three advancedoptically active fragments, within one of which,the synthon (2S,4R)-(30) is a part.

2.1.6 Porcine Pancreatic Lipase

The Sharpless enantioselective epoxidationof allylic alcohols is one of the most reliableabiotic asymmetric reactions (KATSUKI ndM A R T I N , 1996;KATSUKI nd SHARPLESS, 1980;

BEHRENSnd SHARPLESS, 1985). Porcine pan-creatic lipase catalyzed hydrolysis of racemicepoxy esters ruc-(36) provides a synthesis ofboth enantiomers and is able to accept a largevariety of structures (Fig. 15) (MARPLESndROGER-EVANS, 1989; LADNERand WHITE-SIDES, 1984). These hydrolyses proceed withhigh enantiospecificity with long alkyl estergroups R , and are simpler to perform thanSharpless epoxidation. Porcine pancreatic lip-ase is active at watedorganic solvent interfac-es,so the solubility of the substrate in water is

not essential.




( 2 S , 4 R ) - 3 0 33

+34

35 Monensin '

Fig. 14. Synthesis of monensin.

2.2 Enzyme Catalyzed Reduction

Reactions

2.2.1 Yeast Alcohol

Dehydrogenase

2.2.1.1 Reduction of Ketones

and AldehydesOH

An investigation of yeast reductions of sim-ple ketones in the 1960's by MOSHER MACLEOD t al., 1964) showed that in most casesthe (S)-configuration alcohols were obtainedby hydride addition to the R e-face of the car-bony1 group. The stereoselectivity can b e ratio-nalized by Prelog's model which is based on

the differences in size of the groups, howeverthese ar e not always congrue nt with Cahn-In-gold-Prelog priorities (see C hap ter 9, this vo-lume; SERVI,990).

A,.\\\R'

37

Buffer"'5OzCR

'ht**fsK-36 O2CR

R' R ee %

2.2.1.2 Reduction of P-Keto EstersEt 98

H "C3H7 9 2

H nC4H9 96 The enantioselectivity of the reduction ofP-keto acids by yeast can be modulated bychanging the aikoxyl moiety of the ester. For

C3H7

example, comparison Of reduction Of the PO-tassium salt (%a) with the methyl (38b) and

''C3H7 >90

Fig. 15. Porcine pancreatic lipase catalyzed hydro-lysis.



360 10 Synthetic Application s of Enzym e-Catalyzed Reactions

butyl (3&) esters shows a decline in enan tio-meric excess as the size of the alkoxyl moietyincreases (HIRAMA t al., 1983). Moreoverconversion to the carboxylic acid salt increases

the aqueous solubility of the substrate andproduct relative to the corresponding estersand the work-up is simplified. Filtration of th ereaction and extraction with organic solvents,removes the biomass and nonpolar com-pounds, respectively.Treatment with diazome-thane and a second organic extraction yieldsthe carboxylic acid fraction as the correspond-ing methyl esters. The versatile synthon (39b)was used in the total synthesis of compactin(Ma) which was first isolated in 1976 from themolds Penicillium citrinum and l? brevi com-pactum and show ed marked inhibitory activityagainst HMG-CoA reductase, an enzyme in-volved in th e biosynthetic pathway to choles-terol. Mevinolin (4Ob) is even more potent andis used in treatment of patients with danger-ously high cho lestero l levels (Fig. 16) (HIRAMAan d UEI,1982).

The reduction of ethyl acetoacetate (41)(Fig. 17) with bakers’ yeast to ethyl (S)-3-hy-droxybutanoate (42) has bee n extensively in-vestigated an d reported in an Organic Synthe-

ses preparation (SEEBACHt al., 1985). It wasused in an extraordinary synthesis of the maincomponent of the pheromone of the solitarybee, Andrena wikella (44). n a virtuoso dem-onstration of synthetic mastery, methyl ace-toacetate was alkylated sequentially with twoequivalents of the iodide (43). Thus the 11car-bon atoms of the pheromone (44) nd a m ole-

LCoZEt1

Yeast1StepsJ

Steps

J

Fig. 17. Synthesis of a pAndrena wilkella.

43

eromone component of

cule of carbon dioxide were derived solelyfrom two m olecules of ethyl and on e of methylacetoacetate (MORIand TANIDA,981;TENGOe t al., 1990).

The tri-isopropylsilyl derivative (47) wasused in a synthesis of the important carbape-

‘OzR 38

a R = K

bR=Me

c R = ”Bu40

a R = H ; Compactin

b R = Me; Mevinolin

3 9 , % e e

a 99

b 92

c 8 1

Fig. 16. Chiral total syn thesis of compactin.




nem intermediate (49).The chiral center wasused to direct the stereochem istry of a ketene-imine cycloaddition to gene rate the useful C-4keto-substituted azetidinone (48) (Fig. 18)

(TSCHAEWt al., 1988) in 90% overall yield.Ethyl (S)-3-hydroxybutanoate (42) was alsoused as a source of the rem ote stereo center ingriseoviridin (50), a mem ber of the streptogra-min family of antibiotics (Fig. 19) (MEYERSan d AMOS,980).

Many microorganisms such as Alcaligeneseutrophus and Zoogloea ramigera I-16-M uti-lize (R)-poly-3-hydroxybutanoate s an ener-gy reserve, which is stored as granules withinthe cells. The abu ndan ce of this materia l with-in th e cells is extraordinary. For exam ple, etha-nolysis of 50 g of Z. ramigera gave 33 g of th emonomer (R)-(42) MORI,1989).

Ethyl (R)-3-hydroxybutyrate (R)-(42) hasbeen utilized in the synthesis of stegobinone(54) - the pheromone of the drugstore beetle(Fig. 20) (M OR Iand EBATA, 986a).Diastereo-selective methylation of the dianion of (R)-(42) anti to th e alkoxide furnished the m ethylgroup destined to be at C-3 of the pheromone(54) (FRATER, 979a, b). The stereochemistryof putative C -2 was achieved by M itsunobu in-

version with 2,4-dinitrobenzoic acid to give acrystalline product. Methyl (R)-( )-P-Hy-droxyisobutyric acid (52) is manufactured byP-hydroxylation of iso-butyric acid with Can-dida sp., followed by esterification.

All four possible stereoisomers of 5-hy-droxy-4-methyl-3-heptanone56), the phero-mone of the weevil of the genus Sitophilus,were prepared from methyl (R)-3-hydroxy-pentanoate (55) which was manufactured byP-hydroxylation of m ethyl penta noa te or pen-

C 0 2 E t (9 -42

5 0 Griseoviridin

Fig. 19. Antibiotic synthes is using yeast.

tanoic acid with Candida rugosa (Fig. 21) (M o-RI and EBA TA, 986b). The principal reactionswere diastereoselective methylation and Mit-sunobu inversion as used in the synthesis of

stegobinone (54). Similarly diastereoselectiveallylation of methyl (3R)-3-hydroxybutanoate(I?)-(&) and ethyl (3S)-3-hydroxybutanoate(S)-(42) (Fig. 22) furnished (S ) - and (R)-la-vandulol esters (57), respectively, which wereused in the synthesis of optically active acyclicC45 and C50 carotenoids (VONK R A M E RndB A N D E R ,982). (For oth er exam ples of the us-es of 3-hydroxybutanoate esters see SEURLINGand SEEBACH,977; FRATER,979a, b, 1980;FRATERt al., 1984.)

OH TIPS2-Ph -

02 O M F OMe

46 47 48 49

Fig. 18. P-Lactam intermediate synthesis.



362 10 Synthetic Applica tions of Enzyme-Catalyzed Reactions

‘BuMe2SiQ ’ -54

Fig. 20. Pheromone synthesis.

Fig. 21. Stereoisomersof Sitophilus weevil pheromone component.

(R)-46>97%ee

Carotenoid precursors

(3-42 > 97% ee ), (R)-57

Fig. 22. Routes to carotenoids.

2.3 Oxidation Reactions

2.3.1 Horse Liver Alcohol Dehy-

drogenase

Horse liver alcohol dehydrogenase(HLADH) is probably the most intenselystudied oxidoreductase. It accommodates awide range of substrates, but is highly stereose-lective. HLADH is normally used for the re-duction of aldehydes and ketones which is thethermodynamically preferred direction, how-ever, with a suitable cofactor regeneratingsystem, oxidations are also possible.

Oxidation of the diol (58) proceeds withp r o 3 enantioselectivity t o give the hydroxy al-dehyde (59) .This undergoes in situ hemiacetal




formation to give the lactol (60) which is fur-ther ox idized to the lactone (S)-(3b).Using theconditions indicated 5.85 g (68% yield) of thelactone was prepa red in a single reaction (Fig.

23) (DRUECKHAMMERt al., 1985).It is a phy-tyl chain synthon (FISCH LI,980).Similar ster-eoselectivity is exhibited by the HLADH cata-lyzed oxidation of glycerol (61) to L-(-)-glyc-eraldehyde (62) (BALLY and LEUTHARDT,1970).

2.3.2 Pseudomonas putida

The conversion of an unactivated car-bon-hydrogen bond to a carbon-oxygen bondcan b e accomplished using ozone, peracids, orhigh oxidation state iron rea gents such as theGif system. However these rea gents have lowselectivity (usually for methine carbons) anddo no t tolerate most functional groups. TheBarton reaction of steroidal 20-nitrites andBreslow’s remote functionalization tem platesare regioselective due to proximity effects, but

5 8 Coenzymes:

NAD’, FMN,

FMN reductase,

catalase,pH 9. 4ooHLADH 0OH

are not generally applicable. In contrast, manyenzymes are capable of effecting such reac-tions regio- and stereoselectively in the pre-sence of most functional groups (JOHNSON,

1978; DAVIES t al., 1989; DRAUZ nd WALD-

A major a rea of interest has been the P-hy-droxylation of carboxylic acids and esters. ( S ) -

(+ -P-hydro xyisobutyric acid (64)s manufac-tured in 48% yield by oxidation of the pro-S-methyl group of isobutyric acid (63) by Pseu-domonas putidu (Fig. 24) (GOODHUE ndSCHAEFFER,971). It has been used as a syn-thon in syntheses of a-tocopherol (66) (vita-min E) (COHEN t al., 1976), R)-muscone and,unnatural (S)-muscone (67) (BRANCA ndFISCHLI,977).

The ansa macrocyclic maytansinoids (69)have antitumor activity (KUP CH AN t al.,1977),and have been th e focus of many ph ar-macological and synthetic efforts. The “nor th-eastern region” (68) synthon containing fourchiral centers was prepared in gram quan titiesfrom (S ) - (+ -P-hydroxyisobutyric acid (64)

(Fig. 25). Bo th c enters destined to becom e C-5and C-7 were sequentially converted to alde-hydes and alkylated without epimerization at

C-6. The two “new” chiral centers at C-4 andC-5were introduced by S harpless chiral epox-idation, whereas that at C-7 resulted from dia-stereoselective addition to an aldehyd e (ME Y-ERS and HUDSPETH,981).

Furth er examples of the im portance of (64)as a chiral synthon can be found throughoutthe literature, including the synthesis of cal-cimycin (31) (Fig. 11) (EVANS t al., 1979) andother polyether and macrolide antibiotics(JOHNSONt al., 1979; JOHNSONnd KISHI,1979).

MA”, 1995).

(9-3b 0% yield, 87% ee 60 2.3.3 Lipoxygenases

Lipoxygenases are non-heme iron contain-ing dioxygenases which catalyze the reactionof an ally1 mo iety with oxygen to give a hydro-peroxide. Overall the process is equivalent to

HO OH HO an e ne reaction w ith oxygen acting as the elec-trophile. The methylene group participating inthe reaction must be located between two al-kene groups and hence in principle lipoxygen-

ation of arachidonic acid (70) (Fig. 26) could

O HHLADH(q- CHO

61 62

Fig. 23. Enantiotopic oxidation of prochiral meso-

alcohols.



364 10 Synthetic A pplication s ofEnzyme-Catalyzed Reactions

-Pseudomonas putida

48%yield, >99% ee C02H

64steps J

3

J JI Steps

H

I

66 Vitamin E , a-tocopherol ( R ) - 6 7 R 1 = Me, R2= H( 9 - 6 7 R = H , R2= M e

Fig. 24. Enantiotopically specific hydroxylationof isobutyric acid and uses.

‘BuMesiOH O steps_ 9teps_ M&H 23

7 OEEC02H

(3-(+)-64 OHCSEt

68 EE =Ethoxyethyl

69 R = H or acyl; Maytansines

Fig. 25. A biotransformation routine to maytansinoids.

Potato

Lipoxygenase

15%yield

1 1 12 1 5

70 Arachidonic acid 71 (S)-S-HpETE1COzH

72 Leukotriene A

LeukotrienesB4, C4, 4 nd E&-

Fig. 26. The leukotr ienes from arachidonic acid via lipoxygenase.




occur at carbons 5 ,8, 9,1 1,1 2, or 15. Soybeanlipoxygenase catalyzes the formation of (S)-15-HPETE (15-hydroperoxyeicosatetraenoic

acid) and potato l ipoxygenase (S)J-HPETE

(71). he p r o 3 hydrogen a t C-7 is lost stereo-specifically in this reaction and indeed thus farall enzyme catalyzed lipoxygenations occursuch that the hydroperoxide group is intro-duced anti to the eliminated hydrogen (COR EYet al., 1980; COR EY nd L ANSBU RY,983).

2.4 Carbon-Carbon Bond Forma-

tion Reactions

2.4.1 Farnesyl Pyrophosphate Syn-thetase

The enzymes of terpene biosynthesis arerarely employed in biotransformations, never-theless they have tremendous potential. Bothenantiomers of 4-methyldihomofarnesol (75)(Fig. 27) were prepared by coupling the iso-

pentenyl pyrophosphate homologs (74) withtritiated dihomogeranyl pyrophosphate (73)catalyzed by partially purified farnesyl diphos-

phate synthetase from pig liver. Cleavage ofthe pyrophosphate groups by alkaline phos-phatase gave the 4-methyldihomofarnesols(S)-(75) (810 pg), (R)- (75) 590 pg). The radioactivity of the tritium label was used to guidethe purification, which otherwise would have

been extremely difficult on this minute scale(KOYAMAt al., 1980, 1987; KOBA YASHIt al.,1980).

2.4.2 Aldolases

Asymm etric carbon-carbon bond formationbased on catalytic aldol addition reactions re-mains one of the most challenging subjects insynthetic organic chemistry. Many successfulnonbiological strategies have been developed(EVANS t al., 1982; HEATHCOCK,984), butmost have drawbacks - he need for stoichio-metric auxiliary reagents and the need formetal-eno late com plexes for stereoselectivityare two. However, examples of preparativescale aldolase catalyzed reactions abound inthe literature.

High yields of condensation products areobtained with lower primary and secondaryaldehydes, but as expected a tertiary carbonatom a o the aldehyde stops the reaction (EF-

chains or branching aliphatics lead to loweryields. Th e reaction to lerates a wide rang e offunctionality (EFFEN BER GER nd STRAUB,

1987; O'CON NELLnd ROSE,1973). It is worthnoting the mono-functionalization of the dial-dehydes since the high yields obtained(94-98% ) wou ld be very difficult using classi-cal chemistry without the use of protectinggroups.

FENBERGER and STRAUB, 1987), and long

1 Farnesyl pyrophosphate synthetase ( R ) - 7 5

2 Alkaline pho spha tae

OH

73

12% yield

( 9 - 7 5Fig.27. Juvenile hormone 1 synthesis.



366 10 Synthetic Applications ofEnzym e-Catalyzed Reactions

2.4.3 Acyloin Condensation

Considerable attention in fluorine chemis-try has been focused on the search of chiral

synthetic tools for th e asy mm etric synthesis offluorinated bioactive molecules (QU ISTAD tal., 1981). Yeast catalyzed reaction has be enused a s a m eans of preparing chiral trifluoro-methyl compounds (KITAZUM End ISHIKAWA,1984).Ferm entation of trifluoroethanol and ana,p-unsaturated ketone (77) (Fig. 28) withyeast, gave trifluoroketols (78) in yields of26-41%, but w ith an ee of ove r 90%. Theseacyloin-type reactions re sult in valuable chiralsynthons, as they do not necessarily corre-

spond to those easily obtained from the chiralpool of natural products.

2.4.4 Cyanohydrin Formation

Oxynitrilase enzymes, among others, cata-lyze th e asymm etric addition of hydrogen cya-nide to the carbonyl group of an aldehyde(79), to form a chiral cyanohydrin (80) (Fig.29). These are versatile starting materials forthe synthesis of several useful synthons such as

a-hydroxy acids (81), aminoalcohols (82), andacyloins (83) (BECK ER nd PFEIL, 1966; seeChapter 6, this volume). Since only a singleenantiom er is produce d during the add ition toa prochiral substr ate, the availability of differ-ent enzym es that give either (R)-or (S)-cya-nohydrins is imp ortant.

2.5 Multiple Enzyme Reactions

Most enzymes ar e specific with respect to

the ty pe of reaction they catalyze, enabling

CNOxyni rilase

HCN

(85-99ee’s ypical)

RCHO c H+ OH

R

80

79 45-100%yield

CF3CH20H qR0

7 7 R = Me, Et

Yeast

$ 0

F 3 c 4 R 78

O H

Fig. 28. Yeast catalyzed acyloin condensation.

them to operate independently on their ownsubstrate in the presence of oth er enzym es andtheir substrates. This allows multiple, sequen-tial, synthetic transformations to be carriedou t in “one pot reactions” (Fig. 30).

Many am ino acids can be prepa red by lyasecatalyzed addition of amm onia to th e requisitealkene. Aspartic acid (loa) is produced for th emanufacture of aspartame (13) (cf. Fig. 4) bythe aspartase catalyzed addition of ammoniato fumaric acid. Using a whole cell system boththis step and decarboxylation were used to

prepare L-alanine (85) (Fig. 30) (TAKAMATSUet al., 1982). Similarly a com bination of aspa rt-ase, aspartate racemase, and D-amino acidaminotransferase converts fumaric acid toD-alanine (YAMA UCHIt al., 1992).

WHITESIDESnd W ONG WONG t al ., 1983)have com bined enzyme catalyzed and conven-tional steps for the prepa ration of unusual sug-ars (Fig.30).Aldolase catalyzed condensations

COOH

--+OH 81

R

CHzNH;!

H+OH 8 2

R

Fig. 29. Routes to versatile cyanohydrins.R



3 Alicyclic Biotransforrnation Products 367

U*

L-lactaldehydeAldolase

rGlycerol

HO OH HO OP

8 89"=El@=

87

D-LactaldehydeAldolase

H + , Heat

Pyridine-RO

9 2 Furaneol

91 R = p J H o= H

Fig.30. Examples of multiple enzyme transformations.

of dihydroxyacetone phosphate (87) with L- orD-lactaldehyde lead to 6-deoxysorbose (89)

and the blood group related monosaccharide6-deoxyfucose (91), respectively. Either ofthese is readily converted into the flavor prin-ciple furaneol (92) - a caramel flavor compo-nent.

3 Alicyclic Biotransforma-tion Products


Reactions


A wide range of cyclic (and acyclic) meso-diesters are hydrolyzed by PLE. In the hydrol-

ysis of monocyclic cis-1,2-diesters (93) a rever-

sal of stereospecificity is observed as one goesfrom six-membered rings to three membered(Fig. 31) (SABBIONIt al., 1984; MOHRet al.,1983). The cyclopentyl diester marks the"crossover" point with only a 17% ee. As withthe acyclic half esters, either the ester or thecarboxylic acid may be reduced selectivelywith lithium borohydride or borane, respecti-vely, and the products lactonized. For example,oxidative ring cleavage of the lactone (96)

gives the dicarboxylic acid, which after esterifi-

cation can be regioselectively cyclized with po -tassium t-butoxide to the cyclopentenolactone(97) (Dieckmann cyclization). This has beenused in the synthesis of prostacyclin (98) (GAISet al., 1984) and brefeldin A (99) (GAISandLUKAS, 984) (Fig. 32).

The diacetate (100) is hydrolyzed with highenantioselectivity by PLE or PPL (Fig. 33) togive the hydroxy ester (101). Moreover, thecorresponding diol is acetylated by ethyl ace-tate under PPL catalysis to give ent-(l01)

(HEMMERLEnd GAIS,1987). Similarly PLEcatalyzed hydrolysis of the diacetate (102)



368 10 Synthetic A pplication s of Enzyme-Catalyzed Reactions

Major hydro lysis s ite

C02Me n = 3 , 4

C02Me n = 2 , 3

93

n Yield % We

1 90 >97 ( 1R,2S )

2 98 >97 ( 1R,2S )

3 98 1 7 ( 1 S , 2 R )

4 98 >97 ( 1 S 2 R )

Fig. 31. Pig liver esterase specificity.

gives the hydroxy ester (103) which is a keyintermediate in the 3-component synthesis ofprostaglandins, (+)-biotin (210) Fig. 64), andA-factor (WA NG t al., 1984).

3.1.2 Lipases

3.1.2.1 Arthrobacter LipaseA useful application of Arthrobacter lipase

is the resolution of the alcohol moiety of the

meso-94 95 99%Yield

>99%ee

pyrethroid insecticides. The crude mixture ofhydrolyzed (105) and unhydrolyzed (106)products was heated with mesyl chloride fol-lowed by aq ueous base to selectively invert the

stereochemistry of the alcohol (105) (Fig. 34).In this manner, only one enantiomer was ob-tained from racemic starting material (104).

3.1.2.2 Enzymic Ester Resolutions

Enantiomeric distinctions are observed inhydrolyses of esters of chiral alcohols using hy-drolytic enzymes. The synthesis of L-menthol(- -(108) through enzyme catalyzed hydroly-sis of its racemic est er (+ )-(107) is of commer-

cial intere st (Fig.35) (MOROE t al., 1970).

3.1.2.3 Porcine Pancreatic Lipase

The asymmetric hydrolysis of cyclic meso-diacetates using PP L is complementary to thePL E catalyzed hydrolysisof the correspondingmeso-l,2-dicarboxylates.n fact the cyclopen-tane derivative, obtained with low ee's usingPL E (Fig. 31), is produced with 86% ee w ith

PPL (KASEL t al., 1985; LA UM ENnd SCHNEI-DER, 1985).When a range of cyclic diesters (93)and (94) were subject to hydrolysis by PP L an dPLE . the reactions with PP L terminated wh en

0e02C96 97

>85% Total Yield

/ 1

/

"C5Hll 98 Prostacyclin 99 Brefeldin A

Fig. 32. Enzymic synthesis and uses of chiral bicyclic lactones.




AcO OAc

100

AcbAcO

1 0 2

0

-8LE or PPL

HZO

HO OAc

101

rap107

Trichaderma

Species

OH

PLE

H 2 0

AcOA

103 83%yield

82%->96% ee (+)- lo7 (-)- lo8

Fig. 33. Examples of the use of pig liver esterase. Fig. 35. Resolution of a L-menthol ester.

Arrbrobacter

AcO

104 105 106

I 1 MsC1, Et3N

Insecticides-ent-105 106

Fig. 34. Pyrethroid insecticides using lipase resolution.

only one ester group had been cleaved, where-as those catalyzed with PLE proceeded tocleave the “second” ester group unless the re-action was carefully monitored (LAUMENndSCHNEIDER,985; cf. SABBIONIt al., 1984).PLE and PPL hydrolyses of esters are best runat pH 7-8 (maintained with a pH-stat) to avoid

concomitant, base catalyzed hydrolysis. Thesemild conditions were exploited in the PPL cat-alyzed hydrolysis of the methyl ester of thehighly sensitive prostaglandin precursors (109)and (112) o the corresponding acids (110) and(113) (Fig. 36, PORTER t al., 1979; Fig.37,LINet al., 1982, respectively).



370 10 Synthetic Applica tions ofEnzyme-Catalyzed Reactions

3.1.3 Yeast

Although yeast are well known for reducingketones to alcohols however they also act as

esterases. Bakers' yeast was used to hydrolyze10,ll-dihydro-PGA, methyl ester (115)with-out reduction of the ketone group or the 13,14-alkene bond (Fig. 38) (SIHet al., 1972).

109 R = M e110 R = HIh , 92% yie ld

PPL, pH 8, 37T,

Br

e,O2R

-OH

r

0

-OH

r

tC 4 H

Fig. 36. Use of porcine pancreatic lipase in pros-taglandin synthesis.

112 R = M e PPL, p H 8 ,37"C,

113 R = H_1 h, 87% yield

? Z R

-O H

t*a

H

H

-i3H 114

Fig. 37. Synthesis of prostaglandin endoperoxide

analogs.

3.2 Enzyme Catalyzed Reduction

Reactions

3.2.1 Yeast

In analogy to the yeast reduction of P-ketoesters (Sect. 2.2.1.2), cyclic P-keto esters (117)reliably yield the (2R,3S) products (118)(BROOKSt al., 1984).Reductions involving cy-clic substrates are often more diastereo- andenantioselective than the corresponding acy-clic compound. The hydroxy esters (118)wereconverted to unsaturated ketals which under-went diastereoselective addition and alkyla-tion to give (2S)-2-hydroxycyclohexane 6-al-

kyl esters (119) nd (120) with three contigu-ous stereogenic centers for use in naturalproduct synthesis (Fig. 39) (SEEBACHnd HER-RADON, 1987). The hydroxy ester (118) hasbeen used in an approach to the southern re-gion (121)of avermectin (122) (Fig. 40)(BROOKS t al., 1982). Avermectins and mil-bemycins are potent parasitic and anthelminticagents A BAR RE^ and CAPPS, 986).

BROOKSt al. (1982,1983,1984) investigatedthe yeast reduction of a wide range of simple

cyclic P-diketones. Reduction of racemic cy-clopentadiones, e.g., (123)gave predominantly(2S,3S)-3-hydroxycyclopentanones,.g., (W) ,whereas cyclohexadiones, e.g., (127)gave pre-dominantly (2R,3S)-3-hydroxycyclohexanones,

e.g., (l28).The other enantiomer of the start-ing material, e.g.,(123)and (l24)was either re-covered or reduced to the other 3-hydroxy-cycloalkanone diastereoisomer pair. With larg-

115 R = M e116 R = H10,l-Dihydro-PGAl

Bakers' yeast

Fig. 38. Ester hydrolysis using yeast.




1 1 7 1 1 8 1 1 9

Fig. 39. Diastereoselective elaboration of P-hydroxyesters.

1 2 0

1 2 1 0

oc22 Avermectin Az6

Fig.40. The use of yeast in an approach to avermectin.

er mem bered rings (7-9) the stereochemicaloutcom e was less predictable, nevertheless t heenantiomeric excesses were still high, albeitthe yields were lower. The alcohol (124)was

used to prepare an intermediate in the synthe-sis of anguidine (126) BROOKSt al., 1982,1983). The synthesis of zoapatano l (129)requires a (2S,3R)-3-hydroxycyclohexanone

which was prepared from both the major(2R,3S)-(l28) and the m inor (2S,3S)-diastereo-isomers (Fig. 41) (BROOKSt al., 1984).

Yeast has also be used in large-scale reac-tions, as illustrated in t he reduction of the oxo-

isophorone (130) (Fig. 42) to the carotenoidprecursor (131),which has been performed on13 kg of the starting enedione (LEUENBERGER

et al., 1979).

3.2.2 Rhizop us arrhizus

Rhizopus arrhizus induced reduction of thetrione (132) (Fig. 43) is regiospecific and enan-

tiospecific with hydride addition being d irect-ed to the Re-face of the pro-R-carbonyl groupto give the (2R,3S)-stereoisomer (133) (VEL-

LUZ, 1965). The triones (132) can be preparedby treatment of 2-methylcyclopenta-l,3-dionewith base and an a@-unsaturated ketone. Inthe presence of fu rther base they cyclize by anintramolecular aldol reaction to give bicyclicdione intermediates for steroid synthesis(Robinson annulation). The need for reduc-tion is questionable because treatment of thetriones (132) with L- or D-proline effects an

asymmetric aldol reaction to give the bicyclic



372 10 Synthetic Applications of Enzyme-Catalyzed Reactions

123 124 70% yield, 125 126 Anguidine

>98% ee

HS

+-east___)

0

127 128 76% yield, 35% ee

Fig. 41. Routes to anguidine and zoapatanol.

130 131

129 Zoapatanol

0 3

Oxoisophorone 83% yield,

100% ee

Fig. 42. Synthesis of a carotenoid precursor.

ketones as single enantiomers (Hajos-Parrishreaction).


3.3.1 Horse Liver Alcohol

Dehydrogenase

Unless protecting groups are used, discrimi-nation between unhindered primary and sec-ondary alcohol functions in diols such as (134)

and (136) (Fig. 44) is difficult to achieve innonenzymic single step reactions. However,with HLADH only those hydroxyl groups thatcan locate at the oxidoreduction site will beoxidized, so the primary and secondary func-

tions can be discriminated on a regional basis.

Fig. 43. Steroid intermediate from microbial re-

duction.

The initially formed hydroxyaldehyde (l37)undergoes further oxidation via the hemi-acetal (138) in another enantioselectiveprocess to yield lactone (+)-(139) of interestas a synthon for prostaglandin analogs. The

other enantiomer of the bicyclic lactone (139)




OH

HLADH

50% oxidation*

4Y

OH

134

HLADH

50%oxidation____t

OH

136

+

4H OH

135 (-1-134

.\\\OH

4""HO + c-7OH

137 (-)-I36

I tqo..-

OH 0

138 (+)-139

Fig.44. Prostaglandin synthons using HLADH.

which is the precursor of natural prostaglan-dins, has be en obtained by chem ical oxidationof the enantiomer recovered from theHLADH-catalyzed oxidation of rw(136)

(IRWINnd JONES,1977; NEWTONand Ro-

BERTS, 1982).

3.3.2 Microbial Enzymes

The m ost useful enzymes for use in organicsynthesis are those which accept a broad ran geof substrates, while being able to act stereospe-cifically on each. Microbial enzymes have nar-rower structural specificity tolerances thanmam malian ones, however the larger selectionof microorganisms available compensates forthis. Th e identification of an enzy me capableof catalyzing a spec ific reaction on a pa rticular

substrate is performed as for any chemical

transformation - by searching the literaturefor analogies. For exam ple, the known conver-sion of cinerone (140) to cinerolone (141) (Fig.45) (TABENKINt al., 1969) was used to chooseAspergillus niger as a suitable microorganismfor the stereo specific hydroxy lation of th e a,@-

unsaturated ke tone (142) to th e prostaglandin

synthon (143) (KUROZUMIt al., 1973).Fragrance and flavor chemicals have an

enormous world market. A considerable num-ber of these substances are monoterpenoids,sesquiterpenoids, and similar structures. Veryoften the special properties of terpenoids andtheir derivatives depend on the absolute con-figurations of the m olecules,so the synthesis ormodification of these substances dem and reac-tions with high stereospecificity and methodsof introducing chiral centers. Biotransforma-tions are useful in this respect (ROSAZZA,982;KIESLICH,976) with m any applications having



374 10 Synthetic Applications of Enzym e-Catalyzed Reactions

AsperigiIIus niger

140 Cinerone 141 Cinerolone

(CH2),COOH Asperigjllus niger--c + Prostaglandins(!k 42 67% yield

Fig. 45. Stereospecific hydroxylations.

Penicillium8 igitaturn

46%yield>99% el? A

OH

(+)-( ~ - 1 4 4 145 a-TerpineolLimonene

(Orange odour)Fig. 46. Stereospecific modification of terpenoids.

been described (KRASNOBAJEW,984; SCHIND-LER and SCHMID,982).

For example, limonene (144) (Fig. 46) is areadily available and inexpensive naturalproduct. Specific hydration at the double bondof the isopropenyl substituent using Penicilli-urn digitaturn resulted in complete transforma-

tion of (+)-(R)-limonene (144) to a-terpineol(145) (KRAIDMANt al., 1969). Even racemiclimonene affords pure a-terpineol. With mi-croorganisms other than Penicillium digitaturn,limonene may undergo attack at the 1,Zdou-ble bond to form 1,2-dihydro-l ,Ztrans-diols. Anumber of fungi are useful in this biotransfor-mation, with Corynespora cassiicola and Di-plodia gossydina being the most appropriatestrains.This provides an economical method tothese glycols, useful starting materials in thesynthesis of menthadienol, carvone, and other

compounds.

3.3.2.1 Pseudomonas putida

GIBSONt al. (1970) reported the enantiose-lective oxidation of benzene (146)o cis-cyclo-hexadienediol (147a) (Fig. 47) and of tolueneto cis-dihydrotoluenediol(l53) Fig. 49).Thesecompounds are often referred to (incorrectly)as benzene and toluene cis-diol. Many arenesand aromatic heterocycles have been shown toyield diols through microbial techniques (Su-

It is important to note that these compoundscannot currently be made in quantity by con-ventional abiotic chemistry. Hence they offerunparalleled potential for use in natural prod-uct and other syntheses.The microbial conver-sion is facile and high yielding. It is then easyto make derivatives of benzene-cis-diol in astandard radical-type polymerization, such asthe polymer (148) (Fig. 47), which on heatinggives polyphenylene (149) by eliminating twomolecules of H-OR (BALLARD t al., 1983).

Polyphenylene itself is intractable and not eas-ily fabricated, but this method allows produc-tion of polyphenylene films, fibers, coatings,and liquid crystals.

Benzene-cis-diol (147a) has also been usedin the synthesis of both enantiomers of pinitol(152) (Fig. 48). Esterification of the two hy-droxyl groups with benzoyl chloride, pyridine,and DMAP furnished the dibenzoate (147c)which underwent epoxidation anti to the estergroups. Treatment with acidified methanol ef-fected ring opening of the epoxide (150) re-

giospecifically adjacent to the alkene bond to

GIYAMA and TRAGER,986).




0-aIl-[l@]-

GIo n o n

I I146 147

a R = H 148 149b R = A c

Fig. 47. The preparation of polyphenylene.

147c 150

-0,

151

Fig. 48. Synthesisof (+)-pinitol.

give the methyl ether (151).The final two hy-droxyl groups were then installed by osmyla-tion (once more anti to the ester groups), fol-lowed by deprotection to give ruc-pinitol ruc-(152). The individual enantiomers were pre-pared by a parallel route involving resolutionby esterification with an enantiomerically pureacid chloride (LEYand STERNFELD,989).Es-sentially identical methodology was used inthe preparation of (+)- and (-)-pinit01 from1-bromobenzene-cis-diol (HUDLICKY t al.,

1990).The diol derived from toluene (153) (Fig.

49) was ketalized with 2,2-dimethoxypropaneand pTsOH (85% yield) and then ozonolyzedat -60°C in ethyl acetate to give the ketoaldehyde (154) (60-70% yield). Aldol ring clo-sure catalyzed by alumina gave the @-unsat-urated ketone (155), a known prostaglandinintermediate (HUDLICKYt al., 1988).

Hydrogenation of the toluene derived diol(153) was virtually nonselective, however, thediols (156) and (157) were easily separated.

Treatment with periodate gave the dialde-

QH

2 MeOH.

-t3N

OH

152 ( + ) - P i n i t 0 1

hydes (158) which underwent intramolecularaldol condensation to give the cyclopentenylcarboxaldehydes (159), which are potentialsynthons for bulnesene and kessane sesquiter-penes (HUDLICKYt al., 1988).

Benzene-cis-diol(147a) was also used in thesynthesis of D-myo-inositol-l,4,5-triphosphate(162) (Fig. 50). The key reactions are diaster-eoselective epoxidation and regioselectivering opening of the epoxides, which resemblethose used in the previous syntheses of pinitol

(HUDLICKYt al., 1991).Pseudomonas putida also catalyzes the oxi-

dation of styrene, chlorobenzene and deriva-tives to the corresponding cis-diols, whichwere used for synthesis of natural products(HUDLICKYt al., 1989;HUDLICKYnd NATCH-us, 1992; ROUDEN nd HUDLICKY,993). Theacetone ketals of D- and L-erythrose, and ~ 4 -bonolactone which are frequently used in nat-ural product synthesis (HANESSIAN,983),were obtained from chlorobenzene (HUD-LICKY and PRICE, 990;HUDLICKYt al., 1989)

in up to 50% yield.



376 10 SyntheticApplicationsof Enzyme-Catalyzed Reactions

157 24% yield en -15 en - 159

Fig. 49. Uses of toluene cis-glycol synthons.

B n O a I_Bn B n o p ~ ~Bn H 2 0 3 P o q 1P03H2

-- -- --HO*

O P 0 3 H 20

146 160 161 162

Fig. 50. Synthesisof myo-inositol phosphates.

proved to >95% by recrystallization. Ozonol-ysis of the double bond yields 1,4-dialkylcyclo-pentanes or tetrahydrofurans which have beenused in the synthesis of carbocyclic nucleo-tides, aristeromycin (169) and neplanocin A(170) (ARITA t al., 1983) or conventional nu-

pO1ycyclic Biotransfor-mation Products

4.1 HydrolydCondensation

Reactions cleotides such as showdomycin, 6-azapseudou-rine, and cordycepin (ARITA t al., 1984).


PLE catalyzed hydrolysis of the diester(165) (Fig. 51,X=O or CH,) gives a quantita-tive yield of the monoester (166), with an

enantiomeric excess of 77% which can be im-


Acetate and dihydrocinnamate esters are

hydrolyzed at almost identical rates using hy-



4 Polycyclic Biotransforrnation Products 377

1Os042 Me2C0, CuS04

k02Me 164

1 mCPBA % 0 2 R

169 170

Aristerornycin Neplanocin A

~ o ~ M ~67 R = M eILE168 R = H

Fig. 51. Enantioselective hydrolysisand uses of tricyclic.

1hymotrypsin

Fig. 52. Hydrolysis using a-chymotrypsin.

droxide ion (BENDER t al., 1964), however a-chymotrypsin regiospecifically cleaves dihy-drocinnamate esters from simple mixed ace-tate-dihydrocinnamate diesters (Fig. 52)(JONES nd LIN,1972). For example, the diest-er (171) s hydrolyzed to give the lp-acetate(172) n 53% yield with sodium hydroxide,with 35% of the diol also isolated. With a-chy-motrypsin, the lp-acetate is obtained in quan-titative yield, although the rate of hydrolysis isslower. The “normal” substrate for chymotryp-

sin is the cleavage of the C-terminal amidegroup of aromatic amino acids in peptides.Hence in the hydrolysisof the diester (171) hedihydrocinnamyl group is acting as a surrogatefor this group.

4.2 Enzyme Catalyzed ReductionReactions

4.2.1 Yeast

The potent biological activity of prostaglan-dins has spawned enormous synthetic efforts(BINDRAnd BINDRA,977;SCHIENMANNndROBERTS,982;NEWTONnd ROBERTS,982).The majority of the activity resides in the nat-ural enantiomer and hence early syntheses of

racemic prostaglandins initially sufficed, how-ever, these were quickly supplanted by synthe-ses of enantiomerically pure materials. In oneapproach, the racemic bicyclic ketone (173)was reduced with yeast to give the diastereoi-somers (174) nd (175) erived from differentenantiomeric series (Fig. 53). These were sep-arated by column chromatography and thenreoxidized to the ketone and converted to thebromohydrins (176) n one step. Each of thesewas then converted to natural prostaglandins,e.g., (177)by different routes (NEWTON ndROBERTS,980,1982;ROBERTS,985).



378 10 Synthetic Applica tions ofEnzym e-Catalyzed Reactions

0

8ac-1731 Yeast

H H Hh@

(1R,5S,GS)-174 (lS,5R,65)-175

IcNHBrH 2 0 , M e 2 C 0

(1S3S) 176 (lR,5R)-176

Fig. 53. Use of yeast in prostaglandin synthesis.

4.2.2 Microorganism Reductions

As an extension t o th e reduction of cyclic p-ke to esters (Sect. 3.2.1) com pounds with a sec-ond, or third, ring are also reduced in the samemanner with high selectivity (BROOKS t al.,1987). Early interest in th e reduction of cyclic

178

53 % yield MicroorganismI

M 179

#MeO

180 Estradiol-3-methyl ether

Fig. 54. Microorganism reductions in steroid syn-thesis.

p-diketones centered around the conversionof diones such as (178) (Fig. 54) into chiralintermediates, such as the hydroxy ketone(179) (KOSMOL t al., 1967) using m icroorgan-isms such as Bacillus spp. and Saccharomycesspp. The intermediates could then be used in

the total synthesis of steroids, such as estradi-01-3-methylether (180) (DAI nd ZHOU, 985;REHMnd REED, 984).

Hydroxysteroid dehydrogenases (HSDH)catalyze the regiospecific reduction of ketones(and oxidation of hydroxy groups) of steroids.Enzymes selective for reduction or oxidationat positions 3 ,7, 12 , and 20 (Fig. 55) have beenreported (BONARA t al., 1986). In most casesHSDHs have high regio- and diastereoselec-tivity which cannot be achieved with conven-tional abiotic reagents. For example 12-dehy-drocholic acid (181) undergoes reduction at



4 Polycyclic Biotransformation Products 379

7a-HSDHI

3a-HSDH.

Fig. 55. Regiospecific keto ne reduction.

the 7-position (182) and oxidation at the 3-PO-sition (183) in 97-99% yield without affectingthe 1 Zk eto group (CARREA t al., 1984). Itshould be noted that HSDH is a functionaldescription usually derived from substrate

screening. In many cases the biologically sig-nificant substrates for these enzymes are un-known a nd a re unlikely to b e steroids. How ev-er, the ability to reduce large hydrophobic sub-strates suggests that these enzymes will havebroad applicability.

Thermoanaerobium brockii is a thermophil-ic organism which thrives best a t high temp er-atures. Its enzymes have adapted to this re-gime and hence m ay be sufficiently robust forindustrial applications. The alcoh ol dehyd ro-genase from Thermoanaerobium brockii

(TBA DH ) reduces methyl ketones most effi-

HeoHBADH

rac-173 175 95%ee

Fig. 56. Preparation of optically active synthons fo rprostaglandin synthesis.

ciently and hence iso-propanol or acetone a refrequently used as the donor or acceptor for

coupled redox reactions. However bicyclic ke-tones, e.g., (173) (Fig.56) are also efficiently re-

duced w ith Re-face delivery of hydride. In con-trast to the corresponding yeast reduction(Fig. 53) the other enantiomer of the ketone(173) is only reduced . This is an advantage b e-cause it is easier to separate a ketone (173)from an alcohol (175) than to separate two al-cohols (174) and (175) (ROBERTS,985).

Ferm entative reduction of the racemic dike-tone (184) with Aureobasidium pullulans isstereoselective for the Re-faces of both c arbo -nyl groups (184) (Fig. 57). The rrans-1,4-diol

product (185) was isolated in 33% yield ac-companied by a cis-1,4-diol derived from ent-

(184) and other diastereoisomers. The trans-1,4-diol product (185) has a C,-axis of sym me-try (like the diketone (184))and hence addi-tions to the alkene give a single stereoisomer.This was exploited in a syn thesis of compactin(40a) (Fig. 16) (WANG t al., 1981).


drogenase

HLADH reduces all simple cyclic ketonesfrom four- to nine-membered, and parallelsthe rate of reduction by sodium borohydridein isopropanol (VANOSSELAERt al., 1978). Itexcels over traditional chemical methods incombining regio- and enantiomeric specific-ities (JONES, 1985). H L A D H catalyzes the re-duction of many cis- and rrans-decalindiones.These highly symm etrical compounds a re po-tentially useful chiral syntho ns. Th e reductions

are specific for pro-R ketones only (186), to



380 10 Synthetic A pplication s of Enzyme-Catalyzed Reactions

184 185 33%yield

Fig. 57. The first route to compactin.

40a Compactin

ow9%LADHield :w -7% yield .$overall

186 187 188 Twistanone

HLADH

25%Yield0

189

Fig. 58. HLADH-catalyzed reductions.

give enantiomerically pure keto alcohols (187)(Fig. 58). Even with more symmetrical diones,for example, (189),which lacks a ring junctionprochiral center, the reductions are stereospec-ific.The decalin (187, =H) has been convert-ed to (+)-4-twistanone (188) (DODDS nd

J ON ES , 1982,1988).These representative hydroxy ketones are

not readily available using traditional chemi-cal methods, and they are clearly broadly use-ful synthons for steroidal, terpenoid and relat-ed structures.

NAKAZAKInd NAEMURANAKAZAKIt al.,1983;NAEMURAt al., 1984,1986) have shownthat HLADH accepts a variety of cage shapedmolecules as substrate, and reduces them withenantioselectivity. Such rigid hydrocarbonsand derivatives are of special interest as test

cases for chiroptic theories.

190


4.3.1 Microorganism Oxidation

Except in the area of steroid research, little

work on the use of oxygenases to functionalizenonactivated carbon centers has been done.These provide useful synthons for the prepara-tion of medicinally important compounds.Sixty-one cultures were screened for theircapacity to hydroxylate cyclohexylcyclohex-ane (191) Fig. 59) (DAVIESt al., 1986). Twospecies (Cunninghamella blakesleeana andGeotrichum lucrispora) gave the diequatorialdiol (192) s the major product, for use in thesynthesis of the carboxylic acid (l93),an ana-log of the chemotactic agent, leukotriene B3

(194).




R e R --) H02C(CH2) 3&? C8H17

OHeotrichlum lacrispora orCunninghamella blakesleena

H o 2 c ( ; ; w ;I -191 R = H

192 R=OH

OH 194 F8H17

Fig. 59. Preparatio n of leuk otriene analogs.

Fig. 60. Transformations of progestero ne an d lithocholic acid.

4.3.2 Hydroxylation of Steroids

The hydroxylation of nonactivated centers

in hydrocarbons is a very useful biotransfor-mation, since it has very few counterparts intraditional organic synthesis. Intense researchon the stereoselective hydroxylation of al-kahes began in the late 1940s in the steroidfield, when progesterone (197)was convertedto lla-hydroxyprogesterone (198) (Fig. 60),thus halving the 37 steps needed using conven-tional chemistry, and making (198) availablefor therapy at a reasonable cost. This regiose-lective oxidation became commercially impor-tant for the manufacture of cortisone, since

11P-hydroxy configuration is required for the

optimum biological activity of hydrocortisoneand the corticosteroids, but difficult to obtainusing traditional chemistry. Nowadays, virtual-ly any center in a steroid can be regioselective-

ly hydroxylated by choosing the appropriatemicroorganism (DAVIESt al., 1989).The high-ly selective hydroxylation of lithocholic acid(195) n the 7P-position was achieved usingFusariurn equiseti (SAWADAt al., 1982).Theproduct ursodeoxycholic acid (1%) (Fig.60) scapable of dissolving cholesterol and cantherefore be used in the therapy of gallstones.There is a great deal of research activity in thisarea due to the immense industrial impor-tance, most notably by KIESLICH,ONES, HOL-LAND, nd CRABB.




4.4 Carbon-Carbon Bond

Formation Reactions

4.4.1 Yeast Cyclases

Some enzymes involved in the biosynthe-sis of steroids have recently been used inorganic synthesis. 2,3-Oxidosqualene-lanoste-

rol cyclase, from bakers' yeast (Saccharornyces

cerevisiae) catalyzes the synthesis of a numberof lanosterol analogs from the corresponding2,3-oxidosqualene derivatives (MEDINAandKYLER, 988). In the cyclization of the vinylderivative (199) Fig. 61), the cyclization cas-

cade is followed by hydrogen and methylmigrations as usual, but in addition the vinylgroup also undergoes a 1,Zshift from C-8 toC-14 (200). The structure was confirmedby conversion to the alcohol (201) MEDINA tal., 1989) which is a natural product and an in-hibitor of HMG-CoA reductase (GRUNDY,1988).

2,3-O?<ido-

squalene

cyclase

201 R = C H z O H

Fig. 61. Use of lanosterol cyclase.

4.4.2 Reactions Involving Acetyl

CoA

Coenzyme A (CoA) thioesters are involvedin the biosynthesis of steroids, terpenoids, mac-rolides, fatty acids, and other substrates (BILL-HARDTet al., 1989;PATEL et al., 1986).However,these enzymes can only be used practically in or-ganic synthesis if the CoA thioester is recycled,due to its high cost (BILLHARDTt al., 1989).

4.5 Exploiting Combinations

of Enzyme Specificity

4.5.1 Combinations

of Dehydrogenases

The degree of stereochemical controlachievable with nonenzymic methods in asym-metric synthesis has made great improvementsin recent years. Still,no chemical chiral reagentcan yet approach the abilities of enzymes tocombine several different specificities in a sin-gle step reaction.

Reduction of the racemic ketone (202)by

yeast is Re-face selective, consequently bothenantiomers are reduced to (S)-alcohols (Fig.62).The desired alcohol (1 S,7R)-(203) was re-oxidized to the ketone (7R)-(202) nd utilizedin the synthesis of 4-demethoxydaunorubicin(205), a member of the adriamycin family ofanthracycline antibiotics. The unwanted (7s)-epimer was similarly reoxidized to the ketone(7S)-(202), acemized with pTsOH in aceticacid at 110°C (TERASHIMA and TAMATO,1982), and resubjected to yeast reduction.

4.6 Multiple Enzyme Reactions

Interest in multienzyme synthesis has inten-sified, as it has enormous potential, for themanufacture of complex pharmaceuticals. Forexample, the readily accessible Reichstein'scompound S (206)undergoes llp-hydroxyla-tion with Curvularin lunata, but with othermicroorganisms dehydrogenation also occursto give prednisolone (207) Fig. 63) (MOSBACH

et al., 1978).




0 0

&JA "'OH

I 2 0 2OMe

rac-202

Yeast

OH

DMSO, PY,rt , 2h

Friedels-Crafteaction *fpy\

( 7 R ) - 2 0 2

0 + 204

N H 2

205 4-DemethoxydaunorubicinOMe 2 0 3 32% yield

Fig. 62. Use of alcohol dehydrogenases.

206 Reichstein's Compound S

11p-Hydroxylase

I A' - Dehydrogenase

1

207 Prednisolone

Fig. 63. Multienzyme synthesis.

210 (+)-Biotin

Fig.64. Enantioselective hydrolysis using pig liveresterase.




5 Heterocyclic Biotrans-

formation Products


Reactions


The general hydrolysis of meso-diesters hasalready been discussed in Sect. 3.1.1. The ex-amp le shown in Fig.64 shows the enantioselec-

tive hydrolysis of a rneso-diester (208) whichaffords an intermediate for the synthesis of(+)-biot in (210) (IRIUCHIJIMAt al., 1978).Bi-otin functions as a cocarboxylase in a nu mb er

of biochemical reactions. Initial experimentsshowed that the pro -(S) group of the bis-ester(208) was selectively cleaved, although the halfester (209) was only obtained in 38% ee. Incontrast, the bis-propyl ester affords the corre-sponding monoester in 85% yield, and with75% ee. The diacetate of the diol correspond-ing to (208)has been hydrolyzed by PPL to themono-alcohol, by preferentially cleaving the

Fig. 65. Peptide synthesis.

0 r Isopenicillin N @ BH 3 N mNzLSy nthetase + H3Nlyy

I co2

0 % Oco2

2150

co2

214 ' 0

co2Epimerase

217 R = H2 sygenase218 R=OH

Fig. 66. Formation of p-lactam derivatives.



5 Heterocyclic Biotransformation Products 385

pro-(R) acetoxy group, in 70% yield an d 92%ee. This provides an alternative route to (+)-

biotin (WANG nd SIH,1984).

5.1.3 Lipases

5.1.3.1 Pseudomonas Lipases

5.1.2 S(L-a-Aminoadipy1)-L-cysteinyl-D-valine (AC V)Synthetase

In the biosynthesis of non-ribosomal pep-tides (e.g., peptide antibiotics) the carboxylgrou p of amin o acids is activated by esterifica-tion with ATP with displacement of pyrophos-pha te to give aminoacyl adenylate (211). This

is transferred to an enzyme bound thioester,which then reacts with the amino group ofano ther aminoacyl thioester to form a pe ptidebond (Fig. 65) (LIPMA NN,973).

The pep tide chain is thus extended from N -to C -terminus on a multienzyme template. A nexample of this is seen in the biosynthesis of S(L-a-aminoadipy1)-L-cysteinyl-D-valineACV)(214) catalyzed by ACV synthetase (fromStrepfomyces cluvuligerus) (Fig. 66) (BAN KO tal., 1987). The enzyme accepts many non-pro-tein am ino ac ids as well as hydroxy acids. Since

ACV is a precursor of penicillins and cepha-losporin, the enzymatically forme d ACV ana -logs may b e conv erted to th e corresponding p-lactam derivatives. AC V (214) is cyclized toisopenicillin N (215) in a rema rkable rea ctionwhich occurs in a single step with retention ofconfiguration at both carbon centers partici-pating in th e cyclization. Epimerization givespenicillin N (216), which undergoes ring ex-pansion to give desacetoxycephalosporin C(217) and allylic hydroxylation to desacetyl-cephalosporin C (218) (BANKO ,987).

The lipases isolated from different Pseu-domonus species (PSL) are highly selective, es-pecially for the h ydrolysis of the esters of sec-ondary alcohols, and the corresponding re-verse reactions (BOLANDt al., 1991).The lowcost and high selectivity and stability of PSLma ke it a very useful reagent for organic syn-thesis (Fig. 67) (K AN t al., 1985a,b).

All commercially available Pseudomonussp. lipases possess a stereochem ical prefe rencefor the (R)-configuration at the reaction

cente r of seco ndary alcohols. Pseud omo nuslipase (P-3C) was foun d t o be highly selective(E >100) for the hydrolysis of 3-hydroxy-4-phenyl p-lactam derivatives, in an approac h tothe synthesis of the C-13 side chain derivativeof taxol (BR IEN A t al., 1993 ). Th e selectivityrema ined high for the hydrolysis of the 3-acet-oxy derivatives, with the ring nitrogen free orprotected.

5.1.3.2 Mucor Species Lipases

The lipases from Mucor species such as M .miehei and M . avunicus have been used in syn-thesis (C HA N t al., 1988). Both enzymes seemto possess a stereochemical preference similarto that of Pseudo mon as lipase.

M . miehei lipase was used in th e resolutionof methyl-trans-p-phenylglycidate 223) viatransesterification in hexane:iso-butyl alcohol(1 :1) (Fig. 68) (YEEet al., 1992).The unreact-ed substrate (2R,3S)-(223) and product(2S,3R)-(224) were obtained in 95% ee. Both

P-Bloc kers

219 220 2 2 1

R = 'Bu, 'Pr; R1 = Me, Pr, n C ~ H l l

Fig. 67. Synthesis of P-blockers.

2 2 2



386 10 SyntheticApplications of Enzym e-Cataly zed Reactions

ph402Me

rac -223M AP- 10

ROH

1u.

C02M e ph

'C02R'

Ph ( 2 R , 3 S ) - 2 2 3 ( 2 S , 3 R ) - 2 2 4

42% yield 43%yield

Paclitaxel, Taxol@

Fig. 68. Use of lipasesin taxol synthesis.

enantiomers were converted to N-benzoyl-(2R,3S)-3-phenyl isoserine (225), the C-13side chain of the antitumor agent taxol. Thehindered iso-butyl alcohol was used to avoidthe reverse transesterification.

A kinetic resolution for the preparation ofthe anti-inflammatory agent ketorolac (227)was achieved with M. miehei lipase catalyzedhydrolysisof the racemic methyl ester (Fig. 69)(FULLINGnd SIH,1987).Since the ester is eas-ily racemized at pH. 9.7, both enantiomers areeventually converted to the @)-product.

Enantiocomplementary to the process is theuse of protease N and some other microbialproteases to produce the (S)-enantiomer(227),which is reported to be more potent in

animal studies.

5.1.4 Penicillin Acylase

It has been shown that penicillin catalyzedhydrolysis of the phenylacetamido group ofpenicillinG (228) Fig. 70), does not affect thesensitive p-lactam ring (BRIENA t al., 1993).This is now an important part of the procedurefor the industrial production of 6-aminopenic-

illanic acid (229) LAGERLOFt al., 1976).

5.1.4.1 Transacylations Using Peni-

cillin Acylase

Transacylation reactions can be induced, asrepresented in the penicillin and cephalospor-in fields (Fig. 71). This methodology is advan-tageous in that no protecting groups are re-quired. The protease from Xunthomonas citricondenses 6-aminopenicillanic acid (229) ndD-phenylglycine methyl ester (230a) r its p -

hydroxy derivative (230b) o give high yieldsof ampicillin(231a) nd amoxicillin (231b), e-spectively (Fig. 71) (&TO et al., 1980). Similar-ly 6-aminocephalosporanic acid (232) s con-verted into cephalexin (233a) CHOIet al.,1981).

Trypsin mediated exchange of threonine (asits ester derivatives) for the terminal alanineresidue of porcine insulin is the basis of a com-mercial process for the production of insulin

for human use.



drogenase

As has been noted previously (Sects. 2.3.1and 3.3.1) a broad range of meso-diols can beenantiospecifically oxidized using HLADH to

lactones via the corresponding hemiacetals



5 Heterocyclic Biotransformation Products 387

Fig. 69. Preparationof ketorolac.

Y

I 2 2 8

PenicillinAcylase

2 2 9 6-Aminopenicillanic acid

Fig. 70. Hydrolysisof penicillin G.

