Embed Size (px)
Citation preview
Chapter V | |
L A R G E - S C A L E O R G A N I C S Y N T H E S I S
U S I N G C E L L F R E E E N Z Y M E S
George M. Wh i tes ides
1
2
3
lntroduct ion 901
Syntheses Ut i l iz ing Cel l -Free
Considerat ions in DevelopingEnzymic Catalysis 905
Enzyme lsolation 905EnzymeStabi l izat ion 906Enzymic Act iv i ty 91 ICofactor Requirements 9l6Reactor Design 918
Alternat ives and Prospectives
Enzymes: Examples gO2
New Processes Ut i l iz ing Cel l -Free
, t t )
I,la;,t
920
I N T R O D U C T I O N
The effectiveness with which cell-free enzymes can be applied as catalysts inpract ical organic synthet ic procedures depends both on scale and react ion type.Cell-free enzyrne preparations have been widely employed in a variety of typesof small-scale syntheses. and used in several large-scale degradative transforma-t ions not involv ing cofactors. The part ia l hydrolysis of mi lk proteins by renninand of soi l proteins by protease-containing detergents. and the chi l lproof ing ofbeer by paparn. provide examples of the lat ter c lass. Enzymes are not commonlyuti l ized in large-scale reactions that require cofactors or that carry out reactionsrrore conrplex than hydrolysis or isomerizat ion. The border between the regionsof synthesis in which enzymic catalysis has been successful ly used and those inwl i ich i t has not has been def ined pr i rnar i ly by econornics [1] . ln syntheses ofsmall quantit ies of rnaterials fbr research or laboratory use. leasibil i ty andconvenience are usual ly more important than enzynre cost: for example. the factthat i t is possible to jo in deoxyr ibonucleot ide chains enzymatical ly. but d i f f lcul t
Work suppor ted by ' NSI : (RANN) under Gran t No . G I 34284 .
9 0 1
9 O 2 L A R G F S C A L E E N Z Y M I C S Y N T H E S I S
or impossible to join them by conventional synthetic chenrical techniques, rendersthe economics of the production and uti l izat ion of the appropriate l igase ntoot
[2 ] . In the synthes is o f la rge quant i t ies o f a substance, by cont rast , econonics
becomes an important consideration. and the factors defining the cost of areaction sequence requir ing cel l- free enzymic catalysis production and isolat ion
of the enzyme, enzyme operating l i fet ime and specif ic act ivi ty, cofactor replace-ment , reagents , suppor ts , and equipment pecul iar to enzymic reactors may besuch that alternative synthetic procedures ut i l iz ing chen'r ical and/or rnicrobiolo-gical transformations may appear more attract ive than a cel l- free enzyrnic route.
The purpose of this chapter is to i l lustrate brief ly examples or representativereactions for the large-scale synthesis or processing of organic materials in whichcel l- free enzymic synthesis presently plays a role, to discuss considerationsimportant to the successful design of new processes (part icularly processesinvo lv ing t ransformat ions more complex than those now employed) . and to out -l ine cri teria useful in identi fying situations in which cel l- free enzynric catalysismight o f fer advantages over o ther synthet ic s t ra teg ies. Cer ta in o f the cons idera-t ions impor tant in us ing enzynt ic cata lys is in large-sca le appl icat ions par t icu lar -ly those invo lv ing the iso la t ion and s tab i l iza t ion o f enzymes are cent ra l ro thr ' r ruse in synthesis at any scale, although the requirements for successful large-scalcut i l i za t ion are a lmost a lways more s t r ingent than those for smal l -sca le reacr i r - r r rs :others-cofactor regeneration, reactor design are unique to large-scale proc..s' . ' : .We are concerned here wi th the quest ion o f the apphcat ion o f enzr , 'n res ro r l r r .synthesis of relat ively small quanti t ies of materials. this issue is discuss. 'd rrrdeta i l e lsewhere in th is book (Par t I . Chapters IV and VI ) . Rarhc. r . r rc ' r i r l lcons ider the d is t inc t prob lems that must be so lved in synt l - res iz ing lar .gc t lu l r l -t ies (arb i t rar i ly def ined as ranging f rom one or severa l molcs t t r u .nr r r rere i . i isca le) o f a substance us ing enzymic cata lys is . Incorporat ion o t ' th . . tcchr l r r l r r i , :requ i red to so lve these prob lems in to the body of synthet ic orsanrc chenl r t r rrequ i res a b lending of chemica l procedures wi th those t ' ror r . r cnzr nro logr rnJmicrob io logy. These b io log ica l techniques are unfanr i l ia r to nr ( )s t s \n1 l r r ' r r .organic chemists. Nonetheless, despite the fact that synthesis requir ing cel l- t-r. . . .enzymic cata lys is may be technica l ly more complex than nrore fant i l ia r l1pt .s . lchemica l or fermentat ion syntheses, the se lect iv i ty o t fered by enzymrc cata l l ,s is .part icularly for the many biological ly, medicinal ly. and nutr i t ional ly inrporr lnrc lasses of mater ia ls conta in ing carbohydrates, pept ides. and nuc leot ides. ca l ln( ) r .in many cases, be equaled by conventional organic or 1-ermentation synthesis.