(Fig. 72) (JAKOVICt al., 1982). Formation ofthe lactol or lactone boosts the enantiomericexcess by protecting the less reactive hydroxylgroup from oxidation. In support of this asser-tion, oxidation of the corresponding trans-compounds gives essentially racemic products.The lactone (235)was produced on a largescale (JAKOVICt al., 1982) in high yield andexcellent enantiomeric excess. It providedboth chiral centers of the iridoid aglyconemethyl sweroside (236) (HUTCHINSONndIKEDA, 984). Equally good results were ob-

tained in the oxidation of the cyclobutane diol(237)which was used in a synthesis of grandi-sol (239) major component of male boll wee-vil sex pheromone (JONES et al., 1982). Similar-ly the cyclopropyl lactone (241) Fig. 72) wasreadily transformed into (+)-(lR,2S)-cis-methylchrysanthanenate (242) providing anattractive route to the pyrethroids (JONES,1985). Further examples of such biotransfor-mations in synthesis can be found in macrolide(COLLUMt al., 1980b) and prostaglandin syn-

thesis.

5.2.2 Dihydrofolate Reductase

Dihydrofolate reductase catalyzes the in vi-

vo interconversion of folate and tetrahydrofo-late by NADPH. The reduction of dihydrofolicacid to chiral tetrahydrofolic acid was investi-gated using enzymic and nonenzymic means(REESet al., 1986). The results of the enzymicroute far exceeded the chemical methods, pro-

viding pure stable tetrahydrofolate derivativeswhereas traditional chemical routes gave min-imal enantiomeric excesses.The technique wasapplied to the synthesis of enantiopure 5-for-myltetrahydrofolate for use in cancer “rescue”therapy for patients undergoing cancer chemo-therapy with methotetrexate.

5.2.3 Yeast

Yeast biotransformation of the furanylacrolein (243) provided two intermediates




0

s

C 0 2 H

23 1

a R = H; Ampicillinb R = OMe; Amoxicillinl 3 0

a R = H

b R = O HR (>80 M Yield)

Xanthomonas citri protease

Fig. 71. Transacylation reactions.

CO2H

2 3 3 a R = H; Cephalexin

(244)and (246)for the synthesis of a-tocophe-rol (66) (Fig. 73). Fermentation supplementedwith pyruvate provided the diol (244) by acy-loin condensation, whereas standard reductivefermentation effected reduction of both the al-kene bond and the aldehyde group to give thealcohol (246).Cleavage of the furan ring pro-vided the functionality to conjoin the frag-ments (FUGANTInd GRASSELLI,985a).


Formation Reactions

5.3.1 Rabbit Muscle Aldolase

Rabbit muscle aldolase (RAMA) has beenused in the synthesis of numerous oxygen het-erocycles (BEDNARSKIt al., 1989). The en-zyme accepts a wide range of aldehydes, but is

virtually specific for dihydroxyacetone phos-

phate (248) as the nucleophilic component(Fig. 74). The stereochemistry at the carbonsengaged in the newly formed bond is alwayssyn-3,4-(3S) which is equivalent to D-rhreo.RAMA catalyzed aldol reaction of N-acetyl-aspartate P-semialdehyde (249) gave the diol(250) with a maximum of 40% conversion(37% yield). Reduction of the 2-keto groupwith sodium or tetramethylammonium triace-

toxyborohydride gave predominantly the de-sired 2,3-anri-diol. The N-acetyl group wascleaved with 6 N hydrochloric acid, to give theamine which was converted to a ketone (251)(13% overall yield) by transamination with so-dium glyoxylate. Surprisingly enzyme cataly-zed transamination with a range of enzymeswas less effective than the abiotic reaction(TURNERnd WHITESIDES,989).

Products from RAMA catalyzed aldol reac-tions have been used in efficient approaches toC-glycosides(254) and cyclitols (255) (Fig. 75)

(SCHMIDnd WHITESIDES,990) and the bee-



5 Heterocyclic Biotransforrnation Products 389

OH 80%yield

100 % ee

\\+ O0

meso-234 2 3 5 2 3 6OMe

OH. d \ W

""IfdH HLADH

OH 90%yield

100 % ee0

meso-237 238 23 9 ( 1 R,ZS)-Grandisol

\ P O U HLADH \ 6--CO, Me

..\OH 71% yield *

100 % ee0

meso-240 ( - ) - ( l R , 2s)-241 ( + ) - ( l R , 2 9 - 2 4 2

Fig. 72. Use of lactone biotransformation products.

tle pheromone (+ -em-brevicomin (259)(SCHULTZt a l., 1990).

5.4 Nucleotide Chemistry

5.4.1 Phosphorylation

Many structurally specific phosphate hydro-lyzing enzymes are known, and have been

widely used in organic synthesis. Of pa rticularsynthetic importance is their use in selectivephosphate bond formation without the needfor pro tecting groups, for example, in the selec-tive mono- and pyrophosphorylations ofmonosaccharide moieties (SABINA t al., 1984;LADNERnd W HITESIDES,985).

5.4.2 Gene Synthesis

Enzyme mediated phosphorylation is ofgreat value in the nucleic acid field. In polynu-

cleotide synthesis enzyme catalyzed phos-phate bond formations provide solutions tothe problems of con trolled couplings of oligo-nucleotide intermediates. Gene synthesisrelies heavily on this technique, as seen in thestructural genes for yeast analine tRNA and E.

coli tyrosine suppressor tRNA (KHORANA,1976).

5.4.3 Base Exchange

Enzyme catalyzed base exchange has prov-ed t o be of value in other areas of nucleotidechemistry, such as in the preparation of pos-sible antiviral agents (MO RISAWAt al., 1980).The antiviral or antitumor activity of 9 - p - ~ -arabinofuranosylpurines has generated inter-est in such nucleotides, and although they a reobtain able by a sugar-base coupling reactionthere remain definite practical limitations,such as low yields through a laborious andtime consuming process.



390 I0 Synthetic Applications of Enzyme-Catalyzed Reactions

244 \

Fig. 73. Synthesis of natural a-tocopherol.

6 AromaticBiotransformation

Products

6.1 HydrolysidCondensation

Reactions


a-Chymotrypsin is a versatile enzyme ca-pable of showing enantiomeric specificityon abroad range of racemic ester substrates (Fig.76). The use of chymotrypsin as an enzymiccatalyst is advantageous in that its stereospeci-ficity is predictable using a simple model. Thereactive ester stereoisomers are those whosechiralities parallel those of the natural L-aminoacids, and the unreactive enantiomers can allbe recycled via chemical racemization.


6.2.1 Hydroxylation

Selective hydroxylation of aromatic com-pounds is a difficult task in preparative organ-ic chemistry, particularly when the compoundsto be hydroxylated (or their products) are op-

tically active. In such cases the reaction shouldbe conducted rapidly and mildly to prevent ra-cemization and decomposition. KLIBANOV

showed that under certain conditions horse-radish peroxidase (HRPO) can be used forfast, convenient, and selective hydroxylationsin yields of up to 75%. L-DOPA(263)was pro-duced from L-tyrosine (262), and L-epineph-rine (adrenaline) (265) from L-( -)-phenyl-ephrine (264) using HRPO catalyzed hydroxy-lation (Fig. 77) (KLIBANOV t al., 1981).

The preparation of 2-arylpropionic acids byhydroxylation and oxidation of cumenes withCordyceps sp. (SUGIYAMAnd TRAGER,986)is possible, where the cumene is hydroxylated

not on the benzylic carbon as expected, but se-



6 Aromatic Biotransformation Products 391

H O O P

249 2 48

RAMA 37%yieldIM e 0 2Awop

OH 250

1 NaHB(OAc)32 GN HC1

3 TransaminationIHH C 0 2 -

O H

25 1 3 - W x y - D - a r a b i n eheptulosonic acid

Fig. 74. Synthesisof 3-deoxy-~-arabino-heptuloso-nic acid using R A M A .

lectively at the pro-(R) methyl group. For ex-ample, the naphthalene derivative (266) istransformed by Cunningh amella milituris into(S)-naproxen (267) n 98% ee (Fig. 78) (PHIL-LIPS et al., 1986).

The hydroxylation of alkaloids often modi-fies their biological activity and makes avail-

able compounds that are otherwise difficult tosynthesize. For example, acronycine (268), anantitumor alkaloid, is hydroxylated in the 9-position in 30% yield by Cunninghamella ech-inulutu (Fig.78 ) (BETTS t al., 1974).

6.2.2 Oxidative Degradation

Hydroxylation of a substrate by an organismis often a prelude to its utilization as an energysource, which may involve complete conver-sion to CO,. Processes which effect substantial

JH52 HO OP

2 4 8

254 255

H O OP

248

259 (+)-exeBrevicomin

Fig. 75. ?he use of R A M A in carbocycle and

pheromone synthesis.




Chymotrypsin

RC02Me4 RC02Me RCOzH

H+ racemization

rac-260 D-260 L-261

Fig. 76. Aromatic chiral acids and esters.

HRPO, 0 2 , WC ,

Dihydroxy-fumaric acid

OH

262 R = H; L-Tyrosine263 R = OH: L-DopaI

75% yield

R

264 R = H; L-Phenylephrine265 R = OH: L-Epinephrine2

50%yield

Fig. 77. Enzymic hydroxylation in drug synthesis.

chemical modification but which do not lead to

complete degradation are particularly valuable.

Treatment of ruc-(270)with a Rhodococcus sp.,

results in complete degradation of one enan-

M e 009Cunninghamella militaris

266 R = C H ,

267 R = C 0 2 H ; (S)-Naproxen

0 OMe

Cunninghamella echinulata

268 R = H; Acronycine

269 R=OHJ

Fig. 78. Hydroxylation of cumenes and prepara-tion of an antitumor alkaloid.

5

O n H e p

EIC-20

I17% Yield

hodococcus sp

BPM 1 6 1 3

)&02H \

(4 -271 Ibuprofen

Fig. 79. Ibuprofen from stereoselective degrada-

tion.

tiomer, however oxidation of the other enan-

tiomer terminates at the carboxylic acid stage to

give the anti-inflammatory drug ibuprofen ( R ) -

(271) (Fig. 79) (SUGAI nd Mom, 1984).



7 References 393

Bakers' yeast MeNH2

ph a( Tnprnh? H, / W * Ph&

d N - M el-face co2

27 2 CH3A 0 2 H 273 274 (-)-y-Ephedrine

++--KgHO Bakers' yeast

Ph

a OH

2 75 2 7 6 2 7 7 Frontalin

Fig. 80. The synthesis of ( -)-ephedrine and frontalin.


Formation Reactions

6.3.1 Acyloin Condensation

The formation of phenyl acetyl carbinol(273) (Fig.80) from benzaldehyde (272) by fer-menting yeast was first observed in 1921(CROUT t al., 1991).

The enzyme system involves pyruvic acid,decarbo xylation of which provides the C,unit(an acyl anion eq uivalent) which is transferredto the Si-face of th e aldehy de to form a n ( 3 R) -a-hydroxy ketone (acyloin) (273).

There is remarkable tolerance by the en-zyme system with respect to changes in thestructu re of th e aldehyde . a$-unsaturated ali-

phatic and aromatic aldehydes undergo alky-lation and reduction of the acyloin to give(2S,3R)-diols (276) (FRONZAt al., 1982).Theyields of chiral diols are poor (10-35%), butthis is offset by the ease of reaction and thecheapness of starting reagents used. SERVIndFUGANTIFUGANTInd SERVI,988;FUGANTI,1986; FUGANTInd GRASSELLI,985b) haveexploited these com pounds for numerous syn-theses of chiral natural products. A n interest-ing example is provided by the synthesis of(-)-frontalin (277) which is a pheromone of

several bark beetles (Fig. 80). The two chiral

centers created by the yeast reduction a re usedto induce stereoselective addition of a Grig-nard reagent and subsequently both are des-troyed by a pe riodate cleavage. Thus onlythree carbons and no chiral centers from thebiotransforma tion product a re carried through

to (-)-frontalin (277) (Fig. 80) (FUGANTItal., 1983).

7 References

ARITA,M., ADACHI,., ITO, Y.,SAWAI, . , OHNO,M.(1983), Enantioselective synthesis of the carbocy-clic nucleoside (-)-aristerom ycin and ( -)-

neplanocin A by a chemicoenzymatic approach,J.Am . Chem. SOC.105,4049-4055.

ARITA,M.,ADACHI, . , I T O , ~ . ,AWA I, .,OHNO, .,SHIBATA,. (1984), Synthetic studies on biologi-cally active natural products by a chem icoenzy-matic approach, Tetrahedron40,145-152.

BALDWIN,.E.,CHRISTIE,. A .,HABER,.B., KRUSE,L. I. (1976),Stereospecificsynthesis of penicillins.Conversion from a peptide precursor, J. A m .Chem.SOC.8,3045-3047.

BALLARD,D. G., COURT IS, ., SHIRLEY,. M.,TAYLOR,. C. (1983), A biotech route to poly-phenylene, J. Chem. Soc., Chem. Commun.,954-955.

BALLY, ., LEUTHARD T,. (1970), Die Stereospezi-fitat der Alkoholdehyd rogenase,Helv. Chim.Acta53,732-738.




BANKO, ., DEMAIN, . L., WOLFE,S. (1987), AC Vsynthetase: A multifunctional enzyme with abroad substrate specificity for the synthesis ofpenicillin and cephalosporin precursors, J . A m .Chem. SOC.

109,2858-2860.BARR ETT, . G . M., CA PPS,N. K. (1986), Syn theticapproaches to the avermectins : Studies on thehexahydrobenzofuran unit, Tetrahedron Le tt. 27,

BECKER, ., PFEIL,E. (1966), Continuous synthesisof optically active a-hydroxynitrilesJ Am. Chem.SOC.88,42994300.

BEDNARSKI,. D., SIMON ,E. ., BISCH OFB ERG ER,N .,FES SN ER, .-F., K IM, M.-K. e t al. (1989), Ra bbitmuscle aldolase as a catalyst in organic synthesis,J . Am . Chem. SOC.111,627,635.

BE HR EN S, . B., SHARPLESS, . B. (1985), Selective

transform ations of 2,3-epoxy alcoho ls and re latedderivatives - strategies for n ucleophilic attack atcarbon-3 or carbon-2, J . Org. Chem. 50, 5687-5696.

BEND ER, . L., KEDZY,. J., GU NTE R, . R. (1964),The anatomy of an enzymatic catalysis. a-Chymo -trypsin, J.Am . Chem. SOC.86,3714-3721.

BEITS, R. E ., WALTERS, . E. , RO SAZ ZA,. P. 1974),Microbial transformations of antitumor com-pounds. 1. Conversion of acronycins to 9-hy-droxy-acronycine by Cunningham ella echinulata,J . Med. Chem .17,599-602.

BILLHARDT,. M., STEIN,P., WHITESIDES,. M.

(1989), Enzym atic methods for the preparation ofacetyl CoA and analogs,Bioorg. Chem . 17,l-12.BINDRA,. S., BIND RA, . (1977),Prostaglandin Syn-

thesis. New Yo rk Academic Press.BOLAND, ., FRO BL, ., LOR ENZ , . (1991), Estero-

lytic and lipolytic enzymes in organic synthesis,Synthesis,1049-1072.

BONARA,., CAR REA , ., PASTA,., RIVA,S. (1986),Preparative scale regio- and stereospecific oxi-doreduction of cholic acid and dehydrocholicacid catalyzed by hydroxysteroid dehydrogen-ases,J. Org. Chem .51,2902-2906.

BRANCA,Q., FISCHLI, . (1977), Eine chiral-oko-

nomische Totalsynthese von ( R ) - und (S)-Mus-kon via Epoxysulfoncyclofragmentierung,Helv.Chim. Acta. 60,925-944.

BRIENA, ., CRICH,. Z., SIH,C. J. (1993), Chem oen-zym atic synthesis of the C-13 side chain of taxolOptically active 3-hydroxy-4-phenyl p-lactamderivatives, . Org. Chem. 58,1068-1075.

BROOKS,D. W., GROTHAUS,. G., PALMER,. T.(1982), Synthetic studies of trichothecenes, anenantiose lective synthesis of a C-ring, precurso rof anguidine, Tetrahedron Lett. 23,41874190.

BROOKS, . W., GROTHAUS,. G., MAZDIYASNI,.(1983),Total synthesis of th e trichothecene myco-

toxin anguidine, J. Am. Chem. SOC.105, 4472-4473.

5571-5574.

BROO KS, . W., M AZDIYA SNI,., CHAKRABARTI,.(1984), Chiral cyclohexanoid synthetic precursorsvia asymmetric microbial reduction of prochiralcyclohexanediones, Tetrahedron Lett. 25, 1241-

1244.BROOKS, . W.,WILSON, ., WEBB ,M. (1987), Differ-ent enzymatic reactions of an enantiomeric pair:Simultaneous dual kinetic resolution of a ketoester by bakers’ yeast, J. Org. Chem. 52, 2244-2248.

CARREA, ., BONARA , ., CREMO NESI,., LODI,R.(1984), Enzym atic preparation of 12-ketochenod-eoxycholic acid with NADP regeneration, Bio-technol. Bioeng . 26,560-563.

CHAN, ., COX,P. B., ROBERTS,.M. (1988), Conver-gent stereocontrolled synthesis of 13-hydroxy-9Z,llE-octadecadienoic acid (13-HODE), J.

Chem. Soc., Chem. Commun.,971-972.CHEN,C.-S., FU JIMOT O,., SIH, C. J. (1981), Bifunc-tional chiral synthons via microbiological meth-ods. 1.Optically active 2,4-dimethylglutaric acidmonom ethyl esters, J . Am . Chem. SOC.103,358&3582.

CHIBATA,., TOSA, ., SATO, ., MORI,T. (1976), Pro-duction of L-amino acids by aminoacylase adsor-bed on D EAE-Sephadex, Methods Enzym ol. 44,

746-759.CHOI,W. G., LEE,S. B., R w , D. D. Y. (1981), Cepha-

lexin synthesis by partially purified and immobi-lized enzymes,Biotechnol. Bioeng. 23,361-371.

COHEN,N.,EICHEL,W. F., LO PRESTI, . J., NEUKOM,C., SAUCY, . (1976), Synthetic studies on ( 2 R ,4’S, 8’R)-a-tocopherol. An approach utilizingside-chain synthons of m icrobiolog ical origin,J . Org. Chem. 41,3505-3511.

COLLUM, . B., MCDONALD11,J. H., STILL,W. C.(1980a), The polyether antibiotic m onensin. 1.Strategy and degradations,J . Am . Chem.SOC.102,

COLLU M, . B., MCD ONA LD11, J. H., STILL,W. C.(1980b), The polyether antibiotic monensin. 2.Preparation of intermediates, J . Am . Chem. SOC.

102,2118-2120.

COREY,E. J., TRYBULSKI,. J., MELVIN,L. S.,NICOLAOU,. C., SECRIST,. et al. (1978a), Totalsynthes is of ery throm ycins. 3. Stereoselectiveroutes t o intermediates corresponding to C( 1) toC(9) and C(10) to C(13) fragments of erythrono-lide B, J . Am . Chem. SOC.100,4618-4620.

COREY,E . J., KIM, S., Yoo, S., NICOLAOU,. C.,MELVIN,L. S. et al. (1978b), Total synthesis of

erythromycins.4. Total syn thesis of erythronolideB, J . Am . Chem. SOC.100,4620-4622.

COREY, . J.,ALBR IGHT,. O.,BARTON,A.., HASHI-MOTO, S. (1980), Chem ical and enzymic synthesesof 5-HPETE, a key biological precursor of slow-

reacting subs tance of anaphylaxis (SRS ), and 5-H E T E , . Am . Chem. SOC.102,1435-1436.

2117-2118.



7 References 395

FRATER, . (1979a), Uber die Stereospezifitat dera-Alkylierung von P-Hydroxycarbonsaureestern,Helv. Chim. Acta. 62,2825-2828.

FRATER,G. (1979b), Stereospezifische Synthese von

CROUT,D. H. G., DALTON, . , HUTCHINSON,. W., (+)-(3R,4R)-4-methyl-3-Heptanol. as Enan-tiomer eines Pheromons des kleinen Ulmen-splintkafers,Helv. Chim .Acta. 62,2829-2832.

FRATER, . (1980), Uber die Stereoselektivitat dera-Alkylierung von (1R,2S)-(+ -cis-2-hydroxycy-clohexancarbonsaureethylester, elv. Chim .Acta.63,1383-1390.

MULJ-ER, u.7 GUNTHER,w. (1984), Thestereoselective a-alkylation of chiral P-hydroxyesters and applications thereof, Tetrahedron 40,1269-1277.

CORE Y, . J., LANSBU RY,.T. (1983), Stereoch emica lcourse of 5-lipoxygenation of arachidonate by ratbasophil leukemic cell (RBL-1) and potato en-zymes,J . Am . Chem.SOC.105,40934094.

MIYAGOSHI,. (1991), Studies on pyruvate de-carboxylase acyloin formation from aliphatic,aromatic and heterocyclic aldehydes, J . Chem.

Soc., Perkin. Trans.I,1329-1334.DAI,W. M., ZHOU, w.s. (1985), New synthesis of two

optically active steroid C D ring synthons bymicrobial asym metric reduction, Tetrahedron 41,44754482.

DAVIES, . G., DAWSON,M. J., LAWRENCE,. C.,MAYALL,., NOBLE,D. et al. (1986), Microbialhydroxylation of cyclohexylcyclohexane;Synthe- G . ~UGANT1,., L.,

Lett .27,1089-1092.DAVIES, . G., GRE EN, . H., KELLY,D. R., ROBERTS,

S. M. (1989), Biotransformations in preparativeorganic chemistry. London: A cademic Press.

eospecific enzyme-catalyzed reductions of cis-

and trans-decalindiones to enantiomerically purehydroxy-ketones; an efficient access to (+)-4-twistanone,J. Chem. Soc. , Chem. Comm un.,1080-1081.

DODDS, . R., JONES,. B. (1988), Enzymes in o rgan-

FRATER,

sis of an analog of leukotriene B J, Tetrahedron FANTOW G.9 SPREAFICO,F- (l982), Synthesis Of

enantiom erically pur e forms of N-acyl derivativesOf C-methyl Of the aminodeoxy War

L-acosamine from non-carbohydrate precursors,J . Org. Chem.47,3289-3296.

ation of carbohydrate-like chiral synthons, n: En-zymes as Catalysts in Organic Synthesis (SCHNEI-DER, M. P., E d.) NA TO A S1 Series, Vol. 178. D or -drecht: Reidel-Gruyter.

FUGA NTI, ., GRA SSELLI,. (1985a), Enzymes inOrganic synthesis, in: CIBA Foundation S ympo-sium I I I , pp. 112-127. Lo ndon : Pitman .

DODDS, ,R., JONES,. B. (1982), Selective and s ter- FUGANT1, c' (1986), Bakers' yeast-mediated prepar-

ic synthesis' 38' Preparations Of enantiomericallyFUGANTI,., GRAS SELLI,. (1985b), Stereoch emistry

pure chiral hydroxydecalones via stereospecific

and synthetic applications of products of fer ne n-horse liver dehydrogenase-catalyzed tation of +unsaturated aromatic aldehydes byreductions of decalindiones, J . Am . Chem. SOC. bakers' yeast, in: Enzymes as catalysts in Organic110,577-583. Synthesis (CLARK,., PORTE R, ., Eds.). London:

DRAUZ, ., WALDMANN,. (1995), Enzyme Cataly- Pitman.sis in Organic Synthesis, A Comprehen sive Hand- mGANTI, c., SERw, s. (1988), Bioflavour '87book.Weinheim: VC H. (SCHEIER,., Ed.). Berlin: de Gruy ter.

v.w.,WoNG* C.-H. FUGANTI,., GRASSELLI,., SER VI, . (1983), Synthe-(1985), FMN reductase catalyzed regeneration of sis of (- -frontalin from the (2S,3R)-diol pre-NAD(P) for use in enzymatic synthesis, J . 0% pared from a-methylcinnamaldehyde and fer-Chem. 50,5387-5389. menting bakers' yeast, J . Chem. SOC., Perkin

EFFENBERGER,., STRAUB, . (1987), A novel, con- T ~ ~ ~ ~ . I ,41-244.venien t Preparation of dihYdroxYacetone Phos- FULLING, ., SIH,C. J. (1987), Enzymatic second order

phate and its use in Enzym atic aldol reactions, asymmetric hydrolysis of ketorolac esters: in situTetrahedron Le tt. 28,1641-1644. racemization,J.Am . Chem.SOC.109,2845-2846.EV AN S, .A ., SACKS,c.E., mSCHICK,W.A.,TABER, GA IS,H.-J., LUKAS, . L. (1984), Enantioselective

T. R. (1979), Polyether antibiotics synthes is and and enantioconve rgent syn theses of buildingabsolute configuration of the ionoph ore A-23187, blocks fo r the total synthesis of cyclopentanoidJ. Am . Chem. SOC.101,67894791. natural products, Angew. Chem . (Int. Edn. Engl.)

EVAN S, . A., NELSON,. V.,TA BER ,T. . (1982), Ster- 23,142-143.eoselective aldol condensation, Top. Stereochem. GAIS,H.-J., LIED,T., LUKAS, . L. (1984), Asym met-13,l-115. ric synthesis of a nov el enantiomerically pure

FABER,K. (1995), Biotransformations in Organic prostaglandin building block, Angew. Chem. (Int.Chemistry,2nd Ed n. Berlin, New York: Springer- Edn . Engl.) 23,511-512.Verlag. GIBSON , . T., HEN SLEY , ., YO SHIO KA ,., MABRY ,

FISCHLI,. (1980),Modern S ynthetic Metho ds, Vol. 2 T. J. (1970), Form ation of (+ -cis-2,3-dihydroxy-

(SCHEFFOLD,., Ed.). Frankfurt: Salle-Sauerlan- l-methylcyclohexa-4,6-diene rom toluene byder. Pseudomon as putida, Biochemistry 9,1626-1630.

DRUECKHAMMER,. G.,



396 I0 Synthetic Applications ofEnzyme-Catalyzed Reactions

GLASS,. D. (1981),Enzymes as reagents in the syn-thesis of peptides, Enzyme Microb. Technol. 3,

GOODHUE,. T., SCHAEFFE R,. R. (1971), Prepara-tion of (L)-( +)-/3-hydroxyisobutyric acid by bac-terial oxidation of isobutyric acid, Biotechnol.Bioeng. 13,203-214.

GREENSTEIN,. P. 1954), The resolution of racemica-hydroxy amino acids, Adv. Protein Chem. 9,121-202.

GREENZAID,., JENCKS,W. P. 1971), Pig liver ester-ase reactions with alcohols, structure-reactivitycorrelations and the acyl-enzyme intermediate,Biochemistry 10,121&1222.

GRUNDY,. M. (1988),HMG-CoA reductase inhibi-tors for treatment of hypercholesterolemia, N.Eng.J. Med. 319 ,2433 .

HAMADA,M., TAKEUCHI,., KONDO, ., IKEDA,Y.,NAGANAWA,. et al. (1970),A new antibiotic, ne-gamycin,J.Antibiot. 23,17&171.

HANESSIAN,. (1983), Total synthesis of naturalproducts: T he ‘chiron’ approac h. Oxford: Per-gam on Press.

HEATHCOCK,. H. (1984), The aldol addition reac-tion, Asym m. Synth. 3,111-212.

HEMMERLE,., GAIS,H.-J. (1987),Asymm etric hy-drolysis an d esterification catalyzed by esterasesfrom porcine pancreas in the synthesis of bothenantiomers of cyclopentanoid building blocks,Tetrahedron Lett. 28,3471-3474.

HIRAMA, ., UEI,M. (1982),Ch iral total synthesis ofcompactin,J. Am . Chem.SOC.104,42514253.HIRAMA,., GARVEY,. S., Lu, L. D. L., MASAMUNE,

S. (1979), Use of the E-vinyloxyborane derivedfrom S-phenyl propanethioate for stkreospecificaldol-type condensation. A simplified synthesis of

the Prelog-Djerassi lactonic acid, TetrahedronLett. 41,3937-3940.

HIRAMA, ., SHIMIZU,., IWASHITA,. (1983), En-antiospecific syntheses of t ri funct ional ( l 9- 3- h~ -droxy es ters by bakers’ yeast reduction, J. Chem.SOC.,Chem. Commun., 599400 .

HUDLICKY,.,NATCHUS,. (1992),Chemoenzymat-

ic enantiocontrolled synthesis of (- -specionin,J. Org. Chem .57,4740-4746.HUDLICKY,., PRICE ,. D. (1990), Microbial oxida-

tion of chloroaromatics in the enantioselectivesynthesis of carbohydrates - L-ribonic y-lactone,Synlett, 159-160.

HUDLICKY,., LUNA,H. , BARBIERI,., KWART,L. D.

(1988),Enantioselective synthesis through micro-bial oxidation of arenes. Efficient preparation of

terpene and prostanoid synthons, J. Am . Chem.SOC.110,4735-4741.

HUDLICKY,., LUNA,H., PRICE, . D., RULIN,F.(1989), An enantiodivergent approach to D- and

L-erythrose via microbial oxidation of chlorobe n-zene, Tetrahedron L ett.30,40534054.

1-8.

HUDLICKY,., PRICE,. D., RULIN, ., TSUNODA ,.(199O), Efficient and enantiodivergent synthesisof (+)- and (-)-pinitol, J. Am . Chem. SOC.112,

HUDLICKY,.,LUNA,H., OLWO,H. F.,ANDERSEN,.,NUGENT,T.,RICE,. D. (1991),Biocatalysis as th estrategy of choice in the exhaustive enantiom eri-cally controlled synthesis of conduritols, J. Chem.SOC.,Perkin Trans.I , 2907-2917.

HUTCHINSON,. R., IKEDA,. (1984),A general, en-antiospecific synthesis of cyclopentanoid mono-terpenes (iridoids), The total synthesis of ( -)-1-0-methylsweroside aglucone, J . Org. Chem. 49 ,

IRIUCHIJIMA, ., HASEGAWA,., TSUCHIHASHI,.(1978), Structure of a new phenolic glycoside,chesnatin, from ch estnut galls,Agric. Biol. Chem .

IRWIN,A.., JONES, . B. (1977),Asymm etric synthesisvia enantiotopically selective horse liver alcoholdehydrogenase-catalyzed oxidations of diols con-taining a prochiral center, J . A m . Chem. SOC.99,556-561.

ISOWA,Y., OHMORI, ., SATA,M., MORI,K. (1977a),Th e enzymatic synthesis of prot ected valine-5-an-giotensin I1 amide-1, Bull. Chem. SOC. Jpn. 50,2766-2772.

ISOWA, ., OHMORI, ., ICHIKAWA,., KURITA,H.,SATO,M., MORI,K. (1977b), the synthesis of pep-tides by means of proteolytic enzymes, Bull.

Chem. SOC. pn . 50,2762-2765.JAKOVIC,~.. ,GOODBRAND,H..,LoK,P.K., ONES,J.

B. (1982),Enzym es in organic synthesis. 24. Prep-arations of enantiomerically pure chiral lactonesvia stereospecific horse liver alcohol dehydro-agenase-catalyzed oxidations of monocyclicmeso-diols, J.Am . Chem. SOC.104,46594665.

JOHNSON, R.A. (1978),Oxidation in organic chemis-try, Part C (TRAHANOVSKY,. S. , Ed.). New York:Academic Press.

JOHNSON,M. R., KISHI, . (1979), Cooperative effectby a-hydroxy and ether oxygen in peroxidationwith a peracid, Tetrahedron Let t. 45,43474350.

JOHNSON,M. R .,NAKATA,T., I S H I ,~ .1979), Stereo-and regioselective meth ods fo r the synthesis of

three consecutive asymmetric units in manynatural products, Tetrahedron Lett. 45, 4343-4346.

JONES, J. B. (1985), Enzym es in organic synthesis, in:CIBA Foundation Symposium III (CLARK, . ,

PORTER,., Eds.), pp. 3-21. London: Pitman.JONES, J. B. (1986), Enzymes in organic synthesis,

Tetrahedron42,3351-3403.JONES, J. B. (1993),Probing th e specificity of syn thet-

ically useful enzymes, Aldrichimica Acta 26,

105-1 12.

JONES, J. B., LIN,Y. Y. (1972), Evaluation of some ofthe factors involved in the selective hydrolysis of

9439-9440.

2837-2838.

46,1907-1910.



7 References 397

aromatic ester protecting groups by a-chymo-trypsin, Can.J. Chem.50,2053-2058.

JONES,. B., FINCH,M.A. W., JAKOVIC,.J. (1982), En-zymes in organic synthesis.26. Synthesis of enan-

tiomerically pure grandisol from an enzyme-gen-erated chiral synthon, Can. J. Chem. 60,

2007-201 1.KAN,K., MIYAMA, ., HAMA GUCHI,., OHA SHI, .,

WATANABE,. (1985a), Synthesis of (S)-@block-er from (S)-5-hydroxymethyl-3-tert-butyl-2-oxaz-

olidinone or (S)-5-hydroxymethyl-3-isopropyl-2-oxazolidinone,Agric. Biol.Chem.49,207- 210.

KAN,K., MIYAMA, ., HAMAGUCHI,., OHA SHI, .,WATANABE,. (1985b), Stereochemical inversionof (R)-5-hydroxymethyl-3-tert-butyl-2-oxazolidi-

none or (R)-5-hydroxymethyl-3-isopropyl-2-ox-

azolidinone to the corresponding (S)-isomer,A g-

ric. Biol. C hem .49,1669-1674.KARADY, ., AMATO, . S., REAMER,R. A., WEIN-STOCK,. M. (1981),Stereospecific conversion ofpenicillin to thienamicin, J. Am . Chem. SOC.103,6765-6767.

KASEL,W., HULTIN,. G., JONES,. B. (1985), Prepar-ations of chiral hydroxy ester synthons via stereo-selective porcine pancreatic lipase-catalyzed hy-drolyses of meso-diesters,J. Chem. SOC., hem.Commun.,1563-1564.

KATO,K., KAWAH ARA,.,TAKAHASHI,T.,GARASI,.(1980), Enzymatic synthesis of amoxicillin by thecell bound a-amino acid ester hydrolase of Xan-

thomonas citri, Agric. Biol. Chem. 44,821-825.KATSUK I, ., MA RTIN, . S. (1996). Catalytic asym-metric epoxidation of allylic alcohols, Org. React.48,l-130.

KATSUKI,., SHARPLESS, . B. (1980), The first prac-tical method for asymmetric epoxidation, J. A m .Chem. SOC.102,5974-5976.

KHORANA,. G. (1976),Total synthesis of the struc-tural gene for the precursor of a tyrosine suppres-sor transfer RNA from Escherichia coli, J. Biol.Chem.251,565-586.

KIESLICH, . (1976), Microbial transformations ofnon-steroid cyclic compounds. Stuttgart: Thieme.

KITAZUME,., ISHIKOWA,. (1984), Introduction ofcenter of chirality into fluorocompounds by mi-crobial transformation of 2,2,2-trifluoroethanol,Chem. Lett., 1815-1818.

KLIBANOV,A..,BERMAN,.,ALBERTI, . N. (1981),Preparative hydroxylation of aromatic com-pounds catalyzed by peroxidase, J. Am. Chem.SOC.103,6263-6264.

KOBAYASHI, ., KOYAMA,T., OGURA,K., SETO,S.,RITER, E J., BRUGGERMANN-ROTGANS,. E. M.

(1980), Bioorganic synthesis and absolute config-uration of faranal, J. Am. Chem. SOC. 102,6602-6604.

KOBAYASHI,., IIMORI,T., IZAWA,T., Omo, M.

(1981), Ph3P-(PyS),-CH3CN as an excellentcondensing system for p-lactam formationfrom p-amino acids, J. A m . Chem. SOC. 103,2406-2408.

KONDO,S., SHIBAHARA,., TAKAHASHI,., MAEDA,K., UMEZAWA, .,OHNO,M. (1 971), Negamycin, anovel hydrazide antibiotic,J.Am . Chem. SOC.93,6305-6306.

KOSM OL, ., KIESLICH, ., VOSS ING,., KOCH,H.-J.,PETZO LD, ., GIBIAN ,H. (1967), Mikrobiologi-sche stereospezifische Reduktion von 3-Meth-oxy-8.14-seco-l.3.5(10).9-ostratraen-14.17-dion,

Liebigs Ann. C hem. 701,198-205.KOYAMA, ., SAITO,A., OGU RA , ., SETO,S. (1980),

Substrate specificity to farnesyl pyrophosphatesynthetase, application to asymmetric synthesis, .Am . Chem.SOC.102,36163618.

KOYAMA,.,

OG UR A, . , BAKER,EC.,

JAMIESON,.

C., SCHOOLERY,. A. (1987), Synthesis and abso-lute configuration of 4-methyl juvenile hormone1 4-MeJH 1) by a biogenetic approach: A combi-nation of enzymatic synthesis and biotransforma-tion,J.Am . Chem.SOC.109,2853-2854.

KRAIDMAN,.,MUKHERTEE,. B., HILL, . D. (1969),Cyclic vicinal glycols as microbial transformationproducts of some monocyclic terpene hydrocar-bons,Bacteriol. Proc.63.

KRASNOBAJEW,. (1984), Microbial transformationsof terpenoids, in: Biotechnology 1st Edn., Vol. 6a.Biotrunsformations (REHM,H.-J., REED,G.,Eds.).

Weinheim: Verlag Chernie.KULLMAN,W. (1981), Protease-mediated peptidebond formation,J. Biol. Chem.256,1301-1304.

KULLMAN,. (1982), Enzymatic synthesis of dynor-phin (la),. Org. Chem.47,5300-5303.

KUPCHAN,. M., KOMODO, ., BRANFMAN,. R.,SNEDEN, . T., COURT,W. A . et al. (1977), Themaytansinoids. Isolation, structural elucidationand chemical interrelation of novel ansa macro-lides,J. Org. Chem.42,2349-2357.

KUROZUMI,., TORA,T., OSHIMOTO,. (1973), Prep-aration of 2-(4-hydroxy-l-oxocyclopent-2-ene)-

heptanoic acid, an important prostaglandin syn-

thon, Tetrahedron Lett. 49,49594960.LADNER,W. E., WHITESIDES, . M. (1984), Lipase-catalyzed hydrolysis as a route to esters of chiralepoxy alcohols, .Am . Chem.SOC.106,7250-7251.

LAD NER , . E., WHITESIDES,. M. (1985), Enzymat-ic synthesis of deoxy ATP using R N A as startingmateria1,J. Org. Chem.50,1076-1079.

LAGERLOF, ., NATHORST-WESTFELOT,., EKSTROM ,B., SJOBERG,. (1976), Production of 6-aminope-nicillanic acid with immobilizedEscherichia coli,Methods Enzymol. 44,759-768.

LAUMEN , ., SCHNEIDER, . (1985), Enantioselec-tive hydrolysis of cis-1,2-diacetoxycycloalkane i-

methanols: Enzymatic preparations of chiralbuilding blocks from prochiral meso substrates,



398 10 SyntheticApplications of Enzym e-Catalyzed Reactions

Tetrahedron Lett. 26,2073-2076.LEUENBERGER, . G., BOGUTH,W., BARNER,R.,

SCHMID,M., ZELL,R. (1979), Large-scale reduc-tion of oxoisoph orone in the syn thesis of caro ten-

oids, Helv. Chem. Acta 6 2,4 55 46 5.LEY,S. V., STERNFELD,. (1989), Microbial oxidationin synthesis: Preparation of (+)- and (-)-pinit01from benzene, Tetrahedron 45,3463-3476.

LIN, C. H., ALEXANDER,. L., CHICHESTER,. G.,GORMAN, . R., JOHNSON, . A. (1982), 10-nor-9,ll-secoprostaglandins. ynthesis, structure andbiology of endoperoxide analogs, J . Am. Chem.

LIPMANN,. (1973), Nonribosomal polypeptide syn-thesis on polyenzyme templates, Acc. Chem. Res.6,361-367.

MACLEOD,., PROSSER,.,FIKENTSCHER,L.,LANYI,

J., MOSHER, . S. (1964), Asymm etric reductions.XII. Stereoselective ketone reductions by fer-menting yeast, Biochemistry 3,838-846.

MARPLES, .A., ROGER-EVANS,. (1989), Enantiose -lective lipase-catalyzed hydrolysis of esters of e poxy secondary alcohols: A n alternative to Sharp-less oxida tion , Tetrahedron Lett. 30,261- 264.

MEDINA,. C., KYLER,K. S. (1988), Enzym atic cy-clization of hydroxylated surrogate squalenoidswith bakers’ yeast, J . Am. Chem. SOC.110,4818-4821.

MEDINA,. C., GUAJARDO, ., KYLER,K. S. (1989),Vinyl group rearrangement in the enzymatic cy-

clization of squ alenoids: Synthesis of 30-oxyster-ols,J.Am. Chem. SOC.111,2310-2311.MEYERS, . I., AMOS, . A. (1980), Studies directed

toward the total synthesis of streptogramin anti-biotics. Enantiospecific approach to the nine-membered macrocycle of grisoviridin, J . Am.Chem. SOC.02,870472.

MEYERS, . I.,HUDSPETH,. P. (1981), Enantioselec-tive synthesis of C,-C,, fragment (north eastzone) of maytanisoids with four chiral centers(4S,5S,6R,7S), Tetrahedron Lett. 22,3925-3928.

MOHR,P.,WAESPE-SARCEVIC,., TAMM, . , CAW-RONSKA, ., GAWRONSKI,. K. (1983), A Study of

stereoselective hydrolysis of sym metrical diesterswith pig liver esterase, Helv. Chim. Acta. 66,

MORI,K. (1989), Synthesis of optically active phero-mones, Tetrahedron 45,3233-3298.

MORI,K.,EBATA,T.1986a),Synthesis of (2S ,3R,l’R)-stegobinone, the pherom one of the drugstorebeetle with stereocontrol at C-2 and C-2’, Tetra-hedron 42,4413-4420.

MORI,K., EBATA, . (1986b), Syn thesis of all of thefour possible stereoisomers of 5-hydroxy-4-meth-yl-3-heptanone (sitophilure). The aggregationpheromone of the rice weevil and the maize wee-

vil, Tetrahedron 42,4421-4426.MORI,K., IWASAWA,. (1980), Preparation of both

SOC.04,1621-1628.

2501-251 1.

enantiomers of threo-2-amino-3-methyl-hexanoicacid by enzym atic resolution and their conversionto optically active forms of threo-4-methylhep-tan-3-01, a pheromone component of the smaller

Europ ean elm bark beetle, Tetrahedron 36,2209-2213.MORI,K.,TANIDA, . (1981), Syn thesis of three ster-

eoisomeric forms of 2,8-dimethyl-1.7-dioxaspi-

ro(5S)undecane, the main component of the ce-phalic secretion of Andrena wilkella, Heterocycles

MORISAWA,., UTAGAWA,., MUOSHI, ., YOSHINA-GA, F., YAMAZAKI,., MITSUGI, . (1980), A newmethod for the synthesis of some g-P -~-a rab ino -furanosylpurines by a combination of chemicaland enzymatic reactions, Tetrahedron Lett. 21,479482.

MOROE, ., HATTORI,S., KOMATSU, ., YAMAGUCHI,Y. (1970), Chem. Abstr. 73,33900d.MOSBACH, ., OHLSON,S., LARSSON,. D. (1978),

Steroid transfer by activated living immobilizedArthrobacter simplex cells,Biotechnol. Bioeng. 20,

NAEMURA,., KATOH,T., HIKAMATSU,., NAKAZA-KI, M. (1984), The preparation and chirotopicalproperties of cage-shaped pentacyclic hydrocar-bons , Chem. Lett., 1371-1374.

NAEMURA,., FuJII,T., CHIKAMATSU,. (1986), Se-lective and stereospecific horse liver alcohol de-hydrogenase-catalyzed reduction of cage-shaped

meso-diketones. An efficient access to opticallyactive D,-trishomocubane derivatives, Chem.Lett., 923-926.

NAKAZAKI, ., CHIKAMATSU,.,FuJII,T., SASAKI, .,Ao,Y. (1983), Stereoch emistry of ho rse liver alco-hol dehydrogenase-mediated oxidoreduction of

2-brendanone type cage-shaped tricyclic ketonesand the related stereoisomeric alcohols, J . Org.Chem. 48,43374345.

NEWTON,R. F., ROBERTS,. M. (1980), Steric con-trolled prostaglandin synthesis involving bicyclicand tricyclic intermediates, Tetrahedron 36,2163-2196.

NEWTON, . F., ROBERTS,. M. (1982), Prostaglan dinand thromboxanes: An introductory text. Lon-

don: Butterworths.O’CONNELL,. L., ROSE, . A. (1973), Affinity label-

ing of phosphoglucose isomerase by 1,2-anhydro-hexitol-6-phosphates, J . Biol. Chem. 248,

OKANO, ., IZAWA,., OHNO,M. (1983), A generalapproach of trans-carbapenem antibiotics. Enan-tioselective synthesis of key intermediates for(+)-PS-5, (+)-PS-6 and (+)-thienamycin, Tetra-hedron Lett. 24,217-220.

OYAMA,., NISHIMURA,.,NONAKA,~.,IHARA,~.,

HASHIMOTO,. (1981), Syn thesis of an aspa rtam eprecursor by immobilized thermolysin in an or-

15,1171-1179.

1267-1284.

2225-2231.



7 References 399

ganic solvent,J. Org. Chem .46,5241-5242.PATEL, . S., CONLON,. D., WALT,D. R. (1986), En-

zyme-catalyzed synthesis of L-acetylcarnitineandcitric acid using acetyl coenzyme A recycling,J .Org. Chem .51,2842-2844.

PHILLIPS,.T., MATCHAM,. W. J., BERTOLA, . A.,MARX,A. F., KOGER,H. S. (1986), Eur. PatentA pp l .EP 205,21517.

PORTER, . A., BYERS,.D.,HOLDEN, . M., MENZEL,D. B. (1979), Synthesis of prostaglandin H2,J.Am.Chem. SOC.101,43194322.

QUISTAD,. B., CERF,D.C., SCHOOLEY,. A., STAAL,G. B. (1981), Fluoromenalonate acts as an inhibi-tor of insect juvenile hormone biosynthesis, Na-ture 289,176-177.

REHM,H.-J., REED,G. (Eds.) (1984),Biotechnology1st. Edn., Vol. 6a Biotransformations (KIESLICH,

K., Ed.). Weinheim: Verlag Chemie.REES,L., VALENTE,., SUCKLING,. J., WOOD,H. C.S. (1986),Asymmetric reduction of dihydrofolateusing dihydrofolate reductases and chiral boron-containing compounds, Tetrahedron42,117-136.

ROBERTS,.M. (1985), Enzymes in organic synthesis,in: CIB A Foundation Symposium I I I , pp. 31-39.London: Pitman.

ROSAZZA,. P. (1982),Microbial Transformations ofBioactive Compounds, Vol. 1. Boca Raton, FL:CRC Press.

ROUDEN,., HUDLICKY,. (1993), Total synthesis of(+ )-kifunensine, a potent glycosidase inhibitor,

J.Chem.Soc., Perkin Trans.I,1095-1097.SABBIONI,., SHEA,M. L., JONES,. B. (1984), Prepara-tions of bicyclic chiral lactone synthons via stereo-specific pig liver esterase-catalyzed hydrolyses ofmeso-diesters Ring-sized induced reversalof sterec-specificityJ Chem.Soc., Chem. Comm un.,236-238.

SABINA,. L., HOLMES,. W., BECKER, . A. (1984),The enzymatic synthesis of 5-amino-4-imidazole-carbocamide riboside triphosphate (ZTP), Sci-ence 223,1193-1195.

SALTZMANN,. N., RATCLIFFE,., CHRISTENSEN,.G., BOUFFARD,. A. (1980), A stereocontrolledsynthesis of (+)-thienamycin, J. Am . Chem. SOC.

102,61614163.SAWADA,.,KULPRECHA,., NILUBOL,., YOSHIDA,S., KINOSHITA,.,TAGUCHI, . (1982), Microbialproduction of ursodeoxycholic acid from lithio-cholic acid by Fusarium equiseti M41, Appl . En-viron. Microbiol. 44,1249-1252.

SCHIENMANN,., ROBERTS,S. M. (1982),Recent Syn-thetic Routes to Prostaglandins and Thrombox-anes. London: Academic Press.

SCHINDLER,., SCHMID, . D. (1982), Fragrance oraroma chemicals-microbial synthesis and enzy-matic transformation - a review, Proc. Biochem.2, October 2-8.

SCHMID, ., WHITESIDES,. M. (1990), A new ap-proach to cyclitols based on rabbit muscle aldo-

lase (RAMA),J. Am . Chem. SOC.112,967&9671.SCHULTZ, ., WALDMANN,., VOGT,W., KUNZ,H.

(1990), Stereospecific C-C bond formation withrabbit muscle aldolase- chemicoenzymatic syn-thesis of (+ )-exo-brevicomin,Tetrahedron Lett.31,867-868.

SEEBACH,., HERRADON,. (1987), Diastereoselec-tive elaboration of the carbon skeleton of P-hy-droxy esters from yeast reductions, TetrahedronLett .28,3791-3794.

SEEBACH,., SUTTER, . H., WEBER,R. H., ZUGER,M. F. (1985) Ethyl (S)-3-hydroxybutanoate, Urg.Synth.63,l-7.

SERVI,. (1990), Bakers’ yeast as a reagent in organ-ic synthesis,Synthesis, 1-25.

SEURLING,., SEEBACH, . (1977), Synthese von vierchiralen, elektrophilen C,- und C,-Synthesebau-

steinen aus Hydroxycarbonsauren, Helv. Chim.Acta 60,1175-1181.SIH,C. J., SALOMAN,. G., PRICE,P., SOOD, ., PER-

LUZZOTTI, G. (1972), Total synthesis of prosta-glandins 111. 11-Deoxyprostaglandins, Tetrahe-dron Lett.24,2435-2437.

SUGAI, ., MORI,K. (1984), Preparation of (R)-ibu-profen and related carbocyclic acids using micro-bial oxidation of an aromatic hydrocarbon byRhodococcus sp. BPM 1613,Agric. Biol. Chem.48,2501-2504.

SUGIYAMA,.,TRAGER,W. F. (1986), Prochiral selec-tivity and intramolecular isotope effects in the cy-

tochrome P-450 catalyzed w-hydroxylationof cu-mene,Biochemistry 25,7336-7343.TABENKIN,., LEMAHIEN, . A., BERGER,., KIER-

STEAD, R. w. (1969), Microbiological hydroxyla-tion of cinerone to cinerolone, Ap pl. M icrobiol.17,714-717.

TAKAMATSU,., UMEMURA,., TOSA,T., SATO,T.,YAMAMOTO,., CHIBATA, . (1982), Productionof L-alanine from ammonium fumarate using twotypes of immobilized microbial cells,Eur.J. Appl .Microbiol. Biotechnol. 15,142-152.

TENGO,., AGREN, ., BAUR,B., ISAKSSON,., LIUE-FORS,T. et al. (1990),Andrena wilkella male bees

discriminate enantiomers of cephalic secretioncomponents,J. Chem. Ecol. 16,429441.TERASHIMA,.,TAMOTO,. (1982), An efficient syn-

thesis of optically pure anthracycline intermedi-ates by the novel use of microbial reduction, Tet-rahedron L ett.23,3715-3718.

TSCHAEW,. M., FUENTES,. M., LYNCH,. E., LAS-WELL,W. L., VOLANTE, . P., SHINKAI,. (1988),An efficient synthesis of 4-benzoyloxyazetidi-none: An important carbapenem nucleus, Tetra-hedron Lett.29,2779-2782.

TURNER, . J., WHITESIDES,. M. (1989), A com-bined chemical-enzymaticsynthesisof 3-deoxy-6

arabino-heptulosonic acid 7-phosphate,J .

A m .Chem. SOC.111,624627.



400 10 Synthetic A pplications ofEnzyme-Catalyzed Reactions

VA N OSSELAER,. A., LEMIERE,G. L., LEPOIVRE,.A., ALDERWEIRELOT,. C. (1978), Enzy matic invitro reduction of ketones, Bull. SOC.Chim. Belg.

VO N KRAMER, ., WANDER,. (1982), C45-und Cs0-Carotinoide: Synthese von ( R ) - und (S)-Lavan-

dulol, Helv. Ch im. Acta. 65,293-301.VELLUZ, . (1965), Recent advances in the total syn-

thesis of s teroids,Angew. Chem. (Int. Edn. Engl.)

WANG,Y.-F., IH ,C. J. (1984), Bifunctional chiral syn-thons via biochemical methods 4. Chiral precur-sors to (+)- biotin and (-)-A-factor, TetrahedronLett .25,4999-5002.

WANG, .Y.,Hsu , C.T., SIH,C. J. (1981),Totalsynthe-sis of (+)-compactin (ML-236B),J. Am . Chem.SOC.103,653M539.

WANG,Y.-F., IZAWA, ., KOBAYASHI,., OHNO,M.(1982), Stereocontrolled synthesis of (+ )-negam-ycin from an acyclic hom oallyam ine by 1,3-asym-metric induction, J . Am. Chem. SOC. 104,646545466.

WANG,Y.-F., CHEN,C . 4 , GIRDAUKAS,., SIH,C. J.

87,153-154.

4,181-200.

(1984), Bifunctional chiral synthons via b iochern-ical methods. 3. Optical purity enhancement inenzymic asymmetric catalysis, J. Am . Chem. SOC.

WONG,C.-H.,WHITESIDES,. M. (1994),Enzymes inSynthetic Organic Chemistry,Tetrahedron Organ-ic C hem istry Series Vol. 12. Trowbridge: Pergam -on .

WONG,C. H., MAZENOID,. P.,WHITESIDES,. M.(1983), Chem ical and enzymatic syntheses of 6-deoxyhexoses. Conversion to 2,5-dimethyl-4-hy-droxy-2,3-dihydrofuran-3-onefuraneol) and an-alogs, . Org. Chem .48,3493-3497.

YAMAUCHI,., CHOI,S. Y., OKADA,H ., YOHDA,M.,KUMAGAI,. et al. (1992),Properties of aspartateracemase, a pyridoxa l5 ‘-phosphate-independentamino acid racernase, J. Biol. Chem. 267,

YEE,C., BLITHE, . A., MCNABB, . J., WALTS,A. E.(1992), Bioca talytic resolution of tertiary a-sub-stituted carbocyclic acid esters: Efficient prep ara-tion of a quaternary asymm etric carbon center, J.

Org.Chem.57,3525-3527.

106,3695-3697.

18361-18 365.



The Future of Biotransformations



11 Catalytic Antibodies

GEORGEMICHAEL LACKBURNARNAUDARCON

Sheffield,UK

1 Introduction 404

1.1 Antibody Structure and Function 404

1.2 The Search of a New Class of Biocatalyst 404

1.3 Early Examples for Catalytic Antibodies 406

1.4 Methods for Generating Catalytic Antibodies 407

2.1 Transition State Analogs 410

2.2 Bait and Switch 412

2.3 Entropic Trapping 415

2.4 Substrate Desolvation 419

2.5 Supplementation of Chemical Functionality 420

3 Spontaneous Features of Antibody Catalysis 421

3.1 Spontaneous Covalent Catalysis 421

3.2 Spontaneous Metal Ion Catalysis 422

4 How Good are Catalytic Antibodies? 423

5 Changing the Regio- and Stereochemistry of Reactions 426

5.1 Diels-Alder Cycloadditions 4265.2 Disfavored Regio- and Stereoselectivity 427

5.3 Carbocation Cyclizations 430

6.1 Resolution of Diastereoisomers 432

6.2 Cleavage of Acetals and Glycosides 4326.3 Phosphate Ester Cleavage 435

6.4 h i d e Hydrolysis 438

7 Reactive Immunization 439

8 Potential Medical Applications 441

2 Approaches to Hapten Design 410

6 Difficult Processes 431

8.1 Detoxification 441

8.2 Prodrug Activation 4428.3 Abzyme Screening via Cell Growth 445

9 Industrial Future of Abzymes 446

10 Conclusions 446

11 Glossary 447

12 Appendix 451

12.1 Catalog to Antibody Catalyzed Processes 45112.2 Key to Bibliography 479

13 References 481





404 11 Catalytic Antibod ies

1 Introduction

This review seeks to deal with most of the

imp ortan t developm ents in the field of catalyt-ic antibodies since the initial successes wereannounced over a dozen years ago (POLLACKet al., 1986;TRAMONTANOt al., 1986).The de-velopm ent of catalytic antibo dies has requiredcontributions from a num ber of scientific dis-ciplines, which traditionally have n ot w orkedin concert. Thus, while this chapter describescatalytic antibodies, their reactions, and theirmechanisms from a biotransformations view-point, it will also provide a review that de-man ds only a basic biochemical knowledge ofantibody structure, function, and production.Sufficient details of these matters have beensupplied to mee t the nee ds of expe rt and non-expert readers alike. In Sect. 11, there is a glos-sary of most of the immunological terms usedin this review, in language familiar to chem ists.While this survey is not fully comprehensive , itseeks to focus on the m ost significant parts ofa subject which, in a little over a decade, hasachieved much m ore than most critics expect-ed at the outset. A moderately complete sur-

vey of the literatu re is prese nted in th e form ofan appendix (Sect. 12),which lists over 100 ex-amples of reactions catalyzed by antibodies,the haptens employed, and their kinetic pa-rameters.

1.1 Antibody Structure and

Function

One of the most important biological de-fense mechanisms for higher organisms is the

immune response. It relies on the rapid gener-ation of structurally novel proteins that canidentify an d bind tightly to foreign substancesthat would otherwise be dam aging to the par-ent organism. These proteins are called immu-noglobulins and constitute a protein super-family.In their simplest form they ar e made upof four polypeptide chains: on e pair of identi-cal short chains and o ne pair of identical longchains, interconnected by disulfide bridges.The two light and two identical heavy chainscontain repeated homologous sequences ofabout 110 amino acids which fold individually

into similar structural domains, essentially abilayer of antiparallel P-pleated sheets. Thisgives the structure of a n IgG immunoglobulinmolecule whose core is formed from twelve,

similar structural domains: eight from the twoheavy chains and four from the two lightchains (Fig. 1) (BURTON,990). Notw ithstand-ing this a ppar ent hom ogeneity, the N -terminalregions of antibody light and heavy chainsvary greatly in the variety an d num ber of theirconstituent amino acids and thereby providebinding regions of grea t diversity called hyper-variable regions. Th e variety of pro teins sogenerated approaches lolo in higher mam-mals.