2 S Y N T H E S E S U T I L I Z I N G C E L L - F R E E E N Z Y M E S : E X A M P L E S
Existing processes that util ize enzymes for the production of discrete chernicalcompounds on a large scale have in common several features that sintpl i fy thciroperation: l i t t le or no enzyme purif icat ion is required; immobil izat ion is
2 E X A M P L E S 9 0 3
achieved simply using inexpensive supports; the reactions catalyzed are uncom-
pl icated [3 ] . React ions u t i l i z ing L-amino ac id acy lase, penic i l l in amidase, and
aspartase i l lustrate types of enzymic catalysis that are actual ly used in large-
sca le work and that prov ide examples o f types of enzyrn ic t ransformat ions that
can be appl ied re la t ive ly eas i ly in chemica l synthes is .
L-Amino acid acylase is used in a process for resolving racemic amino acids
[4] A racemic mixture of n- and L-rV-acylarnino acid (obtained by chemical
synthesis) is passed through a column containing enzyme inrmobil ized by
adsorpt ion on DEAE ce l lu lose. The L-enant iorner is hydro lyzed to f iee amino
acid. Separation of the D-acylarnino acid from the free L-ar.nino acid is easi ly
accornp l ished. and t l ie D-acy larn ino ac id is subsequent ly ep imer ized and re turn-
ed to the co lumn.
ep imer izat ion
NICRC
*N
IR.C
Ac NHAcI
lO2H + D-RCHCO2II
II
Lamino ac id Iacy lase
IY
, NHAc
to. - o-nlHco, H
H,^
HC
II
IIIF{:
IC,
L -
L- HR
Penici l l in amidase isprocedures for the
used to remove the side chain of penici l l in G in certain
product ion o f 6-aminopenic i l lan ic ac id (6-APA) [5 ] . Th is
ol l
P I ICH2CHN - s r
r\
co2 H
penic i l l in-_--------------
a m i d a s e
I { ,N\r--r-S r
"J-'r {,,6-APA
of semisynthetic peni-
fumarate ion, yieldingmaterial is in turn used as a starting material for a variety
cil l ins. Aspartase catalyzes the addition of ammonia toopt ical ly act ive L-aspart ic acid [6,7] .
904 L A R G E . S C A L E E N Z Y M I C S Y N T H E S I S
C0 2 - N H 4 * f o ; N [ { o .
.'s'artasc > r4, n-orc/ \nnr .H;N
-O2C
The most successful technique used to immobilize aspartase is one that, al-though not formally "cell-free," offers great promise for many types of enzyme-catalyzed syntheses. The enzyme is not isolated. Instead, an intact microorgan-ism containing the enzyme is incubated with acrylamide and appropriate cross-linking agent and radical init iator, and the entire suspension polymerized [6] .The resulting polyacrylamide gel containing immobilized, polymer-fi l led cells,is broken up, and transferred to a column. The resulting preparation showsgood stabi l i ty and act iv i ty. (A i0 X 100-cm colurnn, containing packed cel ls ofEscherichia coli in polyacrylamide, can generate about 700 g of aspartic acid perhour) [7] .