The essential property of the immune

system is its ability to respond to single or m ul-tiple foreign molecular species (antigens)through rapid diversification of the sequencesof these hype rvariable regions by processes in-volving mutation, gene splicing, and RNAsplicing. This initially provides a v ast nu mb erof different antibodies which subsequently ar eselected and amplified in favor of those struc-tures w ith the strongest affinity for on e partic-ular antigen.

1.2 The Search for a New Class of

Biocatalys

Fifty years ago, LINUSPAULINGlearly setout his theory that enzymes achieve catalysisbecause of their complementarity t o the tran-sition state for the reaction being catalyzed(PAULING,948). With hind sight, this co nceptcould be seen a s a logical extension of the newtransition state theory that had been recentlydeveloped to explain chemical catalysis

(EVANSand POLANYI, 935; EYRING,935).The basic proposition was that the rate of a re-action is related t o the difference in Gibb s freeenergy (AGO ) etween the ground state of re-actant(s) and the transition state for that reac-tion. For catalysis to b e manifest, either the en-ergy of the transition state has to be low ered(transition state stabilization) o r the energy ofthe sub strate has to be elevated (substrate de-stabilization). PAULINGpplied this concept toenzyme catalysis by stating that an enzymepreferentially binds to and thereby stabilizes

the transition state for a reaction relative to



1 Introduction 405

Fig. 1. Scheme to show the structure of the peptide com ponen ts of an IgG immunoglobulin:the two light (L) and two heavy (H) polypeptide chains; the disulfide bridges (-S-S-)

connecting them ; four variable regions of the light (V,) and heavy (V,) chains; an d the eight“constan t” regions of the light (C,) and heavy (CH1,CHz,CH3)chains (sh aded rectangle).The hypervariable regions that achieve antigen recognition and binding are located within

sixpolypeptide loops, thre e in the V, and three in the V, sections (shaded circle, top left).These can be excised by protease cleavage to give a fragment antibody, Fab (shadedlobe, top right).

the ground state of substrate(s) (Fig. 2). Thishas become a classical theory in enzymologyand is widely used to explain the way in whichsuch biocatalysts are able to enhance specificprocesses with rate accelerations of up to lo’’over background (ALBERYand KNOWLES,

1976,1977;ALBERY,993 for a review).PAULING pparently did not bring ideas

about antibodies into his concept of enzymecatalysis, although there is a tantalizing photo-graph in a volume of PAULING’Sectures ca.1948 which shows on a single blackboard a car-toon of an energy profile diagram for the low-ering of a transition state energy profile and al-so reference to an immunoglobulin (PAULING,1947).And so it fell to BILL ENCKS n his mag-isterial work 1969 on catalysis to bring togeth-er the opportunity for synthesis of an enzyme

by the use of antibodies that had been engi-

neered by manipulation of the immune system(JENCKS,1969).

“One way to do this (i.e., synthesize an en-z y m e ) is to prepare an antibody to a haptenicgroup which resembles the transition state ofa given reaction”.’

The practical achievement of this goal washeld up for 18 years, primarily because of thegreat difficulty in isolation and purification ofsingle-species proteins from the immune rep-ertoire. During that time, many attempts todemonstrate catalysis by inhomogeneous (i.e.,polyclonal) mixtures of antibodies were madeand failed (e.g., RASOand STOLLAR, 975;

In making this statement, JENCKSwas apparentlynot aware of PAULING’Sdea (JENCKS,personal com-

munication).



11 Catalytic An tibod ies

-- rogress ofreaction -Fig. 2. Catalysis is achieved by lowering the free energy of activation for a p rocess, i.e., the catalyst must bindmore strongly to the transition state (TST) of the reaction than to either reactants or products. Thus:AAG: s - A A G ~ ~ ~ : ~nd AAGcak..

SUMMERS,983). The problem was resolved in1976 by KOHLER nd MILSTEIN’S developmentof hybridoma technology, which has m ade it

possible both t o screen rapidly the “com plete”immune repertoire and to produce relativelylarge am ounts of o ne specific monoclonal anti-body species in vitro (KOHLER nd MILSTEIN,1975;KOHLER t al., 1976).

While transition states have been discussedin terms of their free energies, there have bee nrelatively few attem pts to describe their struc-ture at atomic resolution for most catalyzedreactions. Transition states are high energyspecies, often involving incompletely form edbonds, and this m akes the ir specification very

difficult. In some cases these transient specieshave be en studied using laser femtochemistry(ZEWAILnd BERNSTEIN,988;ZEWAIL,997),and predictions of some of their geometrieshave bee n m ade using molecular orbital calcu-lations (HOUK et al., 1995). Intermediatesalong the reaction coordinate a re also often ofvery short lifetime, though some of their struc-tures have b een studied unde r stabilising con-ditions while their existence and general na-ture can often be established using spectro-scopic techniques or trapping experiments(MARCH, 992a).

The Hammond postulate predicts that if ahigh energy intermediate occurs along a reac-tion pathway, it will resemble the transition

state nearest to it in energy (HAMMOND,955).Conv ersely, if th e tran sition sta te is flanked bytwo such intermediates, the on e of higher ener-gy will provide a closer approximation to thetransition state structure. This assumption pro-vides a stron g basis for th e use of mimics of un-stable reaction intermediates as transitionstate analogs (BARTLETTnd LAMDEN,986;ALBERGt al., 1992).

1.3 Early Examples of Catalytic

Antibodies

In 1986, RICHARDLERNERand PETERSCHULTZndependently reported antibody ca-talysis of the hydrolysis of aryl esters and ofcarbonates respectively (POLLACKt al., 1986;TRAMONTANOt al., 1986). Such reac tions a rewell-known to involve the formation andbreakdown of an unstable tetrahedral inter-mediate (2), and so this can be deemed $0 beclosely related to the transition state (TS+) ofthe reaction (Fig. 3).



1 Introduction 401

Transition states of this tetrahedral naturehave now been effectively mimicked by arange of stable analogs, including phosphonicacids, phosphonate esters, a-difluoroketones,

and hydroxymethylene functional groups (JA-COBS, 1991).LERNER'Sroup elicited antibod-ies to a tetrahedral anionic phosphonate hap-ten2 (3) (AE32.9) while SCHULTZ'Sroup iso-lated a protein with high affinity for p-nitro-phenyl cholyl phosphate (5) (Fig.4, AE 3.2).

" 0 L X ,0 R'

2 Tetrahedralintermediate

0-Products

Fig. 3. Th e hydrolysis of an aryl este r (1) (X =CHz)or a carbonate (1) (X =0)proceeds through a tetra-hedral interm ediate (2) which is a close mode l of th etransition s tat e for the reaction. It differs substanti-ally in geometry and charge from both reactan ts andproducts.

It might be helpful to the reader to indicate thatth e pyridine-2-6-dicarboxylate omponent of (3)was designed for a further purpose, neither used no rneeded for the activity described in the presentscheme.

A E =Appendix En t ry

1.4 Methods for Generating

Catalytic Antibodies

It is appropriate at this stage in the review toconsider the stages in production of a catalyticantibody and to put in focus the relative rolesof chemistry, biochemistry, immunology, andmolecular biology. Nothing less than the fullintegration of these cognate sciences is neededfor the fullest realization of the most difficultobjectives in the field of catalytic antibodies. Inbroad terms, the top section of the flow dia-gram for abzyme production (Fig. 5) involveschemistry, the right hand side is immunology,the bottom sector is biochemistry, and molecu-

lar biology completes the core of the scheme.

ChemistryAt the outset, chemistry dominates the selec-tion of the process to be investigated. The tar-geted reaction should meet most if not all offollowing criteria:

(1) have a slow but measurable spontane-ous rate under ambient conditions;

(2) be well analyzed in mechanistic terms;

(3) be as simple as possible in number ofreaction steps;(4) be easy to monitor;(5) lead to the design of a synthetically ac-

cessible transition state analog (TSA)of adequate stability.

As we shall see later, many catalytic antibodiesachieve rate accelerations in the range lo 3 tolo6.It follows that for a very slow reaction, e.g.,the alkaline hydrolysis of a phosphate diesterwith k, , ca. lo-'' M-' s-',direct observation

of the reaction is going to be experimentallyproblematic. Given that concentrations of cat-alytic antibodies employed are usually in the1-10 pM range, it has been far more realistic totarget the hydrolysis of an aliphatic ester, withkOHca. 0.1M-' s - ' under ambient conditions.

The need for a good understanding of themechanism of the reaction is well illustratedby the case of amide hydrolysis. Many earlyenterprises sought to employ TSAs that werebased on a stable anionic tetrahedral interrne-diate, as had been successful for ester hydroly-

sis. Such approaches identified catalytic anti-



408 11 Catalytic Ant ibod ies

Abzyme Cond i t ions K m 4 kcat4 Ki 3

Identity

6D 4 X pH 8; 25 "C [email protected] s-l 0.16 fl

5 6

Abzyme Cond i t ions Km 6 kcat6 Ki 5

Identity

MOPC167 pH7; 30°C 208fl 0.007 s- l 5 p.M

Fig. 4. LERNER'Sroup used phosphonate (3) as the hapten to raise an antibody which was capable of hy-drolyzing the ester (4). SCHULTZound that naturally occurring antibodies using phosphate ( 5 ) as their an ti-gen could hydrolyze the correspondingp-nitrophenyl choline carbona te (6). Parts of haptens (3) and ( 5 ) e-

quired for antibody recognition have been em phasized in bold).

aof Target

Organic Synthesis

SCREEN Ifor binding

ANTIBODIES

Fig. 5. Stages in the prod uctio n of a catalytic antibody.



1 Introduction 409

bodies capable of ester hydrolysis but not ofamide cleavage! There is good evidence thatfor aliphatic amides, breakdown of the te tra-hedral intermediate is the rate determining

step. This step is catalyzed by protonation ofthe nitrogen leaving group a nd hence this fea-ture m ust be pa rt of the TS A design.

The importance of minimizing the numberof covalent steps in the process to b e catalyzedis r ather obvious. Single- and two-step process-es dom inate the abzyme scene. However, thereis substantial evidence that some acyl transferreactions involve covalent antibody interm edi-ates and so must proceed by up to four cova-lent steps. Nonetheless, such antibodies werenot elicited by intentional design but rather

discovered as a consequence of efficientscreening for reactivity.

Direct monitoring of the catalyzed reactionhas most usually been carried out in real timeby light absorption of fluorescent emissionanalysis and some initial progress has beenmade with light emission detection. The lowquantities of abzymes usually available at thescreening stage put a premium on th e sensitiv-ity of such methods. However, som e work h asbeen carried out of necessity using indirect

analysis, e.g., by H PL C o r NM R.Finally, this are a of re search might well havesupported a Journal of UnsuccessfulAbzymes .It is common experience in the field that somethree out of four enterprises fail, and for noobvious reason. It is therefore imperative thatchemical synthesis of a TSA should not be therate determining step of an abzyme project.The average performance target is to achievehapte n synthesis within a year: one or two ex-amples have employed TSAs that could befound in a chem ical catalog, the most syntheti-

cally-demanding cases have perforce em-ployed multi-step routes of considerable so-

phistication (e.g., A E 13.2). Lastly, the TS A hasto survive in vivo for a t least two days to elicitthe necessary antigenic response.

ImmunologyThe interface of chemistry and immunologyrequires conjugation of multiple copies of theTSA to a carrier protein for production ofantibodies by stand ard m onoclonal technolo-gy (KOHLER nd MILSTEIN,1976). One such

conjugate is used for mouse immunization and

a second one for ELISA screening purposes.The carrier proteins selected for this purposeare bovine serum albumin (BSA, RMM67,000), keyhole limpet hemocyanin (KLH ,

R MM 4 . o 6),and chicken ovalbumin (RMM32,000). All of th ese are b asic prote ins of highimmunogenicity and with m ultiple surface ly-sine residues that ar e widely used as sites forcovalent attachment of hapten. Successfulantibody production can take some threemonths and should deliver from 20 to 200monoclonal antibody lines for screening, pref-erably of IgG isotype.

Screening in early work sought to identifyhigh affinity of the antibody for the T SA , usinga process known as ELISA. This search can

now be performed more quantitatively byBIAcore analysis, based on surface plasmonresonance methodology (LOFAS and JOHN S-SON, 990).A subsequent development is thecatELISA assay (TAWFIK t al., 1993) whichsearches for product formation and hence theidentification of abzymes that can generateproduct.

Methods of this nature are adequate forscreening sets of hybridom as but not for selec-tion from much larger libraries of antibodies.

So,most recently, selection m ethods employ-ing suicide substrate s (Sec t. 6.2) (JAN DA t al.,1997) or DNA amplification methodology(FENNIRIt al., 1995) have be en brought intothe repertoire of techniques for the directidentification of antibodies that can turnovertheir substrate. However, the time-consumingscreening of hybridomas remains the mainstayof abzyme identification.

BiochemistryA fam ily of 100 hybridom a an tibodie s can typ-

ically pro vide 20 tight binders and these needto be assayed for catalysis. A t this stage in theproduction of an abzyme, the benefit of a sen-sitive, direct screen for product formationcomes into its o wn. Following identification ofa successful catalyst, the antibody-producingcell line is usually recloned to ensure purityand stabilization of the clone, then protein isproduced in larger amount (ca. 10 mg) andused for determination of the kinetics andmechanism of the catalyzed process by classi-cal biochemistry. Digestion of such protein

with trypsin or pa pain provides fragment a nti-



410 I1 Catalytic Antibo dies

bodies (Fabs), that contain only the attenu atedupper limbs of the intact IgG (Fig. 1). It isthese com ponents that have bee n crystallized,in many cases with the substrate analog, prod-

uct, or TS A bound in the combining site, andtheir structures determined by X-ray diffrac-tion.

Molecular Biology

Only a few abzymes have reached the stagewhere m utagenesis is being employed in orde rto improve their performance (MILLER t al.,1997). Likew ise, HILVERTs the first to havereached the stage of redesign of the h apten toattem pt the production of antibodies with en-hanced performance (KASTet al., 1996). So,

the circle of production has now been com-pleted for a t least one exam ple, and chem istrycan start again with a revised synthetic target.

2 Approaches to Hapten

Design

O ne can now recognize a variety of strate-gies in addition to the earliest o nes deployedfor hapte n design. Some of these were presen t-ed originally as discrete solutions of th e pro b-lem of abzym e generation, but it is now recog-nized that they need not be mutually exclusiveeither in design or in application. Indeed, morerecent work often brings two or more of themtogeth er interactively. They can b e classifiedbroadly into five categories for the purposes ofanalysis of their principa l design elements. Thesequence of presentation of these here is in

part related to the chronology of their ap pear-ance on the abzym e scene.

(1) Transition state analogs.(2) Bait and switch.(3 ) Entropy traps.(4) Desolvation.( 5 ) Supp lementation of chem ical function-

ality.

2.1 Transition State Analogs

As has clearly bee n shown by th e m ajorityof all published work on catalytic antibodies,the original guided m ethodology, i.e., the de-sign of stable transition state analogs (TSAs)for use as haptens to induce the gen eration ofcatalytic antibodies, has served as the bedrockof abzyme research. Most work has been di-rected at hydrolytic reactions of acyl species,

perhaps because of the broad knowledge ofthe na ture of reaction m echanisms for such re-actions and the wide experience of deployingphosphoryl species as stable mimics of un-stable tetrahedral intermediates. More than 80examples of hydrolytic antibodies have beenreported, including the 47 examples of acylgroup transfer to water (entries under 1-5 ofthe Appendix).

Most such acyl transfer reactions involvestepwise addition of the nucleophile followedby expulsion of th e leaving group with a tran-

sient, high energy, tetrahedral intermediate




sepa rating these processes. The fastest of suchreactions generally involve good leavinggroups and the addition of the nucleophile isthe rate determining step. This broad conclu-

sion from much detailed kinetic analysis hasbeen en dorsed by compu tation for the hydrol-ysis of me thyl acet ate (TERAISHI t al., 1994).This places the energy for product formationfrom an anionic tetrahedral intermediate some7.6 kcal m ol-l lower than for its reversion toreactants. So, for the gene ration of antibodiesfor the hydrolysis of aryl esters, alkyl esters,carbonates, and activated anilides, the designof hapten has focused on facilitating nucleo-philic attack , and with considerable success.

The tetrahedral intermediates used for this

purpose initially deployed phosphorus (V )systems, relying o n the strong polarization ofthe P=O bond (arguably more accuratelyrepresented as P + -0 -) . The range has includ-ed m any of the possible species containing a nionized P-OH group (Fig. 6). O ne particularlygood feature of such systems is that theP+ -O - bond is intermediate in length(1.521 A) between the C -0 - bond calculated' for an anionic tetrahedral intermediate(0.2-0.3 A shorter) and for the C . . - O reak-

ing bond in the transition state (some 0.6 Alonger) (TERAISHI t al., 1994). Oth er tetrahe -dral systems used have included sulfonamides(SHEN,1995) and sulfones (BENE DE'ITI t al.,1996), secondary alcohols (SHOK AT t al.,1990), and a-fluoroketone hydrates ( K ~ A z u -

It is clear that phosphorus-based transitionstates have had the g reatest success, as shownby the many entries under 1-5 of the Appe n-dix. This may be a direct result of their anionicor partial anionic character, a fea ture not gen-

erally available for the oth er species illustratedin Fig. 6, thou gh a-difluorosulfonam ides mightreasonably also share this fea ture as a result oftheir enh anced acidity.

No t su rprisingly, most of the catalytic anti-body binding sites examined in structural de-tail have been found to contain a basic residuethat provides a coulombic interaction withthese TSAs, for which the pro totype is the nat-ural antibody M cPC603 to phosphoryl choline,where the phosphate anion is stabilized bycoulombic interaction with ArgH5* PADLANt

al., 1985). In particular, X-ray structures ana-

M E e t al., 1994).

Phosphate Phosphorothioate

diester diester

Phosphonate Phosphonamidate

monoester

Phosphinic acid Sulfonamide

Sulfone Ketone hydrate

Fig. 6. Transition state analogs for acyl cleavage re-

action via tetrahedral transitions states.

lyzed by FUJII(FUJIIet al., 1995) have shownthat the protonated in catalytic anti-bodies 6D 9,4B 5,8D ll , and 9C10 (A E 1.8) is

capable of forming a hydrogen bond to theoxyanion in the transition state for ester hy-drolysis.

In similar vein, KNOS SOW as identifiedHisH35 ocated proxima te to the oxyanion ofp-nitrophenyl methanephosphonate in thecrystalline binary complex of antibodyCNJ206 and TSA, a system designed to hydro-

lyzep-nitrophenyl ac etate (c.f. A E 2.7) (CHA R-BONNIER e t al., 1995).A third exam ple is seenin SCHULTZ'S tructure of antibody 48 67 ,which hydrolyzes methyl p-nitrophenyl car-bonate (A E 3.lc) .The hapte npni t rop heny l4-carboxybutane-phosphonate is proximate toArgL96 and also forms hydrogen bonds to

TyrH33, nd T ~ T ~ ~ ~Fig. 7) (PAITEN etal., 1996).

Clearly, the oxyanion h ole is now as signifi-cant a feature of the binding site of such acyltransfer abzym es as it is already for esterases

and peptidases - and not w ithout good reason.



412 11 Catalytic Antibodies

TyrL94..

H‘

Fig. 7. Binding site details for antibody 48G7 com-

plexed with hapten p-nitrophenyl 4-carboxybuta-nephosphonate (PAITENet al., 1996). NB Aminoacid residues in antibodies are identified by theirpresence in the light (L ) or heavy (H) chains with anumber denoting their sequence position from theN-terminus of the chain.

KNOSSOWas analyzed the structures of three

esterase-like catalytic antibodies, each elicited

in response to the same phosphonate TSAhapten (CHARBONNIERt al., 1997). Catalysis

for all three is accounted for by transition state

stabilization and in each case there is an oxy-

anion hole involving a tyrosine residue. This

strongly suggests that evolution of immuno-

globulins for binding to a single TSA hapten

followed by selection from a large hybridoma

repertoire by screening for catalysis leads to

antibodies with structural convergence. Fur-

thermore, the juxtaposition of X-ray structures

of the unliganded esterase mAb D2.3 and its

complexes with a substrate analog and withone of the products provides a complete de-

scription of the reaction pathway. D2.3 acts at

high pH by attack of hydroxide on the sub-

strate with preferential stabilization of the

oxyanion anionic tetrahedral intermediate, in-

volving one tyrosine and one arginine residue.

Water readily diffuses to the reaction center

through a canal that is buried in the protein

structure (GIGANT t al., 1997). Such a clear

picture of catalysis now opens the way for site-

directed mutagenesis to improve the perform-

ance of this antibody.

2.2 Bait and Switch

Charge-charge complementarity is an im-

portant feature involved in the specific and

tight binding of antibodies to their respectiveantigens. It is the amino acid sequence and

conformation of the hypervariable (or comple-

mentarity determining regions, CDRs) in the

antibody combining site which determine the

interactions between antigen and antibody.

This has been exploited in a strategy dubbed

“bait and switch” for the induction of antibody

catalysts which perform p-elimination reac-

tions (SHOKAT t al., 1 9 8 9 ; T ~ o ~ ~t al., 1995),

acyl-transfer processes (JANDA t al., 1990b,

1991b;SUGA t al., 1994a;Lr and JANDA,995),

cis-trans alkene isomerizations (JACKSONnd

SCHULTZ, 991), and dehydration reactions

(UNOand SCHULTZ,992).

The bait and switch methodology deploys a

hapten to act as a “bait”.This bait is a modified

substrate that incorporates ionic functions in-

tended to represent the coulombic distribution

expected in the transition state. It is thereby

designed to induce complementary, oppositely

charged residues in the combining site of anti-

bodies produced by the response of the im-

mune system to this hapten. The catalytic abil-ity of these antibodies is then sought by a sub-sequent “switch” to the real substrate and

screening for product formation, as described

above.

The nature of the combining site of an anti-

body responding to charged haptens was first

elucidated by GROSSBERGnd PRESSMAN

(GROSSBERGnd PRESSMAN,960),who used a

cationic hapten containing a p-azophenyltri-

methylammonium ion to make antibodies with

a combining site carboxyl group, essential for

substrate binding (as shown by diazoacet-amide treatment).

The first example of “bait and switch” for

catalytic antibodies was provided by SHOKAT

(SHOKAT t al., 1989), whose antibody 43D4-

3D12 raised to hapten (7) was able to catalyze

the p-elimination of (8 ) to give the trans-enone ( 9 ) with a rate acceleration of 8.8.104

above background (Fig. 8;AE 8.2). Subsequent

analysis has identified a carboxylate residue,

G ~ u ~ ~ ~s the catalytic function induced by the

cationic charge in (7) (Fig. 6) (SHOKAT t al.,

1994).




12

Abzym e Identity Conditions

43D4-3D12 pH 6; 37“C

182 pM 0.003 s- ’ 0.29 pM

Fig. 8. Using th e “bait an d switch” principle, hapte n(7) licited an antibody, 43D4-3D12, which catalyzedthe p-elimination of (8) to a trans-enone (9 ) .Thecarboxyl function in (7) s necessary for its attach-ment to the carrier protein.

A similar “bait and switch” approach hasbeen exploited for acyl-transfer reactions(JANDA et al., 1990b, 1991b).The design of hap-

ten (10) incorporates both a transition statemimic and the cationic pyridinium moiety, de-signed to induce the presence of a potentialgeneral acidbase or nucleophilic amino acidresidue in the combining site, able to assist incatalysis of the hydrolysis of substrate (11)(Fig. 9,AE 2.6).

Some 30% of all of the monoclonal antibod-ies obtained using hapten (10) were catalytic,and so the work was expanded to survey threeother antigens based on the original TSA de-sign (JANDA et al., 1991b).The carboxylate an-

ion in (12)was designed to induce a cationic

14= succinimidyl

Fig. 9. The original hapten (10) demonstrated theutility of the “b ait and switch” strategy in th e gener-ation of antibodies to hydrolyze t he ester s ubstrate(11).Three haptens (U),W), (14) were designed to

examine fu rther the effectiveness of point charges in

amino acid induction. Both charged haptens (U),(13) prod uce d antibodies that catalyzed the hydroly-sis of (11) whereas the ne utral hapten (14) enerat-ed antibodies which bound the substrate unp roduct-ively.

combining site residue, while the quaternaryammonium species (13) combines both tetra-hedral mimicry and positive charge in thesame locus. Finally, the hydroxyl group in (14)was designed to explore the effects of a neutral

antigen.




Three important conclusions arose from thiswork.

(1 ) A charged functionality is crucial for

catalysis.(2) Catalytic antibodies are produced from

targeting different regions of the bind-ing site w ith positive and negative hap-tens (although more were gene rated inthe case of the cationic hap ten usedoriginally).

(3) The com bination of charge plus mimic-ry of the transition state is required toinduce hy drolytic esterases.

Esterolytic antibodies have also been pro-

duced by S UG A sing an alternative “bait andswitch” strategy (AE 2.1) (SUGA t al., 1994a).A 1,2-am inoalcohol function was designed forgenerating not only esterases but also amidas-es. In orde r t o elicit an anionic combining sitefor covalent catalysis, three haptens were syn-thesized, one contained a protonated amine(15) and two featured trimethylammoniumcations (16), (17) (Fig. 10).The outcome wasinterpreted as suggesting that haptens contain-ing a trimethylammonium group were too

sterically demanding, so that the induced an-ionic amino acid residues in the antibody bind-ing pocket w ere too distant to provide nucleo-philic attack at the carbonyl carbon of sub-strate (18).An alternative explanation may bethat coulombic interactions lacking any hydro-gen bonding capability will not be sufficientlyshort-range for the purp ose intended.

Th e use of secondary hydroxyl groups in thehaptens (15) nd (16) was designed to mimicthe tetrahedral geometry of the transitionstate (as in JANDA’S ork), while the third hap-

ten (17) replaced the neutral OH with an an-ionic phospha te group, designed to elicit a cat-ionic combining site residue to stabilize thetransition state oxyanion. However, this func-tion in (17) may have proved to o large to in-duce a catalytic residue close enough to th e de -veloping oxyanion, since weaker catalysis wasobserved relative to haptens (15) and (16)(kcatlkUncat2.4 . o3 , 3 .3 . o3, and -1 o3 fo r(15), (16),and (17), respectively) (Fig. 10).

In orde r to achieve catalysis employing bothacid and basic functions, an alternative zw itter-

ionic hap ten was proposed in which the anion -

O N H 3

I

OZN H

15

@NMe3

1

O2N H

16

O N Me 3

1

HO

18

19

R = position of linker

Fig. 10.Three haptens (W), (16), (17) containing 1,2-

aminoalcohol functionality were investigated as al-ternatives fo r esterase an d amidase induction. Halfof the antibodies raised against hapten (15) wereshown t o catalyze th e hydrolysis of ester (18),here-by establishing the necessity fo r a compact haptenicstructure. Hapte n (19) long with (16) was employedin a heterologous immunization program to elicitboth a general and acid base function in the anti-body binding site.




ic phosphoryl core is incorporated alongsidethe cationic trimethylammonium moiety (cf.17) (SUGA t al., 1994b). Th e difficulty in syn-thesizing such a target hapten can be over-

come by stimulating the immune system firstwith the cationic and then with the anionicpoint charges using the two structurally relat-ed haptens (16) and (19),respectively. Such asequential strategy has been du bbed “he terol-ogous immunization” (Fig. 10) and this resultsof this strategy were compared with thosefrom the individual use of haptens (16) and(19) in a “hom ologous immunization” routine.Of 48 clones produced as a result of the ho-mologous protocols, 7 were found t o b e cata-lytic, giving rate enhancements up to 3-103.By con trast, 19 of the 50clones obtained usingthe heterologous strategy displayed catalysis,the best being up t o two orders of magnitudebetter.

A final example of the ba it and switch strat-egy (THORNt al., 1995) focuses on th e base-prom oted decomposition of sub stituted benzi-soxazole (20) to give cyanophenol(21) (Fig 11,AE 8.4). A cationic h apten (22) was used tomimic the transition state geometry of allreacting bonds. It was anticipated that if the

benzimidazole hapten (22) induced the pres-ence of a carboxylate in the binding site, itwould be ideally positioned to m ake a hydro-gen bond to the N-3 proton of the substrate.The resultant abzym es would thus have gener-al base capability for abstracting th e H-3in thesubstrate.

Two monoclonals, 34E4 and 35F10, werefound to catalyze the reaction with a rate ac-celeration greater tha n lo8, while the p resenceof a carboxylate-containing binding site resi-due was confirmed by pH-ra te profiles and co-

valent modification by a carbodiimide, whichreduced catalysis by 84% .

The b ait a nd sw itch tactic clearly illustrates. that a ntibodies are capab le of a coulom bic re-

sponse tha t is potentially orthogonal to the useof transition state analogs in engendering ca-talysis. By variations in the hap ten employed,it is possible to fashion antibody combiningsites that contain individua l residues t o deliverintricate m echanism s of catalysis.

HI

HO,C2 l4 22

Fig. 11.The use of a cationic hapten (22) mimics thetransition state of the base-promoted decompositionof substituted benzisoxazole (20) to cyanophenol(21) nd also acts as a “bait” to induce the presenceof an anion in the combining site that may act as a

general base. ’

2.3 Entropic Trapping

Rotational EntropyA n important com ponent of enzym e catalysisis the control of translational and rotationalentropy in the transition state (PAGE andJENCKS,1971).This is well exem plified for u ni-molecular processes by the enzyme chorismatemutase, which catalyzes the rearrang eme nt of

chorismic acid (23) into prephenic acid (24)




(Fig. 12). This reaction proceeds through a cy-clic transition state having a pseudo-diaxialconformation (25) (ADDADI t al., 1983).Withthis analysis, BARTLETTdesigned and syn-

thesized a transition state analog (26) whichproved to be a powerful inhibitor for the en-zyme (BARTLETT nd JOHNSON,985). X-raystructures of mutases from Escherichia coli(LEE, et al., 1995),Bacillus subtilis (CHOOKtal., 1993, 1994), and Saccharomyces cerevisiae(XUE and LIPSCOMB,995) complexed to (26)show completely different protein architec-tures although the bacterial enzymes have sim-ilar values of k,,,/k,,,,, (3. lo6) and of Ki for(26). It appears that these enzymes exert theircatalysis through a combination of conforma-

OH

23 Chorismate

0

o'cd

+coo

0

o'cd9 :pcoo

OH

25 Transition state

Fig. 12. Chorismate mutase.

-OH

24 Prephenate

OR

26 Transition state analog

tional control and enthalpic lowering. Sup-porting this, HILLIER as carried out a hybridquantum mechanical/molecular mechanicscalculation on the B. subtilis complex with sub-

strate (23). He concluded that interactionsbetween protein and substrate are maximalclose to the transition state (25) and lead to alowering of the energy barrier greater than isneeded to produce the observed rate accelera-tion (DAVIDSONnd HILLIER, 994).

SCHULTZmployed TSA (26) as a hapten togenerate antibodies to catalyze this same iso-merization reaction (23)-+(24) (Fig. 12)(JACKSONt al., 1988). His kinetic analysis ofone purified antibody revealed that it increas-es the entropy of activation of the reaction by

12 cal mol-' K -l (Tab. 1, Antibody 11F1-2E11,AE 13.2), and gives a rate enhancementof lo4.He suggested that this TSA induces acomplementary combining site in the abzymethat constrains the reactants into the correctconformation for the [3,3]-sigmatropic reac-tion and designated this strategy as an "entrop-ic trap".

HILVERT'Sroup used the same hapten (26)with a different spacer to generate an antibodycatalyst which has very different thermody-

namic parameters. It has a high entropy of ac-tivation but an enthalpy lower than that of thewild type enzyme (Tab. 1, Antibody 1F7, AE13.2) (HILVERTet al., 1988; HILVERTandNARED, 988). WILSON as determined an X-ray crystal structure for the Fab' fragment ofthis antibody in a binary complex with its TSA(HAYNES,t al., 1994) which shows that aminoacid residues in the active site of the antibodycatalyst faithfully complement the compo-nents of the conformationally ordered transi-tion state analog (Fig. 13) while a trapped wa-

Tab. 1. Kinetic and Thermodynamic Parameters for the Spontaneous, Enzyme Catalyzed and AntibodyCatalyzed Conversion of Chorismic Acid (23) into P rephenic Acid (24)

Catalyst Relative AG z AH: ASz K , 23 k,,, 23 Ki 26

Rate [kcal mol-'1 [kcal mol-'1 [cal mo l-' K-'1

Spontaneous" 1 24.2 20.5 - 2.9Chorismate mutaseb 3. o6 15.9 15.9 0 45pM 1.35s- ' 75nM

Antibody 1F7" 250 21.3 15.0 - 2 49 pM 0.023 min-' 600 nM11F1-2Elld 10,OOO 18.7 18.3 - 1.2 260 pM 0.27 min -' 9 pM

a At 25 "C; E.coli enzym e at 25 "C; pH 7.5,14"C ; pH 7.0,lO"C.



a\NH

H - - -

4


b

Fig. 13. Schematic diagrams of X-ray crystal structures show the hy drogen bonding (dashed lines) and elec-trostatic interactions between the transition state analog (26) (in grey) with relevant side chains of a antibody1F7 (HAYNESt al., 1994) and b the active site of the E. coli enzyme (LEEet al., 1995).

ter molecule is probably responsible for theadverse entropy of activation. Thus it appearsthat antibodies have emulated enzymes in

finding contrasting solutions to the same cata-lytic problem.Further examples of catalytic antibodies

that are presumed to control rotational entro-py are AZ-28, that catalyzes an oxy-Cope[3.3] -sigmatropic rearrangement (AE 13.1)(BRAISTEDnd SCHULTZ,994; ULRICH t al.,1996) and 2E4 which catalyzes a peptide bondisomerization (AE 13.3) (GIBBS, t al., 1992b;LIOTTA, et al., 1995). Perhaps the area for thegreatest opportunity for abzymes to achievecontrol of rotational entropy is in the area of

cationic cyclization reactions (LI,et al., 1997).The achievements of the LERNER roup in thisarea (AE 15.1-15.4) will be discussed later inthis article (Sect. 5.3).

Translational E ntrop y

The classic example of a reaction that de-mands control of translational entropy is sure-ly the Diels-Alder cycloaddition. It is acceler-ated by high pressure and by solutions 8M inLiCl (BLOKZIJLnd ENGBERTS,994; CIOBA-NU and MATSUMOTO,997; DELL, 1997) and

proceeds through an entropically disfavored,

highly ordered transition state, showing largeactivation entropies in the range of -30 to-40 cal mol-l K-' (SAUER, 966).

While it is one of the most important andversatile transformations available to organicchemists, there is no unequivocal example of abiological counterpart. However, two cell freefractions from Alternaria solani have beenpurified (KATAYAMAt al., 1998) which are ca-pable of converting prosolanapyrone I1 to sol-anapyrones A and D (OIKAWA t al., 1999).This reaction can be rationalized as an oxida-tion followed by an intramolecular Diels-Al-der reaction (OIKAWAt al., 1998,1997).In ad-dition there are many natural products, such as

himgravine (BALDWINt al., 1995) and themanzamines (BALDWINt al., 1998) which areplausibly formed biosynthetically by intra-molecular Diels-Alder reactions ( CHIHARAand OIKAWA,998),albeit no enzyme has beenimplicated in these latter processes so far.Hence, attempts to generate antibodies whichcould catalyze this reaction were seen as animportant target.

The major task in producing a "Diels-Alder-ase" antibody lies in the choice of a suitablehaptenic structure because the transition state

for the reaction resembles product more close-




ly th an re actan ts (Fig. 14). The reaction p rod-uct itself is an inappropriate hapten because itis likely to result in severe product inhibitionof th e catalyst, thereby preventing turno ver.

Tetrachlorothiophene dioxide (TCT D) (27)reacts with N -ethylmaleimide (28) to give anunstable, tricyclic intermediate (29) that ex-trudes SO, spontaneously to give a dihy-drophthalimide as the bicyclic adduct (30)

(RAASCH,980).This led to th e design of h ap-ten as a bridged dichlorotricycloazadecenede-rivative (31) which closely mimics the high en-ergy intermediate (29) ,while being sufficientlydifferent from the product (30) o avoid thepossibility of inhibition by end-product (HIL-

Several antibodies raised to the hapten (31)

accelerated the Diels-Alder cycloadditionbetween (27) and (28) (Fig. 14). Th e most effi-cient of these, 1E9, perform s multiple turn-overs showing that product inhibition has beenlarge ly avo ided. Co mp arison of k,,, with thesecond order ra te con stant for the uncatalyzedreac tion (k,,,,,=0.04 M -' m in-', 25°C) givesan effective molarity, EM: of 110 M (A E 17.1)(HILVERTt al., 1989).This value is several or -ders of magnitude larger than any attainable

concentration of substrates in aqueous solu-tion, therefore the antibody binding site con-fers a significant entropic advantage over thebimolecular Diels-Alder reaction.

A number of further examples of Diels-Al-der catalytic antibodies have been described(A E 17.2-17.5) and they must necessarily ben-efit from the same entropic advantage overspontaneous reactions, albeit without HIL-

VERT'S clever approach to avoiding productinhibition. The ir success in achieving control ofregio- and stereochemistry will be discussed

later (Sect. 5.1).Of grea ter long-term significance is the con-

trol of translational entropy for antibody cata-lyzed synthetic purposes. BENKOVIC'Sescrip-tion of an an tibody ligase cap able of joining anactivated am ino acid (e.g., 32) to a second ami-no acid t o give a dipeptide and t o a dipeptide

VERT, et al., 1989).

The EM is equivalent to the concentration of sub-strate that would be ne eded in the un catalyzed reac-tion to achieve the sam e rate as achieved by the an-tibody ternary complex (KIRBY, 980).

27 1 2 8

0 I *9

ClVCl

31

Fig. 14. The Diels-Alder cycloaddition of TCTD

(27) and N-ethylphthalimide (28) proceeds throughan unstable intermediate (29) which extrudes SO2

spontaneously to give the dihydrophthalimide ad-duct (30).Hapten (31) was designed as a stable mi-mic of (29) that would be sufficiently different fromproduct (30) to avoid product inhibition of the anti-body catalyst.




2.4 Substrate Desolvatione.g., 33) to give a tripeptide with only lowproduct inhibition is particularly significant(Fig. 15,AE 18.4) (SMITHRUDt al., 1997).Anti-body 16G3 can achieve 92% conversion of

substrates for tripeptide formation and 70%for tetrapeptide synthesis within an assay timeof 20 min. A concentration of 20 pM antibodycan produce a 1.8 mM solution of a dipeptidein 2 h. The very good regiocontrol of the cata-lyzed process is shown by the 80:1 ratio forformation of the programmed peptide (34)compared to the unprogrammed product (35)

whereas the uncatalyzed reaction gives a 1 1ratio.

P h j + 32

,3-Indolyl)

! H

33C P

I 16G3

,3-Indolyl)

Ph

35

Fig. 15.Antibody 16G3 catalyzed peptide bond for-

mation predominantly yields the peptide (34)(34:35, 80:l), whereas the uncatalyzed reaction

gives a 1 1ratio of the two peptides.

A classic example of rate acceleration bydesolvation is the Kemp decarboxylation of 6-

nitro-3-carboxybenzisoxazole36).The chan-ge from water to a less polar environment caneffect a 107-fold rate acceleration which hasbeen ascribed to a combination of substratedestabilization by loss of hydrogen bonding tosolvent and transition state stabilization in adipolar aprotic solvent (KEMP et al., 1975).Both HILVERTnd KIRBY ave sought to gen-erate abzymes for this process (AE 9.1) (LE-WIS et al., 1991;SERGEEVAt al., 1996)HILVERTgenerated several antibodies using TSA (37)

and the best, 25E10, gave a rate acceleration of

23.2. lo3 for decarboxylation of (36), ompar-able to rate accelerations found in other mixedsolvent systems but much less than for hexa-methylphosphoric triamide (lo8, Fig. 16). Inparticular, it is of some concern that the K , forthis antibody is as high as 25 mM, which re-flects the tenuous relationship between thehapten design and the substrate/transitionstate structure. Unfortunately, apparently bet-ter designed TSAs, e.g. (38) SERGEEVAt al.,1996), fared worse in outcome, possiblythrough the absence of a countercation in thebinding site. This may offer an opportunity forprotein engineering to induce the presence ofan N,N,N-trimethyl-lysine residue in the activesite to provide a non-hydrogen bonding saltpair.

Selenoxide syn-eliminations are anothertype of reaction favored by less polar solvents(REICH, 979).The planar 5-membered, pericy-clic transition state for syn-elimination of (40)was mimicked by the racemic proline-basedcis-hapten (39) to give 28 monoclonal antibod-

ies (AE 8.5) (ZHOU t al., 1997). Abzyme SZ-cis-42F7 converted substrate (40) exclusivelyinto trans-anethole (41) with an enhancementratio (ER) of 62 (R=Me, X =NOz) and with alow K , of 33 pM. Abzyme SZ-cis-39C11 gavea good acceleration, k,,, 0.036 min-', k,, , /K,2,400 M-' min-' (substrate 40, R = H = X )comparable to the rate in 1,2-dichloroethanesolution. Unexpectedly, the catalytic benefitappears to be mainly enthalpic for both theantibody and for the solvent switch, as shownby the data in Tab. 2.




J0,O 37

H

/Me0

39

Fig. 16.TSAs (37), 38)for Kemp decarboxylation of6-nitro-3-carboxybenisoxazole 36) nd for selen-

oxide elimination (39).

2.5 Supplementation of Chemical

Functionality

Several antibodies have been modified toincorporate natural or synthetic groups to aidcatalysis (POLLACKt al., 1988). POLLACKndSCHULTZeported the first example of a semi-synthetic abzyme by the introduction of animidazole residue into the catalytic sitethrough selective modification of the thiol-containing antibody MOPC315 (POLLACKndSCHULTZ,989). This yielded a chemical mu-tant capable of hydrolyzing coumarin ester(48) with k,,, 0.052 min-l at pH 7.0,24 "C. In-corporation of the nucleophilic group alonewas previously shown to accelerate hydrolysisof the ester by a factor of lo4 over controls(POLLACKt al., 1988).

The process of modification is shown in Fig.17. Lys-52 (42) is first derivatized with 4-thio-butanal and then an imidazole is bondedthrough a disulfide bridge into the active site(47). This can now act as a general basehucle-ophile in the hydrolysis of (48), as was estab-lished first by the pH-rate profile and then bycomplete deactivation of the antibody by di-

ethyl pyrocarbonate (an imidazole-specific in-activating reagent).The first success in sequence-specific pep-

tide cleavage by an antibody was claimed byIVERSONIVERSONnd LERNER, 989) usinghapten (49) which contains an inert Co(II1)-(trien) complexed to the secondary amino acidof a tetrapeptide (Fig. 18). This approachaimed towards the elucidation of monoclonalantibodies with a binding site that could simul-taneously accommodate a substrate moleculeand a kinetically labile complex such as

Zn(II)(trien) or Fe(III)(trien), designed toprovide catalysis. Much early work by BUCK-INGHAM and SARGESON had shown that suchcobalt complexes are catalytic for amide hy-drolysis via polarization of the carbonyl group,through nucleophilic attack of metal-boundhydroxide, or by a combination of both pro-cesses (SUT~ONnd BUCKINGHAM,987; HEN-DRY and SARGESON, 1990).

Peptidolytic monoclonal antibody 287F11was selected for further analysis. At pH 6.5,cleavage of substrate (50) was observed with a

variety of metal complexes. The Zn(II)(trien)



3 Spontaneous FeaturesofAntibody Catalysis 421

Tab. 2. Parameters for the syn-Eliminationof Selenoxide.(39)R=X =H) in Water, DCM,and Catalyzed byAntibody SZ-cis-39C11

Catalyst AG ; AH: AS kcatlKm kcat ( k o d ER

[kcal mol-'1 [cal mol-' K-'1 [M-' min-'1 [min-l]Water 26.3 f0.15 26.3f 0.15 + 0.014f .47 1.6E-5b

(DCM) 21.8f0.5 20.3f0.5 -4.8f1.7 4.4E-2 2,750SZ-cis-39C11 22.2f1.2 19.7f1.2 -7.8f4.1 2,400 3.5 E-2 2,200

a 25 "C; here and elsewhere exponents are expressed as, e.g.,E-5 .. . o-'.

complex was the most efficient with 400 urn-overs per antibody combining site and a turn-over number of 6.1Op4 -' (Fig. 18).While this

approach is undoubtedly ingenious, there aresome doubts about its actual performance.Thesite of cleavage of the substrate is not, as wouldbe expected from the design of the hapten,between the N-terminal phenylalanine andglycine, but rather it is between glycine and theinternal phenylalanine.

In an equivalent approach, MAHYhas gen-erated a peroxidase-like catalytic antibodywhich binds an iron(II1) ortho-carboxy substi-tuted tetraarylporphyrin, achieving oxidation

rate up to 540 min-l with ABTS as a cosub-strate (DE LAUZONt al., 1999).A major achievement in augmenting the

chemical potential of antibodies has been inthe area of redox processes. Many examplesnow exist of stereoselective reductions, partic-ularly recruiting sodium cyanoborohydride(AE under 22). A growing number of oxida-tion reactions can now be catalyzed by ab-zymes, with augmentation from oxidants suchas hydrogen peroxide and sodium periodate(AE under 21).

3 Spontaneous Features of

Antibody Catalysis

Despite the attention given to the pro-grammed relationship of hapten design andconsequent antibody catalytic activity, thereare now many examples where the detailed ex-

amination of catalysis reveals mechanistic fea-

tures that were not evidently design featuresof the system at the outset. Such discoveriesare clearly a strength rather than a weakness

of the abzyme field, and two of these out-turnsare described below.

3.1 Spontaneoh CovalentCatalysis

The nucleophilic activity of serine in the hy-drolysis of esters and amides by many enzymesis one of the classic features of covalent cataly-sis by enzymes. So it was perhaps inevitable

that an antibody capable of catalyzing the hy-drolysis of a phenyl ester should emerge hav-ing the same property. SCANLANas providedjust that example with evidence from kineticand X-ray structural analysis to establish thatthe hydrolysis of phenyl (R)-N-formylnorleu-cine (51)proceeds via an acyl antibody inter-mediate with abzyme 17E8 (AE 2.3) (ZHOU tal., 1994; Fig. 19). The antibody reaction has abell-shaped profile corresponding to ionizablegroups of pK, 10.0 and 9.1. Based on X-rayanalysis, they appear to be respectively LysH9'

and probably TyrHIO1.This system is deemed toactivate SerH99 s part of a catalytic diad with

(52). In addition to the kinetic andstructural evidence for this claim, a trappingexperiment with hydroxylamine generated amixture of amino acid and amino hydroxamicacid products from substrate (51) n the pres-ence of antibody.

In a similar vein, antibody NPN43C9 ap-pears to employ a catalytic histidine, asa nucleophilic catalyst in the hydrolysis of a p -

nitrophenyl phenylacetate ester (AE 2.8)

(GIBBSt al., 1992a; CHEN, t al., 1993).



422 11 Catalytic An tibodie s

H3N@ 3.2 Spontaneous Metal Ion

I Catalysis

JANDAand LERNERried to show that a met-al ion or coordination complex need not be in-cluded within the hapten used for the induc-tion of abzymes in order that they could (1)bind a metallo-complex and thereby (2) pro-vide a suitable environment for catalysis (WA-DE et al., 1993). The pyridine ester (54) wasscreened as a substrate for 23 antibodies raisedagainst the hapten (53) (Fig. 20). Antibody84A3 showed catalytic activity in the hydroly-

(yJLys52:2(1) O Z N a

(2 ) NaCNBH,

( 3 ) Dithiothreitol

f(CHz)FHO

43-SH

CL,,H

d%7

qn

sis of (54) only in the presence of zinc,-with-arate enhancement of 12,860 over the spontan-eous rate and 1,230 over that seen in the pres-ence of zinc alone. Other metals, such as Cd" ,Co2+,Ni2+,were inactive.The affinity of 84A3for the substrate was high (3.5 pM ) whereasthe affinity for zinc in the presence of substratewas much lower (840 pM). his is far weakerthan any affinity of real potential for the re-cruitment of metal ion activity into the catalyt-ic antibody repertoire (plasma [Zn"] is17.2 pM). However, we can anticipate that ap-plication of site-directed mutagenesis and

computer assisted design will be targeted o nthis problem with real expectation of success.LERNER'Sroup has shown that the catalytic

performance of antibody 38C2 in catalyzing analdol reaction is improved in the presence ofone equivalent of Pd(l1) (FINNet al., 1998).38C2 binds Pd specifically( K d - 1pM ) and anallosteric effect is observed. This study sug-gests that the metal does not bind into the cat-alytic site, but induces a conformationalchange promoting a closer binding to the sub-strate.

Fig. 17. A semi-synthetic abzyme. Selective derivatization of lysine-52 inthe heavy chain of MOPC315 (42) gives an imine which is reduced to anamino disulfide with sodium cyanoborohydride. Reduction to the thioland a series of disulfide exchanges gives an imidazole derivatized abzymecapable of improved hydrolysis of the coumarin ester (48) (k,,, -0.052min- ).



4 How Good are Catalytic Antibodies? 423

Fig. 18. A metal complex 49 used as hapten to raise antibodies capable of incorporating metal cofactors to

facilitate the cleavage of 50 at the position indicated.

In view of the great general importance ofmetalloproteinases, it seems inevitable thatfurther work will be directed at this key areaeither by designed or opportunistic incorpora-tion of metal ions into the catalytic apparatusof abzymes.

SerHg9 ti+6isH3’

4 How Good are Catalytic

Antibodies?

52

Fig. 19. The hydrolysis of (R)-N-formylnorleucine(51) proceeds via an acyl antibody intermediate withabzyme 17E8(52). (ZHOUet al., 1994).

Fig. 20. The pyridine ester (54) was screened as asubstrate for 23 antibodies raised against the h apten

(53).

In the first years of abzyme research, acylgroup transfer reactions have appeared in amajority of examples. Many of these end-eavors have been based on mimicry of a highenergy, tetrahedral intermediate that lies alongsuch reaction pathways (Sect. 2.1) and which,though not truly a “transition state analog”,provides a realistic target for production of astableTSA.Many of these haptens were basedon four coordinate phosphoryl centers.

In 1991, JACOBSnalyzed 18 examples ofantibody catalysis of acyl transfer reactions as

a test of the P~ULINGoncept, i.e., deliveringcatalysis by TS+ stabilization. The range of ex-amples included the hydrolysis of both aryland alkyl esters as well as aryl carbonates. Insome cases more than one reaction was cata-lyzed by the same antibody, in others the samereaction was catalyzed by different antibodies(JACOBS,991).

Much earlier, WOLFENDENnd THOMPSON

(WESTERICK and WOLFENDEN, 1972; THOMP-

SON, 1973), established a criterion for enzymeinhibitors working as TSAs. They proposed

that such activity should be reflected by a line-




ar relationship between the inhibition con-stant for the enzyme Ki and its inverse secondorder rate constant, K,lk,,,, for pairs of inhib-itors and sub strates that differ in structure on-

ly at the TS A hb str at e locus. That has beenwell validated, inter alia, for phosphonate in-hibitors of thermolysin (BARTLETTnd MAR-LOWE, 1983) and pepsin (BARTLETTnd GIAN-GIORDANO, 1996). In orde r to apply such a cri-terion to a range of catalytic antibodies, JA-COBS assumed firstly that the spontaneous+hy-drolysis reaction proc eeds via the same TS+ asthat for the antibody mediated reaction andsecondly that all corrective factors due t o me-dium effects are constant. By treating the hy-drolyses reactions a s a pseudo-first-order pro-

cess, one can de rive a sim ple relationship withapproximations of K Ts an d K s to provide amathematical statement in terms of Ki, K,,k,,,, and k,,,,, (Fig. 21) (WOLFENDEN, 1969;JENCKS, 1975; BENKOVICt al., 1988; JACOBS,

1991).A log-log plot using Ki, K,, kat, and k,,,,

data from the 18separate cases of antibody ca-talysis exhibited a linear correlation over fourorders of magnitude and with a gradient of0.86 (Fig. 22)5. Considering the assumptions

made, this value is sufficiently close to unity t osuggest that the antibodies do stabilize thetransition state for their respective reactions.

However, even the highest kcatlkuncatalue oflo6 in this series (TRAMONTANOt al., 1988)barely compares with enhanc eme nt ratios seenfor weaker enzyme catalysts (LIENHARD,

1973).The fact that many values of K J K i fall be-low the curve (Fig. 22) suggested that interac-tions between the antibody and the substrateare largely passive in terms of potential cata-lytic benefit. This conclusion expo ses a limita-tion in the design of haptens, if solely based onthe transition state concept. It is well knownthat enzymes utilize a wide rang e of devices toachieve catalysis as well as dynamic interac-tions to guide substrate towards the transitionstate, which is then selectively stabilized. How -

ever, as has been illustrated above, the originalconcept of transition state stabilization hasbeen augmented by a range of further ap-proaches in the generation of catalytic anti-bodies and with considerable success.

It may also be worth mentioning here that manyearly estimates of Kd or the affinity of the antibodyto their TSA were u pper limits, being ba sed on inhi-

bition kinetics using concentration sof antibody thatwere significantly higher than the true K , being de-termined.

Fig. 21.A thermodynamic cycle linked to transition state theory gives an equation relating the enhance mentratio for a biocatalyz ed process to the ratio of equ ilibrium constants for the com plex between the biocatalystand (i) substrate and (ii) the transition state for the reac tion.These two v alues can be estimated as K , and Kifor the TSA,respectively.



4 How Good are Ca talytic Antibod ies? 425

6 ,

0Fig. 22. JACOBS’correlation be-tween the enhancem ent ratio(k,,tlk,,,,t) an d the relative affi-nity for th e TSA with respect tothe substrate (K,IKi) (JACOBS,1991).Th e slope is an unweight-

I I I 1

1 2 3 4 5 6 7

ed linear regression analysis. log ( k t 1 uncat)

E

Reaction Catalyzed- 1 v 1

4

3.1 .“ I

1

0

0

1 rP

y = 0.462~ 1.801 r =0.597

-1 0 1 2 3 4 5 6 7 8

Ester Hydrolysis

Ether Cleavage

Claisen

Decarboxylation

Elimination

Miscellaneous

Ester Synthesis

Amide Synthesis

DieldAlder

Fig. 23. The Stewart-Benkovic plot of rate enhanc eme nt vs relative binding of substrate andTSA fo r 60 abzyme catalyzed reactions (STEWART nd BENKOVIC,995).The theoretical unitslope (----)diverges from the calculated linear regression slope (-) for these data (its equati-

on is shown).



426 I 1 CatalyticAntibodies

with a correlation coefficient for the linear fitof only 0.6 and with a slope of 0.46 very differ-ent from the “theoretical slope” of unity. Per-haps of greatest significance are the many pos-

itive deviations from the general pattern.These appear to show that antibody catalysiscan achieve more than is predicted from catal-ysis through transition state stabilizationalone.

5 Changing the Regio-

and Stereochemistry of

Reactions

The control of kinetic vs thermodynamicproduct formation can often be achieved bysuitable modificaton of reaction conditions. Afar more difficult task is to enhance the forma-tion of a disfavored minor product to the detri-ment of a favored major product, especiallywhen the transition states for the two process-es share most features in common. This goal

has been met by antibodies with considerablesuccess,both for reaction pathways differing inregioselectivity and also for ones differing instereoselectivity.In both situations, control ofentropy in the transition state must hold thekey.

5.1 Diels-Alder Cycloadditions

In the Diels-Alder reaction between an un-symmetrical diene and dienophile up to eight

stereoisomers can be formed (MARCH, 992b).It is known that the regioselectivity of theDiels-Alder reaction can be biased so thatonly the four ortho adducts are produced (Fig.24) through increasing the electron-withdraw-ing character of the substituent on the dieno-phile (DANISHEFSKYnd HERSHENSON,979).However, stereochemical control of the Diels-Alder reaction to yield the disfavored exo-

products in enantiomerically pure form hasproved to be very difficult. GOUVERNEURndco-workers were interested in controlling theoutcome of the reaction between diene (55)

si-face re-face

R‘

R’

rim approach

R’ R1

si-face re-face

Fig. 24. Enantio- and diastereo selectiv ity of theDiels-Alder reaction for ortho-approach.

and N,N-dimethylacrylamide (54) (Fig. 25)(GOUVERNEURt al., 1993).

They had shown experimentally that the un -catalyzed reaction gave only two stereoiso-mers: the ortho-endo (cis) (5 7 ) and the ortho-

ex0 (trans) (58) adducts in an 85 :15 mixture.

This experimental observation was under-pinned by ab initio transition state modelingfor the reaction of N-butadienylcarbamic acid(59) with acrylamide(a),hich showed thatthe relative activation energies of the orrho-

endo and ortho-ex0 transition states were ofconsiderably lower energy than the meta-endo

and meta-exo transition structures (not illus-trated) (Tab. 3). The design of hapten was cru-cial for the generation of catalytic antibodiesto deliver full regio- and diastereoselectivity.Transition state analogs were therefore de-

vised to incorporate features compatible with



5 Changing the Regio- and Stereochemistry of Reactions 427

OYNH

04COzNa 55

O Y N H OYNH

C02Na f57 COzNa f58

O Y HOH 59

Fig. 25. The Diels-Alder cycloaddition between the

diene (55) and the dienophile (56) ields two diaste-

reoisomers (57) and (58).Attenuated substrate ana-

logs (59) and (60)were used in molecular orbital cal-

culations of this reaction.

either the disfavored endo (63) or favored ex0

(65) transition states (Fig. 26) (AE 17.5).HILVERTeveloped a new strategy to mini-

mize product binding to the abzyme (HILVERT

et al., 1989). Based on the knowledge that thetransition state for Diels-Alder processes isvery product-like, he designed haptens (64)and (66) to mimic a high energy, boat confor-mation for each product, thereby ensuringavoidance of product inhibition.’IIyo of the monoclonal antibodies pro-

duced, 7D4 and 22C8, proved to be completelystereoselective, catalyzing the endo and theex0 Diels-Alder reactions, with a k,,, of3.44. and 3.17. min-’, respectivelyat 25°C. That the turnover numbers are low

was attributed in part to limitations in transi-tion state representation: modeling studies hadshown that the transition states for both theex0 and endo processes were asynchronouswhereas both TSAs (64) and (66) were basedon synchronous transition states (GOUVER-

In a further experiment, compounds (67)

and (68)were perceived as freely rotating hap-tens for application as TSAs for the sameDiels-Alder addition (Fig. 27). As expected,

each proved capable of inducing both endo-and exo-adduct forming abzymes. It can benoted that (67)produced more “exo-catalysts”(6 out of 7) whereas (68) avored the produc-tion of “endo-catalysts” (7 out of 8) though it isdifficult to draw any conclusion from this ob-servation (AE 17.5) (YLI-KAUHALUOMAt al.,1995;HEINEt al., 1998).

NEUR et al., 1993).

5.2 Disfavored Regio- and

Stereoselectivity

Reversal of Kinetic Control in a Ring ClosureReactionWhen several different outcomes are possiblein a reaction, the final product distribution re-flects the relative free energies of each transi-tion state when the reaction is under kineticcontrol (SCHULTZ and LERNER, 1993).BALDWIN’Sules predict that for acid catalyzedring closure of the hydroxyepoxide (70) thetetrahydrofuran product (69) arising from 5-exo-tet attack will be preferred (Fig. 28)

(BALDWIN,976). By raising antibodies to the



428 11 Catalytic Antibo dies

Tab. 3. Calculated Activation Energies of the Transition Stru cture s Relative to Rea ctan ts for the Reaction of

N-(1-Butadieny1)-carbamicAcid (59) with Acrylam ide (60)~

Transition State Geom etry Calculated Activation E nerg y [kcal mol-’1

RHF/3-21G 6-31G.13-21G

ortho-endo 27.3 40.80ortho-ex0 28.84 42.70meta-endo 29.82 42.88meta-exo 30.95 43.94

endo-Haptenndo-Transition state

( = + L 2

R’ r61 62

63 64

em-Transition state exo-Hapten

R2 = CONMe2 65

Fig. 26. Haptens (64) nd (66) ere designed as analogs of the favor ed e n d o

1 HXR2165

sition state s, respectively.

66

- (63)or disfavored exo-(65) ran-

.C0NMe2

H

67 R = (CHz),COzH

68R =4-carboxyphenyl

Fig. 27. An alternative strategy for eliciting Diels-Aldera se antibodies has employed th e freely rotat-ing ferrocenes (67) and (68) s TSAs.

charged hapten (72), JANDA and co-workersgenerated an abzyme which accelerated 6 - e x 0

attack of the racemic epoxide to yield exclu-sively the disfavored tetrahydropyran product(71) and in an enantiomerically pure form(AE 14.1) (JANDAt al., 1993).

This work reveals that an antibody can se-lectively deliver a single regio- and stereo-chemically defined product for a reaction inwhich multiple, alternative transition states areaccessible and can also selectively lower theenergy of the higher of two alternative transi-

tion states (NAand HOUK,996).



5 Changing the Regio- an d Stereochirnistry of Reactions 429

Fig. 28. The monoclonal antibody 26D9, generatedto the N-oxide hapten (72), catalyzed the 6-exo-tetring closure of (70) regioselectively t o yield the dis-favored tetrahydrop yran product (71). This is a for-mal violation of BALDWIN'Sules which predict aspontan eous 5-exo-tet process to generate tetrahy-drofuran derivative (69).

Syn-Elimination of p-FluoroketonesThe base-catalyzed p-elimination of HF fromthe ketone (73) normally gives the thermody-namically more stable (E)-alkene (74) via astaggered conformation in the transition state(Fig. 29). Hap ten (76) was designed t o enforcethe syn-coplanar conformation of the phenyland benzoyl functions in the transition statean d so catalyze the disfavored syn-elimination

of (73) to give the (Z)-a,@-unsaturatedketone

73ISpontaneous I Disfavored

anti-Elimination

syn-EliminationI I

Fig. 29. The elimination of HF from th e P-fluoroke-

tone (73) is catalyzed by antibody 1D 4, elicited t ohapten (76), to form the disfavored (Z )-a$-unsatu -rated ketone (75). This contrasts the spontaneousprocess in which a n anti-elimination reaction yieldsthe (E)-a,P-unsaturated ketone (74). The syn-eclipsed conformation of (76) is shaded.

(75) (Fig. 29). Preliminary estimations of the

energy difference between the favored anddisfavored processes are close to 5 kcal mol-l(CRAVA'ITt al., 1994), though this value is ex-ceeded in the antibody catalyzed rerouting ofcarbamate hydrolysis from ElcB to BA,2(Sect. 8.2,AE 4.3) (WENTWORTHt al., 1997).Antibody 1D4, raised t o hapten (76) and usedin 15% DMSO at pH 9.0, gave exclusively the(Z)-alkene (75) with K , 212 pM an d k,,,2.95.1Op3 min-l. Un der the same conditions,kobs s 2.48-1OP4 m in- l for formation of (74)and imm easurably slow for the (undetectable)

formation of (75) (AE 8.1).




5.3 Carbocation Cyclizations

The cationic cyclization of polyenes givingmulti-ring carbocyclic compounds with many

sterically defined centers is one of th e m ore re -markable examples of regioselective and ster-eoselective enzyme control which has provid-ed a major challenge for biomimetic chemistry(JOHNSON,968). This system provides an ex-cellent opportunity for the application of re-gio- and stereoc ontro l by catalytic antibodies.

LI and co-workers have discussed the use ofcatalytic antibodies t o c ontrol the reactivity ofcarbocations (LIe t al., 1997).A t an entry level,antibody 4C6 was raised to hapten (77) andused to conv ert acyclic olefinic sulfonate e ster(78) into cyclic alcohol (79) (98%) with verylittle cyclohexene (2% ) produced (Fig. 30,A E15.1) (Lr, et al., 1994).

H

77

78

4C6(9 8 % yield)

Me,PhSib 9

Fig. 30.The N-oxide hapten (77) was used to elicitm A b 4C6 which catalyzed the cyclization of th e

allylsilane (78) to form the cyclohexanol(79).

Moving closer to a cationic transition statemimic, HASSERO DTnd co-workers used theamidinium ion (80) as a TSA for cyclization ofthe arenesulfona te ester (81) (Hasserod t et al.,

1996). One antibody raised to this hapten,HA1-17G8, catalyzed the conversion of sub-strate (81) into a mixture of the 1,6-dimethyl-cyclohexene (82) and 2-methylene-1-methyl-cyclohexane (83) (Fig. 31) ( A E 15.3). By con-trast, the uncatalyzed cyclization of (81)

formed a mixture of 1,2-dimethyl-cyclohexa-nols and a little 1,2-dimethylcyclohexene. vi-dently, the antibody both excludes water fromthe transition sta te and also controls the loss ofa pro ton following cyclization.

P N ’

a82

17G8

a83

Fig. 31. Antibody HA1-17G8 raised against TSA(80) catalyzed the cyclization of (81) o give (82) and

(83).




0

While cyclopentanes have also been pro-duced by antibody catalyzed cyclization (A E15.2) (LI et al., 1996), much the most strikingexam ple of cationic cyclization by antibod ies is

the formation of the decalins (85),(a),nd(87) (Fig. 32). The trans-decalin epoxide (88)

( t l lZ100 h at 37°C) was employed as a m ixtureof two enantiomeric pairs of diastereoisomersas a TSA to raise antibodies, among which wasproduced HA5-19A4 as the best catalyst forcyclization of substrate (84) (A E 15.4) (HAS-

Sufficient substra te (84) was transformed t ogive 10 mg of mixed products. Th e olefinicfraction (70%) was predominantly a mixtureof the three decalins (85), (%),and (87) (2 :3 :1

ratio) and formed along with a mixture of cy-clohexanol diastereoisomers (30%). More-over, the decalins were formed with enantio-meric excesses of 53,53,and 80%,respectively.It is significant tha t the (Z)-isomer of (84) isnot a substrate for this antibody.

The ionization of the arenesulfonate is thefirst step catalyzed by the antibo dy which gen-erates a carbocation as part of a process thatshows an ER of 2,300 with K , 320 pM.The re-sulting cation can the n either cyclize to decal-

ins in a concer ted process (as in transition sta te89) or in two stepwise cyclizations. The forma-tion of significant amounts of cyclohexanolsseem s to favor the latter of possibilities. Mostinterestingly, n hibition studies suggest that th eisomer of the haptenic mixture that elicitedthis antibody has structure (88), herefore, lo-

cating the leaving group in an axial position.This is contrary to the STORK-ESCHENMOSERconcept of equatorial leaving group andpresents a challenge for future examination(STORK nd BURGSTAHLER,955; ESCHEN-

It is an exciting prospect that catalysts of thisnature may lead to artificial enzymes capableof processing natural and unnatural polyiso-prenoids to generate various useful terpenes.

SERODT et al., 1997).

MOSER et al., 1955).

6 Difficult Processes

As scientists studying catalytic antibodies

have gained mo re confidence in the power of

HA5-19A4pH 7.0(biphasic)I

CO,H

88

I

Fig. 32. Formation of isomeric decalins (85)-(87) bycyclization of a terpenoid alcohol catalyzed by anti-body HAS-19A4 raised to hapten (88).The transi-tion state (89) has the leaving group in the equatori-al position, as favored by the STO RK -ESCH ENM OSE Rhypothesis.




abzymes, h eir attention has turned from reac-tions having moderate to good feasibility tomore demanding ones. Their work has on th eone hand tackled more adventurous stereo-

chemical problem s and on the other hand theyare seeking to catalyze reactions whose spon-taneous rates are very slow indeed. Examplesof both of these areas a re discussed in this sec-tion.

6.1 Resolution of Diastereoisomers

Antibodies generally show very good selec-tivity into the recognition of their antigens anddiscriminate against regio- or stereoisomers of

them. This results from a combination of th einherent .chirality of proteins and the refinedresponse of the immune system (PLAYFAIR,1992).In extension, this cha racter suggests thata catalytic antibody sh ould be cap able of simi-lar discrimination in its choice of substrate a ndthe transition state it can stabilize, as deter-mined by th e h apten used for its induction. A salready shown above, the murine immunesystem can respond to a single member of amixture of stereoisomers used for immuniza-

tion (Sect. 5.3). Such discrimination has beenexemplified in antibody catalyzed enantiose-lective ester hydrolysis (JANDAet al., 1989;POLLACK t al., 1989; SCHULTZ, 989) andtransesterification reactions (WIRSCHING tal., 1991;JACOBSENt al., 1992).

One study has made use of abzyme stereo-selectivity to resolve the four stereoisomers( R , R ' , S , S ' ,R , S ' , an d S , R ' ) of 4-benzyloxy-3-fluoro-3-methylbutan-2-01 (94)-(97) throughthe antibody m ediated hydrolysis of a d iaster-eoisomeric mixture of their phenacetyl esters

(90)-(93) (KITAZUMEt al., 1991b).A ntibodieswere raised separately to each of four phos-phonate diastereoisomers (98)-(lOl), corre-sponding to th e fou r possible transition statesfor the hydrolysis of the four diastereoisom er-ic esters (Fig. 33) (A E 1.12). Each antibody op-erated on a mixture of equal parts of the fourdiastereoisomers as substrate t o give each al-cohol in ca. 23% yield, with >97% eelde, andleaving the three other stereoisomers un-changed. Following successive incubation withthe four antibodies in turn, the m ixture of dia-

stereoisomers could effectively be separated

completely (Tab. 4). In a similar vein, KITA-ZUME also resolved the enantiomers of 1 , l J -trifluorodecan-2-01 with 98.5% ee (A E 1.11)(KITAZUMEt al., 1991a).

6.2 Cleavage of Acetals and

Glycosides

The synthesis, modification, and degrada-tion of carbohydrates by antibody catalyzedtransformations are subjects of active investi-gation. Preliminary studies have reported anti-body hy drolysis of mod el glycoside substr ates( Y u e t al., 1994) while the regio- and stereose-lective deprotection of acylated carbohydrates

has been achieved using catalytic antibodies

I 90-931hCHZCOzH

' F98-101

Fig. 33. Four stereoisomeric alcohols (94)-(!W) wereseparated by selective hydrolysis of their respectivephenacetyl esters using four antibod y catalysts, eachraised in response to a discrete stereoisomeric phos-

phonate hapten (98)-(101).




Tab. 4. Kinetic Parameters for those Antibodies Raised against Phosphonates (98)-(101) which Effect the

Resolution of the Fluorinated Alcohols (94)-(97). The Configuration of the Diastereoisomerically Pure Pro-

duct from Each Antibody Catalyzed Process was Shown to Correspond to that of the Antibody-Inducing

Hapten

Product Alcohol Hapten Configuration k,,," [min-'1 K, [pM] Product ee/de [% I

(98) 2R, 3R (+) 0.88 390 99.02s , 3s ( - ) 0.91 400 98.52R,3S (+) 0.94 410 98.5

('O0)101) 2S, 3R (- ) 0.86 380 98.0

(99)

a At 25 "C.

with moderate rate enhancements. Antibodiesraised to the TSA (102) were screened for

their hydrolytic properties of the diester (103).One antibody, 17El1, used in a 20% concen-tration with respect to (102), ffected hydroly-sis exclusively at C-4. This process was fastenough to make spontaneous acyl migrationfrom C-3 to C-4 of no significance: i.e., noC3-OH product was detected. [This migrationreaction is generally fast compared to chemi-cal deacylation (Fig. 34. AE 1.7) (IWABUCHItal., 1994)l. In this context, the use of an anti-body to cleave a trityl ether by an SN1processmay have further applications (AE 7.1) (IVER-SON et al., 1990).Also, the objective of utilizingabzymes in the regioselective removal of aspecified protecting group has been extendedto show that an antibody esterase can havebroad substrate tolerance (AE 1.18) (LI et al.,1995b).

The research towards the demanding glyco-sylic bond cleavage is progressing well, albeitslowly.As a first step,REYMONDas describedan antibody capable of catalyzing the acetalhydrolysis of a phenoxytetrahydropyran, al-

though it is slow, with k,,, 7 . 8 ~ 1 0 - ~-l at24"C, and has a modest ER of 70 (AE 7.4B)(REYMONDt al., 1991).

A classic approach to the problem has beento raise antibodies against TSAs related by de-sign to well-known inhibitors of glycosidases.Piperidino and pyrrolidino cations have highaffinity for pyranosidases and furanosidases(WINCHESTERnd FLEET, 1992; WINCHESTERet al., 1993) and can also be envisaged as com-ponents of a "bait and switch" approach toantibody production. Thus, SCHULTZ as de-

scribed the hydrolysis of the 3-indolyl acetal

(105) y antibody AA71.17 raised to transitionstate analog (104) Fig. 35) (AE 7.2) (Yu,et al.,

1994). Antibody AA71.17 has a useful K , of320 pM but is rather slow in turnover, k,,,

0.015 min-l. By contrast, two other haptens

AcHN

mTo( 102

SH

Ac""yJJ&Fq \

AcHN

103

Abzyme Identity Conditions

1 7 E l l pH 8.2; 20 "C

Km103 k, , ,l03 Ki lo 2

6.6 pM 0.182 min-' 0.026 pM

Fig. 34.Antibody 1 7 E l l raised against the TSA

(102) was screened for its ability to hydrolyze diester

(102) and, used in a 20% concentration with respect

to (102), effected hydrolysis exclusively at C-4.




Br,

CO2H 104

BrQ IIA71.17

pH 5. 5

Br.

HO

HO 108

105

HO

106 107

Fig. 35. Antibody AA71.17 raised against hapten(104)catalyzes the hydrolysisof the aryl acetal(lO5)(Y u et al., 1994).

based on a guanidino and a dihydropyran in-hibitor did not elicit any antibodies showingglycosidase activity.

Further studies led the same team to gener-ate glycosidase antibodies by in vitro immuni-

zation. They raised antibodies to the chair-liketransition state analog (108) derived from 1-

deoxynojirimycin, a classic glycosidase inhibi-tor. They isolated antibody 4f4f which showsk,,, 2.8. lop3min-' and K , 22pM,espective-ly, for the hydrolysis of substrate (109) (Yu etal., 1998) (Fig. 36).

JANDAhas described the production of a ga-Iactopyranosidase antibody in response tohapten (112) Fig. 37).This was designed to ac-commodate several features of the transitionstate for glycoside hydrolysis: notably a flat-

tened half-chair conformation and substantial

I

NO 2

HO

110

111

Fig. 36. Antibody 4f4f raised to (108) catalyzes thehydrolysisof (109) (Yu et al.,1998).

sp2 character at the anomeric position. Some100clones were isolated in response to immu-nization with (112) and used to generate acDNA library for display on the surface ofphage (AE 7.3) (JANDA t al., 1997). Ratherthan proceed to the normal screening for turn-over, JANDA then created a suicide substratesystem to trap the catalytic species.

HALAZY ad earlier shown that phenols

with ortho- or para-difluoromethyl substitu-ents spontaneously eliminate HF to form qui-nonemethides that are powerful electrophilesand that this activity can be used to t rap glyco-sidases (HALAZY t al., 1992). So, glycosylicbond cleavage in (1W) results in formation ofthe quinonemethide (114) that covalentlytraps the antibody catalyst. By suitable engi-neering a bacteriophage system, JANDA wasable to screen a large library of Fab fragmentantibodies and select for catalysis. Fab 1B cata-lyzed the hydrolysis of p-nitrophenyl P-galac-

topyranoside with k,,, 0.007 min-' and K,,,




HO Phosphate Monoesters

The achievement of this reaction is particular-ly impo rtant in light of the fact that ph osphoryltransfers involving tyrosine, serine, or threo-

nine play crucial roles in signal transductionpathways that control many aspects of cellularphysiology. The production of a n abzymewould provide a crucial tool for the investiga-tion and m anipulation of such processes.

SCHULTZ'Sroup employed an a-hydroxy-phosphonate hapten (115) and subsequentlyisolated 20 cell lines of which 5 catalyzed thehydrolysis of the model substrate p-nitrophen-01 phosp hate (116) bove background (Fig. 38)

k N e N HI

CO-linker

*

HO OH

112

YHCO-linker

113

NH & \ 0

115

0

co-linker 114

Fig. 37. Fragment antibody FablB is selected by sui-cide selection with substrate ( I D) from a library of

antibodies generated to hapten (lU) .The suicide in-H,N

termediate is the o-quinonemethide (114). a>."16

OZN

0'4 38E1

530 pM , corresponding to a rate enhancem entof 7 - o4.M oreo ver, this activity was inhibitedby hapten (112)with Ki 15 pM. By contrast,

o+ /

Ho/p' o0the best catalytic antibody, 1F4, generatedfrom hapten (112) by classical hybridomascreening showed k,,, min-l and K ,330 pM, a rate enhanc em ent of only 100.

Clearly, this work offers an exciting methodfor screening for antibodies that can lead to

suicide product trapping and also appears to Abzyme Identity Conditionsoffer a general approach to antibodies withglycosidase activity. 38E1 pH 9.0

D O H 117OZN

Km116 k,,,"116 K i l l s

155 pM 0.0012 min-' 34 pM

"At30"C

6.3 Phosphate Ester Cleavage

The mechanisms of phosphate ester cleav-age vary significantly betwee n monoesters,

GER, 1989). Eac h of these is a target for anti-body catalyzed hydrolysis and progress hasbeen reported for all three cases.

diesters, and triesters (THATCHER and KLU- Fig. 38. Antibody 38E1, generated from a a-hy-

droxyphosphonate hapten (115), catalyzed the hy-

drolysisof p-nitrophenyl phosphate (116) to p-nitro-

phenol (117).



436 11 Catalytic Antib odies

(SCANLAN t al., 1991). Antibody 38E1 wascharacterized in more detail and kinetic para-meters were afforded by LINEWEAVER-BURKEanalysis. This antibody exhibited 11 turnovers

per binding site with no change in V,,, andthus acted as a tru e catalyst. M oreover, exam-ination of su bstrate specificity show ed that ca-talysis was entirely selective for pa ra-substitut-ed species (A E 6.6).

Phosphodiester CleavageThe phosphodiester bond being one of th emost stable chemical linkages in nature, itscleavage is an obvious and challenging targetfor antibody catalysis. In an a ttem pt to mod el ametal-independent mechanism, a nucleotide

analog (118) featuring an 0-phosphorylatedhydroxylamine moiety was chosen by SAKU-RA I and co-workers (SAKURAIt al., 1996)(Fig. 39).

This hapten design aims to mimic the geo-metry and spatial constraints in the pho sphatelinkage, to retain the stereoelectronic configu-ration of the phosp horus atom , and t o act as asimple model of a dinucleotide. For that pur-pose, the retention of the phospha te backbon eseeks to facilitate the form ation of an oxyan-

ion hole in which the electrophilicity of thephosphorus center is increased in the boundsubstrate, while the positive charge on the ha p-

d oP"

? O

/ \ 0

11s

Fig. 39. Hapten for generating catalytic antibodies

for phosphodiester cleavage (SAKURAIt al.,1996).

ten is designed to induce a n anionic amino ac-id in the abzyme binding site, either t o act as ageneral base to activate a nucleophilic watermolecule, or as a nucleophile operating direct-

ly at the phosphorus center. More details areawaited from this work.

The classic case of assisted hydrolysis ofphosphate diesters is neighboring group par-ticipation by a vicinal hydroxy l group, specifi-cally the phosphate ester of a 1,2-diol, whichprovides a rate acceleration greatly in excessof lo6 (WESTHEIMER,968). While vanada tecomplexes of 1,Zd iols have bee n explored aspentacoordinate species for inhibiting en-zymes, they a re toxic and too labile for use inmurine immunization (CRANS, t al., 1991).JANDA has solved this problem by using ofpentacoordinate oxorhenium chelates (WEI-NER et al., 1997).By emp loying hap ten (119) asseparate diastereoisomers, 50 monoclonalantibodies were generated and screened fortheir ability to hydrolyze uridine 3 -(p-nitro-phenyl phosphate) (l20a) (Fig. 40) (A E 6.4).The m ost active of the thr ee antibodies show-ing catalysis, 2G1 2, had k,,, 1.53.10-3 s-' at25 "C and K, 240 pM giving k,,,lk,,,,, 312. Amore favorable expression for protein cata-

lysts working at substrate concentrations be-low K, is k,,,/(K;k,,,,,) (RADZICKAndWOLFENDEN,995), which is 1.3 -10 6M-' andcom pares favorably to the value for RN ase Aof 10'' M-' for the same substra te. The TS AantL(119) proved to be a powerful inhibitorfor 2G 12 with Ki estimated at 400 nM .

Evidently, this is a system with scope alikefor improvement in design and for broade r ap-plication. It is clearly one of t he most success-ful exam ples of antibody catalysis of a d ifficultreaction.

M ore recently, the sam e team ha s used a baitand switch strategy in order to improve thissystem. A nucleophilic residue activating the2'-hydroxyl is induced by a cationic trimethyl-amm onium group, while the 3 '-hydroxyl issubstituted by ap-nitro phen yl phosphoester ina substrate-like shape (l2Ob). Two antibodieswere isolated of which MA 7T.F-1 was foundto obey classic M ichaelis-Menten kinetics withthe values k,,, =0.44 min-' and K , =104 pM.This antibody is 5 times more efficient thanthe one studied before (WENTWORTHt al.,

1998).




Phosphotriester H ydrolysisCatalysis of this reaction was first exhibited byantibodies raised by ROSENBLUMnd co-work-ers (ROSENBLUMt al., 1995). More recently,

LAVEYand JANDA ave explored the genera-tion of abzymes able to catalyze the break-down of poisonous agrochemicals (LAVEYndJANDA,996a).25 Monoclonal antibodies wereraised against the N-oxide hapten (121) ofwhich two were found to be catalytic.The hap-ten was designed to generate antibodies forthe hydrolysis of triester (122) using the "baitand switch" strategy: cationic charge on the ni-trogen atom targeted to induce anionic aminoacids to act as general base catalysts; partialnegative charge on oxygen to favor the selec-tion of antibody residues capable of stabilizingnegative charge in the transition state (Fig.41).

Antibody 15C5 was able to catalyze the hy-drolysis of the phosphate triester (122) with avalue of k,,, =2.65. min-' at 25 "C. Later,a second antibody from the same immuniza-tion program was found to hydrolyze the ace-tylcholinesterase inhibitor Paraoxon (123)with k,,,=1.95.10-3 min-l at 25°C (A E 6.2)(LAVEY and JANDA,1996b). Antibody 3H5

showed Michaelis-Menten kinetics and was

strongly inhibited by the hapten (121). t ex-hibited a linear dependence of the rate of hy-drolysis on hydroxide ion suggesting that 3H5effects catalysis by transition state stabilizationrather than by general acidhase catalysis.

linker, r ,NH

0

119

.o

. .: %

- OR 120

a R = O H

o , . ~ o ~ p ~ o ~b R =0Me l

Fig. 40.Hapten anti-(119) was used to generate 25mA bs from which 2G12 proved to catalyze the hy-drolysis of the phosp hate diester (l20a). (l20b) waslater used as a hap ten in a bait and switch strategy t oobta in antibodies hydrolyzing (l20a).

121 122 123

Abzyme Substrate Conditions Km l o 3 . ,,," K , 121Identity

~ ~~

15C5 122 p H 8.1 87 pM 2.7 min- ' -

3H5 123 p H 9.15 5.05 mM 2.0min- ' 0.98 pM

a Measured at pH 8.25 and 25 "C

Fig.41.The N-oxide (121) was used as hapten to raise mA bs to catalyze the hydrolysis of bo th triesters (122)and (123) .




Phosphate ester hydrolysis is one of themost demanding of reactions for catalyst engi-neering. The progress made so far with catalyt-ic antibodies is extremely promising and ap-

pears to be competitive with studies using met-al complexes if only because they can deliverturnover while metal complexes have for themost part to solve the problem of tight productbinding.

6.4 Amide Hydrolysis

While ester, carbonate, carbamate, and ani-lide hydrolyses have been effectively catalyzedby antibodies, the difficult tasks of catalyzing

the hydrolysis of an aliphatic amide or a urearemain largely unsolved. Much of this problemcomes from the fact that breakdown of a TI*is the rate determining step, as established bymuch kinetic analysis and, more recently, bycomputation. TERArsHI has computed thatC-N bond cleavage for the TI- for hydrolysisof N-methylacetamide (or aminolysis of aceticacid) lies some 18 kcal mol-' above that forC-O(H) bond cleavage (TERAISHIt al., 1994).Clearly, protonation of nitrogen is an essential

step in the breakdown process of such inter-mediates and for anilides that is hardly a prac-tical proposition under ambient conditions. Todate only three investigations of this problemhave shown any success.

A group at IGEN raised antibodies to thedialkylphosphinic acid (124) Fig. 42). Thesewere screened for their ability to hydrolyzefour alkyl esters and four primary amides atpH 5.0,7.0, and 9.0. Just one out of 68 antibod-ies, 13Dl1, hydrolyzed the C-terminal carbox-

amide with stereospecificity for the (R)-sub-strate (l2Sa) (rendered visible by the attach-ment of a dansyl fluorophore to supportHPLC analysis of the course of the reaction)

(AE 5.1) (MARTINt al., 1994).At pH 9.0 and37 "C, mAb 13Dll showed K , 432 pM and k,,,1.65.10-'s-*, corresponding to a half-life of42 d.This activity was fully inhibited by hapten(124)with Ki 14 pM. Unexpectedly, the dansylgroup proved to be an essential component ofthe substrate. Even more surprisingly, he anti-body did not catalyze the hydrolysis of the cor-responding methyl ester (l25b).

More recently, JANDAnd co-workers haveproduced catalytic antibodies for hydrolysis ofa similar primary amide (GAO et al., 1998).

They raised antibodies to the trigonal boronicacid hapten (126) n order to catalyze the hy-drolysis of the primary amide (l27a) andmethyl ester (l27b) Fig.43). Using antibodiesobtained by standard hybridoma techniques,they built a combinatorial library and selectingwith a boronic acid probe and phage display,they were able to isolate Fab-BL25 catalyzingthe hydrolysis of the L-enantiomer of (l27a).

This Fab fragment showed Michaelis-Mentenkinetics ( K , 150 pM and k,,, 0.003 min-')

with a half-life for the substrate of 3.9 h, morethan two orders of magnitude better than for13Dll . It is interesting to note that no catalyt-ic antibodies were obtained by the standardimmunization route.

Whereas most amide substrates for catalyticantibodies have been activated by the use ofaromatic amines (AE 5.3, 5.4, 5 . 5 ) , BLACK-BURN and co-workers chose to explore hydrol-ysis of an aliphatic amide activated throughhalogenation in the acyl moiety. Chloram-

a X = N H ,

bX=OMe

Fig. 42. One out of 68 antibodies raised to the dialkylphosphinic acid hapten (W), ydrolyzed the carboxa-rnide (l25a), but not the ester (l25b).



7 Reactive Immunization 439

OHAcHN

127

a X = N H 2

b X = O M e

Fig. 43. Antibodies raised to the trigonal boronic acid hapten (126) atalyzed the hydrolysis of

the L-enantiomerof the primary amide (l27a).

phenicol(128) was selected as substrate on ac-count of its dichloroacetamide function andthe tetrahedral intermediate for hydrolysiswas mimicked by the neutral sulfonamide(129)and the zwitterionic “stretched transitionstate analog” aminophosphinic acid (130) (Fig.44). Antibodies produced to each of these hap-tens proved too weak to hydrolyze chloram-phenicol (128) at a rate sufficiently abovebackground (kOH1.3.10-3 M- ls -l ) for fur-ther study. However, the switch to a more reac-

tive amide substrate, trifluoramphenicol (131)( k , 6.10-’ s-’, k,, 6.3.10-* M-l s-l at37 “C), enabled useful data to be obtained forantibody 2B5. This showed Michaelis-Mentenkinetics with k,,, 2 . s-l and K , 640 pM atpH 7.0,37”C.The use of kcatl(K,.k,,,,,) gavean ER of 5,200.The relatively high K , may bea consequence of switching the dichloroacet-amide moiety in the hapten for the trifluoro-acetamide group in the substrate and so couldbe improved by appropriate modification tothe hapten. The low rate of turnover achieved

clearly indicates the difficult task ahead forantibody cleavage of a peptide based on tetra-hedral intermediate mimicry alone (SHEN,1995; DATTA et al., 1996).

A new strategy to achieve cleavage of anamide bond has used an external cofactor (ER-

SOY et al., 1998).Antibodies were raised to anarylphosphonate diester. In this case the aryl-propylamide is attacked by any of three nucle-ophile cofactors catalyzing a N + O acyl trans-fer.This generates an ester with a high sponta-neous rate of cleavage. In this way, aryl amidehydrolysis was achieved with k,,, of

1.3-13.3.10-5 min-’ and K , of 77-370 pM forthe substrate and 14-136 pM for the three co-factors.

By contrast, the reverse reaction, that ofamide synthesis,has proved to be a good targetfor antibody catalysis and a range of differententerprises have been successful (AE 18.1-18.4).It would appear here that little more isneeded than a good leaving group and satisfac-tory design of a TSA based on an anionic tetra-hedral intermediate (BENKOVICt al., 1988;

JANDAt al., 1988a; HIRSCHMANNt al. 1994;JACOBSENnd SCHULTZ,994).

7 Reactive Immunization

An alternative approach for the induction ofcatalysis in antibody binding sites is a strategynamed “reactive immunization” (WIRSCHINGet al., 1995).This system uses haptens of inter-

mediate chemical stability as antigens. Afterthe first stimulation of the mouse B cells togenerate antibodies, one of the products of in

vivo chemical transformation of the originalhapten is then designed to act as a second im-munogen to stimulate further maturation ofantibodies that will be better able to catalyzethe desired reaction. The system seems well-designed to achieve the benefits of a neutraland a charged hapten within the same familyof monoclonal antibodies.

An organophosphate diester (132),was cho-sen as the primary reactive immunogen. Fol-



440 I 1 Catalytic A ntibodie s

REACTIVE IMMUNOGEN

Ar = 4-(methylsulfonyl)phenyl

a R = represents position of linker to

b R = H in free haptenwhich carrier protein is attached

0

12 8Chloramphenicol

Spacer-C ONH &CHC12 // ‘

12 9

OH OH

Spacer-CONH CHCI,

13 0

0

13 1Trifluoramphenicol

Fig. 44.Antibodies produced to each of haptens(129), (130) proved too weak to hydrolyze chlor-amphenicol (128) t a rate sufficiently above back-ground . However, the mo re reactive substrate, tri-fluoramp henicol(l31) enabled useful data to be ob-

tained for antibody 2B5 (SHEN,1995;DATTAt al.,1996).

lowing spontaneous hydrolysis in vivo, it be-comes a stable mono ester transition state ana-log (133),which in turn gives a new challengeto the immune system (Fig. 45) (AE 2.14).Cross-reactivity has been dem onstra ted as be-ing an advantage in this process since heter-ologous immunization w ith both diary1 ester(132) and the corresponding monoaryl estergave cross-reacting serum with enhanced af-

13 21pontaneous

in vivo

H 13 3

STABLE IMMUNOGEN - TS A

134

SUBSTRATE

Fig. 45. Antibodies raised simu ltaneously against thereactive (132) and stable immunogen (133) shownabove were capable of efficient “tur nov er” of the re-lated aryl ester substrate (134) (Ab 49H4:K,,,=300pM;kc,,=31 min-’ at 22°C).

finity for the monoaryl TSA.This further pro-motes th e induction of am ino acids capable of

acting as nucleophiles o r general ac id ba se cat-alysts in th e active site. In practice, reactive im -munization with (l32a) enerated 19 monoclo-nal antibodies, 11of which were able to hydro-lyze substrate (132b).The most efficient ab-zyme, SP 04 9H 4, was analyzed kinetically us-ing radioactive substrates. It was establishedthat 49H4 had undergone reactive immuniz-ation, since it was able to hydrolyze the arylcarboxylate aryl (134) very effectively withK , =300 pM; k,,, =31 min-’.




A similar approach has been used by LER-NER and BARBASo induce catalytic antibodiesmimicking Type I aldolases. The reactionscheme is shown below: the aim here was to in-

duce an enamine moiety which can achieve ca-talysis through lowering the entropy for bi-molecular reaction between ketone substrateand aldol acceptor. Compound (135) is a 1,3-diketone which acts as a trap for the “criticallysine”, forming the vinylogous amide (138),which can be monitored by spectrophotome-try at 318 nm (Fig. 46) (AE 16.2) (LERNERndBARBAS, 996). Screening for this catalyticintermediate by incubation with hapten facili-tated the detection of two monoclonal anti-bodies with k,,, 2.28.10-’ M-’ min-’. Fur-

thermore, k,,,/(K;k,,,,,) is close to lo9,mak-ing these antibodies nearly efficient as the nat-urally occurring fructose 1,6-bisphosphatealdolase. Studies on the stoichiometry of thereaction by titration of antibody with acetylac-etone indicated two binding sites to be in-volved in the reaction.

The stereochemical control shown by theseantibodies was remarkable, and so it has beensuggested that it might be possible to programan antibody to accept any aldol donor and ac-

ceptor. In the event, the antibodies generatedin this program accepted a broad range of sub-strates including acetone, fluoroacetone, 2-bu-tanone, 3-pentanone, and dihydroxyacetone.

Finally, the process of reactive immuniza-tion leads to the concept of using mechanism-based inhibitors as haptens, actively promotingthe desired mechanism by contrast to theirconventional use as irreversible enzyme inhib-itors.

8 Potential Medical

Applications

8.1 Detoxification

The idea that abzymes might be used thera-peutically to degrade harmful chemicals inman potentially offers a new route to treat-ment for victims of drug overdose. LANDRY’S

group has generated antibodies to catalyze the

RmT 135

HZN-Lys-Ab

it

Ly s .Ab Lys - Ab

H

H138 139

R = p-(HO2C(CHz)&ONH)C,H,-CHzCH,-

Fig.46.The m echanism by which an essential Lys re -sidue in the antibody combining site is trapped usingthe 1,3-diketone (135) o form the covalently linked

vinylogous amide (138).

hydrolysis of the benzoyl ester of cocaine(140) ielding the ecgonine methyl ester (141)and benzoic acid, products which retain noneof the stimulant or reinforcing properties ofthe parent drugs. The transition state analogused in this experiment was the stable phos-phonate monoester (142) which led to a rangeof antibodies of which 3B9 and 15A10 werethe most effective (Fig. 47) (AE 1.3) LANDRY

et al., 1993).




CONH-linker

PhC02HI0

140 141 142

Abzyme Identity Conditions Km kc,," K ,142

3B 9 pH 7.7 490 pM 0.11 min-' < 2 pM2.3 min-' -15A10 pH 8.0 220 pM

a Temperature not defined

Fig.47.Hapten (142)was used to raise an antibody 3B 9 capable of the hydrolysis of cocaine (140) o the al-cohol (141) hereby effecting cocaine detoxification.

In a more recent study, the same group ex-amined the efficiency of antibody 15A10in vi-vo (METSet al., 1998).They studied the effectof mAb 15A10 on seizure and lethality in a rat

model of overdose and its effect on cocaine in-duced reinforcement in a rat model of addic-tion. They showed how this mAb while effec-tively increasing the degradation rate of thedrug, protects against its lethal effect andblocks its reinforcing effects. The authors sug-gest the use of a humanized mAb to transposethis system to humans.

8.2 Prodrug Activation

Many therapeutic agents are administeredin a chemically modified form to improve fea-tures such as their solubility characteristics,ease of administration, and bioavailability(BOWMANnd RAND, 988). Such a "prodrug"must be designed to break down in viv o releas-ing the active drug, sometimes at a particularstage of metabolism or in a particular organ.The limitation that this imposes on the choiceof masking function could be overcome by theuse of an antibody catalyst for activating theprodrug which could, in principle, be concen-trated at a specified locus in the body. Such se-

lectivity could have implications in targetedtherapies.

Antibody mediated prodrug activation wasfirst illustrated by FUJII ho raised antibodies

against phosphonate (144) to hydrolyze a pro-drug of chloramphenicol(l45). Antibody 6D9was shown to operate on substrate (145) to re-lease the antibiotic (128)with an ER of 1.8.103(AE 1.8) (Fig. 48) (MIYASHITAt al., 1993).Furthermore, FUJII howed unequivocally thatantibody catalyzed prodrug activation is viableby demonstrating inhibition of the growth ofB. subtilis by means of the ester (145) onlywhen antibody mAb 6D9 was present in thecell culture medium. No product inhibitionwas observed by chloramphenicol at 10 mM

supporting the multiple turnover effect seen inthe growth inhibition assay.

CAMPBELLnd co-workers also succeededwith this type of strategy by producing anti-body 49.AG.659.12 against a phosphonateTSA (147), designed to promote release of theanti-cancer drug, 5-fluorodeoxyuridine from aD-Valyl ester prodrug (148) (Fig. 49, AE 1.10,CAMPBELL,t al., 1994).This catalyst was ableto bring about inhibition of the growth of E.coli by the release of the cytotoxic agent,5-fluorodeoxyuridine in vitro .




nmNHCom

16D9

128Chloramphenicol

Abzym e Identity Conditions~ ~~

6D9 p H 8.0 30°C

Km145 k,,, 145 K,144

64 M 0.133min-' 0.06 pM

Fig. 48. The monoclonal, 6D9, raised against phos-phonate (144) atalyzed th e hydrolysis of on e possi-ble regio-isomer (145) f a phenacetyl ester prod rugderived from chloramphenicol(l28).

FJ.

OH 147

n 0

OH 148

Abzym e Identity Conditions

49.AG.659.12 p H 8.0 37"C

Km148 k,,,148 Ki 147

218 pM 0.03 min-' 0.27 pM

Fig. 49.Prodrug (148) s an acylated derivative of th eanticancer drug 5-fluorodeoxyuridine. Antibody49.AG.659.12, raised against phosphon ate (147)wasfound to activate the prodrug (148) n vitro, therebyinhibiting the growth of E . coli. Kinetic parametersare listed.

Much the most developed example of pro-drug activation comes from our own laborato-ry. The cytotoxicity of nitrogen mustards is de-pendent on substitution on the nitrogen atom:electron-withdrawing substituents deactivateand electron-releasing substituents activate abifunctional mustard. Thus, antibody catalyzedcleavage of a carbamate ester of a phenolicmustard can enhance its cytotoxicity to estab-lish the carbamate as a viable prodrug for can-cer chemotherapy (BLAKEY,992; BLAKEYtal., 1995).So the target for prodrug activation




is defined as an aryl carbam ate whose nitrogensubstituent is either an aryl (149a) or alkyl(149b) moiety (Fig. 50).

Aryl carbamates are known to cleave by an

ElcB process with a high dependency on thepK, of the leaving phenol ( p - =2.5). By con-trast, aryl N-methylcarbam ates are hydrolyzedby a BA,2 process with a much lower depen-dency on leaving group (po=0.8) (WILLIAMSand DOUGLA S, 972a, b). Given the electron-releasing nature of the nitrogen mustard func-t ion (a ca . -0.5), th e kinetic advantage of anti-

C 1 d 149

a R' = 3.5-dicarboxyphenyl1

b R = 2 -g lu tw l

c1

H

15 0

151

Fig. 50. Carba mate prodrugs (149a, b) are targets forabzyme cleavage to release a mustard (150) f en-hanced cytotoxicity. ElcB hydrolysis of aryl carba-mates involves the anion (151).TSAs (152a) and(152b) were used t o generate antibodies to catalyze

a BA,2 mechanism for hydrolysis whose kinetic be-havior was evaluated w ith ester (153).

body hydrolysis via the BA,2 pathway coupledto th e proved ability of antibodies to stabilizetetrahedral transition states led to the formu-lation of TSAs (152a) an d (152b). Siting the

linker in the locus of th e nitrogen mustard wasdesigned (1) to minimize potential alkylationof the antibody by the mustard function and(2) to support mechanistic investigations byvariation of the p-substituent of the aryl carba-mate with little or no change in K,, both fea-tures that were realized in the outcome. Bothof these TSAs gen erated large numb ers of hy-bridomas including many catalysts capable ofcarbam ate hydrolysis.

A mechanistic analysis of antibod y DF8 -D5showed it to cleave p-nitrophenyl carbamate(153) with k,,, 0.3 s-', K , 120 yM , an d kc,,/

k,,,,, 300 at 14°C (A E 4.3) (WENTWORTHtal., 1997).This is some tenfold mo re active thana carbamatase antibod y generate d by SCHULTZto a p-nitrophenyl ph osphonate TS A but witha similar E R (A E 4.1) (VANV R A N K E Nt al.,1994). Most sign ificantly, variations in th ep-substituent in substrates for DF8-5 hydroly-sis identified a Ham mett p" value6 of 0.526 toestablish th e BA,2 nature of the reaction. Forthe p-methoxyphenyl carbamate substrate

(." = -0.3) the app arent E R is 1.2.10". Giventha t ther e is a 10' difference in rate for theE lc B and BA,2 processes for the p-nitrophen-yl carbamate (153), the data show that anti-body DF8-D5 has promoted the disfavoredBA,2 process relative to the sp ontan eous E lc Bcleavage by some 13 kcal mol-'. Lastly, it isnoteworthy that DF8-D5 was raised againstthe phosphonamidate ethyl ester (149a) ashapte n, which raises the possibility that it maybe an unexpected product of reactive imm uni-zation (Sect. 7).

The m edical potential of such carbamatasesdepends on their ability to deliver sufficientcytotoxic agent to kill cells. Antibody EA11-D7 , raised against TSA (152b) proved able to

hydrolyze the prodrug (149b) with K, 201 yMand k,,, 1.88 min-' at 37°C (A E 4.2) (WENT-WORTH et al., 1996). Ex vivo studies with thisabzyme and hum an colonic carcinoma (LoVo)cells led t o a marked reduction in cell viabilityrelative to controls. This cytotoxic activity was

A p lu- plot would have an even flatter slope.




reproduced exactly by the Fab derived fromEA11-D7 and was fully inhibited by a stoichio-metric amount of the TSA (152b).Using EA11-D7 at 0.64 pM, ome 70% of cells were killed

in a 1h incubation with prodrug (149b)and theantibody transformed a net 4.18pM of prodrugdelivering more than 2 . IC,, of the cytotoxicagent (150).This performance is, however, wellbehind that of the bacterial carboxypeptidaseCPG2 used by Zeneca in their ADEPT system(BAGSHAWE,990; BLAKEYt al., 1995), beinglo3slower than the enzyme and 4.104 inferiorin selectivity ratio. Nonetheless, it is the firstabzyme system to show genuine medical po-tential and should stimulate further work inthis area.

8.3 Abzyme Screening via Cell

Growth

BENKOVICroposed a new method of select-ing Fabs from the whole immunological reper-toire in order to facilitate a metabolic process(SMILEY nd BENKOVIC,994). A cDNA li-brary for antibodies was raised against hapten(159) and then expressed in an E. coli strain

devoid of any native orotic acid decarboxylaseactivity (Fig. 51).The bacteria were then estab-lished in a pyrimidine-free environment whereonly those bacteria grew which expressed anantibody able to provide pyrimidines essentialfor DNA synthesis, and hence bacterialgrowth. Six colonies expressing an active Fabfragment were found viable in a screen of16,000 transformants (AE 9.2). The remark-able feature of this system is that OCDase,which catalyzes the decarboxylation of oroti-dylic acid to UMP (Fig. 51), is thought to be at

the top end of performance by any enzyme inaccelerating this decarboxylation by some lo*’(RADZICKAnd WOLFENDEN,995).

Such an example illustrates the ability of ab-zymes to implant cell viability in the face of adamaged or deleted gene for a crucial meta-bolic process. The medical opportunities forapplications of such catalysis are clear.

There is sufficient encouragement in theseexamples to show that out of all the prospectsfor the future development of catalytic anti-bodies, those in the field of medical applica-

tions, where selectivity in transformation of

0 0

H 154

j Antibody

t

Ribose-5-P

155

ODCase

I

H 156

158

Ribose-5-P

UM P 157

aDNA Biosynthesis

t 0

0

O IR

15 9

Fig. 51.Pathways for uridylate biosynthesis.PRTase:

phosphoribosyl transferase; ODCase: OMP decar-

boxylase; OMP: orotidine 5‘-phosphate; UMP: uri-

dine 5 ‘-phosphate. Mutants lacking enzymesPRTase or ODCase can complete a route to UMP

provided by an antibody orotate decarboxylase in

conjunction with the naturally occurring uracil

PRTase. Decarboxylation of orotic acid (154) s

thought to proceed through the transition state

(158), or which the hapten (159) was developed

(SMILEYnd BENKOVIC,994).

unusual substrates may be of greater impor-tance than sheer velocity of turnover of sub-strate, may well rank highest.




9 Industrial Future of

Abzymes

In view of the widespread interest in bioca-talysis, it is not surprising that in little over adozen years after catalytic antibodies werefirst reported, a vast literature has developeddocumenting them.The field of research work-ers is international, with groups from four con-tinents showing activity in the area. The poten-tial of “designer enzymes” is already becominga reality in relation to both the chemical indus-try and the pharmaceutical field.

Over 70 different chemical reactions, rang-ing from hydrolyses to carbon-carbon bond

forming reactions, have been catalyzed by ab-zymes and their application to general synthet-ic organic chemistry seems promising. TypicalMichaelis constants (K , ) lie in the range10 pM to 1mM and binding selectivity for theTSA over the substrate is in the range 10- to105-fold.It therefore appears that antibodieshave fulfilled most expectations that theywould be capable of substrate discriminationcomparable to that of enzymes but over a wid-er range of substrate types than foreseen, and

especially effective when programmed to adesignated substrate. The range of reactionsthat may be catalyzed by antibodies appears tobe limited only by a sufficient knowledge ofthe transition state for any given reaction com-bined with synthetic accessibility to a stableTSA or to a suitable “suicide” substrate.

On the other hand, antibodies are generallyable to accelerate reactions by at most lo7times the rate of the uncatalyzed process. Ithas to be said that scientists at large are look-ing for a major step forward in abzymes prop-

erties to achieve rate accelerations up to lo 9that would establish abzymes as a feature ofsynthetically useful biotransformations. At thesame time, it is essential to demonstrate thatproduct inhibition is not an obstacle to thescaled-up use of abzymes.

In relation to syntheses able to deliver usab-le amounts of product, LERNERas shown thatstereoselective reactions can be performed ona gram scale, as in the enantioselective hydro-lysis of a 2-benzylpropenyl methyl ether to thecorresponding (S)-2-benzylpropanol of high

ee (AE 7.4A) (REYMONDt al., 1994). In addi-

tion,JANDAas described an automated meth-od of transposing antibody catalyzed transfor-mations of organic molecules onto the multi-gram scale by employment of a biphasic sys-

tem. The viability of this system was demon-strated by an epoxide ring-closing antibody,26D9, to transform 2.2 g of substrate, corre-sponding to a turnover of 12 7 molecules percatalytic site in each batch process. This provesthat the catalytic antibody does not experienceinhibition by product (AE 14.1) (SHEVLINtal., 1994). It would appear that an improve-ment in abzyme performance of little morethan two orders of magnitude is needed beforecatalytic antibodies can be efficiently put towork in bioreactors and participate in kilo-

gram scale production.Lastly, two technical features of antibody

production may be valuable for the future pro-duction of cheaper abzymes of commercial va-lue. First, the use of polyclonal catalysts, prima-rily from sheep, has had a tough early passagebut now appears to be established for a widerange to transformations (GALLACHERt al.,1990,1992; STEPHENSnd IVERSON,993;Tu-BU L et al., 1994;BASMADJIANt al., 1995;WAL-LACE and IVERSON, 1996).While these catalysts

may not lend themselves to detailed examina-tion by physical organic chemistry, theyshould be able to deliver catalysts at a muchlower cost. Secondly, as science becomes moreenvironmentally friendly and animal experi-mentation is more tightly regulated, approach-es to screening antibodies with in vitro libra-ries may become a more important componentof this domain of investigation. THOMAS asmade a beginning with in vitro immunizationand shown that useful catalysis can be identi-fied (STAHL t al., 1995).While there are some

limitations in this system, notably the relative-ly low substrate affinity of antibodies genera-ted in this way, it is capable of refinement andmay become a useful component of future ab-zyme selection systems.

10 Conclusions

On an evolutionary time scale, abzyme re-

search is just reaching adolescence (THOMAS,



11 Glossary 447

1996), yet already over 80 different antibodycatalyzed chemical reactions have been listedduring its first decade of life. The details un-covered concerning the mechanisms of ab-

zyme catalyzed reactions have been richerthan expected. A wide diversity of purposelydesigned catalysts has been explored, with thepotential impact on the field of medicine andfine chemicals production being implicated.However, the immaturity of antibody catalysishas been exposed by its poor efficiency, whichin spite of intense research efforts to improveall aspects of abzyme generation, continues tohinder wide-scale acknowledgement of its con-tribution to biocatalysis, particularly fromunder the shadow of powerful enzymes.

There now exists sufficient literature aboutcatalytic antibodies not only in terms of theirkinetic behavior (Appendix, Sect. 12) but alsothrough structural information derived fromX-ray crystallographic data (GOLINELLI-PIM-PANEAU et al., 1994;HAYNESt al., 1994; ZHOUet al., 1994) and 3-D modeling of protein se-quences (ROBERTSt al., 1994), that it has be-come possible to speculate on a more generalbasis concerning the scope, limitations, and re-alistic future of the field (STEWARDnd BEN-

KOVIC, 995; KIRBY, 996). In terms of transi-tion state stabilization, catalytic antibodieshave been shown to recognize features of theputative transition state structure encoded bytheir haptens with affinity constants in thenanomolar region whereas it has been estimat-ed that enzymes can achieve transition statecomplementarity with association constants ofthe order of M to deliver rate accelera-tions of up to lo1’ (RADZICKAnd WOLFEN-DEN, 995).The whole subject of binding ener-gy and catalysis has been authoritatively and

critically reviewed by MADER nd BARTLETT,with especial focus on the relationshipbetween transition state analogs and catalyticantibodies (MADER nd BARTLETT, 997).En-

zymes have evolved to interact with every spe-cies along the reaction pathways that they ca-talyze whereas our manipulation of the im-mune system is still relatively simplistic, usinga single hapten to stimulate a full, often multi-step, reaction sequence of catalysis.

The serendipity that may be involved in theisolation of an efficient antibody catalyst isnow well appreciated while recent studies

have shown that non-specific binding proteinssuch as BSA may display catalysis approach-ing the level of abzymes, albeit without anysubstrate selectivity (HOLLFELDER t al.,

1996). All of this serves to emphasize the factthat protein recognition of discrete high ener-gy reaction intermediates need not translateinto efficient protein catalysis. However, theimprovements in hapten design and antibodygeneration strategies described above are be-ing used to highlight more intricate catalyticfeatures. Charged and nucleophilic active siteresidues, substrate distortion (DATTA et al.,1996; YLI-KAUHALUOMAt al., 1996), desolva-tion, and proximity effects have all been nowidentified as components of antibody mediat-

ed catalysis. Using structural informationavailable for an ever increasing number of cat-alytic antibodies, manipulation of the antibodycombining site is now attainable using proce-dures such as chemical modification (POLLACKand SCHULTZ,989; SCHULTZ,989) and site-directed mutagenesis (JACKSONt al., 1991;STEWARTt al., 1994; KAST t al., 1996) to pin-point or to improve the action of abzymes.Thesemi-rational design of antibody catalysts us-ing a combination of such techniques is also

supporting the systematic dissection of theseprimitive protein catalytic systems so as toprovide valuable information concerning theorigin and significance of catalytic mechanismsemployed in enzymes.

If the ultimate worth of antibody catalysts isto be more than academic, then the key mustbe found in their programmability. Here, thecapabilities of abzymes such as their promo-tion of disfavored processes and selectivity forsubstrates and transformations for which thereare no known enzymes may offer prospects

more significant than the further pursuit ofperformance levels comparable to those of en-zymes.

11 Glossary

Abzyme: An alternative name for a catalyticantibody (derived from Antibody-enzyme).Affinity labeling: A method for identifyingamino acid residues located in the binding




site(s) of a peptide. The protein is treated witha ligand which binds to the binding site andreacts with proximal amino acid residues toform a covalent derivative. Upon hydrolysis of

the protein, individual amino acids or shortpeptide fragments linked to the ligand are sep-arated and identified.Antibodies: Proteins of the immunoglobulinsuperfamily, carrying antigen binding sites thatbind non-covalently to the corresponding epi-tope. They are produced by B lymphocytes andare secreted from plasma cells in response toantigen stimulation.Antigen: A molecule, usually peptide, protein,or polysaccharide, that elicits an immune re-sponse when introduced into the tissues of ananimal.B cells: (Also known as B lymphocytes). De-rived from the bone marrow, these cells aremediators of humoral immunity in response toantigen, where they differentiate into antibodyforming plasma cells and B memory cells.Bait and switch A strategy whereby thecharge-charge complementarity between anti-body and hapten is exploited. By immunizingwith haptens containing charges directed atkey points of the reaction transition state,

complementary charged residues are inducedin the active site which are then used in cataly-sis of the substrate.B S A Bovine Serum Albumin. Extracted fromcattle serum and used as a carrier protein.Carrier protein: Macromolecule to which ahapten is conjugated, thereby enabling thehapten to stimulate the immune response.catELISA Similar to an ELISA, except thatthe assay detects catalysis as opposed to sim-ple binding between hapten and antibody. Thesubstrate for a reaction is bound to the surface

of the microtiter plate, and putative catalyticantibodies are applied. Any product moleculesformed are then detected by the addition ofanti-product antibodies, usually in the form ofa polyclonal mixture raised in rabbits. The EL-ISA is then completed in the usual way, withan anti-rabbit “second antibody” conjugatedto an enzyme, and the formation of coloredproduct upon addition of the substrate for thisenzyme. The intensity of this color is then in-dicative of the amount of product formed, andthus catalytic antibodies are selected directly.Conjugate: In immunological terms this usual-

ly refers to the product obtained from thecovalent coupling of a protein (e.g., a carrierprotein) with either a hapten, a label such asfluorescein, or an enzyme.