An obvious advantage of this technique is its simplicity: no enzyme isolationis required. Less obvious advantages also apparent ly at tend i t : the enzyrreimmobi l ized by polymerizat ion of the intact organism is very stable. I t is poss-ib le, a l though not establ ished, that th is stabi l i ty ref lects permeat ion of the cel linter ior by acrylamide. wi th subsequent polymerizat ion to a polyacry ' lanr idegel throughout the cel l matr ix. This gel rnust have pores suff ic ient ly snral l thatany proteases present at t l ie t ime of polyrner izat ion, or re leased by the polypter i -zat ion, are utrable to at tack the aspartase. ln addi t ion, the environnrent providedfor the enzyme its natural matrix with a low concentration of polyacrylarrriclemay discourage confonnat ional changes that lower i ts act iv i ty.
This procedure. in i ts present form, is not appl icable to anumber of 'c lasses ofenzyme catalyzed reactions. Any process that requires diffusion of enzyrles orother tnacromolecules (e.g. , r ibosorrral protein synthesis) wi l l be hinderecl srprohibi ted by the polymer matr ix. The cel l wal l of the rnicroorganisnr nr ight inpr inciple also restr ict d i f fusictn of certain substrates or products tg or l l . r ' r l reinter ior of the cel l . The condi t ion of the cel l wal l in these preparat ions ls . ( ) tknown. The catalyt ic act iv i ty of the aspartase preparat ion. is however. increasc-clby an act ivat iot i procedurc involv ing. inter al ia. heat ing: th is act ivat ion nray serveto disrupt cell merrrbranes sufficiently to pernrit free passage of substrates alclproducts. Acrylamide reacts rapidly wi th several funct ional groups of enzyr lcs.part icular ly the -SH rnoiety of cysteine and other nrercaptan-containing prslc-cules, and is i tsel f capable of modity ing or destroying the act iv i ty of certaipe l l zyn leso l l vc ry shor t c t l t l tac t [8 ] . Po lyacr ,v . ' lan t idc ge ls I iavc . r ' c la t i ve ly pgor 11 . * ,and r t lechat l ical character ist ics, and are not ideal for processes requir ing t heprocessing of very large volumes of solut ion. Final ly. the acceptabi l i ty ro rnan) 'of the wor ld 's food and drug adrninistrat ion bodies of mater ia ls. intendecl tgrhuman use. prepared by a step involv ing in i rnobi l ized whole cel ls is not ent i re l l ,
3 N E W P R O C E S S E S 9 0 5
clear. Nonetheless, for processes that involve relatively simple catalytic transfor-
mation of low-molecular-weight substrates, the immobilization of intact organ-
isms offers a very practical and attractive method of uti l izing enzymic catalysis.
These three examples are processes sufficiently developed to be used in com-
rnercial-scale synthesis. Many examples of smaller-scale syntheses also exist [9],ancl representative examples may be found in Metfutds in L'nzymc.tlogy,
Bittchemical Sltntheses, and similar sources, as well as in other parts of this volume.
3 C O N S I D E R A T I O N S I N D E V E L O P I N G N E W P R O C E S S E S
UTI L IZ I NG CELL -FREE ENZYMIC CATALYSIS
Five principal factors must be considered in designing and developing a practi-
cal synthesis based on enzymic catalysis: (1) the availabil ity, ease of isolation,
and purif ication of the enzyme(s) required; (2) the l ifetime of the enzyme(s)
under operating conditions and the efficacity of stabil ization protocols: (3) the
intrinsic activity of the enzyrne(s), and the efficiency with which enzymic
activit ies can be coupled in reactions requiring multiple enzylxe-catalyzed
steps; (a) the cofactors required for enzymic activity; and (5). the design of
reactors compatible with the transformation to be achieved. We consider each of
these subjects in turn.
E n z y m e l s o l a t i o n
Enzymes are, in principle, available frorn all l iving orgarlislns. For synthetic
work requir ing large quant i t ies of enzymes, certain types of sources are better
than others. The most convenient sources are microorganisms. Yeasts, bacter ia,
and fungi can be grown in quantity using well-established f-ermentation techni-
ques; alternatively, certain comrnonly used lines-particularly Baker's yeast and
E. coli-are commercially available. Starter quantit ies of large trunlbers ot'
standard cultures are available from the Arlerican Type Culture Collection.