Conjugation: The process of covalently bond-ing (multiple) copies of a hapten to a carrierprotein, usually by means of a linkerkpacer todistance the hapten from the surface of thecarrier protein by a chain of about six atoms.ELI SA (Enzyme Linked Immunosorbent As-

say), An immunoassay in which antibody orantigen is detected.To detect antibody, antigenis first adsorbed onto the surface of microtiterplates, after which the test sample is applied.Any unbound (non-antigen specific) materialis washed away, and remaining antibody-anti-gen complexes are detected by an antiimmu-noglobulin conjugated to an enzyme. When thesubstrate for this enzyme is applied, a coloredproduct is formed which can be measuredspectrophotometrically. The intensity of thecolored product is proportional to the concen-tration of antibody bound.Enhancement ratio, ER: Quantified as kc,,/k,,,,,, is used to express the catalytic power of

a biocatalyst. It is a comparison between thecatalyzed reaction occurring at its optimal rate

and the background rate.Entropic trap: A strategy aimed at improvingthe efficiencyof catalytic antibodies, via the in-corporation of a molecular constraint in thetransition state analog that gives the hapten ahigher-energy conformation than the reactionproduct.Epitope: The region of an antigen to whichantibody binds specifically.This is also knownas the antigenic determinant.F a b The fragment obtained by papain hydrol-ysis of immunoglobulins. The fragment has a

molecular weight of -4 5 kDa and consists ofone light chain linked to the N-terminal half ofits corresponding heavy chain. A Fab containsone antigen binding site (as opposed to biva-lent antibodies), and can combine with antigenas a univalent antibody.Fab’: The fragment obtained by pepsin diges-tion of immunoglobulins, followed by reduc-tion of the interchain disulfide bond betweenthe two heavy chains at the hinge region. Theresulting fragment is similar to a Fab fragmentin that it can bind with antigen univalently, butit has the extra hinge region of the heavy chain.



I 1 Glossary 449

substrate and biocatalyst; the smaller the val-ue, the tighter the binding in the complex(FERSHT,985).Library:A collection of antibodies, usually Fab

or scFv fragments, in the range of lo6 to 10’’and displayed on the surface of bacteriophagewhose DNA gene contains a DNA sequencecapable of expression as the antibody protein.Thus, identification of a single member of thelibrary by selection can be used to generatemultiple copies of the phage and sizeableamounts of the antibody protein.Mon oclonal antibody,m A b Describes an anti-body derived from a single clone of cells or aclonally obtained cell line. Its common use de-notes an antibody secreted by a hybridoma

cell line. Monoclonal antibodies are used verywidely in the study of antigens, and as diagnos-tics.Polyclonal antibodies: Antibodies derivedfrom a mixture of cells, hence containing vari-ous populations of antibodies with differentamino acid sequences. Are of limited use inthat they will not all bind to the same epitopesfollowing immunization with a haptenlcarrierprotein conjugate. They are also difficult topurify and characterize, yet have been used

with success in the catELISA system.Positive clones: A phrase usually used to de-scribe those hybridoma clones which bind rea-sonably well to their respective hapten in anenzyme linked immunosorbent assay, therebyeliminating non-specific antibodies raised todifferent epitopes of the haptenkarrier conju-gate.Residues: General term for the unit of a poly-mer, that is, the portion of a sugar, amino acidor nucleotide that is added as part of the poly-mer chain during polymerization.

Site-directed mutagenesis:A change in the nu-cleotide sequence of DNA at particular nucle-otide residues and/or complete codons. This isusually done in order to change the corre-sponding amino acid residue in the protein en-coded by the nucleotide sequence so as to in-troduce or remove functionality from the pro-tein.Single Chain Antibody (scFv): Comprises a V,

linked to a VHchain via a polypeptide linker. Itis thus a univalent functioning antibody con-taining both of the variable regions of the par-

ent antibody.

Hapten: Substance that can interact with anti-body but cannot elicit an immune response un-less it is conjugated to a carrier protein beforeits introduction into the tissues of an animal.

Haptens are mostly small molecules of lessthan 1 kDa. For the generation of a catalyticantibody, a TSA is attached to a spacer mole-cule to give a hapten of which multiple copiescan be linked to a carrier protein.Hybridoma: Cell produced by the fusion ofantibody producing plasma cells with myelo-makarcinoma cells. The resultant hybrids havethen the capacity to produce antibody (as de-termined by the properties of the plasmacells), and can be grown in continuous cultureindefinitely due to the immortality of the mye-

loma fusion partner. This technique enabledthe first continuous supply of monoclonal anti-bodies to be produced.I g G The major immunoglobulin in human ser-um. There are four subclasses of IgG: IgG1,IgG2, IgG3, and IgG4, yet this number variesin different species. All are able to cross theplacenta, and the first three subclasses fix com-plement by the classical pathway. The molecu-lar weight of human IgG is 150 kDa and thenormal serum concentration in man is 16 mg

mL-l.Immunoglobulin: Member of a family of pro-teins containing heavy and light chains joinedtogether by interchain disulfide bonds. Themembers are divided into classes and subclass-es, with most mammals having five classes(IgM, IgG, IgA, IgD, and IgE).k,,,:The rate constant for the formation ofproduct from a particular substrate. k,,, is ob-tained by dividing the Michaelis-Menten pa-rameter, V,,,, by the total enzyme concentra-tion. It is also called the “maximum turnover

number”, because it represents the maximumnumber of substrate molecules converted peractive site per unit time (FERSHT,985).

k,,,/K,: See specificity constant.K L H Keyhole limpet haemocyanin, used forits excellent antigenic properties. It is used as acarrier protein in order to bestow immuno-genicity to small haptens.K,: Is the Michaelis-Menten constant, which isdefined as the substrate concentration atwhich the biocatalyst is working at half itsmaximum rate (V,,,). In practice, K , gives a

measure of the binding affinity between the




Somatic hypermutation: Mutations occurringin the variable region genes of the light andheavy chains during th e formation of memo ryB cells. Those B cells whose affinity is increa-

sed by such mutations are positively selectedby interaction with antigen, and this leads to anincrease in the average affinity of th e antibod-ies produced.Specificity constant: Is defined as kcatlKm.t is

a pseudo-second -order ra te constant w hich, intheory, would be th e actual ra te constant if for-mation of the enzyme-substrate complex wasthe ra te determining step (FERSHT,985).

T S A Transition state analog; frequently a sta-ble an alog of an uns table, high energy, reactionintermediate that is close to related energybarriers in a multi-step reaction.



12 Appendix 451

12 Appendix

12.1 Catalog of Antibody Catalyzed Processes

A . HYDROLYTIC A N D DISSOCIATIVE PROCESSES

1 . Aliphatic Ester Hydr olysis

Reaction I Conditions

Cocaine

389, H 7.7

15A10. pH 8.0polyclonalI

Hapten / Comments IK i

rNo2 7G12

,:,0 OH 3G 2

T' = (CH&CO&

SA

7G12: K, .9 E-2 KM

3G2: Ki 4.7 E-2PM

0 15A10R , = (CHzhNHCO(CH,hCO,H,

R, = Me, R, = H

T S A

3B9: Ki < 2.OEi4 pM

Polyclonal

R, =M e, R, =M e, R, = NH-DTR , = Me, R, =DT, = H

R , = DT. R,=Me, & = H

K m

1.4 E+3

1.3 E+1

5.4 E+O

1.9 E+2

2.2 E+2

N

-

kcat

/ m i d-5.0 E+O

7.0 E-2

3.3 E-2

1 . 1 E-1

2. 3 E+O

N

-

3. 7 E+:

1.7 E+:

5.4 E+:

2.3 E+i

N

-

Entrj

AE

1 . 1

-

1 . 2

1 . 3

-



452 I1 Catalytic Antibod ies

1 F 8 . 0 , 30°C

EtOH 0 0

N (CHd5COzH

HHO

N, (CHz)&GH

HEtO

Y O Q Y0

AcOH " ' O - O y

pH9.0,

25°CBnC02H 42% 6s

I

17E11,pH8.2,2OoC

1 ,F

AcH

it

Hu

T S A

Kl 4.0 E+O pM

T S A

Kl 7.0 E+O pM

0

COzH

T S A

Ki 4. 3 E+O pM

AcH

yHNMeAcHN <

SH 90

T S A

Kj 2.6 E-2pM

2.9 E+Z

1.8 E+2

3.9 E+ 2-5.6E+O

-

2.0 E+

-7.0 E-:

1.OE-:

1.8 E-I

___

N

-g.8 E+ l

).O E+2

!.7 E+ 3

__

-1.4

-1.5

-1.6

-1.7

-



12 Appendix 453

0LoNHcocF30 2

I6D 9 I pH 8.0.30"C

OCsHii

F F

IgG pH 7.0,25"CIR C 0 2 H HOCsHll

F F

OH

49.AG.659.12 pH 8.0,7°C

OH

FGOCHN

0 2

T S A

K, .0 E-2 pM

B

Enantiomers immunized separately

T S A

.4 E+ l

.1 E+ 2

.2 E+2

..3 E+ 2

8.9E+2

1.3 E-1

6.7 E-1

3.0 E- 2

8.9 E-1

8. 6 E-1

.8 E+:

-1.1 E+:-1.7 E+:

N

N

-1.8

-1.9

-. 1 0

~

. l l

-




pH9.0

21°C

2D,0, pH4°C Kinetic resolution30°C Kinetics

80% 88

1H o J = o ~ N o z

NO2

A d O D 0%

2H6 Rester to Ralcohol2H6-I R-ester to R-alcohol21H3 Sesterto Salcohol21~ 3 - I este r to Salc ohd

I = immobilized antibody

2H12E4 pH 8 0,24OC LO2I

2R, 3R

2 s , 3 s

2H6

2H6-I

21H3

21H3-1

T S A

2H6: Ki 2. 0 E+O FM

21H3: Ki 1. 9 E-1 pM

0

T S A

K , 2. 4 E+O pM

1.9 E+2

1.0 E+ 2

1.1 E+ 2

1.8 E+ 2

.3 E+ 3

1.0E+ 3

L.2 E+3

s.9 E+2

!.O E+2

.5 E+ l

-8.8 E-1

9. 1 E-1

9. 4 E-1

8.6 E-1

2.0 E+C

4.6 E+C

4.0 E+C

9.0 E-2

6.0 E- 2

1.9 E- 2

1.4 E+

s.3 E+

1.2 Ec

1.6 E+:

1 . 1 E+:

2.7 E+

.12

. 1 3

-.14

.15



12 Appendix 455

A R = 4-Nitrobenzyl 'OH

B R = 4-Nitrophenyl

p-examples shown below

0 2no-A R = ally1B R = ethylC R = mnitrobenzyl I :i:Iasic)

pH 9.0, 23°C

ROH0 2

I

AcOH HS-5

2. Aryl Ester H ydrolysis

0 2 o m N L C O z H

H5H2-42 pH 7.0,25%

H

N L C O z H1O

OnN

0 A

B

On ' OzC/'# 0

T S A

&o@% 0

0

in vitro Chemical Selection

TSA

Ki (substrateA) 1.2 E+1PM

OTYPIC CATALYSTS

Antibodieselicitedagainst mAbE-2,an anti- acetylcholinesteraseantibody

42.2E+1 pM

Heterologous Immunization

2.8 E+2

3.3 E+l

1.0E+2

3.4 E+3

2.8 E+3

1.5 E+3

6.0 E+2-l.4 E+2

_ __

7.4 E+O

3.4 E+O

3.0 E-2

6.2 E-2

1.1 E-2

4.7 E-1

4.9 E+3-1.3 E+l

-

2.6 E+5

7.2 E+3

3.0 E+l

2.9 E+4

8.8 E+3

1.4 E+6

4.2 E+8

--6.8 E+4

-

-1 . 1 6

-1 . 1 7

-1 . 1 8

1 . 1 9

-

-

2 . 1



456 I 1 Catalytic Antibodies

17E8 1 pH 8.7and 9.5

&oxYNozC3 pH 8.51e 0

+OH DNo2M r n HO

QoA-A 20G9, pH 8.8, 5°C

Bi 20G9. pH 8.5,35"CBi i 20G9 (reverse micelles) - Wo 23

TSA

K , 3.4 E+3 pM

pH 8.;

H3 pH 9.:0 BU

TSA

Ki 5.0E-1 pM

TSA

Kd 2.4 E-2 @vl

TSA

A , Ki 2.2 E-3 FM

Bi, Ki 3.9 E-2 pM

1.1 E+2

2.6 E+2

nr

little

variance

with pH

4.0 E+2

3.6 E+l

1.6 E+2

5.7 E+2

__

2.4 E+O-1.O E+2

2.2 E+2

-1.4 E+2-5.4 E-1

1.9 E+ l

3.9 E+O

-9.7 E+3-1.3 E+4

2.2 E+4

-3.5 E+6

6.9 E+1

1.7 E+4

N

-2 .2

-2.3

-2 .4

2 .5



12 Appendix

X = CH

x = N

X= CH

457

30C6, pH 7.2, 37°C

84A3,pH 7.0, 25°C

27A6, pH 8.3, 37°C

KD2-260, pH6.0,ZO"C7K16.2, pH 7.5,3O0CI

""Q AcOH

OH

NPN43C9, pH 9.3,25"C

Fab-1D, pH 7.2

6D 4 pH8.0,25%IFaJNmoHoQNA

27A6NH

m)P

H o d O

27A6: K i 6.0 E+OpM

Bait and Switch (BS)

ozyy+p?@-0zH

KD2-260

KD2-260:Ki ( 3 0 0 ~ 1 . 2-1@I

0% COzH

7K16.2

7K16.2:Ki 1.4E+ 2 @I

TSA

o z ~ > p , O o NF"43C9

Hb' Fab-1D

O Y H(CHdCQH

TSA

NPN43C9: Kd ( P ~) 1. 0 E+O pM

6D4:Ki 1.6 E-1 pM

T SA

. 1 E+3

.5 E+O

.4 E+ 2

-.9 E+O

.7 E+ 3

8.3 E + l

. I E+ 2

.9 E+O

~

5.0 E-3

(VP)

2.7 E+O

2. 0 E-3

(VP)

2.5 E+O

7.2 E-1

1.5 E+3(estimate

based on

pH rateprofile)

2. 5 E-1

1.6 E+O

-

-1 .O E+6

1. 2 E+3

N

3. 1 E+3

2.3 E+3

2.7 E+4

N

9.6 E+i

-




SOD8 pH 8.O,2S0CIAN+0

flomH

2

H6-32, pH 7.8, 25°C

H5-38,pH 7.0, 25°C

H7-59, pH 7.0, 25°C1H O n JoH

H

0 2

37G2 1 pH8.0,25'C

OH ' HCO(CH2)sCOd

I actone Formation

dNHAC4811 eNHAcPhO pH 7.0

PhOH5°C

0

50D8: , .0 E-2 pM

T S A

H 0 2 e O

H6-32: Ki 3.6 E+ 2fl

H5-38o2 / OH /

NH

H02&O

H5-38: Ki 5.0 E+Ofl0 Me3

H7-59

Qb

HO2

H7-59: Ki 2.3 E+ 2@I

o+ ,OH

% OO2

TSA, Kj 2.5 E- 1 pM

-1.5 E+ 3

-3.5 E+ 2

3.7 E+2

1.1 E+ 3-1.7 E+2

7.6 E+l-

-1.2 E+3

7.1 E-1

1.OE+O

4.9 E-1

2.0 E-1

5.0 E-1-

!.4 E+ 3

t.3 E+ 3

1.6 E+ 3

!.O E+ 3

-.7 E+ 2-



12 Appendix 459

S02MeH

I

49H4 1 pH 8.0,22"C

H S02Me

3. Carbonate H ydrolysis

02ao~o /I A 7K16.2.DH 7.5 . 30°C

"'"QC0 2 MeOH

OH

MOPC167 pH 7.O,3O0C

i 0 2 M e

in viw Ki 1.2E+1 pM

SOnMe R

Reactive Immunization (RI)

Semisvnthetl c Antibodies

Nucleoph ilic thiol groupswere introduced into a2,4-dinitrophenyl igandspec ific antibody binding siteby chemical modification

Ki (DNP-Gly) 2.5 E-1 pM

O H02 A

TSA

Ki 5 .0 E+O pM

1.0 E+2

1.2 E+O

3.3 E+3

6.6 E+2

4.3 E+2

6.8 E+2

2.1 E+2

3. 1 E+1

8.7 E-1

3.1 E-1

1.4E+O

4 .0 E+l

2 . 3 E + l

4.0 E-1

-5.7 E+3

5.0E+4

9 .3 E+2

8. 1 E+2

2.3 E+4

1. 3 E+4

7 .7 E+2

-! .14

!.15

3 . 1

3 . 2

-




A 7K16.2,pH 7.5,30°CB g, pH 8.5, 30°CC 48G7-4A1, pH 8.1, 30°C[soluble ( S ) , immobilizedO)]

(Sheep no. 270), pH 8.0, 25°C

(IVCAT2-6), PH 8.O,3O0C

OH HO-

TSA

Ki 5.0 E+O pM

Sheep no. 27C

NCAT2-f

Shcep no. 270, =OHIVCAT2-6, R = NH(CH,),CH,

TSA

Sheep no. 2 7 0 Kj 9.0 E-3 pM

NCAT2-6: Ki 2 .0 E+2 pM

-3.3 E+ 3

6.6 E+ 2

4.3 E+ 2

6.8 E+ 2-2. 1 E+ 2

3.3 E+O

2.8E+2

-g.9 E + l

b P P )

-

3.1 E-1

1.4 E+C

4.0 E+ 1

2. 3 E+ l

_ __

4.0 E- 1

1.7 E+Z

(basedon 1%

active

protein)

7.2 E + l

-2.1 E-1

b P P )

-9. 3 E+ ;

8.1 E t :

2.3 E+4

1.3 E+4-7.7 E+2

1.5 E+4

5 .5 E+3

4.3 E+ 3

b P P )

-3 . 1

3 . 2

-3 . 3

3 .4



12 Appendix

1.2E+2

8.0 E+l

4.1 E+l

5 .8 E+l

-

4. Carbamate Ester Hydrolysis

-

5.5 E+O

461

OH coz

Y = NO,, Br. F, OMe CozH

DFB-05 pH 6.5, 14%

la""ozH T C O ' "

Y

COzH

5 . Amide Hydrolysis

T S A

K,i 2.0 E+8 M

TSA

2.0E+2

4.3 E+2

-

1. 5 E+O

1.9 E+O-1.8 E+l

5.0 E+O

7.2 E+O

1.9 E+O

-

-

9.9 E-6

2.6 E+:

N-3.0 E+:

1.0Ew

1.0 E+r

1.2E+t

-

1.3 E+2

4.1

-4.2

1.3

5 . 1



46 2 11 Catalytic Antibodies

Ph(CHz)nyly-Phe-fl-AlaGlyOH

287F11 pH 6.5,37 C

O 1

* IPh(CH2)nyOH HZN-Gly-Phe-B-Als-GlyoH

0

ozQNmy(OzH

NPN43C9 pH 9.0.37 'C

NHz HO

Gln16-#-Met'7 VIP

Fab 1 pH 8.5,38 "C

VIP (1-16) VIP (17-28)

Tg

PolvclonalTg-AbITg-fragments

Boc-EAR-MCA

BJP-B6IBE-EAR MCA

8 R = CONHCHzCOzH

Metal com plex cofactor

T S A

Ki 1.0 E+l

TSA

TSA

. .nubod@

Human serum IgG fraction was

found to hydrolyze vasoactiveintestinal polypeptide (VIP).Unknown immunogen

Human serum IgG fractionwas found to hydrolyze

thyroglobulin (Tg).

Unknown immunogen

Bence Jones proteins (BJPs)

(monoclonal antibody light

chains) isolated from the urine ofmultiple myeloma patients, were

found to hydrolyze peptide

methylcoumarin amide

peptide-MCA substrates

N-i.6 E+Z

1.6 E + l

i.4 E+O

__

3.8 E-2-3.9 E-2

.5 E+1

-

3.6 E-2

8.0 E-2

4.5 E-4

3.6 E-1

(based on

1% active

protein)

1.6 E+ l

3.9 E- 3

3.3 E- 2

2 E+5

___

!. 5 E+.5

r.5 E+ 2

1.1 E+ ?

__

N-N

N

-

5.2

~

5.3

5.4

5 . 5

-5 . 6

5 .7

5 .8

-



12 Appendix 463

6. Phosphate Ester Hydrolysis

TX1-4C6 pH 8.5,3OoCI1. N- cvl Serinol Triester Hvdrolvsis

2. Paraoxon Hvdrolvsis

3H 5 OH""'o-?0.';- OEt 0.';- OEt

OEt OE t

0

Electrostatic TS Stabilization

h fluorescence quench) 6.7 E -l pM

( ~ ~ N 0 2 1 5 C 5H5

0

Electrostatic TS Stabilization and

B S

3H5: K , ( p ~.3) 9.8 E-1 pM

TSA

Ki <4.0 E-1 pM

-

1.8 E+l

3.6 E+1

~

1.7 E+l

F.1 E+3

1.3 E+2

2.4E+2

-

1.9 E-

1.0 E-

__

2.7 E-:

2.0 E-:

1.0E+I

3.2 E-;

-

3.6 E+:

9.8 E+:

1.3 E+:

3.5 E+:

j.5 E+?

1.1 E+2

-

6 .1

6 .2

6.3

5.4



464 I1 Catalytic Ant ibod ies

Fab fragment froman IgG

purified from human sera

pH 1.5, 30°C

7. Miscellaneous Hydrolyses

\

37C4, H 6.0polyclonal, pH 7.2

HOHO

FablB HO

Human serum IgG fraction (Fab)

was found to hydrolyze DN A.

Unknown immunogen

RH T O HH

Ki 3.4 E +l FM

37c4

polyclonal

OM*

TSA

37c4: Kd 2.5 E-2

TSA/ BS

Ki 3. 5 E +l FM

in vitro Chemical Selection

K; .5 E+ l uM

__

1.3 E+l

__

. .6E+ 2

--.1 E+I

3 .1 E+l

-3.2 E+ 2

i.3 E+ 2

1.4 E+1

1.2E-3

-1.0 E-1

2.0 E-2(based or

12 %

active

protein)

-1.5 E- 2

7.0E- 3

N

8.0 E+3

-

-2.7 E+

1.3 Et

__

N-7.0 E+

-7 .1

-7 .2

-

7.3



12 Appendix

2.1 E+2

BS and Entropic Trap

465

8 . 1

3.0 E-3 N

1409 OH

OH H

-1ization in Water

12%, >99% ee

D Ketal Hvdrolvsis

Ar

pH5.6, 4% 14D9i

87% ee

HOf

8. Eliminations

Disfavored s m F i r n i w

0

P h 5 FAH 9.0 P h y

37%

A

B

C

R D p D

E

&Lo.

R = CH2NHCO(CH2)&0&I

B S

A: d 1.0E-2 pM

-3. 4 E+Z

1.OE+Z

5.0E+ l

2.3 E+2

2.5 E+l

-5.7 E-3

4.7 E-3

1.2 E-2

1.OE-2

1.5 E-3

-1.5 E+:

1.0 E+1

5.0 E+i

1.3 E+i

1.4E+1

-7.4




1.7 E+2

2.6 E+2

20A2F6p E &H 0

0 2 0 2

2 3 - C o ~ e limination

1.7 E+1

2.3 E+l

21B12.1I pH 7.2, 37%

F2 Elimination

Selenoxide Elimination

Me0mSfDO2

SZ-28F8 pH 8.0, 25°CIMe0 NO2

9. Decarboxylations

21D8, pH 8.0, 0°C25E10, pH 8.0, 20°C1I 02mcN OH

43D4-3D21

20A2F6

0 2

20A2F6: Kl 1.6 E+ l pM

TSA

Ki 2.0 E-1 pMH

C02H

BS

C02H 0 Br

TSA

Ki 8.2 E-2 pM

SOsH 21D8

25E10

Medium Effect

21D8: Kl 6.8 E-3 pM

25E10 K; .4 E-3 uM

1.8 E+2

1.1 E+3

!.4 E+2

.2 E+2

.5 E+O-

1.9 E-1

3.5 E-4

2.4 E-5-1.0E+ l

1.8 E-1-I

8.8 E+

1.2 E+

-9 .1 E+

2 . 1 E+

1.6 E+

-

__

B.2

-8 . 3

-8.4

-3. 5

-

1.9E+4

2.3 E+4I



10. Cycloreversions

9 .2

-. 3

-.4

-

Retro Diels-AlderRe-

HNO

A 300 nm

R I R I Rq

A 15F1-381,R, =OH, R2 = H,

B UD4C3.5,R,= NHCH2C02Me, = H

1.3 E+2

Heterologous Immunization

(with hydroxylated form)

&(pH 9.0) 9.0 E-1 P M

A 6. 5 E+O

B 2.8 E+2

R = Me cis, synR = H trans,syn

15F1-B1: Ki < 1.0 E+O pMm 4 c 3 . 5 : Kd (fluorescencequench)

12 Appendix

2.1 E-4

1.6 E-1

2.8 E-2

7.3 E-2

1. 2 E+O

4.1 E-1

-1 O E+8

1.5 E+4

1.9 E+5

2.3 E+2

2.2 E+2

3.8 E+2

10.1

~

10.2

-



468 11 Catalytic Ant ibod ies

11. Retro-He nry and Retro-Aldol Reactions

2.2 E+2

AcHJy+

4.8 E+O

AcHN(4R. 55) (>95 % de)(4R, 5R) (43% de)

d H

AcHN(45 . 5s) (>95 % de)(45 , 5R) 65 % de)

3. NTRAMOLECULAR PROCESSES

12 . Isomerizations

Reaction / Conditions

. . .&@dvl-Drolvl cistrans isom-

""aHz 0

VTT1E3pH 8 .0 ,4"C

0 2

. .somenzabm

P N O z H7.5Y110-4

25°C0 2 '

Hz

TSA

Ki (app) 2.6 E+O pM

Hapten 1Comments / K

R CO(CHz)&OOH

TSA, Kj 1.0E+1 pM

CO&I

BS, Ki 6. 7 E+O pM

kcatlKm

1. 3 E+2 M-*min-'

(kcat and Km ot

measured separate1

kcat/KmM-lmin-'bPP)

1 . 1 E-1

2.2 E-2

7.8 E-2

2. 2 E-1

kimid

2.5 E-4

M-1

min-1-N

N

2.7 E+

1.5 E+

11.1

. 1 . 2

-Entry

AE

12 .1

-

12.2



12 Appendix 469

&0 ' \9D5H12

13. Rearrangements

QJcdH

lF7. pH 7.5,14"C11F1-2E11, pH7.0. 10°C

02@COp

i

OH

T S A

K, 1.3 E+2 pM

8 CO2H

TSAKd (BIAcore) 2.1 E-1 pkf

\

ph&coR

HO

R = NH(CH&O(Cb)zNHz

T S A

Kj 3.0 E-2-1 .6 E-1 pM

B

" " ~ c o P

+ q ( O H 1F7

0 0

1F7:Ki 6.0 El

H

1 lF1-2E11

0

l lFl-2E11: Kj9.0E+OpM

1.6 E+2

4.2 E+2

1.9 E+1

!.6 E+2

2.3 E+O

2.6 E-3

2.6 E-2

2.3 E-2

2.7 E+O

8.3 E+2

2. 9 E+ 3

5.3 E+

2.5 E+

1.0E+

__

12.3

12.4

.3.1

.3.2



470 11 Catalytic An tibod ies

Ar Qo

HO NH

62C7 1 pH5.8

9NO2

Ab-DNP pH 7.4,23%

NOzI

&CpNO2

COzH

TSA

2 B 4 Ki 1.0E-1@I

r‘HO

TSA

X = Linker

~

1.9 E+2

8.3 E-1

6.7 E+2

1.7 E+5

7 .2 E-3

3.6 E+1

1.3 E-5

1.8 E+1

7.0E+

nr

-1.0 E+I

1.9E+r

3 . 3

3.4

. 3 .5

-



12 Appendix

TX1-4C6, pH 7.0 , (biphasic)TM1 -87D7, pH 7.0 , (biphasic)

14. Epoxide Opening

471

15. Cationic Cyclization

0 6"

Me80% 63% 60%

from R = cis-Me from R = frans-Me from R =H

1

1

uL

HO

Qo

TSA

TSA

TX1-4C6: Ki 1.0E+O pM

TM1-87D7: Ki 1.4E+O

A

B

H

T S A

A: Ki 1.0E+O pM

B: Ki 1.0E+O pM

C: Ki 1.0E+O pM

-

3.6 E+2

2.0E+2

-

1.3 E+;

1.5 E+1

i.8 E+1

I .O E+Z

1.1 E+1

-

9.0 E-1

9.0 E-l

2.0 E-2

2.0 E-2

1.3 E-2

2.1 E-2

1.0E- 2

N

N

N

N

N

. 4 . 1

15.1

.5 .2



472 11 Catalytic An tibod ies

Kln

[pM]

3.6 E+2

4.7 E+2

1.7 E+l

5.4 E+1

HA1-17G8pH 7.0 1 % O H(biphasic)

ct : 1a&J l

kc at

[min-'1

1.4E-4

4.9 E-4

6.7 E-3

4.4 E- 3

HA5-19A4pH 7.0

0 (biphasic)