(See also Appendix 1, Part I for a l is t of other cul ture col lect ions.) In addi t ion.
microorganisms offer an advantage over plant or animal sources that is parti-
cularly valuable when large quantit ies of enzyrnes are required; narnely that
select ive mutat ion and genet ic manipulat ion can of ten be used to develop strains
that produce relat ively large quant i t ies of the enzymes of interest , and enzyme
induction is easier to accomplish with microorganisms than with higher organ-
isms. A final, significant advantage of rnicroorganisms as enzyme sources is that
the convenience with whicii they can be handled l ias. in the past. made them
favorite subjects for mechanistic enzyrnology and biology. Thus, large quantit ies
of data are already avai lable on the preparat ion and propert ies of many interest-
ing enzymes. Unfortunately, some of the microorganisms most commonly
exploited as enzyme sources for mechanistic work, particularly Salmonella
typhimuriurn, are sufficiently pathogenic to make their growth in quantity
9 0 6 L A R G E S C A L E E N Z Y M I C S Y N T H E S I S
inconvenient or hazardous: a l ternat ive sources for these enzvmes can. however .
usual ly be found.
Plant and animal sources of enzymes are less convenient than rnicrobial
sources; they may, however , o f fer the on ly source of enzymes of in terest . P lant
sources, in par t icu lar , may become increas ing ly a t t rac t ive as the techniques for
p lant t issue cu l ture develop [10]I so la t i on o f cnzy rnes l o r l a rge -sca le syn the t i c wo rk p resen ts econo r l i c p r . ob -
i en rs no t e t r coun te red rn sn ia l l - sca le p r cpa ra t i on . I n gcne ' ra l . enzyn re i so la t i o r r
and pur i f ica t ion is an expens ive process. and the less manipu la t ion requi red inreaching a useable preparat ion. the bet ter . The least complex technique con-ceptua l ly is the use of whole ce l ls [6 , I I 13] . A l though th is technique wi l l
ce r t a i n l y be w ide l y cxp lo i t ed when app l i cab le . i t i s no t ye t c l ea r whe the r i t ca r r
be employed for o ther than very s i r t rp le t ransfornrat ions. Ner t lcast conrp lex
i s t o use t l t c c rude p ro te in m ix tu re ob la i ned on b reak ing ce l l s l v i t h a n r rn -
i t t tunr o f pur i f ica t ion. Typ ica l ly . auto lys is . lys is wi th an adc led enz\ ,nre . ( ) r
mechanica l or osmot ic d isrupt ion o f ce l ls y ie lds a rn ix ture o f ce l l contponents
conta in ing, in addi t ion to the enzyme(s) o f in terest , var ious proteascs. nuc le icac ids. l ip ids , carbohydrates, and prote ins that rnay b ind or degrade the productsor s tar t ing rnater ia ls o f the react ion o f in terest , the enzy l les requi red to e f1-ectthe des i red t ransformat ions. or essent ia l cofactors (Scheme I ) . Of ten. however .e i ther no or surpr is ing ly l i t t le fur t l ie r t reatn tent is requi red to obta in Lrseablepreparat ions. Thus, i f the enzyme of in terest is thermal ly s tab le . br ie f heat ing rnuydenature n tany of the o ther prote ins present . Enzyrne i rnrnobi l iza t ion by adsorpt -ion, ent rapment , or cova lent a t tachment o f c rude preparat ions to suppor ts nravdecrease the rnobi l i ty o f t l te proteases suf l - ic ient ly that the1, becorne un i r r rpor-t an t i n de te r t t i i n i ng e r t zyn r i c l i f c t i n res : a l t c rna t i ve l y . add i t i on o l ' p ro t casc i n l r i -b i tors may decrease the i r ac t iv i ty (see be low) . Nuc le ic ac ids can be rer loved b.vprec ip i ta t ion (protamine su l fa te , s t repton iyc in) or enzynt ica l ly degrac led.Ammoniurn su l fa te or organic so lvent f rac t ionat ion serves to renrove sonte o f - theunwanted prote ins f rom the crude preparat ion. These and re la ted e lenrentary 'pu r i f i ca t i on t echn iques a re desc r i bed i n a va r i e t y o f sou rces [ 14 . 15 ] . A ione .they may prove insuf f ic ient or unsat is factory in many cases; the impor tant p i - r in1Io reahze, however . is that i t i s not necessar i ly requi red (or even des i rab le) topur i fy an enzyme to homogenei ty for i t to be use l 'u l in a synthet ic procedurc .and i t is usual ly wel l wor lh the e f for t to exarn ine the act iv i ty and s tab ih ty o f 'c rude preparat ions.