m&m2 : 3 : 1

TSA

IC501.2EtO pM

TSA

Ki 1.4E+O PM

3.5 E+1

3.2 E+2

C. BIMOLECULAR ASSOCIA'ITW AND SUBSTITUTIONPROCESSES

16. Aldol Reactions


70H6pH 7.5

B

o&H -

~~~ ~

Hapten / Comments (Ki)

A

BS

EOlH Retro-ddol

Reactive Immunization

2.5 E-2

2.1 E-2

7.0E+1

2.9E4

N

-15.3

-15.4

-

-Zntr

AE

16.1

-6.2



12 Appendix

H 2 T & o

HO- NH

473

;~D~oor9.3 ,20~c1-10 % vhr acetone

17. Diels-Alder Cycloaddition

OA C 0

H11 I pH8.0,18"C

Dienophile

([Diene]0.61 mhf)

CI

TSA, Entropic Trap

DieneDienophile

TSA,

Kj 1.3 E-1 PM

Diene

TSA

1.8 E+ 3

b P P )

4.9 E+ 3

(aPP)

2.1 E+4

b P P )

1. 1 E+ 3

1.4E+ 2

N

8.3 E+3

1 . 1 E+ 3

1.8 E-4

@PP)

9. 6 E-5

b P P )

4.3 E+O

@PP)

4.0E+l

N

3.3 E+l

5.5 E-2

-1.OE+2

b P P )

1.7 E+2

b P P )

,.1 E+2

@PP)

-3.5 E-1

N

L.8 E+1

N

L6.3

.7 .2

.7.3




2.3 E-2

(app)

6.6 E-2

O=FCAr

309-1G7 pH 7.3

18.1

1 .1 E+ I

(app)

18.2

1.6E+1

Ar = o-C.H.CONHPr

IHO

7D4, pH 7.4.37"C

22C8, pH 7.4, 37%

4D5. pH 7.4,37%1 3 ~ 5 ,n 7.4,37oc

CO2H COzH

18. Acyl Transfer Reactions

0" Hz1;;YO, 23°C

mNHCocH30".hOH

24811 I pH7.0

Dienophile( [ r r m Diene]5.0mM )

R = CH2)5C02H Dienophile

([cis Diene] 5.0 mM)

T S A

Dienophil

1365 em) Dien

Dienophil

HN Ester

o% ([Amine]20 mM )

COzH

TSA, IC50 1.0E+1 pM

3. 1 E+3

@PP)

3.9 E+3

( w p )

__

9.6 E+2

1.7E+ 3

7.0E+Z

7.5 E+3

1.6 E+3

5.9E+ 3

2.7 E+3

1.OE+4--

2.2 E+3

b P P )

4.9 E+3

1.2 E+3

3.4 E- 3

3.2 E-3

3.5 E-3

1.2 E-3-

1.2E+

bPP)

2.6 E+

@PP)

4.8 E+

1.8 E+

4.9 E+l

6.9 E+

-

17.4

7.5

-



12 Appendix 475

B'Q,,, 1 965.1, pH 7.4

0

'2

NC

'N+Ph

H

A 21 H3, pH 9.0

B 21H3. pH 8.5,23"C

(96% octane)

t

CHCHO

Azide

Amine

TSA

$7 Alcohol

vJoster

TSA

Kd 2.4 E-4 fl

A Ester

alcohol

d: ' .~poEster

alcohol

TSA

21H3: Ki (app) 2.0 +O pM

-1.5 E+1

1.5 E+:

-L O E+?

1.6 E+4-1.7 E+i

1.6 E+i

-3.0 E+?

1.3 E+?

1 . 1 E+;

1.3 E+:

-5.9 E- 2

_ _ _

1.3 E+l-1.4E+ l

-2. 1 E+1

3.0 E+O-

-1 .OE+4

-1.9 E+2-5.5 E+4

-N

nr-




15A9, pH 7.5,25"C

19.Amination Reactions

2OAF2F6, pH 7.3,25"C,syn :antiS:l43D4-3D12, pH 6.5,25"C,syn anti1:9

Antisera-ATE3 ~ ~ ~ c o z 'Go L-Phe.

0

17C5-11C2pH 7.0, 25°C

Transamination

(loo PLP)

COzH

2G0

Elimination

(looph4 PLP)

20AF2F6 Ketoni

([NH2OH] 20mM0 NH3

J

0 43D4-3D12 Ketoni

([NHzOH] 20 mM

~LOzH

0

TSA

m H : A Amino acid

HO* mdoxd

L-amino acid Kd 6. 0 E-3 pM

D-tUl lh lO acid:Kd 1.7 E-2

2.7 E+3

(aPP)

9.4 E+2

b P P )

3.9 E+3

1.6 E+2

1.2 E+2

7.1 E+2

4.2 E-1

bPP)

5.0 E+1

bPP)

-

1 . 1 E+l

bPP)

6.7 E+O

@PP)-1.5 E+ l

1.8 E+1

-2.1 E-1

-

8.7

-

9.1

-9 .2

9.3

-



12 Appendix


D" a'04

2884.2 pH 5.5, 23%

f l?00 NalOJ

102

0 2

477

Hapten / Comments ( K )

0 2p p g OSulfide

NaIO4COPH

T S A

&j 5.2 E-2

20. Miscellaneous

P -Ph- NHAC

I

1685 I Nal, 37°C (biphasic)

Ph- Ai+ ' HO-+ R - NHAC

I

564, NaCN I pH 7.0,37"C

&pHO-0 ' CN

Potphyrin + M2+

7G12-A10-61-A12pH 8.0,26"C1Porphyrin-M2*

Sulfonate

( [NaI] 0.15M)

' h ; s c a y y ~0 o

H o d o

NaI

([Sulfonate] 0.75mM )

T S A

Kj ca . l . O E + l p.M

Enone

T f lCN-

T S A

K, 1.1 E+l pM

COzH COzH

-1.3 E+ 2

b P P )

1.5 E+5

b P P )

5.4 E+ l

I 4 E+ 2

1.9 E +l

5.0 E+1

-

-2.8 E-5

b P P )

2.8 E-2

@PP)

2.1 E-2

5 . 2 E -4

8.4 E-5

-

4. 3 E + l

2.5 E+28 . 2 E+OI

3.0 E-i

-2.6E+:

1.7 E+:

-




4.0 E+2

20811, CH,CN

T~r H202,pH 6.6 *

66?bec

2.4 E+l-4GlZ-AlO-Gl -A12

pH 8.0, 10°C

Fe(lll), H202,

0H rnesoporphyrin

22. Reductions

"QiBi

NaBH,CN A5(1 mM) pH 5.0,2ZoCI

de > 99 %

Safranine T [O]

rnAb, flavin. dithionite1SafranineT [R]

R

R

Metal cofactor complex

YA,inke

TSA

COzH COM

Metal cofactor complex

0

Cofactor complex

Kd 8.0E-3

-N

2.6 E+2

bPP)

2.4 E+4

N

enhance

ment1I

-

2. 9 E+:

bPP)

N

-

11.1

1.3

1 .4

-!2 .1

12.2



12 Appendix 479

HO 0

66D2 1 pH5.8.25"C

HO

37839.3

pH 5 .0 ,Z "CR = isopropyl NaCNBH,R = Et

R = B n I0DHRR = Et. 99 % S

A;::\ \ \

0

02R = E t

50 mM NaBH3CN

0.15mM NaBH3CN

T S A

Kd 3.3 E-2

12.2 Key to Bibliography

A Hydrolytic and DissociativeProcesses

1. Aliphatic Ester Hydrolysis1.1SHEN t al., 1992;1.2 TANAKA t a]., 1996;1.3 3B9 LANDRYt al., 1993,15A10 YANGet

al., 1996, polyclonal BASMADJIANt a]., 1995;1.4 NAKATANIt al., 1993; 1.5 IKEDAet al.,1991;1.6 FUJII t al., 1991;1.7 IWABUCHI et al.,1994;1.8 MIYASHITAt a]., 1993;1.9 KITAZUMEet al., 1994;1.10 CAMPBELLt al., 1994;1.11KI-TAZUME et al., 1991a; 1.12 KITAZUME et al.,1991b;1.13 IKEDAnd ACHIWA,997;1.14 JAN-DA et al., 1989,1990a;1.15 POLLACKt al., 1989;1.16 A TAWFIKt al., 1993, B TAWFIKt al.,

1997); 1.17 JANDA et a]., 1994; 1.18 LI et al.,1995b;1.19 IZADYDAR t al., 1993.

2. Ar yl Ester Hydrolysis2.1 SUGA t a]., 1994b;2.2 TAWFIKt al., 1990;2.3 pH 8.7 G u o et al., 1994,pH 9.5 ZHOU t a].,1994;2.4 KHALAFt a]., 1992;2.5 A MARTINt

al., 1991,B DURFORt al., 1988;2.6 30C6 JAN-DA et al., 1990b, 84A3 WADEet al., 1993,21A6JANDA et al., 1991b;2.7 KD2-260 OHKUBOt

al., 1993, 7K16.2 SHOKATet al., 1990; 2.8NPN43C9 GIBBS t al., 1992a,Fab-1D CHEN t

al., 1993; 2.9 TRAMONTANOt al., 1986; 2.10

3.0 E+3

6.0E-11.2E+O 6.0 E+l

kuncat

1 . 1 E-3

min-1

M.1

i2.3

12.4

TRAMONTANOt a]., 1988; 2.11 SUGAet a].,1994a;2.12 JANDAet al., 1991a;2.13 NAPPERt

al., 1987;2.14 WIRSCHINGt al., 1995;2.15 POL-LACK et al., 1988.

3. Carbonate Hydrolysis3.1 A SHOKATt al.,1990,B JACOBSet al., 1987,C SPITZNAGELt al., 1993;3.2 POLLACKt al.,1986; 3.3 "Sheep no. 270" GALLACHERt al.,1991,IV-CAT 2-6 STAHLt al., 1995;3.4 WAL-LACE and IVERSON, 1996.

4. Carbamate Ester Hyd rolysis4.1 VANVRANKENt al., 1994;4.2 WENTWORTHet a]., 1996;4.3 WENTWORTHt al., 1997.

5. Am ide Hydrolysis5.1 MARTINt al., 1994;5.2 IVERSON nd LER-

DET~Iet al., 1996;5.5 GALLACHERt al., 1992;5.6 PAULet al., 1989; 5.7 LI et al., 1995a; 5.8PAUL t a]., 1995.

NER, 1989; 5.3 JANDA et al., 1988b;5.4 BENE-

6. Phosphate Ester Hydro lysis6.1 ROSENBLUMt al., 1995;6.2 15C5 LAVEYand JANDA, 1996a, 3H5 LAVEYand JANDA,1996b;6.3 BRIMFIELDt al., 1993;6.4 WEINERet al., 1997;6.5 SHUSTERt al., 1992,GOLOLO-BO V et al., 1995;6.6 SCANLANt a]., 1991.




7. Miscellaneous Hydrolyses7.1 37C4 IVERSON t al., 1990, polyclonal STE-PHENS and IVERSON, 993;7.2 Y u et al., 1994;7.3 JANDA et al., 1997;7.4 A REYMONDt al.,

1992, 1993, 1994, B REYMONDt al., 1991, CSHABATt al., 1995, D SINHA t al., 1993b, ESINHA t al.. 1993a.

8. Eliminations

KAT et al., 1989,20A2F6 UNO and SCHULTZ,1992;8.3 YOONet al., 1996;8.4 THORN t al.,1995;8.5 ZHOU t al., 1997.

8.1 CRAVA'ITet al., 1994;8.2 43D4-3D21,SHO-

9. Decarboxylations9.1 21D8 LEWISt al., 1991,25E10TARASOWtal., 1994;9.2 SMILEYnd BENKOVIC,994;9.3BJORNESTEDTt al., 1996;9.4 ASHLEY t al.,1993.

10. Cycloreversions10.1 BAHR t al., 1996;10.2 A COCHRANt al.,1988,B JACOBSEN t al., 1995.

11. Retro-Henry and Retro-A ldol Reactions11.1 FLANAGANt al., 1996; 11.2 REYMOND,1995.

B Intramolecular Processes

12. Zsomerizations12.1 YLI-KAUHALUOMAt al., 1996; 12.2JACKSONand SCHULTZ,991;12.3 KHETTALtal., 1994;12.4 UNOet al., 1996.

13. Rearrangements13.1 BRAISTEDnd SCHULTZ,994,ULRICHtal., 1996; 13.2 1F7 HILVERTt al., 1988, HIL-

VERT and NARED, 988,llF1-2Ell JACKSONetal., 1988;13.3 2B4 GIBBS t al., 1992b, RG2-23C7 LIOTTAet al., 1993;13.4 CHEN t al., 1994;13.5 WILLNERt al., 1994.

14. Epoxide Opening14.1 1.JANDA et al., 1993,SHEVLINt al., 1994,2. JANDA et al.. 1995.

C Bimolecular Associative and SubstitutionReactions

16. Ald ol Reactions

16.1 KOCH et al., 1995; 16.2 WAGNER t al.,1995; 16.3 A REYMONDnd CHEN,1995a, B

REYMONDnd CHEN,995b.

17. Diels-Alder Cycloaddition17.1 HILVERTt al., 1989;17.2 BRAISTEDndSCHULTZ,990;17.3 SUCKLINGt al., 1993;17.4trans MEEKELt al., 1995, cis RESMINIt al.,1996;17.5 7D4 and 22C8 GOUVERNEURt al.,1993,4D5 and 13G5 YLI-KAUHALUOMAt al.,1995.

18. Ac yl Transfer Reactions18.1 JANDA et al., 1988a;18.2 BENKOVICt al.,1988;18.3 JACOBSENand SCHULTZ,994;18.4HIRSCHMANNt al., 1994, SMITHRUDt al.,1997;18.5 JACOBSEN et al., 1992;18.6 A WIR-SCHING et al., 1991, B ASHLEY nd JANDA,

1992;18.7 GRAMATIKOVAnd CHRISTEN,996.

19. Amination Reactions19.1 UNO t al., 1994;19.2 TUBUL t al., 1994;19.3 COCHRANt al., 1991.

20. Miscellaneous20.1 LI et al., 1995d;20.2 COOK t al., 1995;20.3COCHRANnd SCHULTZ,990a, for a secondexample see KAWAMURA-KONISHIt al., 1996.

D Redox Reactions

21. Oxidations21.1 HSIEH t al., 1994;21.2 KEINAN t al., 1990;21.3 KOCH et al., 1994; 21.4 COCHRANnd

SCHULTZ,990b.

22. Reductions22.1 NAKAYAMAnd SCHULTZ,992;22.2 SHO-KAT et al., 1988;22.3 JANJICand TRAMONTANO,1989;22.4 HSIEH t al., 1993.

15. Cationic Cyclization15.1 TX1-4C6 LI et al., 1994,TMI 87D7 LI etal., 199%;15.2 LI et al., 1996;15.3 HASSERODT

et al., 1996;15.4 HASSERODTt al., 1997.



13 References 481

13 References

AD DA DI, ., JAFFI, E. K., KNOWLES,. R. (1983), Se-condary tritium isotope effects as probes of theenzymic and nonenzymic conversion of choris-mate to prephenate, Biochemistry 22,4494-4501.

ALB ERG , . G., LAU HO N, . T., NYFE LER , ., FASS-LER, A., BARTLETT,. A. (1992), Inhibition ofEPSP synthase by analogues of the tetrahedralintermediate and of EPSP,J.Am . Chem.SOC.114,3535- 3546.

AL BERY , . J. (1993), Trans ition state theory revisit-ed , Ad v. Phys. Org. Chem.28,139-170.

ALBERY,., KNOWLES,. R. (1976), Evolution ofenzyme function and development of catalyticefficiency,Biochemistry 15,5631-5640.

ALBERY,., KNOWLES,. R. (1977), Efficiency andevolution of enzym e catalysis,Angew. Che m. (Znt.Edn. Engl.) 16,285-293.

ASHLEY,. A., JAND A, . D. (1992), Antibody cataly-sis in low water content media, J . Org. Chem. 57,6691-6693.

ASHLEY,. A., Lo, C.-H. L., MCELHANEY,. P., WIR-SCHING,., JAN DA , . D. (1993), A catalytic anti-body model for PLP-dependent decarboxylases,J. Am . Chem. SOC.115,2515-2516.

BAGSHAWE,. D. (199O), Antibody-directed enzy-me/prodrug therapy (ADEPT), Biochem. SOC.

Trans.18,750-752.BAHR,N., GULLER, ., REYMOND,.-L., LER NER , .

A. (1996), A nitroxyl synthase catalytic antibody,J.Am. Chem. SOC .118,3550-3555.

BALDWIN,. E. (1976), App roach vector analysis: Astereochemical approach to reactivity, J. Chem.SOC.Chem. Commun. 738-741.

BALDWIN,. E., CHESWO RTH,., PARKER,.S., Rus-SELL, . T. (1995), Studies towards a postulatedbiomimetic Diels-Alder reaction for the synthe-sis of himgravine, Tetrahedron Lett. 36, 9551-9554.

BALDWIN,. E., CLARIDGE,. D. W., CULSHAW, . J.,HEUPEL,. A,, LEE,V.et al. (1998), Investigations

into the manzam ine alkaloid biosynthesis hypo-thesis, Angew. Chem. (Znt. Edn. Engl.) 37,2661-2663.

BARTLETT,. A., GIANGIORDANO,. A. (1996),Transition-state analogy of phosphonic acid pep-tide inhibitors of pepsin, J . Org. Chem. 61,3433-3438.

BARTLETT,. A., JOHNSON ,. R. (1985),An inhibitorof chorismate mutase resembling the transition-state conformation, J.Am . Chem. SOC.107,7792-7793.

BARTLETT,P. A,, LAMDEN,. A. (1986), Inhibition of

chymotrypsin by phosphonate and phospho nami-date peptide analogs,Bioorg. Chem. 14,356-377.

BARTLEIT, . A., MARLOWE, . K. (1983), Phosphon-amidates as transition state analogue inhibitors ofthermolysin,Biochemistry 22,461-624.

BASMADJIAN,. l?, SINGH,S., SASTRODJOJO,.,SMITH, . T.,AvoR , K. S. et al. (1995), Gen erationof po lyclonal catalytic antibodie s against cocaineusing transition state analog s of cocaine conjuga-ted to diptheria toxoid, Chem. Pharm. Bull. 43,

BENEDETTI,., BER TI, .,COLOMBATTI,A.,BERT, .,LIND A, ., TONIZZO,. (1996), anti-Sulfonamideantibod ies catalyse the hydrolysis of a heterocy c-lic amide , Chem. Commun. 1417-1418.

BENKOVIC,. J., NAPPER ,A. .,LE RNE R, . A. (1988),Cataly sis of a stereospec ific bimo lecular amidesynthesis by an antibody, Proc. Natl. Acad. Sci.

BJ~RNESTEDT,., ZHON G, ., LERNE R, . A., BAR-BAS 111, C. F. (1996), Copy ing natu re’s mechanismfor the decarboxylation of p-keto acids into cata-lytic antibodies by reactive immunization, J . Am.Chem. SOC.118,11720-11724.

BLAKEY,. C. (1992), Drug targeting w ith monoclo-nal antibodies - a review, Acta Oncologica, 31,

BLAKEY, . C., BU RKE ,. J., DAVIES, . H., DOW ELL,R. I., MELTON, . G. et al. (1995), An tibody-d irec-ted enzyme prodrug therapy (AD EP T) for treat-ment of major solid tumour disease, Biochem.SOC.Trans.23,1047-1050.

BLOKZIJL,., ENGBERTS,. B. F. N. (1994), Enforcedhydrophobic interactions and hydrogen-bondingin the acceleration of Diels-Alder reactions inwater,ACS Symp. Series 568,303-317.

BOW MA N, . C., RAN D,M. J. (1988), Textbook ofPharmacology. Oxford: Blackwell Scientific Pu-blications.

BRAISTED, . C., SCH ULT Z,. G. (1990), A n Antibo-dy-Catalyzed bimolecular Diels-Alder reaction,J . Am . Chem.SOC.112,743Ck7431.

BRAISTED,A. ., SCHULT Z,? G. (1994),An antibody-catalyzed oxy-Cope rearrangement, J . Am. Chem.

BRIMFIELD,. A., LENZ,D. E., MAXWELL, . M.,BROOMFIELD,. A. (1993), Catalytic antibodieshydrolysing organophosphorus esters, Chemico-Biolog ical Znteractions - England 87,95-102.

BURTON, . R. (1990), Antibody: The flexible ad-aptor molecule, Trends Biochem. Sci. 15,64-69.

CAMPBELL,. A., GONG , ., KOC HERSPE RGER ,. M.,YONKOVICH,., GALLOP,M. A., SCHULTZ,. G.(1994), A ntibody-catalyzed prod rug activation,J.

Am . Chem. SOC. 16,2165-2166.CHARBONNIER,.-B., CARPENTER, ., GIGANT,B.,

GOLINELLI-PIMPANEAU,., ESHHAR,Z. et al.(1995), Crystal structure of the complex of a cata-lytic antibody Fab fragment with a transition-sta-

1902-1911.

USA 85,5355-5358.

91-97.

SOC.116,2211-2212.




te analog - structural similarities in esterase-likecatalytic antibodies, Proc. Natl. Acad. Sci. USA

CHARBONNIER,.-P., GOLINELL I-PIMPA NEAU ,., GI-GANT, B., TAWFIK,D. s., CHAP, R. et al. (1997),Structural convergence in the active sites of a fa-mily of catalytic antibodies, Science 275, 1140-1142.

CHE N, .-C. J., DA NO N,T.,ASTRY,., MUB ARA KI,.,JANDA, . D., LERN ER, . A. (1993), Catalytic an-tibodies from combinatonal libraries, J . A m .Chem. SOC.115,357-358.

CHEN, ., REYMOND,.-L., LERNE R, . A. (1994),A nantibody-catalyzed 1,2-rearrangement of carb-on-carbon bonds, Angew. Che m. (Znt. Edn. Engl.)

CHOOK, . M., KE, H. M., LIPSCOMB, . N. (1993),Crystal sturctures of the monofunctional choris-mate mutase from Bacillus subtilis and its com-plex with a transition state analog, Proc. Natl.Acad . Sci. USA 90,8600-8603.

C H O O K , ~ .., GRA Y,.V., KE ,H. M. L., LIPSCOMB ,.N. (1994),The m onofunctional chorismate muta-se from Bacillus subtilis - tructure determinationof chorismate mutase and its complexes with atransition-state analog and prephe nate and impli-cations for the mechanism of enzyma tic reaction,J . Mol. Biol. 240,476500.

CIOBANU,., MATSUM OTO,. (1997),Recent advan-ces in organic synthesis under high pressure, Lie-

bigs Ann. Chem .4,623-635.COCHRAN,. G., SCHULTZ,.G. (1990a),Antibody-catalysed porphyrin metallation, Science, 249,

COCHRAN,. G., SCHULTZ ,. G. (1990b),Peroxidaseactivity of an antibody-heme complex, J . Am.Chem. SOC. 12,9414-9415.

COCHRAN,. G., SUGASAWARA,., SCHULTZ,. G.(1988), Pho tosensitized cleavage of a thym ine di-mer by an antibody,J . Am . Chem. SOC.110,7888-7890.

COCHRAN, . G., PHAM,T., SUGASAWARA,.,SCHULTZ,. G. (1991), Antibody-catalyzed bi-

molecular imine formation,J .

Am. Chem. SOC.113,667M672.COO K, . E.,A LL EN , . A., MILLER , . B., WHISNANT,

C. C. (1995), Antibody-Catalyzed Michael reac-tion of cyanide with an cy,&msaturated ketone, J .

A m . Chem. SOC. 17,7269-7270.CRANS, . C., FELTY,R. A., MILLER,M. M. (1991),

Cyclic vanadium (V) alkoxide: An analogu e of theribonuclease inhibitors, J . Am. Chem. SOC. 113,

CRAVAIT,B. F.,ASHLEY,. A., JAN DA , . D., BOGE R,D. L., LERNER, . A. (1994), Crossing extrememechanistic barriers by antibody catalysis: Syn

elimination to a cis olefin,J .

Am . Chem. SOC.116,

92,11721-11726.

33,1607-1609.

781- 783.

265-269.

6013-6014.DANISHEFSKY,., HERSH ENSO N,. M. (1979), Regio-

specific synthesis of isogabaculine, . Org. Chem.44,1180-1181.

DAITA, A., PARTRIDGE,. J., BLACKBURN,. M.(1996), European network on antibody catalysis1993-1995. Brussels, Luxembourg, CSC-EC-EA-EC, 36.

DAVIDSON,. M., HILLIER ,. H. (1994),Claisen rear-rangement of chorismic acid and related analo-gues: An ab initio molecular orbital study, J .

Chem.SOC.Perkin Trans. I I , 1415-1417.DE LAUZON,., DESFOSSES,., MANSUY ,., MA HY,.-

P. (1999),Studies of the reactivity of artificial per-oxidase-like hemoproteins based on antibodieselicited against a spec ifically designed ortho-car-boxy substituted tetraarylporphyrin, FEBS Lett.

DELL,C. P. (1997), Cyc loadditions in synthesis, Con-temp. Org. Synth. 487-117.

DURFOR,C. N., BOLIN,R. J., SUGASAWARA,. J.,MASSEY, . J., JACOBS,.W., SCHULTZ,. G. (1988),Antibody catalysis in reverse micelles, J. Am .Chem. SOC. 10,8713-8714.

ERSOY,O., FLECK,R., SINSKEY,., MASUMUNE,.(1998),Antibody catalyzed cleavage of an am idebond using an external nucleophilic cofactor, J.Am. Chem. SOC.120,817-818.

ESCHENMOSER,A.,UZICKA,., JEGER, .,ARIG ON I,D. (1955), Eine stereochemische Interpretation

der biogenetischen Isoprenregel bei d en Triterpe-nen, Helvet. Chim.Acta 38,189&1904.EVANS,M. G., POLANYI,M. (1935), Some applica-

tions of the transition state method to the calcu la-tion of reaction velocities, especially in so lution,Trans. Faraday SOC .31,875-894.

EYRING, . (1935), The activated complex and theabsolute rate of chemical reactions, Chem. Rev.

FENNIRI, .,JANDA , . D., LER NER , . A. (1995),En-coded reaction cassette for the highly sensitivedetection of the making and breaking of chem icalbonds, Proc. Natl. Aca d. Sci. US A 92,2278-2282.

FERSHT,A. F. (1985), Enz yme Structure and Mecha-nism, p. 98-154. New York: Freeman.FINN, . G., LERN ER, . A., BARB ASl l , C. F. (1998),

Cofactor-induced refinement of catalytic anti-body activity: A metal-specific allosteric effect, J.Am . Chem. SOC.120,2963-2964.

FLAN AGA N, . E., JACOB SEN,. R., SWEET,E.,SCHULTZ,. G. (19%), Antibody-catalyzed retro-aldol reaction,J. Am . Chem. SOC.118,607843079.

FUJII, ., LERNER, .A., JANDA, . D. (1991), Enan-tiofacial protonation by catalytic antibodies, J .

Am. Chem. SOC.113,852843529.FUJII,., TANAKA,., MIYASH ITA,., TANIMU RA,.,

KINOSHITA, . (1995), Correlation between anti-

443,229-234.

17,65-77.



13 References 483

gen-combining-site structures and functions wit-hin a panel of catalytic antibodies generatedagainst a single transition state analog, J. A m .Chem. SOC.117,6199-6209.

GALLACHER,., JACKSON,. S., TOPHAM,. M., SE-

ARCEY, M.,TuRNER,B. . et al. (1990), Polyclonal-antibody-catalysed hydrolysis of an aryl nitro-phenyl carbonate, Biochem. SOC.Trans. 18, 600-

601.

GALLACHER,., JACKSON,. S., SEARCEY,., BAD-

MAN, . T., GOEL,R. et al. (1991), A polyclonalantibody preparation with Michaelian catalyticproperties, Biochem. J . 279,8714381.

GALLACHER,., SEARCEY,M., JACKSON, . S.,BROCKLEHURST,. (1992), Polyclonal antibody-catalysed amide hydrolysis, Biochem. J . 284,675-680.

GAO,C., LAVEY, . J., Lo,C. H. L., DATTA,., WENT-

WORTH JR.,P., JANDA, . D. (1998), Direct selec-tion for catalysis from combinatorial antibody li-braries using a boronic acid probe: primary am idebond hydrolysis, J. Am. Chem. SOC.120, 211-2217.

GIBBS, .A., BENKOVIC,.A., JANDA,. D., LERNER,

R. A., BENKOVIC,. J. (1992a), Substituent effectson an antibody catalyzed hydrolysis of phenylesters: Further evidence for an acyl-antibody in-termediate,J. Am. Chem. SOC. 14,3528-3534.

GIBBS, .A.,TAYLoR,., BENKOVIC,.J. (1992b),An-tibody-ca talysed rearrangemen t of the peptide

bond, Science 258,803-805.GIGANT,B., CHARBONNIER,.-B., ESHHAR,Z.,GREEN, . S., KNOSSOW,. (1997),X-ray structu-res of a hydrolytic antibody and of com plexes elu-cidate catalytic pathway from substrate bindingand transition state stabilization through waterattack and product release, Proc. Natl. Aca d. Sci.

GOLINELLI-PIMPANEAU,., GIGANT,., BIZEBARD,T.,

NAVAZA,., SALUDJIAN,. et al. (1994), Crystalstructure of a catalytic antibody with esterase-like activity,Structure 2,175-183.

GOLOLOBOV,. V., CHERNOVA,. A., SCHOUROV,.

V., SMIRNOV,.V.,KUDELINA,.A., GABIBOV,A..(1995),Cleavage of supercoiled plasmid DNA byautoantibo dy Fab fragment: Application of theflow linear dichro ism techniqu e, Proc. Natl. Aca d.Sci. US A 92,254-257.

GOUVERNEUR,. E., HOUK,K. N., DE PASCUAL-TE-

RESA, ., BENO,B., JANDA, . D., LERNER,. A.(1993), Control of the en0 and endo pathways ofthe Diels-Alder reaction by antibody catalysis,Science 262,204-208.

GRAMATIKOVA,. I., CHRISTEN,. (1996), Pyridoxal5 ‘-phosphate-dependent catalytic antibody, J.

Biol. Chem .271,30583-30586.

GROSSBERG,. L., B PRESSMAN,. (1960), Nature of

USA 94,7857-7861.

the combining site of antibody against a haptenbearing a positive charge, J. Am . Chem. SOC.82,5478-5482.

GUO, ., HUANG, ., SCANLAN,. S. (1994), Kineticand mechanistic characterization of an efficienthydrolytic antibody: Evidence for the formationof an acyl intermediate, J. Am . Chem. SOC.116,6062-6069.

HALAZY,., BERGES,V., ERHARD, ., DANZIN, .(1992),Difluoromethylphenyl glycosides as suici-de substrates for glycosidases,Bioorg. Chem. 18,330-335.

HAMMOND,. S. (1955),A correlation of reaction ra-tes,J. Am . Chem.SOC.77,334-338.

HASSERODT,., JANDA,. D., LERNER, . A. (1996),Antibody catalyzed terpenoid cyclization,J.A m .Chem. SOC. 18,11654-11655.

HASSERODT,., JANDA, . D., LERNER,. A. (1997),Form ation of bridge -methylated decalins by anti-body-catalyzed tandem cationic cyclization, J .

Am . Chem.SOC.119,5993-5998.HAYNES,. R., STURA,.A., HILVERT,., WILSON,.

A. (1994), Ro utes to catalysis: Structure of a cata-lytic antibody and comparison with its naturalcounterpart, Science 263,646-652.

HEINE,A., STURA, . A., YLI-KAUHALUOMA,. T.,GAO,C. C., DENG,Q. L. et al. (1998),An antibod

resolution,Science 279, No. 5358,1934-1940.HENDRY,?, SARGESON,. M. (1990), Metal-ion pro-

moted reactions of phosphate derivatives,Progr.Inorg. Chem.38,201-258.HILVERT,D., NARED,K. D. (1988), Stereospecific

Claisen rearrangement catalyzed by an antibody,J . Am. Chem.SOC.110,5593-5594.

HILVERT,., CARPENTER,.H., NARED, . D.,AuDI-

TOR, M.-T. M. (1988), Catalys is of co ncerted reac-tions by antibodies: The Claisen rearrangemen t,Proc. Natl. Aca d. Sci. U SA 85,4953-4955.

HILVERT, ., HILL,K. W., NARED,K. D., AUDITOR,

M.-T. M. (1989), Antibody catalysis of aDiels-Alder reaction, J. Am. Chem. SOC.111,

9261-9262.

HIRSCHMANN,., SMITH11, A. B., TAYLOR,. M.,BENKOVIC,. A., TAYLOR,. D. et al. (1994), Pep-tide synthesis catalyzed by an antibody contai-ning a binding site for variable amino acids,Scien-ce 265,236237.

HOLLFELDER,., KIRBY, . J., TAWFIK, . S. (1996),“Off-the-shelf” proteins that rival tailor-madeantibodies as catalysts,Nature 353, M 3 .

HOUK, . N., GONZALES,., LI,Y. (1995), Pencyclicreaction transition states: passions and punctilios,1935-1995, Acc. Chem. Res.28,81-90.

HSIEH,L. C., YONKOVICH,., KOCHERSPERGER,.,

SCHULTZ,. G. (1993), Controlling chemical reac-

tivity with antibod ies,Science 260,337-339.

eno-Diels-Alderase inhibitor complex at 1.95‘l




HSIEH, . C., STEPHANS,. C., SCHULTZ,. G. (1994),An efficient antibody-catalyzed oxygenation re-action, J. Am . Chem.SOC.116,2167-2168.

ICHIHARA,., OIKAWA,. (1998), Diels-Alder typenatural products - Structures and biosynthesis,Curr. Org. C hem. 2,365-394.

IKEDA,K., ACHIWA, . (1997), Antibody-mediatedregio- and enantioselective resolution of a glyce-rol derivative, Bioorg. Med. Chem. Lett. 7, 225-228.

IKEDA,.,WEINHOUSE,. I., JANDA, . D., LERNER,

R. A. (1991),Asymmetric induction via a catalyticantibody,J.Am . Chem.SOC.113,7763-7764.

IVERSON,. L., LERNER,. A. (1989), Sequence-spe-cific peptide cleavage catalyzed by an antibody,Science 243,1184-1189.

IVERSON,. L., CAMERON,. E., JAHANGIRI,. K.,

PASTERNAK,. S. (1990), Selective cleavageof tri-tyl protecting groups catalyzed by an antibody, .

Am . Chem. SOC .112,532Q-5323.IWABUCHI,~.,IYASHITA,.,TANIMURA,., KINOS-

HITA, K., KIKUCHI, .,F~ J I I ,. (1994), Regio- andstereoselective deprotection of acylated carbo-hydrates via catalytic antibodies, J . Am. Chem.SOC.116,771-772.

IZADYAR,., FRIBOULET,., REMY,M. H., ROSETO,

A., THOMAS,. (1993), Monoclonal anti-idiotypicantibodies as functional internal images of enzy-me active sites: Production of catalytic antibodywith a cholinesterase activity, Proc. Natl. Acad.Sci. US A 90,8876-8880.

JACKSON,. Y., SCHULTZ,. G. (1991),An antibody-catalyzed cis-trans isomerization reaction, J. A m .Chem. SOC. l3,2319-2321.

JACKSON,. Y., JACOBS,. W., SUGASAWARA,.,

REICH,S. H., BARTLE-IT, . A., SCHULTZ,. G.(1988), An antibody-catalyzed Claisen rearrange-ment,J.A m. Chem. SOC .110,48414842.

JACKSON,. Y., PRUDENT,. R., BALDWIN,. P.,SCHULTZ,. G. (1991), A mutagenesis study of acatalytic antibody,Proc. Natl. Aca d. Sci. U SA 88 ,5 M 2 .

JACOBS,. W. (1991), New perspectives on catalytic

antibodies,Bio. Technology9,258-262.JACOBS,., SCHULTZ,. G., SUGASAWARA,., POWELL,

M. (1987), Catalytic antibodies,J.Am . Chem.SOC.109,2174-2176.

JACOBSEN,. R., SCHULTZ,. G. (1994), Antibody ca-talysis of peptide bond formation, Proc. Natl.Acad. Sci. US A 91,5888-5892.

JACOBSEN,. R., PRUDENT,. R., KOCHERSPERGER,.,

YONKOVICH,.,SCHULTZ,. G. (1992),An efficientantibody-catalyzed aminoacylation reaction,Science 256,365-367.

JACOBSEN,. R., COCHRAN,. G., STEPHANS,. C.,KING,D. S., SCHULTZ,. G. (1995). Mechanistic

photocleavage,J . Am . Chem.SOC.117,5453-5461.JANDA, . D., LERNER,R. A., TRAMONTANO,.

(1988a), Antibody catalysis of bimolecular amideformation, J. Am . Chem.SOC.110,48354337.

JANDA, . D., SCHLOEDER,., BENKOVIC,. J., LER-

NER, R.A. (1988b), Induction of an antibody thatcatalyzes the hydrolysisof an amide bond, Scien-ce 241,1188-1191.

JANDA, . D.,BENKOVIC,. J., LERNER, . A. (1989),Catalytic antibodies with lipase activity and R orS substrate selectivity,Science 244,437-440.

JANDA, .D.,ASHLEY,.A., JONES,T. .,MCLEOD,.A., SCHLOEDER,. M.,WEINHOUSE,. I. (1990a),Immobilized catalytic antibodies in aqueous andorganic solvents, J . Am. Chem. SOC. 112,8886-8888.

JANDA, . D., WEINHOUSE,. I., SCHLOEDER,. M.,LERNER, . A., BENKOVIC,. J. (1990b), Bait andswitch strategy for obtaining catalytic antibodieswith acyl-transfer capabilities,J. Am. Chem. SOC.

112,1274-1275.JANDA,K. D., BENKOVIC,. J., MCLEOD, . A.,

SCHLOEDER,. M., LERNER, . A. (1991a), Sub-strate attenuation: An approach to improve anti-body catalysis,Tetrahedron47,2503-2506.

JANDA,. D.,WEINHOUSE,. I.,DANON,T.,ACELLI,

K. A., SCHLOEDER,. M. (1991b), Antibody baitand switch catalysis:A survey of antigens capableof inducing abzymes with acyl-transfer pro-perties, J . Am . Chem. SOC.113,5427-5434.

JANDA,. D., SHEVLIN,. G., LERNER, . A. (1993),Antibody catalysis of a disfavored chemicaltransformation, Science 259,490-493.

JANDA, . D., Lo, C.-H. L., Lr, T., BARBAS11, C. F.,WIRSCHING,., LERNER, . A. (1994), Direct se-lection for a catalytic mechanism from combina-tonal antibody libraries, Proc. Natl. Acad. Sci.

JANDA, . D., SHEVLIN,. G., LERNER, . A. (1995),Oxepane synthesis along a disfavored pathway:The rerouting of a chemical reaction using a cata-lytic antibody,J.Am. Chem. SOC.117,2659-2660.

JANDA, . D., Lo, L.-C., Lo, C.-H. L., SIM,M.-M.,WANG, .

et al. (1997), Chemical selection for ca-talysis in combinatonal antibody libraries, Scien-ce 275,945-948.

JANJIC,N.,%AMONTANO,. (1989), Antibody-cat-alyzed redox reaction, J. Am. Chem. SOC. 111,

JENCKS, . P. 1969), Mechanisms for catalysis andcarbonyl and acyl-group reactions, in:Catalysis inChemistry and Enzym ology , pp. 3-6 and 523-537.NewYork: McGraw Hill.

JENCKS,. P. (1975), Binding energy, specificity andenzyme catalysis: the circe effect, Adv. E nzymol.Relat. Areas Mol. Biol. 43.219-410.

USA 91,2532-2536.

9109-9110.

studies of antibody-catalyzed pyrimidine dimer JOHNSON,. S. (1968), Non-enzymic biogenetic-like



I3 References 485

olefinic cyclisations,Acc. Chem. Res. 1,143.KAST, I?, ASIF-ULLAH, ., JIANG, N., HILVERT, .

(1996), Exploring the active site of chorismatemutase by combinatorial mutagenesis and selec-tion: The importance of electrostatic catalysis,Proc. Natl. Ac ad . Sci. US A 93,5043-5048.

KATAYAMA, ., KOBAYASHI, ., OIKAWA, ., HOM A,M., ICHIHARA ,. (1998), Enzy matic activity andpartial purification of solanapyrone synthase:first enzym e catalysing Diels-Alder reaction,Biochim. Biophys. Acta 1384,387-395.

KAWAMURA-KONISHI,., HOSOM I, ., NEYA,S., Su -GANO, s.,FUNASAKI,., SUZUKI, H. (1996), Kineticcharacterization of antibody-catalyzed insertionof a metal ion into porphyrin, J . Biochem. 119,857-862.

KEINAN, ., SINHA, . C., SINHA-BAGCHI,A.,ENORY,E., GHOZI, . C. et al. (1990), Towards antibody-mediated metallo-porphyrin chemistry, PureAppl . Chem.62,2013-2019.

KEMP,D. S., Cox,D. D., PAUL,K. G. (1975),The phy-sical organic chemistry of benzisoxazoles.IV. Theorigins and catalytic nature of the solvent rate ac-celeration for the decarboxylation of 3-carboxy-benzisoxazoles,J.Am . Chem.SOC.97,7312-7318.

KHALAF,A.., PROCTOR,G.R., SUCKLING,.J.,BEN-CE , L. H., IRVINE,. I., STIMSON,. H. (1992), Re-markably efficient hydrolysis of a 4-nitrophenylester by a catalytic antibody raised to a n amm oni-um hapten, J. Chem. SOC.Perkin Trans. I, 1475-

1481.K H E ~ A L ,., DELAUZON,.,DESFOSSES,., USHIDA,S. , MARQUET,. (1994), A catalytic antibody iso-merizing a A s-3-ketosteroid,C. R. Aca d. Sci. Paris,Life Sciences 317,381-385.

KIRBY, . J. (1980), Effective molarities for intra-molecular reactions, Adv. Phys. Org. Chem. 17 ,

KIRBY, . J. (1996), The potential of catalyic antibo-dies,Acta Chim. Scand.50,203-210.

KITAZUME,T.,IN, .T.,YAMAMOTO,T.,YAMAZAKI,T.(1991a), Antibody-catalyzed double stereoselec-tion in fluorinated materials, J. Am . Chem. SOC.

113,8573-8575.KITAZUME,., LIN, . T., TAKE DA, ., YAMAZAKI,.(1991b), Stereose lective synthesis of fluorinatedmaterials by a n antibody, J . Am . Chem. SOC.113,

KITAZUME, ., TSUKAMOTO,., YOSHIMURA, .(1994), A catalytic antibody e licited by a haptenof tetrahedral carbon-type, J . Chem. SOC.Chem.Commun.1355-1356.

KOCH,A.,REYMOND,.-L.,LERNER,R.A.1994),An-tibody-catalyzed activation of unfunctionalizedolefins for highly enantioselective asymmetricepoxidation,J. Am . Chem.SOC.116,803-804.

KOCH, ., REYMOND,.-L.,LERNER, . A. (1995),An-

183-278.

2123-2126.

tibody catalysis of multistep reactions: An aldoladdition followed by a disfavored elimination, J .

Am . Chem.SOC.117,9383-9387.KOHLER, ., MILSTEIN,. (1975), Con tinuous cultu-

res of fused cells secreting antibody of predefinedspecificity,Nature 256,495497.

KOHLER,G.,MILSTEIN,. (1976), Deriva tion of spe-cific antibody-producing tissue culture and tu-mour lines by cell fusion, Eur. J . Immunol. 6,

KOHLER, ., HOW E,. C.,MILSTEIN,. (1976), Fusionbetween immunog lobulin-secreting and non-secreting myeloma cell lines, Eur. J . Immunol. 6,292-295.

LANDRY, . W., ZHAO,K., YANG,G. X.-Q., GLIC K-MAN, M ., GEORGIADIS,. (1993), Antibody-cata-lyzed degradation of cocaine, Science 259,

LAVEY, . J., JANDA,K. D. (1996a), An tibody cataly-zed hydrolysis of a phosphotriester, Bioorg. Med.Chem. Lett. 6,1523-1524.

LAVEY, . J., JANDA,K. D.(1996b), Catalytic antibod ymediated hydrolysis of Paraoxon, J . Org. Chem.61,7633-7636.

LEE,A . Y., STEWART,. D., CLARDY,., GANEM, .(1995), New insight into the catalytic mechanismof chorismate mutases from structural studies,Chem. Biol. 2,195-203.

LERNER, . A., BAREIAS11, C. F. (1996), Using theprocess of reactive immunisation to induce cata-

lytic antibod ies with complex mechanism s: Aldo -lases,Acta Chim. Scand.50,672-678.LEWIS, . , KRAMER,., ROBINSON,., HILVERT,.

(1991), Medium effects in antibody-catalyzed re-actions,Science 253,1019-1022.

Lr,T.,JANDA,K. D. (1995), Syn thesis of a bifunctionalhapten designed to mimic both transition stateand to induce “Bait and Switch” catalysis,Bioorg.Med. Ch em. Lett. 5,2001-2004.

LI, T., JANDA, K. D., ASHLEY,. A., LERNER, . A.(1994), Antibody-catalyzed cationic cyclization,Science 264,1289-1293.

Lr, L., PAUL,S., TYUTYULKOVA,., KAZATCHKINE,

M. D., KAVERI, . (1995a), Catalytic activity ofanti-thyroglobulin antibodies,J . Immunol. 3328-3332.

Lr,T., HILTON,.,JANDA,K. D. (1995b),The potentialapplication of catalytic antibodies to protectinggroup removal: Catalytic antibodies with broadsubstrate tolerance, J . Am. Chem. SOC. 117,

LI,T., HILTON,.,JANDA, K. D. (1995c), R emarkableability of different antibod y catalysts to contro land diversify the product o utcome of cationic cy-

clization reactions, J . Am . Chem. SOC.117,3308-3309.

LI, T., JANDA,K. D., HILTON, ., LERNER, . A.

511-519.

1899-1901.

2123-2127.



486 I 1 Catalytic An tibodies

(1995d), Antibody catalyzed nucleophilic substi-tution reaction at a primary carbon that appear to

proceed by an ionisation mechanism, J . A m .Chem. SOC.117,2367-2368.

LI,T., JANDA, . D., LERNER, . A. (1996), Cationiccyclopropanation by antibody catalysis, Nature379,326-327.

LI,T.,LERNER,. A., JANDA, . D. (1997),Antibody-catalyzed cationic reactions: Rerouting of chemi-cal transformations via antibody catalysis, Acc.Chem. Res.30,115-121.

LIENHARD, . E. (1973), Enzymatic catalysis andtransition-state theory,Science 180,149-154.

LIOTTA, . J., BENKOVIC,.A., MILLER, . P., BENKO-VIC,S. J. (1993), A catalytic antibody for imidehydrolysis featuring a bifunctional transition-sta-te mimic, . Am . Chem. SOC.115,350-351.

LIOTTA, . J., GIBBS, . A.,TAYLOR,.D., BENKOVIC,l? A., BENKOVIC,. J. (1995), Antibody-catalyzedrearrangement of a peptide bond: mechanisticand kinetic investigations, . Am . Chem. SOC.117,4729-4741.

LOFAS,S.,JOHNSON,B. (1990),A novel hydrogel ma-trix on gold surfaces in surface plasmon resonan-ce sensors for fast and efficient covalent immobi-lisation of ligands,J. Chem. SOC.Chem. Commun.15261528.

MADER, . M., BARTLETT,. A. (1997),Binding ener-gy and catalysis: the implications for transitionstate analogs and catalytic antibodies,Chem. Rev.

MARCH,. (1992a), Determination of the presence ofan intermediate, in: Advanced Organic Chemis-try: Reactions Mechanisms and Structure, pp.217-219. New York: John Wiley & Sons.

MARCH,. (1992b), Addition to carbon-carbon mul-tiple bonds, in:Advanced Organic Chemistry: Re-actions Mechanisms and Structure, pp. 839-852.New York: John Wiley & Sons.

MARTIN, . T., NAPPER,A. D., SCHULTZ,. G., REES,A. R. (1991), Mechanistic studiesof a tyrosine-de-pendent catalytic antibody, Biochemistry 30,

MARTIN,M. T., ANGELES,.S.,

SUGASAWARA,.,AMAN,N. I., NAPPER, . D. et al. (1994), Anti-

body-catalyzed hydrolysis of an unsubstitutedamide, . Am. Chem. SOC.116,6508-6512.

MEEKEL,A.A. ., RESMINI, ., PANDIT, . K. (1995),First example of an antibody-catalyzed hetero-Diels-Alder reaction, J . Chem. SOC. Chem. Com-

mun.571-572.METS,B., WINGER, ., CABRERA,.,SEO, S.,JAMDAR,

S. et al. (1998), A catalytic antibody against co-caine prevents cocaine’s reinforcing and toxic ef-fects in rats, Proc. Natl. Aca d. Sci. U SA 95,10176.

MILLER, . P., POSNER,. A., BENKOVIC,. J. (1997),Expanding the 43C9 class of catalytic antibodies

97,1281-1301.

9757-9761.

using a chain-shuffling approach, Bioorg. Med.Chem.5,581-590.

MIYASHITA,., KARAKI, ., KIKUCHI,M., FUJII, .

(1993), Prodrug activation via catalytic anti-bodies,Proc. Natl. Aca d. Sci. U SA 90,5337-5340.

NA,J., HOUK, . N. (1996), Predicting antibody cata-lyst selectivity from optinum binding of catalyticgroups to a hapten, J . Am . Chem. SOC. 18,9204-9205.

NAKATANI,., HIRATAKE,., SHINZAKI,., UMESHI-TA, R., SUZUKI,. et al. (1993), A mode of inhibi-tion of an esterolytic antibody, Tetrahedron Lett.34,49454948.

NAKAYAMA,. R., SCHULTZ,.G. (1992), Stereospe-cific antibody-catalyzed reduction of an a-ketoamide,J. Am . Chem. SOC.114,780-781.

NAPPER, . D., BENKOVIC,. J., TRAMONTANO,.,LERNER, . A. (1987),A stereospecific cyclisationcatalysed by an antibody,Science 237,1041-1043.

OHKUBO,., URATA, ~.,ERI,K., ISHIDA,., SAGA-WA,T. t al. (1993),The catalytic activities of IgGand IgM monoclonal antibodies for the hydroly-sis of p-nitrophenyl acetate, Chem. Lett. 6,1075-1078.

OIKAWA, ., YAGI,K., WATANABE,., HONMA, .,ICHIHARA,. (1997), Biosynthesis of macropho-mic acid: plausible involvement of intramolecularDiels-Alder reaction, J . Chem. Soc., Chem. Com -mun.97-98.

OIKAWA,., KOBAYASHLT.,ATAYAMA,., SUZUKI,Y., ICHIHARA,A. (1998), Total synthesis of(- -solanapyrone via enzymatic Diels-Alder re-action of prosolanapyrone, J . Org. Chem. 63,874843756.

OIKAWA, ., SUZUKI, ., KATAYAMA, ., NAYA,A.,SAKANO,., ICHIHARA,A.1999), Involvement ofthe Diels-Alder reaction in the biosynthesis ofsecondary natural products: the late stages of thebiosynthesis of the phytotoxins, solanapyrones, .

Ch em . Soc., Perkin Trans.I, 1225-1232.PADLAN,. A., COHEN, . H., DAVIES, . R. (1985),

On the specificity of antibody 3-antigen interac-tions - phosphocholine binding to MCPC603 and

the correlation of 3-dimensional structure and se-quence data, Annales de l’lnstitut Pasteur - Im -munology C136,211.

PAGE,M. I., JENCKS, . P. (1971), Entropic contribu-tions to rate accelerations in enzymic and intra-molecular reactions and the chelate effect, Proc.Natl. Acad . Sci. USA 68,1678-1683.

PAITEN,. A., GRAY, . S.,YANG,. L., MARKS, . B.,WEDEMAYER,. J. et al. (1996),The immunologi-cal evolution of catalysis,Science 271,1086-1091.

PAUL, .,VOLLE, . J., BEACH, . M., JOHNSON,. R.,POWELL, . J., MASSEY,. J. (1989), Catalytic hy-drolysis of vasoactive intestinal peptide by hu-man autoantibody,Science 244,1158-1162.



13 References 487

PAUL, ., LI, L., KAL AGA, ., WILKINS-STEVENS,.,STEVENS,. J., SOLOMON,. (1995), Natural cata-lytic antibodies: Peptid e hydrolyzing activities ofBence Jones proteins and VL ragment, J. Biol.Chem.270,15257-15261.

PAULING,. (1947), Chem ical achievement and hopefor the future, Silliman Lecture, Yale UniversityPress.

PAULING,. (1948), Nature of forces between largemolecules of biological interest, Nature 161,

PLAYFAIR,. H. L. (1992), Immunology at a Glance.O xf or d Blackwell Scientific.

POLLACK,. J., SCHULTZ,. G. (1989),A semisynthe-tic catalytic antibody, J. Am. Chem. SOC. 111,

1929-1931.POLLACK,. J., JACOBS,J. W., SCHULTZ,. G. (1986),

Selective chemical catalysis by an antibody,Scien-ce 234,1570-1573.POLLACK,. J., NAKAYAMA,. R., SCHULTZ,. G.

(1988), Introduc tion of nucleoph iles and spec-troscopic probes into antibody combining sites,Science 242,1038-1040.

POLLACK,. J., HSIUN,., SCHULTZ,. G. (1989), Ste-reospecific hydrolysis of alkyl esters by anti-bodies,J.Am . Chem.SOC.111,5961-5962.

RAASCH, . S. (1980),Annelations with tetrachloro-thiophene 1,l-dioxide,J. Org. Chem.45,856-867.

RADZICKA,., WOLFENDEN,. (1995), A proficientenzyme,Science 267,90-93.

RASO, ., STOLLAR,. D. (1975),The antibody-enzy-me analogy. Cha racterisation of antibodies tophosphopyridoxyltyrosine derivatives, Biochem-istry 14,584-591.

REICH, . J. (1979), Functional group m anipulationusing organoselenium reagents, Acc. Chem. Res.

RESMINI, ., MEEKEL,A.. P., PANDIT, . K. (1996),Catalytic antibodies: Regio- and enantioselectivi-ty in a hetero Diels-Alder reaction, Pure Ap pl .Chem.68,2025-2028.

REYMOND,.-L. (1995), (S)-Selective retroaldol re-action and the (R)-selective p-elimination with

an aldolase catalytic antibody,Angew . Ch em. (Znt.Edn. Engl.) 34,2285-2287.REYMOND,.-L.,C H E N , ~ .1995a),Catalytic, enantio-

selective aldol reaction using antibodies against aquaternary amm onium ion with a primary aminecofactor, Tetrahedron Le tt. 36,2575- 2578.

REYMOND,.-L.,C H E N , ~ .1995b),Catalytic, enantio-selective aldol reaction with an artificial aldolaseassembled from a primary amine and an antibo-dy, J. Org. Chem .60,6970-6979.

REYMOND,.-L., ANDA,K. D., LERNER, .A. (1991),Antibody catalysis of glycosidic bond hydrolysis,Angew. Chem. (Int. Edn. E ngl.) 30,1711-1713.

REYMOND,., JANDA, K. D., LERNER, . A. (1992),

707-709.

l2,22-30.

Highly enantioselective protonation catalyzed byan antibody,J.Am . Chem.SOC.114,2257-2258.

REYMOND,.-L.,JAHANGIRI ,G. K., STOUDT,., LER-NER, R . A. (1993),Antibody catalyzed hydrolysisof enol ethers, J.Am . Chem.SOC.115,3909-3917.

REYMOND,.-L.,REBER,.-L.,LERNER, . A. (1994),Enantioselective, m ultigram-scale synthesis witha catalytic antibody, Angew. Chem. (Int. Edn.Engl.)33,475477.

ROBERTS,. A., STEWART,., BENKOVIC,. J., GET-ZOFF, E. D. (1994), Catalytic antibody mo del andmutagenesis implicate arginine in transition-statestabilization,J. Mol. Biol.235,1098-1116.

ROSENBLUM,. S., Lo, L.-C.,LI,T.,JANDA,K.D., LER-NER, R. A . (1995),Antibody-catalyzed phosphatetriester hydrolysis, Angew. Chem. (Int. Edn.Engl.) 34,2275-2271.

SAKURAI, ., WIRSCHING,., JANDA, K. D. (1996),Synthesis of a nucleoside hapten w ith a[P(O)-0-N] linkage to elicit catalytic antibodieswith phosphodiesterase activity, Bioorg. Med.Chem. Lett. 6,1055-1060.

SAUER,. (1966), Diels-Alder reactions: new prepa-rative aspects,Angew. Chem. (Znt.Edn. Engl.) 5,211-220.

SCANLAN,T.., PRUDENT,. R., SCHULTZ,. G. (1991),Antibody-catalyzed hydrolysis of phosphatemonoesters,J.Am. Chem. SOC.113,9397-9398.

SCHULTZ,. G. (1989), Catalytic antibodies, Acc.Chem. Res. 22,287-294.

SCHULTZ,. G., LERNER, . A. (1993),An tibody cata-lysis of difficult chem ical transform ations, Acc.Chem. Res. 26,391-395.

SERGEEVA, . V., YOMTOVA,V., PARKINSON,.,OVERGAAUW,., POMP, . et al. (1996), Haptendesign for antibody-catalyzed decarboxylationand ring-opening of benzisoxazoles, Isr. J. Chem.

SHABAT,., ITZHAKY,., REYMOND,.-L.,KEINAN,.(1999 ,Antibo dy catalysis of a reaction otherw isedisfavoured in water, Nature 374,143-146.

SHEN, .-Q. (1995), Catalytic antibodies for the hy-drolysis of chloramphenicol, Thesis, Sheffield

University.SHEN, ., PRIEB E, ., PATEL, ., RUBO , ., SU,T. et al.(1992), An approach for the generation of sec-ondary structure specific abzymes, TetrahedronLen. 33,3417-3420.

SHEVLIN,. G., HILTON,., JANDA, K. D. (1994),Au-

tomation of antibody catalysis: A practical me-thodo logy for the use of catalytic antibodies in or-ganic synthesis, Bioorg. Med. C hem. Lett. 4,297-302.

SHOKAT, . M., LEUMANN,. J., SUGASAWARA,.,SCHULTZ,. G. (1988), An antibody-mediated re-dox reaction, Ange w. C hem. (Znt. Edn. E ng l.) 27 ,

36,177-183.

1172-1 175.



488 I 1 Catalytic An tibodies

SHOKAT, . M., LEU MAN N,. J., SUGASAWARA,.,SCHULTZ,. G. (1989), A new strategy for the ge-neration of catalytic antibodies, Nature 338,

SHOKAT,. M., KO, M. K., SCA NL AN ,. S.,KOCHER-SPERGER, L.,YONKOVICH,. et al. (1990), Catalyticantibod ies: A new class of transition -state ana lo-gues used to elicit hydrolytic antibodies,Angew.Ch em . (Znt. Edn. Engl.) 29,12961303.

SHOKAT, ., UNO ,T., SCHULTZ,. G. (1994), Me-chanistic studies of an antibody-catalyzed elimi-nation reaction,J. Am. Chem.SOC.116,2261-2270.

SHUSTER ,A. ., GOLOLOBOV,.V., KVASHUK,.A.,BOGOMOLOVA,A.., SMIRNOV,.V., GABIBOV,A.G. (1992), DN A hydrolysing antibodies, Science256,665-667.

SINHA, . C., KEINAN, ., REYMOND,.-L. (1993a),Antibody-catalyzed enantioselective epoxide hy-drolysis,J.Am . Chem.SOC.115,4893-4894.

SINHA, . C., KE INAN, ., REY MON D,.-L. (1993b),Antibody-catalyzed reversal of chemoselectivity,Proc. Natl. Aca d. Sci. US A 90,11910-11913.

SMILEY,. A., BENKOVIC,. J. (1994), Selection of ca-talytic antibodies for a biosynthetic reaction froma combinatorial cDNA library by complementa-tion of an auxotrophic Escherichia coli: Anti-bodies for orotate decarboxylation, Proc. Natl.Aca d. Sci. USA 91,8319-8323.

SMITHRUD,. B., BENKOVIC,. A., BENKOVIC,. J.,TAYLOR,. M.,YA GER , . M. e t al. (1997), Investi-

gations of an antibody ligase,J.Am . Chem. SOC.

119,278-282.SPITZNAGEL,.M., JACOBS,.W., CLA RK , . S. (1993),

Random and site-specific imm obilisation of cata-lytic antibodies, Enzyme Microb. Technol. 15 ,916-921.

STAHL,M., GO LDIE, ., BOURNE ,. €'.,THOMAS,N. R.(1995), Cataly tic antibod ies generated by in vitroimmunization,J.Am . Chem. SOC .117,51665165.

STEPHE NS, . B., IVERSON,. L. (1993), Catalyticpolyclonal antibodies, Biochem. Biophys. Res.Commun.192,1439-1444.

STEWART,. D., BENK OVIC,.J. (1995),Transition-sta-

te stabilization as a measure of the efficiency ofantibody catalysis,Nature 375,388-391.STEWART,. D., ROBE RTS,.A., THOMAS,. R., GET-

ZOFF, E. D., BENKOVIC,. J. (1994), Site-directedmutagenesis of a catalytic antibody: A n arginineand a histidine residue play key roles, Biochem-istry 33,1994-2003.

STORK, ., BURGSTAHLER,. A . (1955), Stereoch em-istry of polyene cyc liatio n, J.Am . Chem.SOC.77,

5068-5077.SUCKLING,. J.,TEDFORD ,., BENCE , .M., IRVINE,.

I., STIMSON,. H . (1993), An antibody with dualcatalytic activity, J . Chem. SOC. Perkin Trans. I ,

269-271.

1925-1929.

SUGA,H., EESOY,O., TSUMURAYA,., LEE, J., S INS -KEY,A.., MASA MUN E,. (1994a), Esteroly tic anti-bodies induced to hap tens w ith a 1,2-am ino alco-hol functionality, .Am . Chem. SOC. 16,487-494.

SUGA,H.,ERSOY, .,WILLIAMS,. F.,TSUMURA YA,T.,MARGOLIES,,. N. et al. (1994b), Catalytic anti-bodies generated via heterologous imm unization,J.Am . Chem. SOC.116,6025-6026.

SUMMERS,. (1983), Catalytic principles of enzymechemistry, Thesis,Harv ard U niversity.

SUTTON,P. A., BUCKINGHAM,. A. (1987), Cobalt(111)-promoted hydrolysis of amino acid estersand peptides and the synthesis of small peptides,Acc . Chem. Res.20,357-364.

TANAKA,, KINOSHITA,., TAN IMUR A,., FUJII, .

(1996), Relax ing substrate specificity in antibody-catalyzed reactions: Enantioselective hydrolysisof N-Cbz-amino acid esters, J. Am. Chem. Soc.

TARASOW,T.., LEWIS , ., HILVERT,. (1994), Inves-tigation of med ium effects in a fam ily of decarb-oxylase antibodies, J. Am . Chem. SOC.116,7959-7963.

TAWFIK,D. S., ZEMEL,R. R., ARAD-YELLIN,.,GREEN,. S., ES HH AR ,. (1990),A simple methodof selecting catalytic monoclon al antibodies thatexhibit turnover and specificity,Biochemistry 29,9916-9921.

TAWFTK, . S., G REEN ,B. S., CHAP,R., SELA,M.,ESHHAR, . (1993), catELISA: A facile route to

catalytic antibodies ,Proc. Natl. Aca d. Sci. U SA 90,373-377.TAWFIK, . S.,LINDNER,. B., CHAP,R., ESHH AR,.,

GREEN,B. S. (1997), Efficient and selective p-nitrophenyl-ester-hydrolyzing ntibodies elicitedby a p-nitrobenzyl hapten, Eur. J. Biochem. 244,619-626.

TERAISHI,K., SAITO,M., FUJII , NAK AM URA ,.(1994), Design of the hapten for the induction ofantibodies catalysing aldol reaction, I . Mol.Graphics 12,282-285,

THATCHER,. R. J., KLU GER , . (1989), Mechanismand catalysis of nucleop hilic substitution in ph os-

phate esters,Adv. Phys. Org. Chem.25,99-265.THOMAS,. R. (1996), Catalytic antibodies - reach-ing adolescence?, Natural Product Reports 13,

THOMPSON,. C. (1973), Use of peptide aldehydes togenerate transition-state analogs of elastase, Bio-chemistry l2,47-51.

THORN, . N., DA NIEL S, . G., AUD ITOR,M.-T. M.,HILVERT,D. (1995), Large rate accelerations inantibody catalysis by strategic use of hapteniccharge,Nature 373,228-230.

TRAMONTANO,., JANDA,K. D., LERNER, . A .(1986), Catalytic antibodies, Science 234, 1566-

1570.

118,2332-2339.

479-512.



13 References 489

as inhibitors of papain, J. Biol. Chem.247,8195-8197.

WESTHEIMER,. H. (1968), Pseud orotation in the hy-drolysis of phosphate esters, Acc. Chem. Res. 1,70-78.

WILLIAMS,., DOUGLAS,. T. (1972a), E lc B mech-anisms. Part 11. Base hydrolysis of substitutedphenyl phosphoramidates, J . Chem. SOC. erkinTrans.II , 1455-1459.

WILLIAMS,., DOUGLAS,. T. (1972b), El cB mech-anisms. Pa rt 111.Effect of ionisation of the amido -N H on the reaction of hydroxide ion with methyl3- and 4-benzamidobenzoates,J. Chem. SOC.Per-kin Trans.II,2112-2115.

WILLNER,., BLON DER, ., DA GAN , . (1994), Re-versible optical-recording by a dinitrophenol an-tibody-catalyzed ring-opening of 6,8-dinitrospiro-

pyran,J. Am . Chem.SOC.16,3121-3122.WINCHESTER,., FLEET,G. W. (1992), Am ino-sugarglycosidase inhibitors - versatile tools for glyco-biologists,Glycobiology 2,199-210.

WINCHESTER,., ALDAHER,., CARPENTER, . C.,DIBELLO,. C., CHOI,S. S. et al. (1993), The struc-tural basis of the inhibition of hum an alpha -man-nosidases by azafuranose analogs of mannose,Biochemical J.290,742-749.

WIRSCHING,.,ASHLEY,. A., BENKOV IC,. J., JANDA ,K. D., LERN ER , . A . (1991), An unexpectedly ef-ficient catalytic antibody operating by ping-pongand induced fit mechanisms,Science 252,680485.

WIRSCHING,., ASHLEY,. A., Lo, C.-H. L., JAND A, .D., LERN ER , . A. (1995), Reactive imm unization,Science 270,1775-1782.

WOLFENDEN, . (1969), Transition state analoguesfor enzy me catalysis,Nature 223,704-705.

XUE,Y. F., LIPSCOMB,. N. (1995), Loca tion of theactive-site of allosteric chorismate mutase fromSaccharomyces cervisiae, and comments on thecatalytic and regulatory mechanisms, Proc. Natl.Acad. Sci. USA 92,10595-10598.

YANG, ., CHU N,., ARAKAWA-URAMO TO,., WANG ,X., GAWINOWICZ,. A. e t al. (1996),Anti-cocainecatalytic antibodies: A synthetic approach to im-

proved antibody diversity,J.Am . Chem.SOC.118,

YLI-KAUHALUOMA,. T., ASHLEY,. A., Lo, C.-H.,TUCKER,., WOLFE,M. M., JANDA, . D. (1995),Anti-metallocene antibodies: A new approach toenantioselective catalysis of the Diels-Alder re-action,J.Am . Chem.SOC.117,7041-7047.

YLI-KAUHALUOMA,. T., ASHL EY,. A., Lo , C.-H.L.,COAKLEY,., WIRSCHING ,., JAN DA , . D. (1996),Catalytic antibodies with peptidyl-prolyl cis-transisomerase activity, J. Am . Chem. SOC. 18, 5496-5497.

YOON, . S., OE I,Y., SWEET, ., SCHU LTZ,. G. (1996),

A n antibody-catalyzed [2,3]-elimination reaction,

5881-5890.

TRAMONTANO,., AMMANN,. A., LERNER,R. A.(1988),Antibody catalysis approaching the activi-ty of enzymes,JA m . C h em .SOC.110,2282-2286.

TUBUL,A., BRUN,P., MICHEL,R., GHARIB, ., DE

REGGI,M. (1994), Polyclonal antibody-catalyzedaldimine formation, Tetrahedron Lett. 35, 5865-5868.

ULRICH, . D., DRIGGERS,. M. G., SCHULTZ,. G.(1996), An tibody catalysis of pe ricyclic reactions,Acta Chim. Scand. 50,328-332.

UNO,T., SCHULTZ,. G. (1992), A n antibody-cat-alyzed dehydration reaction, J. Am. Chem. SOC.

114,6573-6574.UNO,T., GONG, ., SCHULTZ,. G. (1994), Ste reosel-

ective antibody-catalyzed oxime formation, J.

Am . Chem.SOC.16,1145-1146.UNO,T., Ku, J., PRUDENT,. R., HUAN G, ., SCHULTZ,

P. G. (1996), A n antibody-catalyzed isomerizia-tion reaction, J.Am . Chem.SOC. 18,3811-3817.VANVRANKEN,. L., PANO MITRO S,.,SCHULTZ,. G.

(1994), Catalysis of carbamate hydrolysis by anantibody, Tetrahedron Lett. 35,3873-3876.

WADE,W. S., ASHL EY,. A., JAHANG IRI,. T., McEL -HANEY, G., JANDA , . D., LERN ER, . A. (1993), Ahighly specific metal-activa ted catalytic antibody,J.Am . Chem.SOC.115,4906-4907.

WAGNER,., LERN ER, . A., BARBA S11, C. F. (1995),Efficient aldolase catalytic antibodies that use theenamine mechanism of the natural enzyme,Science 270,1797-1800.

WA LLAC E, . B., IVERSON ,. L. (1996), The influen-ce of h apten size and hydroph obicity on the cata-lytic activity of elicited polyclonal antibodies, J.

Am . Chem.SOC.18,251-252.WEINER, . P., WIEMANN,., WOLFE,M. M., WENT-

WORTH JR., P., JAND A, . D. (1997), A pentaco-ordinate oxorhenium(V) metallochelate elicitsantibody catalysts for p hosphodiester cleavage,J.Am . Chem.SOC.19,408&4089.

WENTWORTHR., P., DATTA,A.,BLAK EY, ., BOYLE,T.,PARTRIDGE,. J., BLACK BURN ,.M. (1996),To-wards antibody directed abzyme prodrug thera-py, ADAPT Carbamate prodrug activation by a

catalytic antibody and its in vitro application tohuman tumor cell-killing, Proc. Natl. Acad. Sci.

WENTWORTHR., P., DATTA,A., SMITH,S., MAR-SHALL, A., PARTRIDGE,. J., BLACKBURN,. M.(1997), An tibody catalysis of BAc2 aryl carbama-te ester hydrolysis: A highly disfavored chemicalprocess,J.Am . Chem.SOC.119,2315-2316.

WENTWORTHR.,P., L I U , ~ . , E NT WO RT H,. A., FAN,P., FOLEY,M. J., JAND A, . D. (1998), A bait andswitch hapten strategy generates catalytic anti-bodies for phosphodiester hydrolysis,Proc. Natl.Aca d. Sci. USA 95,5971.

WESTERICK,. O.,WOLFENDEN,. (1972),Aldehydes

U S A 93,199-803.




J.Am. Chern. SOC.118,11686-11687.Yu, J., HSIEH,L. C., KOCHERSPERGER,., YONKO-

VICH , ., STEPHA NS,. C. et al. (1994),Progress to-

wards an antibody glycosidase, Angew. Chem.(Int. Edn. Engl.) 33,339-341.

Yu, J., CH OI, S. Y., MO ON ,K.-D., CHUNG,H.-H.,

YOUN, . J. et al. (1998),A glycosidase antibodyelicited against a chair-like transition sta te ana logby in vitro immunization, Proc. Natl. Aca d. Sci.U S A 95,2880.

ZEWAIL, . H. (1997), Femtochemistry,Adv. Chem.Phys. 101,892 (and other articles in this issue).

ZEWAIL, . H., BERNSTEIN,. B. (1988), Special Re-port - Real time laser femtochemistry - viewingthe transition from reagents to products, Chem.Eng. News 66,24-43.

ZHOU,G. W., Guo , J., HUANG, ., FLETTERICK,. J.,SCANLAN,T.. (1994), Crystal structure of a cata-lytic antibody with a serine protease active site,Science 265,1059-1064.

ZHOU,2.S., JIANG, ., HILVERT,. (1997),An anti-body-catalyzed selenoxide elimination, J . A m .Chem. Soc. 119,3623-3624.



12 Synthetic Enzymes -Artificial Peptides in

Stereoselective Synthesis

PAULA. BENTLEYNew York, NY 10027,USA

1 Introduction 492

2 Cyclic Dipeptides 492

3 Poly-Leucine 499

3.1 Other Applicationsof Oligopeptides 505

4 Use of Linear Di-, Tri- and Tetrapeptides 508

5 Summary and Future Directions 513

6 References 513





492 12 Synthetic Enzymes -Artificial Peptides in Stereoselective Synthesis

1 Introduction

Low molecular weight peptides may havefunctioned as the earliest natural catalystsprior to the evolution of their sophisticatedenzyme relatives (EHLERand ORGEL,1976;FERRISt al., 1996; LEE et al., 1996). Now theyare being revital ized for use in the laboratory.Synthetic applications of polypeptides (EBRA-HIM and WILL S, 997), and polypeptides plusother polymers (Pu,1998) have b een reviewedrecently.

This chapter will review the applications ofsynthetic peptides to stereoselective synthesis,acquainting the reader with the history and

latest developments, as well as speculating onthe futu re directions of this intriguing area.

2 Cyclic Dipeptides

O ne of the greatest challenges for syntheticchemists is to design small molecules whichhave th e catalytic selectivity of enzymes. How-ever the complex and ordered architecture of

an enzyme active site, which is seemingly aprerequisite for selectivity is not easily mim-icked by small acyclic molecules. Conse quent-ly chemists have resorted to cyclic structureswhich are constrained to be highly orderedand which p r i m a f a c ie have properties whichcan b e predicted with som e certainty. This wasINOUE'Sremise for study ing cyclic dipep tides(diketopiperazines). After initial studies withvery limited success (O K U t al., 1979, 1982),he was rew arded in 1981 when th e dipeptide,

cycZo-[(S)-Phe-(S)-HisI13) in benzene cata-lyzed the addition of hydrogen cyanide tobenzaldehyde ( la) giving the (R )-cyanohy drin(2a,mandelonitrile) in 90% ee a t 40% conver-sion (OKU nd INOUE,981).Thisreaction wassubsequently optimized to 97% ee in 97%yield by running the reaction in toluene at