E n z y m e S t a b i l i z a t i o n
The s tab i l iza t ion o f enzyrne preparat ions is a top ic o f cent ra l in rpor tance inenzymic synthesis: even relat ively inaccessible and expensive enzylnes can. inprinciple, be used in a practical synthesis provided the l i fet ime of the enzyrxe
under operat ing condi t ions is long. Unfor tunate ly , in most ins tances, less is
3 N E W P R O C E S S E S 9 0 7
Cell disruption
Proteaseinhibit ion
Precipitation orhydrolysis ofnucleic acids
lnit ial separation
Furtherpurif ication
Scheme 1. Representative Enzynrc Isolation korrdure
known about rnechanisrns o f the processes lead ing to losses in enzymic act iv i ty
than about the mechanism of the enzymic activi ty i tself . Consequently, enzyme
stabi l izat ion is presently as much art as science. I t is possible to point to a
number of experimental protocols that are known to increase the stabi l i ty of
enzymes, and in some cases to rat ional ize these procedures on plausible mechan-
is t ic grounds. In any new case, however , the process of dev is ing a set o f exper i -
mental condit ions result ing in good operating l i fespans for an enzyme remains
large ly empi r ica l [16]
ProteasesA major contr ibutor to the loss of enzymic activi ty in crude enzyme-contain-
ing solut ions (and in many "pure" preparations) is destruction of the enzyme
by proteases. Proteases occur in a wide range of molecular weights. and operate
over a range of va lues of pH [17-19] . The ab i l i ty o f pro teases to pers is t through
CellsII
I nutolysis, lysozyme, mechanical or osmot ic
i disruPtionBroken Cells
l*nl _ ^ zI Phcl ' l rSo2F: PhNHC\I NH'I
I etotamine; streptornycin; DNAse and
I RNAseV
Crude Protein Mixture I
II Catcium phosphate gel precipitation;
I tNH4)2SO4 and/or acetone f iact ional
I ptecipi tat ions.I
Crude Protein Mixture I lI
Y Gel penrteation, ion exclrange, and/ori uri lnlty chrornatography
IPure E,nzyme
9 O B L A R G E - S C A L E E N Z Y M I C S Y N T H E S I S
even careful chrornatographic purif ication procedures 120-221 has led to anonly partially jokhg feeling among enzymologists that for every protein there isan almost identically constituted protease. There are three effective strategies fordealing with protease contamination of an enzytne. First, the protease can beinhibi ted by addi t ion of substances that i r reversibly block or completely inhibi tthe protease active sites. Two broadly effective cornpounds for this purpose arePhCH2SO2F [2.3J (cal led phenyl nrethyl sul lonyl f ' luor ide in biochenr ical c i r -cles) and diisopropylphosphofluoridate (DIFP). Both inhibit serine proteases(e.g. , t rypsin. chymotrypsin); the forrner is less toxic than the lat ter . which isalso a very powerful acetyl cholinesterase inhibitor. Benzamidine and relatedmater ia lsare also useful in inhibi t ing proteases related to thrornbin [2a] . Proteinprotease inhibi tors [25] are too speci f ic to be useful for broad-spectrurn pro-tease inhibit ion. Second, immobilization of the enzyme preparation usuall l 'renders protease act iv i ty unimportant, s ince the solut ion protein-protein en-counters that permit at tack by the proteases are unimportant wi th imntobi l izedenzymes [26] . Third. careful c l t rornatography- part ic ular ly ' af f in i ty chronratog-raphy may reduce protease act iv i ty to insigni f lcant levcls.
A u t o x i d a t r o n : O x i d a t i o n R e d u c t i o n B u f f e r i n g
Many enzymes require -SH or -S-S- rnoiet ies as essent ia l structural or catah' t icelements t8] . Both can be oxidized rapidly under condi t ions commonlv en-countered in euzymic reactors. Al though the detai led rnechanisnts of t l teseoxidat ions are not known, several of their qual i tat ive character ist ics have beenident i f ied. Tirus, the in i t ia l oxidat ion of cysteine to cyst ine occurs nrore rapi t l l r
I(-SH
R_S{I
R-SI
R-S
[.-,elL.-A I+,
RSO.r
RSO I
aI h igh than at low pH. is s t rongly cata lyzed by t races of t rans i t ion rneta ls
Ipa r t i cu la r l y Fe ( l l l ) and Cu ( l l ) ] . and can bc reve rsed w i t h su i t ab le reduc ine
agents . The subsequent ox idat ion o f cyst ine d isu l f ide l inkages (u l t imate l l ' t t r
cyste ine su l fon ic ac id) is usual ly s lower , but is not eas i ly reversed. Autox idat i r rnrnay a l so des t roy p ro te ins by rou tes no t i nvo l v i ng SH g roups [ 2 ]1 .
The s ign i f icance of these observat ions fbr enzynres that are act ive in the fu l lyreduced fomt ( i .e . , w i th funct iona l cyste ines but wi thout in tpor lant cyst inegroups) is c lear : the enzyme should be s tored and used under anaerob ic conc l i -t ions (a nitrogen or argon atrnosphere); the reagents used shoulcl be f ' rec oft rans i t ion-meta l contaminants : a su i tab le reduc ing agent should be present inthe so lu t ion to reverse any advent i t ious ox idat ion. The f i rs t condi t ion is d i t ' t l cu l t
3 NEW PROCESSES 909
to satisfy adequately: I l t l of air may contain enough oxygen to oxidize the
cysteine moieties of I mg of protein. The second is also diff icult to meet.
Although buffers and reagents may be reasonably stripped of transition-metalions by passing them through a resin containing SH groups before use, theenzyme itself usually cannot be. Since many proteins bind transition-rnetal ions
strongly, it is often impractical to try to keep solutions containing enzymes freeof transition metals. Thus, the major burden of protectrng enzymes againstdeactivation by autoxidation usually falls on the reducing agent added to thesolut ion.
The most popular and effective material for this purpose is dithiothreitol(Clelland's reagent, DTT), ordinarily added in concentrations approximately
1000 times greater than that of the enzyme [28] DTT is unfortunately an
expensive material, and the less expensive but less effective Z-mercaptoethanol is
often used in its place.
en ./sl +\s'
OFIHr^)a ...........--L.H S I . _
*rH
SH/ , , n O H
Enr n tY
\s -s.--..r[..on
DTT /t//
SH
Enz +\SH
orlS'^\r,'
Jlnn
The stabi l izat ion of enzymes that require an intact cystine disulf ide l inkage
against oxidation is a problerr that has not been satisfactori ly solved. The
presence in so lu t ion o f a large excess of an added mater ia l conta in ing a d isu l f ide
l inkage should s tab i l ize an essent ia l d isu l f ide in a prote in . but the i r revers ib i l i ty
of oxidation proceeding past the disulf ide oxidation level presents obvious
problems. The coupling of two or rnore enzyme systems operating at dif ferent
ox ida t i on l eve l s has no t been a t t emo ted .