Fo r 18of the 20 am ino acids involved in DNA en -coded pe ptide biosynthesis the C ahn-Ingold-Pre-log designation (S) is equivalent to the Fischer-Rosanoff designation L, which is the predominant

naturally occurring enantiomer. The exceptionsare glycine (achiral) a nd cysteine.

-20 "Cwith two equivalents of hydrogen cya-nide (Fig. 1 andTab. 1,Entry 5 ) (TANAKAt al.,1990).A n im portant feature of this reaction isthat the catalyst must be presen t in the form of

a gel for enantioselectivity to b e achieved.Despite wide ranging studies, variation of

the original ligand's structure (3) (Tab. 1) hasachieved very limited success. Rep laceme nt of(S)-phenylalanine with (S)-leucine gives theopposite enantiomer of the cyanohydrin (ent-

2a) (Tab. 1, Entry 4). If the phenyl ring ofphenylalanine is changed to naphthalene oranthracene (Tab. 1,Entries 11and 12) enantio-

R'8"z 1

HCN, 3

Toluene,-10to -2OOC

3 c y c l e [CS)-Phe-@)-His]1R'

82 2

B R' =H , R ~ =

bR '= H, RZ= F'h-0

97% yield, 97% ee

97% yield, 92% ee

88% yield, >9 8% ee

Fig. 1. Optim al substrate a nd conditions for cyclic di-peptide hydrocyanation.




Tab. 1.Cyclic Dipeptides Used in Stereoselective Hydrocyanation of Benzaldehyde and Derivative l b

Entry Catalyst R1 R* Product ee Y/C Time

["/.I ["/I [hl(Reference)

RtyZ *. 4 cycb[X-(S)-His]

1 4 cycfo-[Gly-(S)-His] H H (R)-2a

(R)-2a

3 4 cyclo-[(R)-Ala-(S)-His] H Me (R)-2a

4 4 cycfo-[(S)-Leu-(S)-His] 'Bu H (S)-2a

5 4 cyclo-[(S)-Phe-(S)-His] Bn H (R)-2a

s 3 (R)-2a

(R)-2a

0

2 4 cycfo-[(S)-Ala-(5')-His] Me H (R)-2b

(R)-2b

3.3 70' 44 (1)

9.9 50' 47 (1)7.5 90" 47(1)

55 85y 5 (3)90 40' 0.5 (4)10.1 90" 42 (1)71 75y 17(2)

26 21y 20 (2 )

97 97" 8 ( 5 )

5 cyclo-[X-(R)-His]

6 5 cyclo-[(R)-Phe-(R)-His] H Bn (S)-2a 93 95' S( 5)

7 4 cyclo-[(S)-Phe-(R)-His] Bn H (S)-2b 11 75y 21 (2)(S)-2b 88 93y 18.5 (2)

8 5 cycfo-[(S)-PhGly-(S)-His] h H 2b 0 3OY 20 (2)9 4 cyclo-[(S)-Qr-(S)-His] H (R)-2b 21 6 l Y 24( 2 )

A10 4 cyclo-[(S)-MeOTyr-(S)- OM e H

Qis]

A

(R)-2b 28 74Y 42 (2)



494

Tab. 1.Continued

12 Synthetic Enzymes -Artificial Peptides in StereoselectiveSynthesis

Entry Catalyst R' R2 Product ee YIC Time

P o 1 [%I [h1(Reference)

11

12

13

14

15

16

17

4 cyclo-[(S)-1-naphthy1Ala-(S)-His]

94 cyclo-[(S)-1-anthracenylAla-(S)-His]

R 1 =

4 cyclo-[(S)-His-(S)-His]

4 cyclo-[(It)-His-(S)-His]

4 cyclo-[(S)-3-thienyl-

Ala-(S)-His]

4 cycZo-[(S)-2-

thienylAla-(S)-His]

4 cycZo-[(S)-2-(5-Me-

thieny1)Ala-@)-His]

18 4 cyclo-[(S)-2-(5-Br-

thienyl)Ala-@)-His]

(R)-2a 0 <lo' - ( 6 )

R ~ = H

(R)-2a 0 <lo' - ( 6 )

R ~ = H

H (R)-2b 2 90Y 42 (2)(R)-2a 2.5 50" 23 (1)

Hf"\\ (S)-2b 10 5OY 21(2)

YN4

H 2a 0 <20' - (6)

H (R)-2a 72 40' - ( 6 )

H 2a 0 <20' - ( 6 )

H 2a 0 <20" - 6)




Tab. 1.Continued

Entry Catalyst R' R2 Product ee Y/C Time

["/.I ["/I [hl(Reference)

% '9 6

.fH H

R' 0

19 6 cyclo-[(S)-(N-Me)-Phe- Me H 2a 0 <lo' - 7)

20 6 ycZo-[(S)-Phe-(S)-(N- H Me 2a 0 <lo' - (7)

(S)-His];R3 H

Me)-His]; R3= H

Me)-His];R3 Me21 6 cycZo-[(S)-Phe-(S)-(2- H H (R)-2a 22 10' -(7)

22 7 yczo-[(S)-(p,p- Me Me (R)-2a 60 20" -(7)

23 7 ycZo-[(S)-(P-Ph)- Ph H (R)-2a 36 20' -(7)

diMe)-Phe-(S)-His]

Phe-(S)-His]~~

24 4 yclo-[(S)-(a-Me)-Phe- Bn Me (R)-2a 15 50' -(7)(S)-His] (R)-2a 99 98' - ( 8 )

25 4 yclo-[(R)-(a-Me)-Phe- Me

(S)-His]Bn (R)-2a 32 9OY - (8)

26 8 (5R)-5-(4-imidazolylmethyl-3-N-phenyl-2,4- (S)-2a 37 90' l (9 )imidazolidinedione

Y, = yield; C, = conversion. References:1OKU t al., 1982;2 JACKSONt al., 1988;3MORI t al., 1989;4 OKUand INOUE,981;5 TANAKAt al., 1990; 6 NOE et al., 1996;7 NOEet al., 1997;8 HULSTt al., 1997; 9 DANDA,

1991.



496 12 Synthetic En zym es -Artificial Peptides in Stereoselective Synthesis

selectivity is completely lost, possibly due tothe disruption of supramolecular assembly.Substitution of the amidic protons believed tobe involved in intermolecular hydrogen bond-

ing (see Fig. 4) by a m ethyl group also leads toa loss of any stereoselectivity (Tab. 1,Entries19 and 20). Incorporation of (S)-a-methyl-phenylalanine (Tab. 1, Entry 24) in place of

(S)-phenylalanine (Tab. 1,Entry 5) is the onlymodification which retains high enantioselec-tivity. Replace me nt of the six memb ered ring,by a five mem bered ring (8),reduces enantio-selectivity, however the natur e of the enan-tiodiscrimination achieved [(S)-cyanohydrin( e n t 3 a ) , (R)-histidinyl residue] suggests thatthe same mode of discrimination is operating

as in the six membered ring case (Tab. 1,En-tries 6 ,7, and 26).

The expansion of substrate range (MAT-THEWS et a]., 1988; KIM and JACKSON, 1994)(Tab. 2) ha s led to a system of grea t practicalvalue (for a review see NORTH,1993) whichhas been applied to the preparation of naturalproducts, drugs and their analogs (Fig. 2)(BROW N t al., 1994). However several limita-tions have been revealed. Arom atic aldehydesbearing electron don ating substituents under-

go hydrocyanation with high enantioselectiv-ity, bu t enantioselectiv ity is redu ced if electronwithdrawing substituents are present. Theenantioselectivity of hydrocyanation of ali-

H

Ar R

Ar =p -oM e-C & , , R = PhCO; Tembamide

Ar =p-OMe-CsH4,R = (E) Ph-CH=CH-; Aegeline

Ar =rn-oPh-CsHd,R =bu;Salbutamol derivative

Ar =p-OH-C6H4,R = CbBz-m,p-OMe;Denopamine

Fig. 2. Natural products and drugs p roduced usingcyclic dipeptide catalyzed hydrocyanation as the keystep (BROWNt al . ,1994).

phatic aldehydes (which mostly react rapidly)is only low to m oderate, and that of conjugateda ld eh yd es ( ~ 3 1 % e) and ketones (<19% e e)is even low er.

The addition of cyanide to imines (9) is theenantiodiscriminating step in the Strecker syn-thesis of a-am ino nitriles ( lo) ,which are pre-cursors of a-amino acids (Fig. 3). Catalysis ofthis reaction by cyclo-[(S)-Phe-(S)-His] (3)

was unsuccessful, which was attributed to fail-ure of the imidazole group to accelerate pro-ton transfer to the imine. Replace me nt of thehistidine residue in the cyclic dipeptide with(S)-a-amino-y-guanidinobutyric cid (which

R'IR29

HCN, 10, MeOH

or M H ,

H75 to 25 OC

11 cyclo[@)-Phe-~-a-am inoyguauidinobutyric acid]I

R2

H-Ns)(

R' H 10

R' =RZ=Aromatic; 82-98%yield, 10->99%ee

R' = Aliphatic, R2 = Aromatic; 88%yield, 75% ee

R' =Aromatic,2 Aliphatic, 71-8155 yield, 10-17%ee

Fig. 3. Cyclic dipeptide catalyzed enantioselectiveStrecker reaction (IYER et al., 1996).

Tab. 2. Substrate Range and Enan tioselectivity for Asymm etric Hydrocyanation of Aldehydes Catalyzed bycyclo-[ S)-Phe-(S)-His)] (3)

Functionality (N o. of Examples) 0-30% ee 3 1 4 0 % e e 61-90% e e 91-100% e e~~ ~ ~

Aromatic aldehydes (43) 9% 28% 46% 16%

Aliphatic aldehydes (18) 50% 44% 5yo -




has a m ore basic side chain group) yielded anextremely efficient catalyst (11). The resultswith a range of substrates show some interest-ing parallels with hydrocyanation. Aromatic

imines with electron donating substituentsgive a-amino nitriles with high enantiomericexcesses, whereas those bearing electron with-drawing substituents, heteroarom atic a nd ali-phatic functionality give low enan tiomeric ex-cesses. Unlike hydrocyanation the Strecker re-action is conducted as a homogeneous reac-tion in methanol, moreover the cyclic dipep-tide (11) gave the opposite enantiomer as themajor product when com pared with the equiv-alent hydrocyanation process, suggesting theinvolvement of different mechanistic factors(IYER t al., 1996).

Cyclic dipeptide hydrocyanation is one ofthe few examples of reactions which exhibitautoinduction (BOLM t al., 1996). In th e hy-drocyanation of m-phenoxy-benzaldehyde(lb, Fig. 1) catalyzed by cyclo-[(R)-Phe-(R)-His] (ent-3)a gradual increase in enantioselec-tivity for formation of the (S)-cyanohydrin(ent-2b) was observed as the reaction pro-gressed. When a small amount of the enantio-pure product of this reaction, the (S)-cyanohy-

drin (ent-2b),was included prior t o addition ofhydrogen cyanide, the optical purity of theproduct re maine d at circa 96% ee throughoutthe course of the reaction. The presence of aninitial trace of (R)-cyanohydrin (2b) gave aslightly lower enantioselectivity, bu t a sim ilarrate of increase in enantioselectivity for for-mation of the (S)-cyanohydrin (ent-2b), overthe duration of the reaction when comparedwith n o pre-addition (DAN DA t al., 1991; cf.KOG UT t al., 1998). M oreover, pre-addition ofachiral 2-cyano-propan-2-01 (acetone cyano-

hydrin) increases the enantioselectivity of fur-fural hydrocyanation from 53 to 73% ee. Pre-addition of (S)-1-phenyl-1-ethanol causes asimilar enhancement of enantioselectivity(72% ee ), but the (R)-enantiomer had only amodest effect (58% ee), which was identical tothat of methanol (58% ee) (KOGUTet al.,1998). These pheno me na are broadly consis-tent with a catalyst in which hydrogen bondingplays a major role and, as might be expected,an excess of th e alcohol causes a dro p in enan-tioselectivity, attributa ble t o the disruption ofhydrogen bonds.

As noted above, the dipeptide catalyst mustbe p resent as a gel in solution for enantioselec-tive hydrocyanation to occur. In early experi-ments the gel tended to dissolve as the reac-

tion proceeded w ith erosion of the enantiose-lectivity. Switching from benzene to toluenelargely solved this problem , an d initial gel for-ma tion is prom oted by a range of prescriptionsfor pretreatment of the dipeptide to giveamorphous non-crystalline material. This isachieved by rapid crystallization of the d ipep-tide from methanol alone or by addition ofether (TANAKA t al., 1990; DAN DA t al.,1991), spray drying (DONG nd PETTY, 1985),interaction w ith supercritical COz (SHVO t al.,1996), or lyophilization (KOGUT t al., 1998).IR observations indicate that during this acti-vation process intermolecular hydrogen bo nds(Fig. 4) are broken, while intramolecular hy-drogen bonds are formed (JACKSONt al.,1992).However, the im portance of intermolec-ular hydrogen bonding for good enantiodis-crimination is clearly undersco red by the poorresults which have been o btained when the cy-clic dipeptide has b een immobilized (KIMandJACKSON,992). The importanc e of hydrogenbonding is also clear from th e fact that me tha-

nol destroys chiral control in the reaction.The mechanism by which cyclic dipeptidesprovide such high enantiodiscrimination has

R' = Bn, R2 Irnidazole(syn)

R' = Imidazole, F?= Bn (anti)

= Hydrogen bonding

Fig. 4. Intermolecular hydrogen bonding of cyclic di-peptides used for enantioselective hydrocyanation( S w o t al., 1996;KOGUT t al., 1998).



498 12 Synthetic Enz ymes -Artificial Peptides

been the subject of considerable investigation.If it is assumed that catalysis involves aug-mented direct addition of cyanide to the alde-hyde (1,Fig. 1) by cyc lo-[(S)-Phe-(S )-His] (3),

then the (R)-cyanohydrins (e.g., 2) must beforme d by a ddition of cyanide to the Si-face ofthe carbonyl group. Localization of the hydro-gen cyanide most plausibly occurs by coordi-nation to th e basic imidazole ring of histidine.The only issues remaining are then the posi-tion of th e aldeh yde and the means by which itis orientated and/or activated to undergo nu-cleophilic attack.

On e proposal is that the transition state re-sembles (14) (Fig. 5) , where th e imidazole ringis rotationally restricted by hydrogen bonding

to the carbonyl group of histidine (KIM andJACKSON,994), while the aldehydic oxygen ishydrogen bonde d t o the am ide group of histi-dine (TANAKAt al., 1990). .rr-Stacking mayalso exist between ligand and substrate, theoverall effect being to cre ate a U -shape mole-cule. Cyanide becomes associated with theimidazole ring by ionic interaction. A n alterna-tive arrangement is supported by NMR dataand has the aldehydic oxygen hydrogen bond-ed to the phenylalanine amide group (15)

(HULS T et al., 1997). NM R and molecularmodeling studies suggest that the pre-transi-tion state mod el (16) does not adopt a U-shapeand that H CN forms a covalent bond to imida-zole (CALLAN Tt a l., 1992).

Interestingly cyclo-[(S)-Leu-(S)-His] (17)

gives the ent-cyanohydrin (ent-2) when com-pared to the product from cyclo-[(S)-Phe-(S)-His] (3) (HOGGet al., 1994). Cyclo-[(S)-Leu-(S)-His] (17) adopts a conformation in solu-tion in which the imidazole ring is folded ov erthe diketopiperazine ring, whereas the pre-

dominant conformer for cyclo-[(S)-Phe-(S)-His] (3, Fig. 6) has the phenyl ring folded overthe diketopiperazine ring (HO GG t al., 1994).This may account for the difference in the sub-strate profile of the two catalysts. Cyclo-[(S)-Leu-(S)-His] (17) performs consistently betterfor aliphatic aldehydes compared to aromaticaldehydes (MO RI et al., 1989), whereas theconverse is true for cyclo-[(S)-Phe-(S)-His]

A radically different mechanism was pro-posed based on the observation that cyclo-[(S)-Phe-(S)-His] (3) in the presence of dioxy-

(3)

in Stereoselective Synthesis

14

- - - - =Hydrogen bonding

@plNo

1 15H

NC- H,

17cyclo-[(s)-Leu-(.S)-His]

Fig. 5. Imidazolekyanide salt mechanism for cyclicdipeptide hydrocyanation and possible conformati-o n of the cyclic dipeptides (TANAKAt al., 1990;KIMand JACKSON,1994).

gen catalyzes the oxidation of benzaldehyde t obenzoic acid. In th e absence of catalyst this re-action proceeds via an aldehyde hydrate,whereas the catalyst can accelerate this reac-



3 Poly-Leucine 499

3

1812a

NH

3- - H'

Fig. 6.Amino1 mechanism for cyclic pep tide cataly-zed hydrocyanation.

tion by attack of benzaldehyde (la) by theimidazole moiety to give an aminol intermedi-ate (18).This could also serve as a substrate fornucleophilic displacement by cyanide to givethe cyanohydrin (2a) (HOGG t al., 1993) (Fig.6). The conformation of cyclo-[(S)-Phe-(S)-His] (3) has been determined in considerabledetail by solution state NMR (HOGG andNORTH,1993), molecular modeling (NORTH,1992), and solid state NMR studies (APPERLEYet al., 1995). These studies indicate that a

strong hydrogen bond exists between the hy-

drogen on N-3 of the imidazole ring and theamidic carbonyl group of the histidine residue.This of course could play a major role in en-hancing the nucleophilicity of N-1 as required

for the aminol mechanism. Whether this in-volves full proton transfer to give the enolicform of the amide (Fig. 6) or a protonatedamide or some other pathway is conjecture atpresent.

Moreover hydrocyanation has second orderkinetics with respect to the cyclic dipeptidesuggesting two molecules of cyclic dipeptideare involved in the rate determining step. It isplausible that the aldehyde is bound to oneimidazole group and the hydrogen cyanide toanother imidazole group with the catalytic

event occurring in the void between two mole-cules of dipeptide (SHVO t al., 1996). If this isthe case, and catalysis involves multiple di-peptide units, it will be ironic, because the in-ital premise for using cyclic dipeptides was thattheir structural rigidity would mimic the hy-drogen bonding arrays of long acyclic poly-peptides. Have the catalysts evolved (by selec-tion for enantioselectivity) towards structureswhich resemble acyclic polypeptides?

3 Poly-Leucine

The first breakthrough in the use of synthet-ic peptides in stereoselective synthesis wasachieved by JULIA JULIA t al., 1980; COLON-NA et al., 1987). Using poly-L-alanine to cata-lyze the epoxidation of chalcone (19) undertriphasic conditions (polypeptide/organic sol-

vent/aqueous phase) the (2R,3S)-epoxide (20)

was obtained in up to 93% ee (Fig.7). By usingpoly-D-alanine the ent epoxide (21) was pre-pared with comparable enantioselectivity. Incollaboration with COLONNAhe reaction wasfound to be applicable to a fair range of a,p-unsaturated ketones (Tab. 3), gave optimal re-sults in toluene and carbon tetrachloride andwhile the rate of the reaction was dependenton the substrate : atalyst ratio, the enantio-meric excess obtained was largely unaffected.These latter two features indicate enzyme-likecatalysis and a low background uncatalyzed

epoxidation rate (JULIA et al., 1982). The



500 12 Synthetic E nzy me s -Artificial Peptides in Stereoselective Synthesis

Triphasic reaction, 1-3 days

Ph

212o t Toluene,NaOH, t

Ph Ph

20 21

Biphasic reaction, 5-3 0 min

Fig. 7. Biphasic and triphasic poly-leucine stereosel-ective epoxidation.

so-called “poly-alanine” used in this work wasprepared by polymerization of the N-carboxy-anhydride of L-alanine initiated by n-butyl-amine, thus the polymer bears a C-terminal N-

butyl amide group. However the catalytic ac-tivity is unaffected by the nature of this groupand even if the C-terminus is used as a site forimmobilization to polystyrene, enantioselec-tive catalytic activity is retained (BANFI t al.,1984). Poly-L-alanine oligomers with an aver-age of 5 ,7,10 and 30 residues were tested ascatalysts, with only 10- and 30-mers making vi-able peptides for epoxidation, although the 30-mer was considered marginally optimal (JULIA

et al., 1980,1982). More than 15 oligopeptideswere prepared and assayed for enantioselec-

tive epoxidation of chalcone (18),but the onlyexamples with practical levels of enantioselec-tion were poly-leucine and poly-isoleucine(COLONNAt al., 1983).

The major disadvantage of triphasic reac-tions is that the reactions are mostly slow (1-3 d) which results in partial degradation of thepoly-alanine and partial loss of activity (JULIAet al., 1982,1983;COLONNAt al., 1983,1984).Despite the slow rate of the reaction it was ex-ploited successfully in the synthesis of flavo-

noids (cf.29,

Fig. 8)(BEZUIDENHOUDT

t al.,1987;AUGUSTYNt al., 1990; VANRENSBURGet al., 1996), although superior results weresubsequently achieved with biphasic condi-tions (NELet al., 1998).

The next developments in the use of poly-peptides as epoxidation catalysts came aboutas a consequence of a large scale synthesis of

Tab.3.The Epoxidation of “Chalcone-vpe” Substrates(19 nd l9a-d)Using Poly-L-Alanine Under Tripha-sic Conditions (JULIA et al., 1980,1982,1983; COLONN At al., 1983,1984)

0

Toluene, HzOz.q NaOH

4.\’ Rz

Rl- RZ Poly-L-alanine, triphasic conditions R’

19 and1 9a-d 20 and 20a- d

Entry Substrate R’ RZ Time

[hl

ee

[%1Yield

1% 1~~ ~

1 19 Ph Ph 37 93 752 19a ‘Pr Ph 42 50 683 19b Ph ‘Pr 52 60 494 19c Ph ‘BU 49 100 245 19d Ph Naphthyl 44 80 77



22 SK&F 104353

y

25 (+)-Goniopypyrone

23 Diltiazem 24 Clausenamide

0

Ftl

FhH

HHOO H

26 (+)-Goniofufurone

AcO 0

27 (+)-Goniotriol

28 TaxolTMISide hain] 29 Dihydroflavonols

Fig. 8. Na tur a l produ c ts and d rugs prep ared us ing poly- leuc ine ca ta lyzed epoxida t ion

SK&F 104353 (22,Fig. 8).A selection of com-mercial homo-polypeptides were investigated

and poly-L-leucine was found to give the mostconsistent results. The catalyst was preparedby placing the N-carboxyanhydride of L-leu-cine into trays to a depth of 6-8 cm in a cham-ber at 70-75% relative humidity. Polymeriza-tion was initiated by water from the air. Pre-swelling the peptide for 6 h under the reactionconditions prior to the addition of the sub-strate and the use of hexane as solvent led tohigher enantioselectivity in shorter times, withthe additional bonus that the recovered cata-lyst retained full activity over 6 cycles (BAURES

et al., 1990;FLISAKt al., 1993).

ITSUNO repared immobilized poly-alanineand poly-leucine by initiating polymerization

of the N-carboxyanhydrides with aminometh-yl groups on the surface of cross-linked micro-porous polystyrene. This provided a more ro-bust catalyst, affording a higher degree of recy-clability (12 cycles) and reproducibility (ITSU-NO et al., 1990). Surprisingly, even peptideswith as few as four residues had substantialenantioselective catalytic activity (83% ee),whereas the corresponding non-immobilizedpeptides gave only poor enantiomeric excessesand yields.

In the last few years, the Roberts group at

Exeter and latterly Liverpool has further ex-



502 12 Synthetic Enzym es -Artificial Peptides in Stereoselective Synthesis

panded the scope of poly-leucine catalyzedepoxidation (for a review see LASTERRA-SANCHEZnd ROBERTS,997). Initially the tri-phasic conditions were extended to an in-

creased range of (E)-@unsaturated ketones(Tab. 4) (LASTERRA-SANCHEZnd ROBERTS,1995; LASTERRA-SANCHEZt al., 1996; KROU-TIL et al., 1996a,b) and subsequently to (Z)-di-substituted, digeminally substituted, and tri-substituted a,P-unsaturated ketones (Tab. 5 )(RAYand ROBERTS,999;BENTLEY,999).

Phase transfer reagents were added to in-crease the rate of the reaction, without loss ofenantioselectivity, suggesting the phase trans-fer properties of poly-leucine were not implic-it in its enantioselectivity. These conditions

were used in the epoxidation of phenyl-P-sty-ryl sulfone, but only achieved 21% ee. This dis-appointing result can be rationalized by in-creased steric hindrance of the sulfone oxy-gens or weaker hydrogen bonds possiblethrough the sulfone oxygens (compared withthe ketone) to the amide of the peptide back-bone. The substrate would be less tightlybound into the chiral environment enhancingthe role of the background reaction.This couldindicate key hydrogen bonding of the sub-

strate to poly-leucine via the ketone in enones(BENTLEY, 999). Alternative oxidation sys-tems to hydrogen peroxidelsodium hydroxidewere found including sodium percarbonate(LASTERRA-SANCHEZt al., 1996) and sodium

perborate/sodium hydroxide (KROUTILt al.,1996a; SAVIZKYt al., 1998). These methodshave some advantages, but the reaction timesare still prolonged (>24 h). The discovery of

biphasic conditions, in which the urea-hydro-gen peroxide complex (UHP) (HEANEY,993)is used with 1,8-diazobicyclo[5.4.0]undec-7-

ene (DBU) (YADAVnd KAPOOR,995) as thebase in THF led to a dramatic decrease in reac-tion time (5-30 min; BENTLEY t al., 1997a).The biphasic conditions were successfully ap-plied in the synthesisofdiltiazem (23) ADGERet al., 1997), clausenamide (24)CAPPIet al.,1998), (+)-goniopypyrone (25), +)-goniofu-furone (26), and (+)-goniotriol (27) CHENand ROBERTS,999), TaxolTM ide chain (28)

(ADGERet al., 1997), and dihydroflavonols(29) NELet al., 1998) (Fig.8).

The standard THF/DBU/UHP biphasic re-action protocol has been significantly im-proved in terms of enantioselectivity, with lim-ited loss of reaction rate. This has beenachieved by dilution of the reagents and in-creasing the amount of poly-leucine relative tothe substrate, while changing the solvent toiso-propyl acetate (Tab. 5) . These conditionswere used to expand the substrate range to en-

compass trisubstituted and digeminally substi-tuted enones (32) BENTLEYnd ROBERTS,n-published data). Substrate (32) offered aninteresting contrast to prior results (e.g., 30f);these “tethered chalcones” (30 and 32) could

Tab. 4. The Epoxidation of Structurally Varied Substrates (19e-j) Using Poly-L-Leucine Under Triphasic

Conditions (LASTERRA-SANCHEZnd ROBERTS,995; LASTERRA-SANCHEZt al., 1996; KROUTIL t al.,

1996a,b)

Toluene, H202,aq NaOH 2..\‘5 R2

R I A4 2 Poly-L-leucine, triphasic conditions * R ‘

19e-j 20e-j

Entry Substrate R1 RZ Time ee Yield

[hl 1%1 [”/.I

1 19e Cyclopropyl Ph 18 77 85

2 1% Ph-C C Ph 96 90 57

Ph (SCH2)2C=CH 115 >94 51

Ph - 76 7619h PhCO

5 19i PhCH=CH 2-Naphthyl 72 >96 78

6 19j Ph PhCH=

CH-

>98-

3 19g



3 Poly-Leucine 503

Tab. 5.The Epoxidation of Digeminal and Trisubstituted Alkenes (Ma-h, 32)Catalyzed by Immobilized Po-

ly-L-Leucine Under Biphasic Conditions.Protocol: Substrate (3Oa-h, 2) 0.24 mmol), PrOAc (1.6 mL), ac-

tivated immobilized PLL (200 mg), urea-hydrogen peroxide complex (UHP, 11 mg) and 1,8-diazobicy-

clo[5.4.0]undec-7-ene (DBU, 30 mg). UHP and DBU were added every 12 h until reaction was complete

(BENTLEY,999)

0

R

Ma-h 'PrOAc, UHP, DB U*

Immobilized poly-L-leucine, biphasic conditions0

Ph

32

Q&R

n

31a-h

Ph

33

Substrate R n Time ee Yield

[hl ["/.I 1% I

30a Ph 2 90 84.5 76

30b p-Br-C,H, 2 72 82.5 81

30c p-NO,-C,H, 2 78 96.5 85

30d Me 2 60 92.0 66

30e 'BU2 192 83.0 6330f Ph 1 48 88.0 72

30h" H 2 7 94.0 64

32 - - 72 0.0 67

3% Ph 3 168 59.0 74

a Reagent additions made after 0.5,1,3 nd 6 h

give an indication as to the orientation thatthe (E)-disubstituted a$-unsaturated ketoneadopts so as to be processed by poly-leucine.

Although initial studies found the range of

bases to be of utility with UHP very limited,further work demonstrated potassium carbo-nate to provide comparable enantioselectivityto DBU, ultimately leading to sodium per-carbonate's (BENTLEY,999) replacement ofUHP and DBU. Development of these condi-tions has provided distinct improvements(RAYand ROBERTS,999; Allen et al., 1999) inthe reaction rate of this economically signifi-cant alternative biphasic reaction.

Poly-leucine is prepared from leucine N-car-boxyanhydride (NCA) (34) (Fig. 9) (KRICHEL-

DORF, 987), which in turn is prepared from

leucine and triphosgene (DALYand POCHE,1988). There is a strong correlation betweenthe purity of NCA as assessed by melting point(should be within 2 "C of literature values;

LeuNCA: 76-78 "C) and the quality of poly-leucine as a catalyst. Good quality NCA is ob-tained by controlling the temperature of thereaction and limiting the exposure of NCA tomoisture at all stages of the process. An extrastep of stirring poly-leucine in aqueous NaOHand toluene followed by washing with selectedsolvents is required to provide superior mate-rial (35) for biphasic epoxidation reactions,when compared to unactivated poly-leucine(ALLEN t al., 1998; BENTLEYt al., 1998; Nu-GENT, npublished data).




a 4 M aq NaOH, toluene

approx. 20 h

b H 2 0 , 1:1, 1.4 (H20:Acetone),

acetone, EtOA c, EtZO

H' F &

-f0

JHN

The means by which poly-leucine achievesits impressive chiral discrimination has recent-ly begun to be addressed by the preparation ofpolypeptides of precisely defined lengths, con-

taining both D- and L-residues attached via aspacer to a polystyrene resin. The homo-deca-peptide of L-leucine (P l, Fig. 10) catalyzes theepoxidation of chalcone to give the expectedepoxide (20) with a moderate enantiomericexcess, whereas the 5 ~ / 5 ~iastereoisomer(P2) gives essentially racemic epoxide (20).The 20-mer homologs (P3-P8) behave differ-ently.The 7 ~ / 1 3 ~P4), and 5 ~ / 1 5 ~P5) diaster-eoisomers give predominantly the enantio-meric epoxide (21), apparently under controlby the D-leucine residues in the N-terminal re-

gion (BENTLEYt al., 1998). It is clear from the1 ~ / 5 ~ / 1 4 ~iastereoisomer (P6) giving pre-dominantly the epoxide (21), nder control bythe D-leucine residues, 2L/SD/13L (P7) givingvirtually racemic epoxide and good selectivityfor the epoxide (20) being restored with the3 ~ / 5 ~ / 1 2 ~iastereoisomer (P8) that the termi-nal trimer is the most important determinantof enantioselectivity (BENTLEY, npublisheddata). Moreover, the terminal amino groupcan be replaced by N,N-dimethylamino-(BENTLEYt al., 1998) or other substituents (-

JULIA t al., 1982) without a substantial effecton enantioselectivity. Cleavage of the peptidefrom the polymer does not change the stereo-control of the N-terminal region, eliminatingits comparatively less hindered environmentas an explanation for these observation.

The C-terminal region also plays a signifi-cant role in determining enantioselectivity, po-tentially providing the correct scaffold allow-ing the N-terminal region to be presented in aneffective orientation. An immobilized 18-mer

of L-leucine is a highly enantioselective epoxi-dation catalyst (Fig. 11, Pl ), whereas the corre-sponding tert-leucine analog (P4) is non-enan-tioselective.As might be anticipated, polypep-tides with 9 terf-leucine or a mixture of D,L-leU-cine residues in the C-terminal region (P2 andP5) and 9 L-leucine residues at the N-terminusare also selective catalysts, but interestinglythe 9 ' ~ / 9 ~atalyst (P3) is also fairly enantiose-lective (BENTLEYt al., 1997b).

In summary, the length of poly-leucine mustbe at least 10 residues for good enantioselec-

tivity, with only 18-20 residues providing opti-



3 Poly-Leucine 505

rious positions in poly-leucine Yo ee

MajorEpoxide

using D- and L-leucine (shownas D and L , respectively).

Fig. 11.The determination of the role of aminoacids in various regions of poly-leucine, usingL-leucine, racem ic leucine, and L-leucine[shown as L, D,L), and 2 , espectively].

mum stereoselectivity and a higher rate(FLOOD,npublished data). Shorter polypep-tides are more enantioselective if immobilized.Three to five residues at the N-terminus arethe most important determinant of enantiose-lectivity, but this can be modified profoundlyby five to fifteen C-terminal residues, althoughthe stereochemistry of these residues does notseem to be important. The terminal aminogroup does not play a key role in determining

enantioselectivity.Other examples of reactions for which poly-

peptides can be used as enantioselective cata-lysts are rather sparse but they do give impor-tant directions for expansion of poly-leucinetechnology. It has been shown that palladiumbinds to the peptide backbone (FRANCIS et al.,1996). Poly-L-leucine has been used in anenantioselective palladium catalyzed carbony-lation to give the (&)-lactone ( 3 7 ) (Fig. 12) andgave better chiral induction than a range ofchiral alcohols (e.g., menthol, 1,l-bi-2-naph-

thol, and diethyl tartrate) and phosphines (BI-

66 5 91 80 45 67 11 89

20 20 20 21 21 21 21 20

No. f

residues0

9

18

%ee >90 81 64 3 0

NAPand CHIRAPHOS). Poly-D-alaninegavethe (S)-lactone (ent-37) ,but in only 8% enan-tiomeric excess. It has been speculated that thereaction goes via the complex ( 3 8 ) ,althoughno explanation was proffered for invoking theunusual protonated amide (ALPERand H A -MEL, 990).

3.1 Other Applications ofOligopep ides

Stereoselective electrochemical reductionand oxidation reactions have been carried outusing electrodes coated with a range of homo-polypeptides. Citraconic acid ( 3 9 ) (ABEet al.,1983; NONAKAt al., 1983; ABE and NONAKA,1983) and 4-methylcoumarin (41) (ABEet al.,1983) were reduced on poly-L-valine coatedgraphite electrodes with moderate success(Fig. 13). Amazingly, the authors reported that

both poly-D- and poly-L-valine coated elec-



506 12 Synthetic Enzym es -A rtificial Peptides in Stereoselective Synthesis

OH

0 2 , PdCl2, CUCl2.

PLL, HCl, THF,

18h,1tI37 49%yield, 61%ee

4Fig. 12. Stereoselective carbonylation with poly-L-leucine.

trodes reduced citraconic acid (and also mesa-conic acid) to (S)-methylsuccinic acid (40)(25% and 5% ee, respectively). This seeminglyimpossible result was probably due to differ-

ent degrees of polymerization of each of thepolypeptide samples, since although they hadopposite optical rotations these were not iden-tical in magnitude. More fruitful studies in-volved the oxidation of phenyl sulfides (43) tothe corresponding sulfoxides(44). artially re-usable platinum electrodes were prepared bycovalently coating with poly-pyrrole and thendipping in a solution of poly-L-valine. Al-though the enantioselectivity of the electrodesdiminished with each cycle of use, the activitycould be almost wholly restored by redipping

in the solution of poly-L-valine.In some cases

A39

Graphite electrode coated

with poly-L-valine,

Na& P04 , ethanol, pH 6

41

40 25% ee 42 43% ee

Fig. 13. Stereoselective reduction with poly-L-valine.

Tab. 6.Electrochemical Enantioselective Oxidationof Sulfides t o Sulfoxides with Poly-L-Valine Coate dElectrodes (KOMORI nd NONAKA,983,1984)

Pt electrode coated

with poly-pyrrole ..and poly-L-valine

'%. ,o

P h N S ' R * Ph+R"Bu4BF4,aq MeCN

J

43a-f 44a-f

Entry Substrate R ee Yield[% I [Yo]

-1 43a Me 12 43b "B u 18 -

3 43c 'B u 44 694 43d 'Pr 77 565 43e 'BU 93 456 43f 'C,H,, 54 -

good enantioselectivity was observed (43e)(Tab.6) (KOMORInd NONAKA,983,1984).

The selective epoxidation of the terminal al-kene bond of the hexaene squalene, is one ofthe "landmark" achievements of organic syn-thesis.The result was rationalized as due to the

polyene chain forming a coiled conformation



3 Poly-Leucine 507

in polar media which minimizes contact of thenon-polar chain with the solvent and shieldsthe internal alkene bonds from reaction withthe epoxidation reagent (VAN TAMELEN,

1968).This idea was extended by preparing ho-mochiral peptide derivatives of (E,E)-farnesicacid with the anticipation that the peptidechain would induce a helical conformationwhich would render epoxidation both regio-and enantioselective.

Initial studies using hexa-L-phenylalanylmethylester farnesate (45a; Fig. 14) gave poorresults, possibly because the material had lowsolubility and aggregation to give p-pleatedsheet structures occurred. Insertion of a-ami-no-iso-butyric acid (Aib) residues into thepeptide chain increased solubility and conse-quently improved the results obtained. Thebest results were achieved with the N-terminalfarnesoyl derivative of the nonapeptide; [L-

Phe-L-Phe-Aib],OBz (45b). Bromohydroxyla-tion followed by base treatment gave the (R)-

epoxide (46b) but with only 25% chiral in-duction, whereas epoxidation with m-chloro-peroxybenzoic acid gave the (S)-enantiomer(ent-46b) (12% ee) plus bis-epoxide. The enan-tiocomplementary results with bromohydrox-

ylationhase and peracid are consistent withapproach to the less hindered face of an alkenebond in a right-handed a-helix (BUDTet al.,1986).

, RN.H

45a,bINBS, g1yme:water (5:1), 0 O C ;

b K2CO3, MeOH, 0 "C

R H

46

a R = (L-Phe)6-OMe

b R = (L-Phe-L-Phe-Aib)3-OBz; 5% ee

Fig. 14. Regio- and enantioselective epoxidation of

farnesamide N-oligopeptides (BUDT t al., 1986).

0 0

0 H29 cp s

Ph NH2

48 Pps

H2, 1200 psi, 47

H - H y 04-9% ee

H - N y 0

50 5 1

Fig. 15.Stereoselective hydrogenation using a rhodi-um polypeptide complex (Ac-Ala-Aib-Ala-Pps-Ala-Ala-Aib-Cps-Ala-Ala-Aib-Ala7).

Hydrogenation reactions have been carriedout using rhodium and a derivatized oligomerof alanine (47) (Fig. 15) with incorporation ofAib to enhance secondary structure. Althoughinitially poor enantiomeric excess was ob-tained (GILBERTSONt al., 1996), an intriguing

combinatorial strategy was adopted from this



508 12 Synthetic Enz ym es -Artificial Peptides in Stereoselective Synthesis

approach (GILBERTSONnd WANG,1996). A63-member library was created, its generic fea-tures were Ac-Ala-Aib-Ala-1-2-3-4-5-Ala-Aib-Ala-NH,. The peptides either had two Pps

(cf.48),wo Cps (cf.49) or one of each. In somecases these residues were at positions 1and 5 ,in others they were adjacent to each other.Theremaining positions were taken with combina-tions of Ala, Aib, Phe, Val, His, and Ile with theobjective of placing the latter four residues inproximity to the metal. When the catalystswere used in the hydrogenation of (50) at 400psi the highest enantioselectivity (51,18% ee),was achieved with Cps-Ala-Ala-Aib-Cps. Ageneral trend found was that the better cata-lysts had Ala at positions 2 and 3 with another

amino acid at 4. It shows some of the potentialfor accurate rational modification of singleamino acids in an oligomer so as to designoligopeptides utilizing variation in side chainsin the assembly of a chiral scaffold.

A noteworthy failure is an attempted ster-eoselective epoxidation of unfunctionalizedolefins using a metalloporphyrin strapped witha peptide (Ac-Ala-Cys-Glu-Gln-Leu-Leu-Lys-Glu-Leu-Leu-Gln-Lys-Cys-Ala-NH,)hroughthe cysteine sulfur atoms (GEIER nd SASAKI,

1997). Despite no enantiodiscrimination beingobserved, the study demonstrates a thoroughrationalization of the design process (GEIERand SASAKI,999; GEIER t al., 1999).

4 Use of Linear Di-,Tri-

and Tetrapeptides

INOUE’S onsiderable contribution to thiswhole field has also embraced linear dipep-tides attached to a Schiff base (52,53)Fig. 16),which in the presence of titanium(IV), cata-lyze enantioselective hydrocyanation (Tab. 7).Good enantioselectivity has been obtainedwith a diverse range of aldehydes (54a-o)(MORI t al., 1991;ABEet al., 1992),with lowertemperatures providing higher enantiodis-crimination. While Ile, Leu, Ala, and phenylglycine have also been studied as part of thecatalyst, Val, Phe, and Trp appear to provide

the best results (NI?TA t al., 1992).

52

a Nap-(S)-Val-(S)-Phe

b Nap-(R)-Val-(S)-Phe

c Nap-(R)-Val-(R)-Phe

53 Nap-(5’)-Val-(S)-Trp

Fig. 16. Dipep tide ligands for titanium alkoxide cata-

lyzed enantioselective hyd rocyanationof

aldehydes(MORI et al., 1991; ABE t al., 1992; NITTA t al.,1992).

An elegant development of this methodolo-gy is the titanium(1V)-catalyzed addition oftrimethylsilyl cyanide (TMSCN) to meso-epoxides using a combinatorially optimized li-gand (interesting examples in the series: ligand56 I-IV and i-iv) for each substrate type(57a-c and 59) (Tab.8,Entries 1,2;7, s) (COLE

et al., 1996; SHIMIZUt al., 1997).The appendage of glycine as R3 as provid-

ed increases of enantioselectivities of up to20% ee compared to R3 OMe with differentligands, for three substrates (Tab.8, Entries 3,4; 10, 11; 12, 13). Interestingly it appears thechirality of specific amino acids do not entirelydetermine which product is the major enan-tiomer in the reaction; the functionality of theside chain is at least partially responsible (Tab.8, Entries 4,5,nd 6 ) (SHIMIZUt al., 1997).

Extension of this work has produced an

even more successful enantioselective addi-



4 Use of Linear Di-,ri- and Tetrapeptides 509

Tab. 7. Enantioselective Hydrocyanation of Aldeh ydes Using Titanium D ipeptide Com plex Catalysts

Ligands 52 or 53 , toluene, HCN, Ti(OEt),or Ti(dPr), HxOH H04* H*

R H R CN Rx CN

54a-o (R)-55a-o (S)-55a-o

54a la 54b 54c OPh 54d 54e 54f

54g 54h 54i 54j 54k

541 54m 54n 540

En try Ligand Substrate Temp Time Product ee Conv.

[“CI PI [Yo1 [Yo1

12345

6

78

9101112131415161718

52a52b52c53535352c52a52a52a52a52a52a52a52a52a52a52a

54a54a54a54a54b54c54d54e

54f

54h54i54j54k54154m

54n540

54g

- 0 4- 0 4- 0 7- 0 3- 0 7.5- 0 7.5- 0, - 0 11,12- 0 1.5- 0 1.5- 0 18- 0 119- 0 22- 0 71- 0 20- 0 46- 0 143- 0 143- 0 69

(R)-55a,86(S)-55a,38(S)-55a,84(R)-55a,88

(R)-55c,86

(R ) -55e , 4(R)-55f, 74

(R)-55b,90

(S)-55d, 86

(R)-55g,81

(R)-55h,89(R)-55i,85

55j, 7055k, 37551,7255m,6055n,60550,68

85848688

8874859999828385742290287857




Tab. 8. Titanium Dipeptide Complex Catalyzed Enantioselective meso-Epoxide Ring Opening with Trime-thylsilyl Cyanide (COLE t al., 1996;SHIMIZUt al., 1997)

Nq.*osiM+

58

Ti(O'Pr),, toluene, 4 'C, TMSCN - N y O s i M e ,

"Pr "Pr Ti(O'Pr),, toluene, 4 "C,TMSCN "Pr "Pr

59 60

Entry Ligand 56 Substrate Product 58 or 60

R' R2 R3 n ee [YO] Yield [%]

12

3

4

5

6

7

89

10

11

12

13

II1

I1

I1

I1

I1

I

I1

I1

111

111

IV

IV

11

111

11

11

...

iiv11

111

111

11

11

11

11

...

...

OMeOEtOMe

GlYGlYGlY

GlY

GlY

GlY

OMe

OEt

OMe

OMe

57a57a57a57a57a57a5%

5%57c57c57c59

59

11111

12

2

3

3

3-

64

75

63

83

10

ent-SSa,58

86

70

52

69

84

58

78

(<39% after 48 h), bu t gradual addition of iso-propanol provided the product in higher yieldand improved enantioselectivity with somesubstrates. It is thought that th e alcohol causesslow in situ release of hydrogen cyanide fromth e TMSCN (KRUEGERt al., 1999).Th e use ofthe diphenylmethyl protecting group enablesthe nitrile group and the amino protectinggroup to be cleaved simultaneously with 6 N

hydrochloric acid to give the corresponding

56

57

72

-

65

65

-

79

69

-

amino acid without racemization. A similartype of catalyst and approach can be seen withanoth er stereoselective Strecker reaction (SIG-MAN and JACOBSEN, 1998) and epoxidation(FRANCISnd JACOBSEN, 999).

Recently MILLER nd coworkers have de-veloped tetrapeptide catalysts for the acyla-tion of alcohols incorporating a 3-( l-imidazo-ly1)-L-alanineresidue (64a shown as an N-acet-

ylated intermediate; Fig. 17). The challenge in



4 Use of Linear Di-, Tri- and Tetrapeptides 511

Tab. 9. Titanium Dipeptide Complex Catalyzed Enantioselective Addition of Cyanide to Imines (StreckerReaction)

62a-g

Ti(O’Pr), (0.10equiv .), toluene, H

TMSCN (2 equiv.), PIOH

(1.5 equiv.) added over 20 h, 4 OC 63a-g

62,63 e f

a R3= R4 = H

b R 3 = C 1 , R = H

c R3 =Br, R4 =H

d R 3= H, R4 = OMe

4

Entry Ligand 61 Substrate Product 63a-g

R1 R* ee [YO] Yield [Yo]

1 OMe H 62a 97 822 c1 c1 62b 93 853 C1 c1 62c 94 934 c1 c1 62d 94 99

5 OMe H 62e 93 80

6 OMe H 62f 90 87

7 Br B r 62g 85 97

creating an “active site” with a short peptide isto ind uce it to fold back up on itself, rather thanform an ex tende d structure. This is achieved inthis case by the comb ination of proline an d Aib(Aib is a seemingly popular choice for synthet-ic peptides) to give a p-turn. The a-methylben-zamide group was incorporated to provide 7r-

stacking with th e acylimidazolium ion. The cat-alyst (64a) showed good enantioselectivity inthe acylation of a range of alcohols (e.g., 65)bearing an adjacent amide group, which pre-sumably orientates the alcohol in the hydro-

gen bonding array of th e catalyst. The im por-

tance of hydrogen bonding was further dem-onstra ted by a solvent optim ization study. Thehighest enantioselectivity was achieved innon-polar, non-Lewis base solvents, whereasprotic solvents resulted in totally non-selectiveacylation. The role of the imidazole group inth e catalyst (64a) was demonstrated by replac-ing it with a phenyl group (i.e., a pheny lalanineresidue), to give an analog (64b)which did no tact as an acylation catalyst (MILLERet al.,1998).

Further investigations studied the replace-

ment of the C-terminal a-methylbenzamide



512 12 Synthetic Enz ym es -Artificial Peptides in Stereoselective Synthesis

Z N \

a R = QBocHN 0 2-( X - - H - N 0

\ ..,,,,, bR=Ph

- R d 64 HObNHAc A C Q ~ N H A C

A c 2 0 , oluen e, CHCl,, 0 "C+

rac-65 (2R73R)-65 (2S.3 8-66 84% ee

Fig. 17.Tripeptide benzam ide catalyzed enantioselective acylation (MILLER t al., 1998).

group with phenylalanine, valine, or glycine togive the tetrapeptide (67) (Tab. 10) and re-placement of the L -proline residue by a D-pro-line residue (68) (Tab. 11).

The diastereomeric catalysts (67) an d (68)give enantiocomplementary products with thelatter being mo re enantioselective in all cases.Evidently the chirality of the proline residuedictates which enantiomer undergoes acyla-

tion, but th e enantioselectivity is also mod ulat-ed mainly by histidine, but to some extent bythe C-terminal residue. For the L-proline basedcatalyst (67), D-amino acids at the C-terminusincrease enantioselectivity, while the converseis true for the D-proline based catalyst (68).Evidently, the u-relationship is advantageousfor selective catalysis over the I-form. NMRan d IR data were used to determine that the

Tab. 10.L-Proline Containing Tetrapeptide Catalysts for the Acylation of Alcohols (COPE LAN Dl al., 1998)

1

Entry

= T N aR}'

OM eN d/

67 2-5 mol%

Ac 20, oluene, CHCl,, 25 "C

rac-65Catalyst 67 S (2R,3R)-65R1 ee [%

(2R,3R)-65 (2S.38-66

(2S,3S)-66 Conv.

e e [ Y O ] ["/I

1 L-Phe 3.0 442 D-Phe 5.7 893 L-Val 3.4 544 D-Val 4.3 655 GlY 3.5 50

3436353938

5671616357



5 Summary and Future Directions 513

Tab. 11.D-Proline Containing Tetrapeptide Catalysts for the Acylation of Alcohols (COPELANDt al., 1998)

HJ--(NHAc 68 2-5 mol% OMe i R1

*

V A q O , toluene, CHC13, 25 “C

rac-65 (2S.38-65 (2R,3R)-66~

Entry Catalyst 68 S (2S,3S)-65 (2R,3R)-66 Conv.

R’ ee [Yo] ee [% I [Yo

1 L-Phe 28 98 73 582 D-Phe 14 89 66 573 L-Val 21 99 63 614 D-Val 9.2 88 55 625 GlY 14 97 57 63

L-proline based catalyst (67) gives a type I1p-turn w ith one hydrogen bond and th e D-pro-

line based catalyst (68) gives a type 11’ p-tu rnwith two h ydrogen bonds. This probably givesa more rigid structure, which is the origin ofthe g reater enantioselectivity of the latter cat-alyst (COPELANDt al., 1998).

While successful examples of pep tides beingused in stereoselective synthesis are currentlylimited in number, the few positive examplesdem onstra te a significant capacity for their vi-able application. The trend towards combina-torially designed ligands, which allow ad apta-tion for individual sub strates, increases th e po -tential importance. To this end peptidesuniquely offer a cheap, simple, enantiopure(with both enantiomers available), diverserange of subunits (amino acids) with the po-

tential for metal binding and hydrogen bond-

ing of reagents and substrates (BUR GER ndSTILL, 995; FRANCISt al., 1996; SEVERINt

al., 1998).The chemistry for prep aration of theligand (peptide) is very well understood withfacility for solid phase synthesis and goodcomprehension of their structure. Such com-patibility offers the exciting possibility of m ul-tidomain peptides facilitating multiple stereo -selective reactions tailored to specific require-ments. Furth er exploitation of th e versatility ofsynthetic peptides as catalysts in stereoselec-tive synthesis is inevitable.

Since the preparation of this chapter therehave been interesting studies using combina-

torial variation of peptides t o optimize serine-protease mimics (DE MU YN CKt al., 2000) andin the search to find means of controlling cy-clizations of epoxy-alcohols to provide the en-ergetically disfavored product (CHIOSIS,000).In addition to this, diphenylglycine has dis-played an interesting capacity for the enantio-selective inclusion of sulfixides (AKAZOMEtal., 2000).



514 12 Synthetic Enzy mes -Artificial Peptides in Stereoselective Synthesis

6 References

ABE,S., NONAKA,T.1983), Electroorgan ic reactionson organic electrodes. Part 5. Durable chiralgraphite electrodes modified chemically with po-ly(L-valine) and poly(N-acryloyl-~-valinemethylester), Chem. Lett.,1541-1542.

AB E ,S., N O N A K A ,., FUCH IGAM I,. (1983), Electro-organic reaction s on organic electrodes. 1. Asym -metric reduction of prochiral activated olefins onpoly-L-valine-coated g raphite, J. Am . Chem. SOC.

105,3630-3632.AB E ,H. , NIITA,H., MORI,A., INOU E, . (1992), A fa-

cile synthesis of optica lly active y-cyano allylic al-cohols using asymmetric hydrocyanation of a#-

alkenyl aldehy des followed by stereospecific [3.3]sigma tropic chirality transfer of the cyanohy drinacetates, Chem Lett . ,2443-2446.

AD GE R, . M., BARKLEY,. V., BERGERON,., CAPPI,M. W., FLOWERDEW,. E. et al. (1997), Impro vedprocedure for JuliCColonna asymmetric epoxi-dation of a#-unsaturated ketones: total synthesisof diltiazem andTa xolTM ide-chain,J. Chem.SOC.,

Perkin Trans. 1,3501-3507.AKAZOME,M., UENO,Y., OOISO,H. , OGUR A,K.

(2000), Enantioselective inclusion of methylphenyl sulfoxides and benzyl methyl sulfoxidesby (R)-phenylglycyl-(R)-phenylglycyl and thecrystal structures of the inclusion cavities, J. Org.Chem.65,68-76.

ALL EN, . V., ROBERT S,. M., WILLIAMSON,. M.(1998), Polyamino acids as man-made catalysts,Ad v. Biochem. Eng. Biotechnol. 63,126-143.

ALLEN,. V., DR AU Z, .-H., FLOOD , . W., ROBERTS,S. M., SKIDMORE,. (1999), Polyamino acid cata -lyzed asymmetric epoxidation: sodium percarbo-nate as a source of base an d oxidant, TetrahedronLett. 40,5417- 5420.

ALPER,H., HAMEL,N. (1990 ), Poly-L -leucine as a nadded chiral ligand for the palladium catalyzedcarbonylation of allylic alcohols, J. Chem. SOC.,

Chem. Comm un. ,135-136.APPERLEY,., NORT H,M., STOK OE, . B. (1995),The

catalytically active conformation of cyclo-[(S)-His-(S)-Phe] as determined by solid state NM R,Tetrahedron:Asymmetry 6,1869-1872.

AUGUS T YN,. A . N., BEZUIDEN HOU DT,. C. B., FER -REIRA,D. (1990), Enantioselective synthesis offlavonoids. Part 1. Poly-oxygenated chalconeepoxides, Tetrahedron46,2651-2660.

BANFI,S., COLONNA,. , MOL INARI, ., JULIA,S.,

GUI XE R ,. (1984), Asy mm etric epoxidation ofelectron-poor olefins. 5. Influence on stereose lec-tivity of th e structure of poly-a-amino acids usedas catalysts, Tetrahedron40,5207-5211.

BAURES,.W., EGGLESTO N,. S. ,FLISAK,. R. , GOM -BATZ,K. et al. (1990), A n efficient asym metric

synthesis of substituted phenyl glycidic esters,Tetrahedron Lett. 31,6501-6504.

BENTLEY,. A . (1999), Th e use of poly-leucine instereoselective synthesis, Thesis, University ofLiverpool.

BENTLEY,. A., BERGERON ,., CAPPI,M. W., HIBBS,D. E. et al. (1997a), Asym metric epoxidation ofenones employing polymeric a-amino acids innon-aqueous media, J . Chem. SOC.,Chem. Com-mun.,739-740.

BENTLEY,. A., KROUTIL,W., LITTLECHILD,. A.,ROBERTS,. M. (1997b), Preparation of polyami-n o acid catalysts for use in Julia asymm etric epox-idation, Chirality 9,198-202.

BENTLEY,. A., CAPPI,M. W., FLOOD , . W., ROBERT S,S. M., SMITH,. A. (1998), Tow ards a mechanisticinsight into the Julia-Colonna asymm etric epoxi-dation of a$-unsaturated ketones using discretelenghts of poly-leucine, Tetrahedron Lett. 39 ,

BEZUIDENHOUDT,. C. B., SWANEPOEL,. , Au c u s -TYN,. A. N., FERRE IRA,. (1987),The first enan-tioselective synthesis of poly-oxygenated a-hy-droxydihydro-chalcones, Tetrahedron Lett. 28,48574860.

BOLM,C., BIENEWALD ,., SEG ER, . (1996), Asym-metric autocatalysis with amplification of chiral-ity,Angew. C hem. (Int. Edn. Engl.) 35,1657-1659.

BRO WN , . F. C., DO NO HU E,. C., JACKSO N, . R.,MCCARTHY,T.. (1994), Syn thetic application s ofoptically active cyanohydrins. Enantioselectivesyntheses of the hydroxyamides Tembamide andAegeline, the cardiac drug Denopamine, andsome analogs of the bronchodilator Salbutamol,Tetrahedron50,13739-13752.

BUDT,K.-H., VATELE,.-M., KISHI,Y. (1986), Termi-nal epoxidation of farnesate attached to helicalpeptides, J.Am . Chem. SOC.108,608CL-6082.

BURGER,M. T., STILL,W. C. (1995), Synthetic iono-phores. Encoded combinatorial libraries of cy-clen-based receptors for Cu2+ an d Co”, J. Org.Chem. 60,7382-7383.

CALLANT, . , COUSSENS,., MATEN,T. V. D., DE

VRIES, . G.,DE VRIES,K. (1992), NM R measure-ments and semi-empirical calculations in a firstapproach to elucidate the mechanism of enan tio-selective cyanohydrin formation catalyzed by cy-clo-(S)-Phe-(S)-His, Tetrahedron: A symmetry 3,401414 .

CAPPI,M. W., CHEN,W. P., FLO OD , . W., LIA O,Y. W.,ROBERTS,. M. e t al. (1998), New procedures forthe Julia-Colonna asymm etric epoxidation: syn-thesis of (+)-C lausen am ide,J. Chem.SOC.,Chem.Commun.,1159-1160.

CHEN,W.-€?,ROBERTS,. M. (1999), JuliCColonnaasym metric epoxida tion of fury1 styryl keto ne as aroute to interme diates to naturally occurring styryllactones,J. Chem. Soc., Perkin Trans. 1,103-105.

9297-9300.



6 References 515

CHIOSIS, . (2000), Study into the cyclization of an

epoxy-alcohol to the energetically disfavored

product by peptides from non-biased combinator-

ial libaries,Tetrahedron Lett. 41,801-806.

COLE,B. M., SHIMIZU,. D., KRUEGER,. A., HAR-

R I ~ ,. P. A,, SNAPPER, . L., HOVEYDA,. H.

(1996), Discovery of chiral catalysts through li-

gand diversity: Ti-catalyzed enantioselective ad-

dition of TMSCN to meso-epoxides, Angew.Chem. (Int.Edn. Engl.) 35,1668-1671.

COLONNA,., MOLINARI,., BANFI, .,JULIA, ., MA-

SANA, J.,ALVAREZ,. (1983), Synthetic enzymes

4. Highly enantioselective epoxidation by means

of polyaminoacids in a triphasic system: influence

of structural variations within the catalyst, Tetru-hedron 39,1635-1641.

COLONNA, s., JULIA,s., MOLINARI, H., BANFI,s.(1984), Catalytic asymmetric epoxidation by

means of polyaminoacids in a triphase system,

Heterocycles 21,548.

COLONNA,. , MANFREDI, ., SPADONI, . (1987),

Asymmetric syntheses catalyzed by natural and

synthetic peptides, in: Organic Synthesis: Mod ernTrends fo r Organic Synthesis (CHIZHOV,.,Ed.),

pp. 275-284. Oxford: Blackwell.

COPELAND,. T., JARVO,. R., MILLER,. J. (1998),

Minimal acylase-like peptides. Conformational

control of absolute stereospecificity, J. Org.Chem. 63,67846785.

DALY,W.H.,POCHE,D. (1988),The preparation of N -carboxyanhydrides of a-amino acids using bis(tri-

chloromethyl)carbonate, Tetrahedron Lett. 29,

DANDA, . (1991), Asymmetric hydrocyanation of

benzaldehyde catalyzed by (SR)-5-(4-imidazolyl-

methyl)-2,4-imidazolidinedione,ull. Chem. SOC.

Jpn. 64,3743-3745.

DANDA, ., NISHIKAWA,H.,OTAKA,. (1991),Enan-

tioselective autoinduction in the asymmetric hy-

drocyanation of 3-phenoxybenzaldehyde cata-

lyzed by cyclo-[(R)-phenylalanyl-(R)-histidyl], J .

Org. Chem. 56,6740-6741.

DE MUYNCK,., MADDER,A.,ARCY, ., DE CLER-

CQ , P. J., PEREZ-PAYAN,. N. et al. (2000), Appli-

cation of combinatorial procedures in the search

for serine-protease-like activity with focus on the

acyl transfer step, Angew. Chem. (Int. Edn.) 39,

5859-5862.

145-148.

DONG,W., PETTY, . (1985), US . Putent 102,4,554.

EBRAHIM,., WILLS,M. (1997), Synthetic applica-

tions of polymeric a-amino acids, Tetrahedron:Asymmetry 8,3163-3173.

EHLER,K. W., ORGEL, . E. (1976), N,N'-carbonyl-

diimidazole induced peptide formation in aque-

ous solution, Biochim. Biophys. Acta 434, 233-

243.

FERRIS,. l?, HILL,A. .,LUI,R., ORGEL, . E. (1996),

Synthesis of long prebiotic oligomers o n mineral

surfaces,Nature 381,5941.

FLISAK,. R., GOMBATZ,. J., HOLMES, . M., JAR-

MAS, A. A., LANTOS,. et al. (1993),A practical,

enantioselective synthesis of SK&F 104353, J .

Org. Chem. 58,6247-6254.

FRANCIS, . B., JACOBSEN,. N. (1999), Discovery of

novel catalysts for alkene epoxidation from met-

al-binding combinatorial libraries,Angew. Chem.(Int. Edn. E ngl.) 38,937-941.

FRANCIS, . B., FINNEY,. S., JACOBSEN,. N. (1996),

Combinatorial approach to the discovery of nov-

el coordination complexes,J.Am . Chem.SOC.118,

GEIER,G.R. I.,SASAKI,T.1997),The design, synthe-

sis and characterization of a porphyrin-peptide

conjugate, Tetrahedron Lett. 38,3821-3824.

GEIER,G. R. I., SASAKI,T. (1999), Catalytic Oxida-

tion of alkene with a surface-bound metallopor-

phyrin-peptide conjugate, Tetrahedron 55, 1859-

1870.

GEIER,G.R. I.,LYBRAND,. P., SASAKI,. (1999), On

the absence of stereoselectivity in the catalytic

oxidation of alkenes with a surface-bound metal-

loporphyrin-peptide conjugate, Tetrahedron 55,

GILBERTSON,. R.,WANG, . (1996), The combinato-

rial synthesis of chiral phosphine ligands, Tetrahe-dron Lett. 37,64754478.

GILBERTSON,. R., WANG, ., HOGE,G. S., KLUG,C.

A., SCHAEFER, . (1996), Synthesis of phosphine-

rhodium complexes attached to a standard peptide

synthesis resin,Orgunometallics 15,4678-4680.

HEANEY, . (1993), Oxidation reactions using mag-

nesium monoperphtalate and urea hydrogen per-

oxide,Aldrichim. Acta 26,3545.

HOGG,D. J. F NORTH,M. (1993), Mechanistic stud-

ies on the asymmetric addition of HCN to alde-

hydes catalyzed bycyclo-[(S)-His-(S)-Phe], etru-hedron 49,1079-1090.

HOGG,D.J. P.,NORTH, ., STOKOE,. B.,TEASDALE,

W. G. (1993), The oxidation of benzaldehyde to

benzoic acid catalyzed by cyclo-[(S)-His-(S)-

Phe], and its implications for the catalytic asym-

metric addition of HCN to aldehydes, Tetruhe-dron:Asymmetry 4,1553-1558.

HOGG,D. J. P., NORTH,M., STOKOE,. B. (1994),A

conformational comparison of cyclo-[(S)-His-(S)-Leu] and cyclo-[(S)-His-(S)-Phe],atalysts