Env i ro n rnen t
A variety of solution variables are capable of influencing the stability ot'enzylnes: pH. temperature; ionic strength, buffer composition. solvent character(and immobi l izat ion support , for immobi l ized enzymes). the nature of inter-faces present in solut ion, and the nature and concentrat ion of other solut ioncomponents. Most of these var iables are presumed to act by inf luencing the ter-tiary structure of the enzyme. No generalizations concerning rnost of these inter-
9 1 0 L A R G E S C A L E E N Z Y M I C S Y N T H E S I S
actions are presently possible: indeed, no general izat ions may exist, since the
relat ive importance of various factors wil l di f fer for dif ferent proteins. I t is,
however, worthwhile to question one assumption that is commonly tacit in
much experimental work, namely, that most enzymes exist in the cel l in apredominantly aqueous environment, and that, once removed f;om the cel l ,
enzymes are l ike ly to be rnost s tab le in water . There is l i t t le quest ion that for
enzymes that are clearly associated with membranes, the presence of l ipids in
so lu t ion may be essent ia l for s tab i l i ty and act iv i ty 1291 .At another ext reme,extracel lular enzymes probably are most stable in water. For the interr-nediate
c lass o f enzymes that ex is t in the in ter ior o f the ce l l , i t i s by no means c learwhat so lu t ion composi t ion would be expected most c lose ly to dupl icate the i r
native environment. The f luid port ions of cel ls may contain high concentrat ionsof many components o ther than water , and dupl icat ion in v i t ro o f the ce l lu larmicroenvironmerrt of an enzyme, even in the uncommon event in which thephys ica l locat ion o f the enzyme wi th in the ce l l is known. cannot present ly be
carr ied out by des ign. Thus. many of the s tandard empi r ica l t r icks used to
stabi l ize enzymes in solut ion may be effect ive by virtue of providing the type of '
mixed aqueous-nonaqueous env i ronment found in the ce l l . In fac t . i t seen)sposs ib le that invest igat ion o f the ba lance of hydrophi l ic and l ipophi l ic characterthat is most e f fec t ive in s tab i l iz ing an enzy lne rn ight prov ide an rnd i rect
approach to in f -er r ing the character o f the ce l l in ter ior .
Br ie f l i s t ing and comment on severa l o f these t r icks is wor thwhi le . Enzyrnes
may denature a t in ter faces [30] ln ter fac ia l denaturat ion may of ten be con-trol led by al lowing an inexpensive. inert protein (cornmonly bovine serun' la lbumin) to adsorb on the sur face of g lasware to be used wi th enzymes or byusing polyethylene or Teflon rather than glass apparatus. Since interf-acialdenaturat ion o f suscept ib le enzymes may a lso occur a t the so lu t ion-a i r (or iner tgas) interface, gases should not be bubbled through an enzyme solut ion. norshould s t i r r ing be so v igorous as to cause s t rong vor tex format ion. ln cont ro lhngthe pH and ion ic s t rength o f enzyme so lu t ions, i t i s impor tant to remember thatbuf fer components [31] and sa l ts used to ad just ion ic s t rength in f luence manysolu t ion proper t ies o ther than those of immediate concern. Thus, so lu t ionsconta in ing, for example, phosphate buf fer or t r is buf fer o f the same pH do nothave the same va lues of sur face tens ions, in terna l pressures, or many otherparameters that might inf luence the conformations of enzymes. The subject of 'hydrophobic interactions involving proteins is discussed in greater detai l inChapter VIII of this volume and elsewhere [32-361. Addit ion of glycerol, sorbitol.polyvinyl alcohol, or other neutral, hydrophil ic polymers (often in high concen-t ra t ions) . s tab i l izes many enzymes [37-391. In a t least one ins tance, th is s tab i l i -zat ion has been correlated with a change in tert iary structure [39] . Polyelectro-lytes may also strongly inf luence enzyme stabi l i ty and activi ty [a0] . Theinf luence of detergents on enzyme stabi l i ty has not been extensively investi-