for asymmetric addition, Tetrahedron 50, 7933-

7946.

HULST, ., BROXTERMAN,. B., KAMPHUIS,., FOR-MAGGIO, F., CRISMA, M. et a]. (1997), Catalytic

enantioselective addition of hydrogen cyanide to

benzaldehyde and p-methoxybenzaldehyde using

cyclo-His-(a-Me)Phe as catalyst, Tetrahedron:Asymmetry 8,1987-1999.

ITSUNO,., SAKAKURA,., ITO, K. (1990), Polymer-

supported poly(amino acids) as new asymmetric

8983-8984.

1871-1880.



516 12 SyntheticEnzymes-Artificial Peptides in Stereoselective Synthesis

epoxidation catalyst of a$-unsaturated ketones,J. Org. Chem.55,60474049.

IYER,M. S., GIGSTAD,. M., NAM DEV,. D., LIPTON,M. (1996), Asym metric catalysis of the Stre cker

amino acid synthesis by a cyclic dipeptide,J.

Am.Chem.SOC.118,491W911.JACKSON,. R., JAYA TILAK E,. S., MAITHEWS,B.R.,

WILSHIRE,. (1988), Evaluation of som e cyclic di-peptides as catalysts for the asymmetric hydro-cyanation of aldehydes, Aust. J. Chem. 41, 203-213.

JACKSON , . R., JA COBS, . A., KIM, H. J. (1992),Spectroscopic characterization of the active formof the “ Ino ue” dipeptides (cyclo-[(S)-Phe-(S)-His-]),Aust.J. Chem.45,2073-2076.

JULIA,S., MASANA,.,VEGA, . C. (1980), Syntheticenzymes - highly stereoselective epoxidation of

chalcone in a triphasic toluene-water-poly[(S)-alanine] system, Angew. Chem. (Int. Edn. Engl.)

JULIA, ., GUIXER,., MASANA,., ROCAS,., COLON-NA, S. et al. (1982), Synth etic enzym es. 2. Cataly ticasymmetric epoxidation by means of polyamino-acids in a triphase system, J. Chem.Soc., Perkin

Trans. 1,1317-1324.JULIA,S., MASANA,., ROCAS,., COLONNA,., AN-

NUNZIATA,., MOL INARI,. (1983), Synthetic e n-zymes. 3. Highly stereoselective epoxidation ofchalcones in a triphasic toluene-water-poly[(S)-alanine] system, Anales De Quimica Serie C-

Quimica Organica Y Bioquimica 79,102-104.KIM,H. J., JACKSON,. R . (1992), Polymer a ttach edcyclic dipeptidedes as catalysts for enantiosel-ective cyanohydrin formation, Tetrahedron:

Asymmetry3,1421-1430.KIM,H. J.,JACKSON,. R. (1994), Substituent effects

on the enantioselective hydrocyanation of arylaldehyde, Tetrahedr0n:Asymmetry ,1541-1548.

KOGUT, . F.,THoEN, . C.,LIFTON, M. A. (1998), Ex-amination and enhancem ent of enantioselectiveautoinduction in cyanohydrin formation by cyclo-

[(R)-His-(R)-Phe], J. Org. Chem.63,4604-4610.KOM ORI,T., ONAK A,T.1983), Electroorganic reac-

tions on organic electrodes. 3. Electrochemicalasymmetric oxidation of phenyl cyclohexyl sul-

fide on poly(L-va1ine)-coated platinum elec-trodes, J.Am. Chem.SOC.105,5690-5691.

KOMORI,T., ONAK A,T.1984), Electroorganic reac-tions on organic electrodes. 6. Electrochemicalasymm etric oxidation of unsymm etric sulfides tothe corresponding chiral sulfoxides on poly(ami-no acids)-coated electrodes, J. Am. Chem. SOC.

106,2656-2659.KRICHELDORF,. R . (1987), a-Aminoacid-N-Car-

boxyanhydridesan d Related Heterocycles. Berlin:Springer-Verlag.

KROUTIL,W., LASTERRA-SANCHEZ,. E., MAD-

19,929-931.

DRELL, S. J., MAYON,?,MORGAN, et a]. (1996a),Development of the Julia asymmetric epoxida-tion reaction. Part 2. Application of the oxidationto alkyl enones, enedion es and unsaturated ketoesters, J. Chem.Soc.,Perkin Trans. 1,2837-2844.

KROU TIL, ., MAY ON,?,LASTERRA-SANCHEZ,. E.,MADDRELL,. J., ROBERTS,. M. et al. (1996b),Unexpected asymmetric epoxidation reactionscatalyzed by polyleucine-based systems, J. Chem.

Soc., Chem. Commun.,845-846.KRUEGER, . A., KUNTZ,K. W., DZIE RB A, . D.,

WIRSCHUM,. G., GLE ASO N,. D. et al. (1999),Ti-catalyzed enantioselective a dditi on of cyanide toimines. A practical synthesis of optically pure a-

amin o acids,J.Am. Chem.SOC.121,4284-4285.LASTERRA-SANCHEZ,. E., ROBERTS,. M. (1995),

Enantiocomplementary asymmetric epoxidationof selected en one s using poly-L-leucine and poly-o-leucine, J. Chem. Soc., Perkin Trans. 1, 1467-1468.

LASTERRA-SANCHEZ,. E., ROBERTS,. M. (1997),A n im portant niche in synthetic organic chemis-try for som e biomimetic oxidation reactions cata-lyzed by polyamino acids, Curr. Org. Chem. 1 ,187-196.

LASTERRA-SANCHEZ,. E., FELFER,., MAYON,.,ROBERTS, . M., THORNTON,. R., TODD,C. J.(1996), Development of the Julia asymmetricepoxidation reaction. Part 1.Application of theoxidation to enones other than chalcones, J.

Chem.SOC.,Perkin Trans. 1,343-348.LEE,D. H., GRANJA,. R., MARTINEZ,.A., SEVERIN,

K., GHAD IRI, . R. (19 96),A self-replicating pep -tide, Nature 382,525-528.

MATTHEWS,. R., JACKSON,. R., JAYA TILA KE,. S.,WILSHIRE,., JACOBS, . A . (1988), Asymm etrichydrocyanation of a range of aromatic and ali-phatic aldehydes, Aust. J . Chem.41,1697-1709.

MILLER, . J., COPELAND, . T., PAPAIOANNOU,.,HORSTMANN,. E. , RUEL,E. M. (1998), Kineticresolution of alcohols catalyzed by tripeptidescontaining the N-alkylimidazole substructure, J.

Am. Chem.SOC.120,1629-1630.MORI,A., I K E D A , ~ . ,INOSHITA,. , INO UE,. (1989).

Cyclo-( (S)-leucyl-(S)-histidyl). A catalyst forasymm etric addition of hydrogen cyanide to alde-hydes, Chem. Lett.,2119-2122.

MORI ,A., NITTA, H., KU DO ,M., INOUE,. (1991),Peptide-metal complex as an asymmetric cata -lyst. A catalytic enantioselective cy anohydrin syn -thesis, Tetrahedron Lett. 32,433311336.

NEL,R. J.J.,VANHEERDEN,. S.,VANRENSBURG,.,FERREIRA,. (1998), Enantioselective synthesis offlavonoids. Par t 5. Poly-oxygenated P-hydroxydi-hydrochalcones, Tetrahedron Lett.39,562>5626.

NITTA,H., Yu, D., KUDO,M., MO RI,A., INOUE, .(1992), Peptide-titanium complex as catalyst forasymm etric addition of hydrogen cyanide to alde-



6 References 517

use of sonochemistry in the asym metric epoxida-tion of substituted chalcones with sodium perbo -rate tetrahydrate, Tetrahedron: Asymmetry 9,

SEVERIN, ., BERGS,R.. BECK,W. (1998), Bioorga-nometallic chemistry-transition metal complexeswith a -am ino acids and peptides, Angew. Chem.(Int. Edn . Engl.) 37,1634-1654.

SHIMIZU, . D.,COLE,B. M., KRUEG ER,.A., KUNTZ ,K. W., SNAPPER, . L., HOV EYDA , . H . (1997),Search for chiral catalysts through ligand diver-sity: substrate-specific catalysts a nd ligand screen -ing on solid phase,Angew. Chem . (Int. Edn. Engl.)36,1704-1707.

SHVO,Y., GAL ,M., BECKER, ., ELGAVI, . (1996),Asymmetric hydrocyanation of aldehydes withcyclo-dipeptides: a new mechanistic approach,Tetrahedr0n:Asymm etry 7,911-924.

SIGMAN, . S., JACOBSEN,. N. (1998), Schiff basecatalysts for the asymmetric Strecker reactionidentified an d optimized form of parallel synthet-ic libraries, J. Am . Chem. SOC. 20,4901-4902.

TANA KA, . , MORI,A. , INOUE, . (1990), The cyclicdipeptide cyclo-[(S)-phenylalanyl-(S)-histidyl] asa catalyst for asymmetric addition of hydrogencyanide to a1dehydes.J. Org. Chem. 55,181-185.

VANRENSBURG,., VANHEERDEN,. S., BEZUIDEN-HOUDT, B. c.B., FERREIRA, . (1996), The firstenantioselective synthesis of trans- and cis-dihy-droflavonols, J. Chem. SOC., Chem. Com mun.,

VANTAMELEN,. E. (1968), Bioorganic chemistry:Sterols and acyclic terp ene epoxides, Acc. Chem.Res. 1,111-120.

YADAV, . K., KAPOOR, . K. (1997), 1,8-Diazobicy-clo[5.4.0]undec-7-ene: a rem arkable base in theepoxidation of a#-unsaturated lactones and oth-er en ones with anhydrous t-BuO OH , TetrahedronLett. 51,85733584.

3967-3969.

2747-2748.

hyde, J.A m . Chem. SOC. 14,7969-7975.NOE ,C. R., WEIGA ND, . , PIRKER, . (1996), Studies

on cyclic dipeptides, I: Aryl modifications of cy-clo-[Phe-His], Monatshefte fur Chemie 127,

NOE, C. R ., WEIGAND , , , PIRKER,S., LIEPERT,.(1997), Studies o n cyclic dipeptides, 11. Methylat-ed modifications of cyclo-[Phe-His], Monatsheftefu r Chemie 128,301-316.

N O N A K A ,.,AB E ,S., FUCHIGAMI,. (1983), Electro-organic reactions on organic electrodes. Part 2.Electrochemical asymmetric reduction of citra-conic and m esaconic acids on optically active po-ly(amino acid)-coated electrodes, Bull. Chem.SOC. pn. 56,2778-2783.

NORTH,M. (1992),A co nform ational study on cyclo-[(S)-phenylalanyl-(S)-histidyl] by molecular mo-deling and NMR techniques, Tetrahedron 48,

NORTH,M. (1993), Catalytic asymmetric cyanohy-drin synthesis, Synlett, 80 73 20 .

OK U, .-I. , INO UE ,. (1981),Asymm etric cyanohydrinsynthesis catalyzed by a synthetic cyclic dipep-tide, J. Chem. SOC., hem. C omm un., 229-230.

OK U, .-I., ITO,N., INOU E,. (1979),Asymmetric cya-nohydrin synthesis catalyzed by synthetic dipep-tides, 1,Makrom ol. C hem. 180,1089-1091.

OK U, .-I., ITO,N., INOUE , . (1982),Asymmetric cya-nohydrin synthesis catalyzed by synthetic dipep-tides, 2, M akrom ol. Chem. 18 3,579-586.

P u , L. (1998), Rece nt developments in asymmetriccatalysis using synthetic polymers with mainchain chirality, Tetrahedron: Asymm etry 9 , 1457-1477.

RAY,P. C., ROBERTS,. M . (1999), Overcom ing d ia-stereoselection using poly amino acids as chiralepoxidation catalysts, Tetrahedron Lett. 40,1779-1782.

SAVIZKY,. M ., SUZ UK I, ., BovE, J. L. (1998), The

1081-1097.

5509-5522.



Index

AA A A D see aromatic-L-amino-acid decarboxylaseA A D see acetoacetate decarboxylaseA a D see aspartate a-decarboxylaseabzymes see also catalytic antibodies- industrial fut ure of 446- screening via cell growth 445- semi-synthetic - 420acetal, hydrolysis of, use of catalytic antibodies

acet ate kinase, phosphorylation by 224acetoaceta te decarboxylase (AAD) 58ff- Clostridium acetobutylicurn 5%

- decarboxylation of acetoacetate , mechanism 60acetoacetic acid, decarboxylation of, abiotic mech-

anism 59- - by acetoacetate decarboxylase 60

acetoin 63fa-ac etola ctate , biosynthesis of 64- decarboxylation of, by acetolactate decarbox-

ace tolac tate decarboxylase 62ff- of Bacillus sp. 63facetolactate synthase 91acetone, production by Clostridium acetobutyli-

acetyl pho sph ate, phosphorylation potential 231aconitase 109ffcis-aconitate, decarboxylation of, by aconitate de-

acon itate decarboxylase 64f- decarboxylation of cis-aconitate 65- of Aspergil lus terreus 64facyl cleavage reaction, transition state analogs

acyl transfer reactions, tetrahed ral interm ediates

- use of catalytic antibo dies 474ffacylamide, production of 287f

acylases, production of non-natural amino acids

- L-stereospecificity 356

432ff, 465

ylase 63

cum 61f

carboxylase 65

411

410

285f

acylation, enantioselective -, tripeptide benzamide

- of alcohols, use of tet rap ept ide catalysts 512facyloin condensation 393- pyruvate decarboxylase mediated - 49- yeast catalyzed - 366adenosine triphosphate (AT P), and phosphodies-

ter formation 232- structure of 222adenylate kinase, phosphorylation by 226Alcaligenes bronchisepticus, decarboxylation of a-

alcohol dehydrogenase, horse liver 362f, 372f,

- yeast, ke to ester reduction 359ff- - ketone reduction 359

alcohols, acylation of, use of tetrapeptide cata-

- conversion to aldehydes, by Pichia pastoris 345- flavor-active -, produ ction of 345faldeh yde-ly ases 45, 89faldehydes, asymmetric hydrocyanation of, cata-

- enantioselective hydrocyanation of, use of titan-

- flavor-active -, produ ction of 345f

aldol addition 7ffaldol reaction 388f- use of catalytic antibo dies 472faldolases 6ff- asymmetric carb onx arb on bond formation 365- classific ation of 7- dihydroxy-acetone phosphate-dependent - 7ff- glycine-dependent - 19f- phosphoenolpyruvate-dependent - 15ff- pyruvate-dependent - 15ff- 2-deoxyribose 5-phosphate aldolase (DERA)

algae, marin e -, production of iodo me than e 190- - source of halometabolites 178alicyclic biotran sform ation products 367ffaliphatic biotr ansf orm ation products 353ff

catalyzed - 510ff

aryl-m-methylmalonates 73

379f

lysts 512f

lyzed by cyclo-[(S)-Phe-(S)-His)] 496

ium dipe ptid e complex catalysts 508f

17ff





520 Index

aliphatic este r, hydrolysis o f, use of catalytic anti-

alkaline pho spha tase , phosphorylation by 226alkaloids, chlorinated - 185

alkenes, epoxidation of, use of poly-L-leucine 503- halog enation of 196falkyl mercury -lyase 143alkynes, halogenation of 197amidases, production of non-natu ral amino acids

- synthesis of pharmaceutical precursors, kilo-

amides, hydrolysis of, use of catalytic antibodies

amination, use of catalytic antibodies 476amines, large-scale resolution of 279f- - involving a Pseudomonas lipase 279f

amin o acid production , engineering novel bio-

- L-lysine 321ff- L-phenylalanine 314ff- L-threonine 321ff- microbial pathways 313ffamino acids, chlorinated - 185- natural -, pro duc tion of 285- non-natural -, 285fD-amino acids, prod uctio n of 326S(L-a-aminoadipy1)-L-cystein-D-valineynthetase

~-2-aminobutyrate, ovel biosynthetic pathway

7-aminocephalosporanic acid, biocatalytic route

1-amino-1-deoxy-D-sorbitol,xidation of, by Glu-

- - protection groups 302- - pyridine formation from 6-am ino-6-deoxy -~-

- - time course 304- - with Gluconobacter oxydans 3046-aminopenicillanic acid (6-APA), production of,

using penicillin G acylase 280aminosorbitol, regioselective oxidation of, with

Gluconobacter oxydans 295amm onia lyases 47, 125Amycolatopsis, vanillin prod uctio n by 339fangiotensin I1 356angiotensin synthesis, use of thermolysin 355fantibiotics, chloroamphen icol 178- chlorotetracycline 178- griseofulvin 178antibodies, affinity for transition state analog,

screen ing of 409- function of 404- galactopyranosidase 434-

glycosidase, generation by in vitro immuniza-

bodies 451ff

285f

gram-scale rou tes 285

438f, 461f

chemical pathways 324ff

385

324f

280

conobacter oxydans 300ff

sorbose 302

tion 434

- production stages 408- sequence-specific pep tide cleavage by 420- structure of 404ff- - antibody 48G7 411fantibo dy catalysis, hydrolysis of 421- spontane ous covalent - 421- spontan eous features of 421ff- spontan eous metal ion 422f- transition state stabilization 423f6 -APA see 6-aminopenicillanic acidL-arginine, d ecarboxylation of, by arginine decar-

arginine decarboxylase 80arginin e kinase 224ff- phosphorylation by 224ffaristeromycin 230- phos phory lation of 230farom a chemicals see also natural flavors- natural -, carboxylic acids 344f- - fermentative pro duc tion of 333ff- - flavor-active alcohols 345f- - flavor-active aldeh ydes 345f- - flavor-active lacton es 340ff- - methyl ketones 347- - phenolic com pound s 346f- - vanillin 334ffaromatic L-amino acid decarboxylase (AAAD),

- decarboxylation of tryp toph an L-tryptophan 83arom atic biotransformation products 390ff

arom atic compounds, chlorinated-

18 8- selective hydro xylation of 390Arthrobacter lipase 368faryl esters, hydrolysis of, tetrahedral interme-

- - use of catalytic antibo dies 406faspa rtam e, manufacture of 125- - comm ercial process 2861aspartame synthesis, enzymatic - 355aspartase 126ff- biotechnology of 128f- covalent modification 127- of Bacillus sp. 126ff-of E. coli 126ffL-aspartate, use of 128

a-aspartate decarboxylase, mechanism of decar-

aspartate a-decarboxylase (A aD ) 72ff- structure 72asp artate 4-decarboxylase 74ffAspergillus niger, stereospecific hydroxylation

Aspergillus terreus, aconitate decarboxylase 64fA T P see adenosine triphosphateazasugars, struc ture of 12- synthesis of, use of rabbit muscle aldolase 10f

boxylase 80

deactiv ation of 84

diate 406f, 455ff

boxylation 75

373f



Index 521

B

Bacillus sp., acetolactate decarboxylase 63- aspartase 126ffbait and switch strategy, generation of catalytic

bake rs’ yeast 31f- acyloin cond ensatio n 23ff- Diels-Alder reaction 30- esterase of 370- Micha el additio n 31f- pyruvate decarboxylase of 23ff- - synthesis of sole rone 27- synthesis of lano stero l analogs 382- transaldolase 20- transketolase 20- 2,3-oxidosqualene lanosterol cyclase 29fbasidiomycetes, production of halometabolites

Beauveria bassiana, synthesis of (R)-2(4-hydroxy-

benzaldehyde, stereoselective hydrocyanation of,

benzaldehyde derivatives, stereoselective hydro-

benzofurans, chlorinated - 187benzoylformate decarboxylase (B FD ) 65ff- of Pseudomonas putida 6%benzylformate decarboxylase (BFD), decarboxyla-

antibodies 412ff

176

phen0xy)proprionic acid 282

use of cyclic dipep tides 493ff

cyanation of 493

tion of [p-(bromomethyl)benzoyl]formate,mechanism 67

Bertrand-Hudson rule 299ffbetapol, production of 289B F D see benzoylformate decarboxylasebiocatalysis, ther mo dyn am ic cycle 424biocatalysts, industr ial use of 278ffbiohalogenation, com merical application of 201f- enzym ology of 192ff- mechanistic aspects of 184ffbiotin, enzymic synthesis of 384fbiotransformation products, alicyclic - 367ff- aliphatic - 353ff- aromatic - 390ff- heterocyclic - 384ff- polycyclic - 376ff- - use of alcoh ol dehy droge nases 382fp-bloc kers, synth esis of 385Brevibacterium flavum, immobilized whole cells,

fum aras e activity in 107ffbrewers’ yeast, pyruvate decarboxylase of 25bromine 184- content of the earth 178bromop eroxidases, isolated enzymes 194- producing cells 194- sequences from Streptomyces venezuelae 182- turnover number 179- X-ray crystal struc ture of 195brompoperoxidases, of Streptomyces aureofaciens

19.5

- X-ray crystal structure of 195butanol, production by Clostridium acetobutyli-

cum 61f

C

C4-dicarboxy lic acids, ma nufac ture of 108C A see carbonic anhydraseCaldariomyces fi m ag o, haloperoxidase of 197- hem e chloroperoxidase of 193caldariomycin, chlorination by a chloroperoxid-

carbamatase, antibody 444f- med ical poten tial 444fcarbamate ester, hydrolysis of, use of catalytic an-

carb ape nem antibiotic 350f- synth esis of 353fcarbazoles, chlorinated - 187carbocation cyclizations, use of catalytic antibo d-

carb ohy drat e analogs, synthesis of 17fcarbohy drate hydro-lyases 115ffcarb ohy drate lyases 115ffcarbohydrate synthesis, use of rabbit muscle

carbo hydra tes, kinetic glycosylation 246f- thermodynam ic glycosylation 246fcarbon-carbon bond form ation reactions Sff,

carb on-c arbo n lyases 44f, 47ffcarbon dioxide, hydration of, by carbonic anhy-

carbon-halide lyases 142fcarbo n-nitrog en lyases 47, 125ffcarbon-oxygen lyases 46,98ffcarbona tes, hydrolysis of, tetrahedral interme -

- - use of catalytic antibodies 406f, 459fcarbonic anhydrase (CA) 99f- hydration of carbon dioxide, mechanism 100carboxykinases 84f- listing of 85carboxylesterase NP, synthesis of (S)-naproxen

carboxylic acids, fermentative production of 344f- hydroxylation of, by Pseudomonas putida 363f- used in flavor compositions, sensoric description

carboxyl-lyases 44,47ffL-carnitine, biocatalytic rou te to 288fcatalysis, fre e energ y of activatio n 406catalytic antibo dies 403ff- - see also abzymes- acyl-transfer reaction s 412f, 474ff- - hap ten design 412f- aldol reactio ns 472f- amination reactions 476

ases 181

tibodies 461

ies 43Of

aldolase 1Of

365ff, 382,38 8,39 3

drase 100

diates 406f

281

of 344



522 Index

- p-elimination reactions, hap ten design 412f- bimolecular associative and substitution proc-

- carbamatase 444f- carbo cation cyclizations 430f- cation ic cyclization 471f- - of polyen es 430f- changing regio- and stereochemistry of reac-

- cleavage of, acetals 432ff- - carbam ate prodrug s 444f- - glycosides 432ff- - phosph ate esters 435- conjugate addition 477- conversion chorismic acid into prephenic acid

- cycloreversions 467- decarboxylations 466f- detox ification by 441f- “Diels-Alderase” 417ff- Diels-Alder cycloadditions 418, 426ff, 473- - regioselectivity 426f- - stereoselectivity 426- disfavored regio- and stereoselectivity 427ff- eliminations 465f- epoxide opening 471- esterolytic -, hap ten design 414- formation of isomeric decalins 431- gen eratio n of, biochemistry 409f- - chemistry 407ff- -

hapten fo r phosp hodiester cleavage 436f- - immunonology 409- - metho ds for 407ff- - molecular biology 410- - suicide sub strate 434f- hap ten design 410ff- - bait an d switch 412ff- - con trol of translational entropy 417ff- - entropic trapping 415ff- hydrolysis of, ace tal 465- - aliphatic ester s 451ff- - amides 438f, 461f- - aryl est ers 406f, 455ff- -

carbamate esters 461- - carb ona tes 406f, 459f- - cocaine 441f- - enol ethe r 465- - epoxide 465- - ether 464- - glycoside 464- - ketal 46.5- - phosp hate esters 463f- - phosphotriesters 437- - trifluoramphenicol 439f- hydro lytic an d dissociative processes 451ff- induction of catalysis, by reactive immuniza-

- intram olecu lar processes 468ff

esses 472ff

tions 426ff

416

tion 439ff

- isomerizations 468f- Ke mp decarboxylation 419f- ketaliza tion in wate r 46.5- med ical applica tions 441ff- nucleophilic substitu tion 477- oxidations 477f- peptide bond form ation 418f- peroxidase-like - 421- porphyrin metalation 477- pro drug activation 442ff- rearrangements 469f- redo x reactions 477ff- reductions 478f- resolutio n of diastere oisom ers 432f- retro-aldol reactions 468- retro-He nrv reaction 468- ring closure reaction, reversal of kinetic con -

trol 427ff- side chains, interactions with transition state

- sub strate desolvation 419f- supplementation of chemical functionality

- syn-elimination, of p-fluoroketones 429- - of selenoxide 419ffcationic cyclization, use of catalytic antibodies

CDP-ascarylose 118ff- biosynthesis of, by Yersinia pseudotuberculosis

C-glycosides, synthesis of, by ra bbit muscle

chalcone, epoxidation of, use of poly-L-alanine

analog 417

42Off

471f

118

aldolase 13

499fchitooligosaccharides, synthesis by transglycosyla-

tion 251chloroamphenicol 178ffchlorinated aromatics, from basidiomycetes 176chlorine 184- content of the earth 178chloromethane, anthropogenic inputs 184- output of volcanoes 184chloroperoxidases, cloning from human eosino-

- halide-dependent reactions 180- isolated enzymes 194- producing cells 194- sequenced from microbes 182- turnover number 179chlorotetracycline 178fchorismate, common aroma tic pathway 316chorismate mu tase 415f- dereg ulation of 317ff- transition state 416- transition state analog 416chorismic acid, conversion into prephenic acid, ki-

- - thermodynamic param eters 416

phils 201

netic param eters 416



Index 523

(+)- t rans chrysanthemic acid, synthesis of, usingan esterase from Arthrobacter globiformis 283

chymotrypsin 356a-chymotrypsin, dihydrocinnamate ester hydroly-

- stereoselectivity of 354cinnamaldehydes, biotransformation products of

citrate dehyd ratase 109citric acid cycle, lyases in 48Citrobacter freundii, tyrosin e-phe nol lyase 95ffclausenamide, preparation of, using poly-leucine

catalyzed epoxidation 501fClostridium acetobutylicum, acetoaceta te decar-

boxylase 59ff- production of, acetone 61f- - butanol 61fClostridium beijerincki, isopropanol production

Clostridium tetanomorphum, Pmethylaspar tase

clozylacon, synthesis of 108cocaine, hydrolysis of, use of catalytic antibodies

cocoa butter equivalents, lipase catalyzed synthe-

compactin, synthesis by yeast alcohol dehydrogen-

conjugate addition, use of catalytic antibodies

corynebacteria, biochemical pathway engineering

- biosynthetic pathway of, L-hornoserine 323- - L-isoleucine 323- - L-lysine 323- - L-methionine 323- - t- threonine 323- overprod uction of amin o acids 322Corynebacterium glutamicum, production of, glu-

tamate 321ff- - lysine 321ff- - threonine 321ffcoryneform bacteria, production of phenylalanine

crea tine kinase 224ff- phos phory lation by 224ffCT P see cytidine triphosph atecum enes, hydroxyla tion of 39Of, 392curcumin, as sub strate for vanillin production 335- hydrolysis of 335cyanohydrin form ation, enzymatic - 366cyclic dipeptides, catalysis of Strecker reaction

- enantiodiscrimination mechan ism 497f- hydro cyan ation by 492ff- - amino1 mechanism 498f- -

autoinduction 497- - imida zolekyan ide salt mechanism 498

sis 376f

26f

62

129

441f

sis 289

ase 360

477

in 321

314

496f

- - mechanism 497ff- - optimal substrate 492- - production of natural products and drugs

- hydroxyanation by, optimal conditions 492- stereoselective hydrocyanation by, of benzalde-

- - of benza ldehy de deriva tives 493ff- use in stereos elective synthesis 492ff- used for enantioselective hydrocyanation, inter-

cyclitols, synthesis of 13P-cyclodextrin 142cyclodextrin glycosyltransferases, bacterial - 267fcyclopropanes, halogenation of 197cycloreversions, use of catalytic antibodies 467cytidine triphosphate (CTP), and phosphodiester

496

hyde 493ff

molecular hydrogen bonding 497

formation 232

DD A G D see dialkylglycine d ecarboxylaseD A H P synthase 315- activity, deregulation of 317D D T , dehydrochlorination of 206- dehydrohalogenation of 205fy-decalactone, form ation of 340ff- - Po xid ati on pathway of ricinoleic acid 342- - by Yarrow ia lipolytica 341f- sensoric description of 341decalins, formation of isomeric -, use of catalytic

decarboxylation, of a-ac etola ctate 63- of acetoa cetic acid 59f- of cis-aconitate 65- of L-arginine 80- of dimethylglycine 87- of fluoro pyruv ate 54, 56- of L-histidine 81- of L-lysine 80- of ma lonic acid 69- of malonyl-CoA 69- of a-methylornithine 79- of L-orn ithine 79- of orotidine-5 '-pho sph ate 82- of oxalo acetate 56- of phen ylpyru vate 86- of L-tyrosine 82- pyruvate decarboxylase mediate d - 49- - catalytic sequenc e 52- use of catalytic antibo dies 466fdehalogenases 181ff- classif ication of 182ff- commerical application of 211f- dehalogenation via epoxide formation 183- dehydrohalogenation 183- glutathione-dependent- 205- halidohydrolase-type 204f

antibodies 431



524 Index

- haloalcohol 205- hydrolytic dehalogenation 183ff- - classif ication of 183f- oxygenolytic deh alog ena tion 182f- reductive dehalog enation 182dehalogenations 202ff- aliphatic - 203- catalyzed by specific enzymes 181- dioxygenase-catalyzed - 208ff- hydrolytic - 183, 210, 204f- monooxygenase-catalyzed - 210- of chlorinated arom atic compounds 207- oxidative - 203f, 208- oxygenolytic - 18 3- reductive - 182, 203,20 7- via epoxide formation 183dehydratases 98fdehydroalanine residues in peptides, formation

dehydrochlorination, of DDT 2 0 6- of lindane 206dehydrohalogenation l83,205f, 211

- of lindane 205f3-dehydroquinate dehydratase (DHQD) lllf1-deoxynojirimycin 296ff- a-glucosidase inhibitor 297- industrial synthesis of 295- prod uctio n of, by chemical synthesis 299- - by extraction from plants 298- -

by ferme ntation 298f- - combined biotechnological-chemicalsynthe-

- sources of 297f- structure of 296ff2-deoxyribose-5-phosphate aldolase 17ff- acetaldehyde condensations 18- from Escherichia coli 17- synthesis of carbohy drate analogs 17f3-deoxy-~-arabino-heptulosonate-phosphate

deoxysugar synthesis 23D H A D see a$-dihydroxy acid dehydratase

D H A P see dihydroxyacetone phosphateD H Q D see 3-dehydroquinate dehydratasediab etes mellitus, treatm ent of, by Miglitol 298dialkylglycine decarboxylase (D A G D ) 86f- decarboxylation catalyzed by 88- transam ination catalyzed by 88diaminopimelate decarboxylase, decarboxylation

dichlorophenol, deg rada tion of 2073,6-dideoxyhexoses, naturally occurring - 1203,6-dideoxysugars 117ffDiels-Alder cycloadditions, co ntr ol of translatio n-

-of tetrachlorothio phene and N -ethylphthalim-

of 134

- of DDT 205f

sis 299ff

synthase see D A H P synthase

of L-arginine 80

al entropy 417

ide 418

- us e of catalytic antibo dies 473fDiels-Alder reaction, ena ntio- and diastereoselec-

- - for ortho-app roach 426- in biosy nthes is 3Off- intramolecular - 30f- use of catalytic antibod ies 426ff“Diels-Alderase”, antibody 417dihydroflavonols, preparation of, using poly-leu-

dihydrofolate reductase 387dihydroxyacetone phosphate ( DH AP ), structure

- synthesis of 8ffdihydroxyacetone phosph ate analogs, structure of

a$-dihydroxy acid dehyd ratase (D H A D ) l l l f- of Salmonalla typhimurium 112diltiazem, com mercial synthesis of, lipase cata-

- preparation of, using poly-leucine catalyzed

dimethylglycine, oxidative decarboxylation of, by

dioxins, chlorinated - 188

- deg rada tion of 207dioxygenase 208ffdipeptide ligands, enantioselective hydrocyanation

disacc haride s, synthesis of, by glycosylation 253f-

by imm obilized glycosyltransferases 258y-dodecalactone, forma tion of, from oleic acid

L- D OP A , synthesis of 95- - hybrid pathway 98L-D OP A decarboxylase see aromatic L-amino acid

dopamine, multi-enzyme synthesis of 98drugs, preparation of, using poly-leucine catalyzed

- production of, using cyclic dipepti de catalyzed

- synthesis of, enzymic hydroxylation in 392

tivity of 426

cine catalyzed epoxidation 501f

of 8

8

lyzed- 279

epoxidation 501f

dialkylglycine decarboxylase 87

of aldehydes 508

343

decarboxylase (AAAD)

epoxidation 501f

hydrocyanation 496

EE. coli, asparta se 126ff- fumarases l O l f f

eliminations, use of catalytic antibo dies 465fenantiomerically pure pharmaceutical interm e-

diates, synthesis of 278ff- - amines 279f- - chiral C 3 building blocks 283- - diltiazem 279- - naproxen 281f- - pyrethroid precursors 282f- -

6-aminop enicillanic acid 280- - 7-aminoceph aIosporanic acid 280f



Index 525

endoglycosidases 255engineering of microbial pathways, for amino acid

production 303ffenol ethe r, hydrolysis of, use of catalytic antibod-

ies 465enolase 112fentropic trapping, rotational entropy 415ff- translational entro py 417ffenzym e catalysis, conce pt of 404ff- entropic trapping, rotational entropy 415ff- transition state stabilization 423f- enzyme-catalyzed reactions, synthetic applica-

enzyme reactions, multiple - 366enzyme specificity, exploiting comb inations 382enzymes see also synthetic enzymes- and carbon-carbon bond formation 5

-in carb ohy drate chemistry 243ffephedrine, industrial synthesis of 24

- production using the Knoll process 53f- synth esis of 393epoxidation, of “chalcone-type” substrates, using

- of digeminal and trisubstituted alkenes, use of

- poly-leucine catalyzed - 501f- - preparation of natural products and pro-

drugs 501f- regio and enantioselective, of farnesamide-N-ol-

igopeptides 507- stereoselective, biphasic - 499f- - triphasic - 499fepoxides, formatio n of, by degrad ation of vicinal

haloalcohols 206f- hydrolysis of, use of catalytic antibodies 465- opening of, use of catalytic antibodies 471Erwinia herbicola, tyrosine -phen ol lyase 95fferythromycin, enzymatic synthesis of 358Escherichia coli, deoxyribose-5-phosphate

- phenylalanine hyper-secreting phe noty pe 321esterases, microbial - 357f- synthesis of (+ ) - t ran s chrysanthemic acid 283ether glycoside, hydrolysis of, use of catalytic anti-

eugenol, degrad ation of, by Pseudornonas sp .

- for ferulic acid production 337exoglycosidases 255

tions of 351ff

poly-L-leucine 502

poly-L-leucine 503

aldolase 17

bodies 464

337f

F

farnesam ide N-oligopeptides, regio- and enantio-

farnesyl pyrophosphate synthetase, from pig liver

fatty acids, chlorinated - 186

selective epox idation of 507

365

FDP aldolase see fructose-1,6-diphosphate

fermentative production, of natural aroma chemi-

ferulic acid, as substrate for vanillin production

- fed-batch formation 338f- produ ction of 336ff- - as inter me diate for vanillin production 336ff- - by Pseudomonas sp. 337- - from eugenol 337flavor-active alcohols, production of 345flavor-active aldehydes, production of 345fflavor-active lactones, fermentative production of

- sensoric description 341fluorine 184fluorine metabolism 199ff- enzymes of 199fffluoroace tate, formation by Streptomyces cattleya

- plant an d microbial metabolite 190fluoroace tone, production by plant hom ogenates

fluorocitrate, biosynthesis of, catalyzed by aconit-

- citrate synthase 199- forma tion by 199pfluoroketones, syn-elimination of, use of catalyt-

ic antibodies 429fluoropyruvate, decarboxylation of, by pyruvate

decarboxylase 54, 56fluo roth reon ine, biosynthesis of 199ffolate an d tetrahy drofolate, interconversion of

N-formyl-1-amino-1-deoxy-D-sorbitol,iotransfor-

- - with Gluconobacter oxydans cells 303ffructose-l,6-diphosphate FDP) aldolase 7ff- - see also rabbit muscle aldolase- aldol condensation 14- resolution of racemic aldehydes 9fucosyltransferases 259fuculose 1-phosphate aldolase, overexpression of

fumarases lOOff- biotechnology of 106ff- class I lOlff- - structure 101- class I1 103ff- - of Therrnus therrnophilus 109- enzymology 104- hydration of acetylene dicarboxylate 102- in imm obilized Breviba cteriurn fravurn cells

- non -natu ral substrates 105f- of E. coli lOlff- role in anaerob ic metabolism 101

aldolase

cals 333ff

338ff

340ff

199f

190

ase 110

387

mation of, at industrial scale 303f

14

107ff



526 Index

- stereochemistry 104f- structures of 104- sub strate specificity 102furans, chlorinated - 187furanyl acrolein, yeast biotransformation of 387

G

GAD see glutam ate decarboxylaseD-galaCtOnate deh ydra tase 119galactopyranosidase 434galactosyltransferases 25%- from bovine colostrum 265fgangliosides 262fGluconobacter oxydans 305- biotransformation of N-formyl-1-amino-1 deoxy-

-cells, fermentative production at industrial

- continuous ferm enta tion of 306- in the synthesis of vitamin C 297- oxidation of 1-amino-1-deoxy-D-sorbitol00ff- - protection groups 302- - pyridine formation from 6 -amino-6-deoxy-~-

- regioselective oxida tion of aminoso rbitol 295ff- D-sorbitol dehydrogenases 306glucose, exog onic oxida tion of 222glucose-6-phosphate, enzymatic synthesis of 228fglucosidase, an d glycosyltransferase, consecutive

glycosylation reactions 268a-glucosidase inhibitors 297f- deoxynojirimycin 297ff- Miglitol 298ff- nojirimycin 297ffglucurate dehydra tase 119glutam ate decarboxylase (GAD) 76ff- mechanism of decarboxylation 77gluta rate diesters, hydrolysis of, use of pig liver

N-glycans, biosynthesis of 263f0-g lyca ns, biosynthesis of 263glyceraldehyde, specific phosphorylation of 230

glycerol, specific phos phory lation of 230glycerol kinase, phosphorylation by 224glycidol butyrate, resolution of, using porcine pan-

glycoconjugates, synthesis of 251ff- - by glycosyltransferases 261ffglyco lipids, synth esis of 262f- - solid-phase me thods 264glycopeptides, enzymatic synthesis, using transgly-

cosylation 252- glycosylation sites, N-linkage 263- - 0-linkage 263- synthesis of 263f

glyco protein s, synthesis of 263fglycosidase antibody 434

D-sorbitol, a t indus trial scale 303f

scale 304ff

sorbose 302

esterase 353f

creatic lipase 283

glycosidases 245ff- availability of 246- imm obilizatio n of 268f- specificity for glycosyl do no r 246-

synthesis of small oligosaccharides 249- use in consecutive glycosylation reactions 268glycosides, cleavage of, use of catalytic antibodies

- synthesis of, multiglycosylated sub strate s 255glycosidic linkages, form ation of 243ff- - use of organ ic solven ts 254glycosphingolipids, biosy nthetic pathw ays 262fglycosyl donors, activated -, commonly used -

- anomeric configuration 247- availability to glycosyltransferases 256fglycosylation see also transglycosylation-

by reverse hydrolysis 253ff- - aqu eou s systems 253f- - use of organ ic solven ts 254- diastereoselective - 248f- disacc haride form ation, regioselectivity of 250- glycosides as acceptors 249- of cell surface s 264- of non -natu ral disaccharides 266- who le-cell systems 266fglycosylation reaction s, isolation of prod ucts 269fglycosylic bon d, cleavage of, use of catalytic anti-

glycosyltransferase activity, assay of 269

glycosyltransferases 256ff- acceptor-donor analogs 264ff- availability of, enzy me s 256- - glycosyl don ors 256f- expression in insect cells 268- expression in yeast 268- heterologous expression of 266f- immobilization of 258, 268f- in glycoconjugate synthesis 261ff- in oligos accha ride synthesis 258ff- use in consecutive glycosylation reaction s 268- using non-nucleotide dono r subs trates 267f(+ -goniofufurone, preparation of, using poly-leu-

(+ )-goniopypyrone, preparation of, using poly-

(+ )-goniotriol, preparation of, using poly-leucine

griseofulvin 178fgriseoviridin, enzy ma tic synth esis of 361guaiacol, formation fro m vanillic acid 346

432ff

247

bodies 433f

cine catalyzed epox idation 501f

leucine catalyzed prep aratio n 501f

catalyzed prep arati on 501f

H

halide methyltransferase 198f- biosynthesis of chlo rom etha ne 198f

halides, physicochemical pro perti es 189halidohydrolases, classification of 205



Index 527

H D C see histidine decarboxylasehem e haloperoxidases 193- mechanism 195heterocycles, chlorinated - 187heterocyclic biotransformation products 384ffhexane squalene 506

hexokinase, phosphorylation by 226histidase 132ff- catalytic activity 134- deamination of labeled L-histidine, mechanism

- deamination of L-histidine, mechan ism 138- inhibitio n of 135- mutan ts 135, 139L-histidine, decarboxylation o f, by histidine decar-

histidine decarboxylase (H D C) 81fhog kidney acylase (H K A ) 356f- stereocontrolled synthesis of a penic illin 356fL-homoserine, aspartate derived amino acids, in

corynebacteria 323horse liver alcohol dehydrogenase, reduction of

fou r- to nine-mem bered cyclic keto nes 379fhorse liver alcohol dehydrogenase (HLADH)

362f- oxida tion reaction s 362f, 372fH S D H see hydroxysteroid dehydrogenaseshydantoinase, productio n of non-natural amino

hydrocyanation 493ff- asymmetric -, of aldehydes 496- stereoselective -, of benz aldehy de 493ff- - of benz aldehy de deriva tives 493ff- use of cyclic dipeptides, amino1 mechanism

- - autoinduction 497- - imidaz olekyan ide salt mechanism 498- - mechanism 497ffhydrogenation, stereoselective - 507hydro lyases 46, 98fhydroxylation, enzymic -, in drug synthesis 390,

hydroxym andelonitrile lyase 90hydroxynitrilase (NHL), synthesis of pyrethroid

(R)-2-(4-hydroxyphenoxy)propionic acid, synthesis

hydroxysteroid dehydrogenases (HSD H), regio-

3-hydroxy-y-decalactone,ormation of 341ff

137

boxylase 81

acids 285f

498f

392

precursors 282

of, use of Beauveria bassiana 282

specific reduc tion of ke ton es 378f

haloalcohol dehalogenases, formation of epox-

haloaliphatics, priority pollutants 203halocompound transforming enzymes, synthetic

importance 179ffhalocompounds 173ff- anthropo genic sources 175f- biogenic origins 184ff- biosynthesis 173ff- biotransformation 173ff- distribution 175- function 175ff- of basidio my cetes 176- of marin e algae 178halogenation, enhancement of bioactive efficacy

haloorganics, recalcitrance of 177haloperoxidases 181ff- assay method s 180- bacterial -, catalytic mechanism 182- basic mechanism 192- classification of 19 3- enantioselectivity of 180f- hem e, mechanism of 193, 195- imm obilization of 202- isolated enzyme s 194- natu ral sources of 192f- non-heme nonm eta l -, mechanism 195- of Caldariomyces fumago 197- pH opt ima 180- producing cells 194- reaction s of 196ff- regiospecific biotranform ations 198- stereoselective biotransformations 198- temperature optima 180- vanadium 193ffHam mond postulate 406haptens, boronic acid, generation of catalytic anti-

- charged, generation of catalytic antibodies

- design -, strategies 410ff- dialkylphosphinic acid, generation of catalytic

antibodies 438- for generating catalytic antibodies, for phos-

pho dies ter cleavage 436f- generation of catalytic antibodies, bait and

switch strategy 412ff- - carrier prote ins for 409- - design of 410ff- - Diels-Alder cycloaddition 417f- - entro pic trapping 415ff- - tetra hed ryl intermediates 410f- - transition state analogs 410ff- heterologous immunization strategy 414f- reactive imm unization 439ff- zwitterionic -, generation of catalytic antibod-

ides 205

179

bodies 438f

412ff

ies 414f

I

I DP C see indolepyruvate decarboxylaseIgG imm unoglobulin, struc ture of 404fimidazoleglycerol pho spha te dehydratase 114imm obilizatio n, of glycosidases 268f- of glyco syltrans ferases 258, 268f



528 Index

- of halope roxida ses 202immunization, heterologous - 414ffindolepyruvate decarboxylase (ID PC ) 87ffindoles, chlorinated - 186- halog enation of 196insect cells, expression of glycosyltransferases 268intramolecular processes, use of catalytic antibod-

ies 468ffiodine 190iodom ethane, productio n by marine algae 190iodoperoxidases, isolated enzym es 194- producing cells 194isocitrate lyase 90isoeugenol, as substrate for vanilline production

L-isoleucine, aspartate derived amino acids, in co-

isomerases, carbon+arbon bon d formation 6isom erization s, use of catalytic antibo dies 468fisopropanol, produ ction by Clostridium beijer-

335

rynebacteria 323

incki 62

K

Ke mp decarboxylation, of 6-nitro-3-carboxybenzi-

ketal, hydrolysis of, use of catalytic antibodies

ketalization, use of catalytic antibodies 465ketoses, synthesis from amino acid synthons 10

Klebsiella pneumoniae, malon ate decarboxylation

- malonyl-CoA decarboxylase 69ffKnoll process, for the manufacture of L-phenyl

soxazole 419

465

mechanism 71

carbinol 53f

L/?-lactam, enzym atic synthesis of 360flactones, flavor active -, production of 340fflanosterol, struc ture of 28lano stero l cyclase 382Lelo ir glycosyltransferases 267

leuk otrie ne analogs 380fligases, carbon-arbon bon d form ation 6lindane, dehydrochlorination of 206- dehydrohalogenation of 205flinear dipeptides, use in stereoselective synthesis

lipases 368ff, 385f- commercial synthesis of diltiazem 279- of Mucor sp. 385f- of Pseudomonas sp. 385- synthesis of, cocoa butter equivalents 289- - pharmaceutical precursors, at kilogram-scale

- -struc tured triglycerides 289lipid mod ification 289

508ff

routes 284

lipids, chlorinated - 186- production of structured - 289lipoxygenases 363fflyases 41ff- acting on poly-D-galacturon ates 123- acting on polysaccharides 122ff- classes 44ff- definition 43- in th e citric acid cycle 48- nomenclature 43- systematic names 6L-lysine, aspartate derived amino acids, in coryne-

- bacter ial produ ction of 321ff- - pathway engineering 321ff- decarboxylation of, by lysine decarboxylase 80lysine decarb oxyla se 79f

bacteria 323

M

malate, routes to 105male ate hydratase, of Pseudomonas pseudoali-

Malomonas rubra, malo nate decarboxylation

- malonyl-CoA decarboxylase 69ffmalo nate decarboxylation, by Klebsiella pneum on-

- by Malonomonas rubra, reaction mechanisms

-synthesis of chiral

-106malonic acid, decarboxylation of, abiotic mecha-

malonyl-CoA, decarboxylation of, by malonyl-

malonyl-CoA decarboxylase, of Klebsiella pneu-

- of Malomonas rubra 69ffmalonyl-CoA decarboxylase (MC D) 68ffmande lonitrile lyase 90mannosyltransferase 260fL-methionine, aspartate derived amino acids, in

methoxycarbonyl ph osphate, phosphorylation po-

methyl ketones, formation by Penicillium roque-

/?-methylaspartase 129ff- addition of amines to substituted fumarates

- addition of ammonia to substituted fumarates

- bearing the Ser-Gly-Asp sequenc e 130- inhibitors of 130- mechanism of action 133- of Clostridium tetanomorphurn 129

a-me thylornithine , decarboxylation of, by orni-

genes 114

mechanism 71

iae, reaction mechanism 71

71

nism 69

Co A decarboxylase 69

moniae 69ff

corynebacteria 323

tential 231

forti 34 7

132

131

thine decarboxylase 79



Index 529

mevinolin, synthesis by yeast alcohol dehydrogen-

Michael addition, by bakers’ yeast 31fmicrobial enzymes, oxidation reactions 373fmicrobial ester ases 357fmicrobial pathways, for amino acid production

microorganism oxidation 380f- prep aratio n of leuk otrie ne analogs 380fmicroorganism reduction, of polycyclic com -

Miglitol 296ff- a-glucosidase inhibitor 298- structure of 296monensin, synthesis of 358fmonooxygenase, synthesis of (R)-2(4-hydroxy-

monosaccharide carboxylic acid dehydratases

- listing of 116f- of Pseudomonas putida 115

mon othiopho sphate, stereospecific phosphoryla-

Mucor sp. lipases, synthesis of heterocyclic com-

mutagenesis, development of phenyl alanine over-

ase 360

313ff

pounds 378ff

phenoxy)proprionic acid 282

115ff

tion of 229f

pounds 385f

producing organisms 315ff

N

(S)-napro xen, synthesis of 281f- - with carboxylesterase N P 281natural arom a chemicals see arom a chemicalsnatural flavors see also arom a chemicals- definitio n of 334natural products, preparation of, using poly-leu-

cine catalyzed epoxidation 501f- production of, using cyclic dipeptide catalyzed

hydrocyanation 496negamycin synthesis 353f- use of pig liver este rase 353fNeuA c aldolase see sialic acid aldolaseN H L see hydroxynitrilase

nicotinamide, produc tion of 287fnitrilases 287fnitrile hydratases, prod uctio n of, acrylamide 287f- - nicotinamide 287fnitriles, pathw ays of hydrolysis of 287nitrogen mustards, cytotoxicity 443- prod rug activation 443fnojirimycin 296ff- a-glucosidase inhibitor 297- struc ture of 296N T P see nucleoside triphosphatesnucleic acid derivatives, hydrolysis of 235fnucleic acids, chlorinated - 187nucleocidin 190

nucleophilic substitution, use of catalytic antibod-

nucleoside kinase, phosphorylation by 226nucleoside triphosphates (NT P), as phosph ate

nucleotide phosph ate metabolism 222

ies 477

source 228ff

0

O A D see oxaloacetate decarboxylaseO C D see oxalyl-CoA decarboxylaseO D C see ornithine decarboxylaseoligopeptides, farnesamide N-oligopeptides, regio-

and enantioselective epoxidation of 507- rhodium polypeptide complex, stereoselective

hydrogenation by 507foligosaccharides, comp lex, assembly of 250f- in vitro enzymatic synthesis 245oligosaccharide synthesis, use of glycosidases 249- use of glycosyltransferases 258ffO P D see orotidine-5 ‘-ph osphate decarboxylaseorganic solvents, use in form ation of glycosidic

organofluorine comp ounds 184- naturally occurring - 189organohalogens, na tural - 184ff- - listing of 185fforganoiodine compounds, isolated from marine

organisms 19OfL-ornithine, decarboxylation of, by ornithine de-

carboxylase 79ornithine decarboxylase (ODC) 78f- of Trypanosoma brucei 78orotidine 5 ‘-phosphate, decarboxylation of, by

protidine 5 ‘-phosp hate decarboxylase 82orotidine-5 ’-phosph ate decarboxylase (OP D ) 82overexpression, of fructose-1-phosphate aldolase

- of rhamnose-1-ph osphate aldolase 14- of tagatose-1,6-diphosphate ldolase 14overp roduc tion of amino acids 315ffoxalate decarboxylase 55foxaloaceate, decarboxylation of, by oxaloacetate

oxaloacetate decarboxylase (O A D ) 55ff- decarboxylation of oxaloacetate 57f- - mechanism 57- of Pseudomonas putida 57fOxalobacter formigenes, oxalyl-CoA decarboxyl-

ase 67foxalyl-CoA decarboxylase (O CD ), of Oxalobacter

formigenes 67foxida tion reaction s 386ff- enzymatic- 386- - synthesis of heterocyclic com poun ds 386ff- use of catalytic antibodies 477foxidative deg rada tion 391f- enzymic -, of aromatic compounds 391f

linkages 254

14

decarboxylase 56ff



2,3-oxidosqualene cyclase 28ff- from pig liver 28ff0x0-acid-lyases 45, 90foxynitrilases 282

PP A L see phenylalanine ammonia-lyaseParkinson’s disease, synthesis of L-DOPA 95PCB see polychlorinated biphenylsP D C see pyruvate decarboxylasepec tate lyases 124fpectic substances, cleavage of 125- structure 124pectin lyases 125penicillin, stereocontrolled synthesis, using kidney

acylase 356fpenicillin acylase, use in transac ylation s 386penicillin G, hydrolysis by penicillin acylase 386fpenicillin G acylase (PGA), production of 6-ami-

Penicillium digitatum, stereospecific modification

Penicillium roqueforti, blue veined cheese flavor

pentachlorophenol, degradation under aerobic

- reductive deh alog ena tion of 207PEP-CL see phosphoenolpyruvate carboxylasespeptide bonds, form ation of, by catalytic antibod-

nopenicillanic acid 280

of terpe noid s 374

compounds 347

conditions 210

ies 418fpeptide cleavage, sequence specific -, by an anti-body 420ff

peptide formation , by thermolysin 355fpeptides see also synthetic enzymes- chlorinated - 185pesticides, halogenated -, distribu tion of 177P G A see penicillin G acylasepharmaceutical intermediates see enantiomerically

pure pharmaceutical intermediatespharmaceutical precursors, kilogram-scale routes

to, involving amidases 285- - involving lipases 284

- -involving pro tease s 284fphenolic compo unds, form ation of 346f

phenols, chlorinated - 188- halog enation of 196phenylacetate catabolism, pathway 66phenylalanine, biosynthetic pathway, rate limiting

- overproducing organisms 315ff- regula tion of biosynthesis 315- secretion from the cell 320fD-phenylalanine, engineering biochemical path-

way, for whole cell biosynthesis 327- novel biosynthetic pathway 325ff

L-phenylalanine, bacterial pro duc tion of 314ff- biosynthetic pathways 314f

steps 320

- feedback inhibition of prephenate dehydratase

phenylalanine ammonia-lyase (PA L) 137ff- deamination of phenylalanine 139f- inhibitors 140- of Rhodotorula glutinis 139ff- substrates 140- - pyridyl cinnamates 141phenylpyruvate, decarboxylation of, by phenyl-

phenylpyruvate decarboxylase 85pherom ones, cockroach aggregation, halo-

organic- 177- com ponent, synthesis by yeast alcohol dehydro -

genase 360- enzy ma tic synthesis of 361f- synth esis of 22phosphate, pK, values 222phosphate ester, hydrolysis of, use of catalytic an-

pho sph ate este r hydrolysis, by nuclease 236- by phosphatase enzym es 236fphosphate monoester, cleavage of, use of catalytic

phosph ate sources, inorganic phosp hate 228- nucleoside triphosphates 228ffphosphodiester, cleavage of, use of catalytic anti-

phosphodiester formation, coupling with adeno-

-coupling with cytidine tripho spha te 232f- coupling with uridine triph osp hate 233

phosphodiesterase, phosphorylation by 226phosphoenolpyruvate 229ff- enzymatic synthesis of 229- phosphorylation potential 229phosphoenolpyruvate carboxylases (PEP-CL) 84f- listing of 85

phosphonates, inhibitors of phosphodiester cleav-

phosphoramidate esters, hydrolysis by alkaline

phosphoryl transfer 222ff-

cofactor regeneration 228- mechanism of 224- phosp hate source 228phosphoryl transfer enzym es 223ff- classif ication of 223ff- listing of 225phosphoryl transfer reactions 223ff- enzy me classes 223ff- two groups of 223phosphoryl transfer reagents, phosphorylation po-

tentials of 227phosphorylation 221ff- enzymes 223ff-

enzymic synthesis of heterocyclic compounds

319

pyruvate decarboxylase 86

tibodies 463f

antibodies 435f

bodies 436f

sine triphosph ate 232

age 236

phosphatase 237

389



Index 531

phosphotriester, hydrolysis of, use of catalytic an-

Pichia pastoris, alcohol conversion to aldehydes

pig liver esterase (PLE), specificity of 367f- stereose lective hydrolysis 353f

I- synthesis of heterocyclic compo unds 384fpig liver 2,3-oxidosqualene c yclase 28fP - 0 bo nds, cleavage of 235ff- formation of 228ff- form ing reaction s, listing of 234fpollutants, recalcitrant - 202poly-alanine, imm obilized -, pr ep arat ion of 5OOfpoly-L-alanine, epox idation of, chalco ne 499f- - “chalcone-type” sub strate s 499ffpoly-leucine, epox idation by 499ff- - preparation of natural products and pro-

- immobilized -, preparation of 501poly-L-leucine, preparation of 503f- stereoselective carbonylation by, palladium cata-

poly-L-valine, stereoselective electrochemical re -

polychlorinated biphenyls (PCB ), degradatio n of

- oxygenolytic cleava ge of 209- patte rns of dechlorination 208polycyclic biotr ansf orm ation products 376ff- use of alcohol dehydrogenases 382f

polyenes, cationic cyclization of, use of catalytic

polynucleotide synthesis, enzyme catalyzed phos-

polyol dehydrogenases 306polyols, oxidation-of 300- - according to the Bertrand-Hudson rule 301polyp eptide s, as epox idation catalysts 499ffporcine pan crea tic lipase (PP L) 368ff- asym me tric hydrolysis 368f- catalyz ed hydrolysis 358fporphyrin metalatio n, use of catalytic antibodies

potato lipoxygenase 365PPL see porcine pancreatic lipaseprednisolone, multienzyme synthesis 382fprep hena te dehydra tase, deregulation of 317ff- feedback inhibition, by L-phenylalanine 319prodrugs, activation of, antibody mediated 442ff- carb am ate, cleavage by abzymes 444f- of chloramphenicol, hydrolysis of 442f- of 5-fluo rodeo xyurid ine, activatio n of 442fprogestero ne, biotransformation of 381propionic acid, production of, by Propionibacter-

prostaglandin synthesis, use of porcine pancreatic

tibodies 437

345

drugs 501f

lyzed - 505f

ductio n by 505f

207

antibodies 430f

phate bond formations 389

477

ium sp. 345

lipase 369f

prostaglandin synthons, use of horse liver alcohol

prostaglandins, chlorinated - 186proteases, synthesis of pharmaceutical precursors,

pseudo-enzymes see synthetic enzymespse udo eph edri ne, industrial synthesis of 24Pseudomonas lipases, synthesis of p-bloc kers 385Pseudomonas pseudoalcalignes, malea te hydrat-

Pseudomonas putida 363f- benzoylformate decarboxylase 65ff- formation of alicyclic compounds by oxidation

- hydroxylation of carboxylic acids 363- monosaccharide carboxylic acid dehydratases

- oxaloacetate decarboxylase 57f- oxida tion of styrene 375Pseudomonas sp., eugenol degrada tion pathway

- prod uction of ferulic acid 337pyrethroid insecticides, using Arthrobacter lipase

resolution 369pyrethroid precursors, synthesis of, use of a

monooxygenase 282- - use of an ester ase 283- - use of oxynitrilases 282pyrophosphate, p K , values 222pyrroles, chlorinated - 186pyruvate decarboxylase (PD C) 23ff, 49ff- activatio n of 53- acyloin cond ensatio n 23ff- biotransformations 53ff- decarboxylation of fluoropyruvate 54, 56- mechanism 49ff- of Saccharomyces cerevisiae 50ff- of yea st 54f- of Zymomonas mobilis 49f, 53ff- structure 49ffpyruvate kinase, phosphorylation by 224

dehydrogenase 372f

kilogram-scale route s 284f

ase 114

reactions 374ff

115

338

Qquinolines, chlorinated - 187

R

rabbit muscle aldolase (RAMA) 7ff- synthesis of azasugars 1Off

- synthesis of carbo hydra tes 10f- synthesis of cyclitols 13- use in synthesis of heterocyclic compounds

rational en trop y, in enzyme catalysis 415ffreactive immunization, induction of catalysis in

rearr ang em ents , use of catalytic antibodies 469f

388f, 391

antibo dy binding sites 439ff



532 Index

redo x reactions, use of catalytic antibodies 477ffreduction reactions, catalyzed by yeast enzymes

- use of catalytic antibo dies 478fretro-aldol reaction, use of catalytic antibodies

retro-He nry reaction, use of catalytic antibodies

rhamnulose-1-phosphate aldolase, overexpression

Rhizopus arrhizus, enzymatic reduction by 371frhodium polypeptide complex, stereoselective hy-

Rhodotorula glutinis, phenylalanine-ammonia

- production of L-phenylalanine 141

370ff

468

468

of 14

drogena tion by 507f

lyase 139ff

sugars, synthesis of, by one-pot multiple enzyme

- - by two sep multiple enzyme catalysis 19sulfides, electrochemical enantioselective oxidation

Symbiobacterium sp., tyrosine-phenol lyase 95ffsyn-eliminations, of selenoxide 41%

synthases, systematic nam es 6synthe tic enzym es 491ff- artificial peptides in stereoselective synthesis

- cyclic dipep tides 492ff- poly-leucine 499ff- use of linear dipep tides 508ff- use of tetra pep tides 511ff- use of tripeptid es 510ff

catalysis 18

of, use of poly-L-valine 506

491ff

Biotechnology Vol 8b Bio Transformation II, 2nd Ed

Documents

Transcript of Biotechnology Vol 8b Bio Transformation II, 2nd Ed