3 N E W P R O C E S S E S 9 1 1
gated. anc l jud ic ious choice o f sur f 'ace act ive agents wi l l p robably prov ide s ign i l - -
icant s tab i l iza t ion fc l r enzymes r to t c lass i f ied as "mei l tbra l te bout rc l " but s t i l l
assoc ia ted weakly wi th rnernbranes. l ip ids , or hydrophobic por t ions o f o ther
prote ins in v ivo. Stud ies o f the sens i t iv i ty o f the act iv i ty o f p-a lan ine carboxy-
pept idase to detergent s t ructure suggests that cons iderab le care may be requi red
to achieve the best balance of hydrophil ic and l ipophil ic interactions and
"st ructure-making" and "s t ructure-break ing" proper t ies fbr such s tab i l iza t i t ln
l4t ̂ 421 .The s to rage s tab i l i t y o f so lu t i ons o f n rany enzyn res i s i nc reased by t l i t ' p res -
cncc o t ' subs t ra t cs . co fac to rs . o r p roduc t s . and bo th s tab i l i t y and ac t i v i t y o f
enzymes may require rnetal ions [43] The fornrer characterist ic rnay ref lect
e i ther s tab i l iza t ion o f ter t ia ry s t ructure or protect ior r o f the act ive s i te reg ion
f rc tn i a t tack by oxygen. rneta l ions, or o ther so lu t ion co lnponents . T l ie f requent
requi rernent fbr s ign i f icant concent ra t ions o f f ree Mg( l l ) or o ther rna in-group
rneta ls for ac t iv i t l , p laces some l i rn i ta t ion on the extent to which the act iv i ty o t -
t rans i t ion rneta l ions can be suppressed by adding chela t ing agents such as
EDTA. The assoc ia t i on cons tan t s be tween EDTA and Cu ( l l ) and Mg( l l ) a re
abou t 10 re and l 0e r ro l / l i t e r . r espec t i ve l y [ 44 ] Thus . i f t he o r i g i na l concen t ra -
t i on o f Cu ( l l ) i n an enzyme so lu t i on i s 10 -a , 4 / . add i t i on o f enough EDTA to
reduce t he concen t ra t i on o f uncon rp lexcd CL r ( l l ) t o 10 -12 X I r . v i l l have ve r y l i t t l c
i n f l uence on t he concen t ra t i on o f f r ee Mg( l l ) [ (Mg ( l l . l ) / (Mg ( l l ) 'EDTA) = l 0 r l .
Add i t i on o f enough more EDTA to reduce t l t e C iu ( l l ) concen t ra t i on t o 10 -14 tU
wi l l . however . s ign i f icant ly per turb the concent ra t ion o f f ree Mg( l l ) [ (Mg( l l ) ) /(Mg ( l I ) .EDTA) = l l . Thus . t he re i s a l owe r l im i t be low w l i i ch i t i s no t p rac t i ca l
to reduce the concent ra t ion o f t rans i t ion-meta l ions in so lu t ion by che la t ion.
Although this l i rrr i t is low. catalysis of autoxidation of enzyt-ne sul lhydryl groups
by these nreta ls nr ight , in pr inc ip le , s t i l l be s ign i t - icant . Fur ther . the r t te ta l -
EDTA cotnp lexes lnay be autox idat ion cata lys ts i t t the i r own r ight . Thus.
a l though addi t ion o f che la t ing agents should (and does) great ly decrease the ra te
of rneta l -cata lyzed autox idat ion. i t may not cor t rp le te ly e l iminate i t . ln fac t . the
use of EDTA in so lu t ions conta in ing DTT may be super f luous. s ince the la t ter is
probably a very s t rong chela t ing agent in i ts own r ight .
Enzymic Ac t i v i t y
The act iv i ty o f an enzytne to be used in a syr t thet ic react ion determines the
amount of enzyme required to achieve synthesis of a given amount of product in
a set t i rne. For cal ibrat ion, I mg of an enzyme having specif ic act ivi ty of about
700 international units (1. U.) wi l l catalyze the formation of I mole of product
per day, when operating at maximum velocity (a specif ic act ivi ty of I I . U. = I
trrnrol substrate transformed/(min) (mg) of protein at V^^*) [45] The specif ic
act iv i ty o f enzymes var ies wide ly . Table 7 .1 l is ts spec i f ic act iv i t ies o f enzymes
catalyzing reactions representative of those that might be useful in organic
r- F-w . f ,
tt- tr-v t$
\o ca* O .\ r +tr-
()!
-6
!? 3 er rlO Y a h ^ 6
c e ; { E1 J ) ^ u O u . 1 )5 0 Y o o ' J b 0 2
o , l 3 y I = 3 €E ! 5 s - ! x!
o S 5 - - d - X3 1 T ; 6 . ( ) 6 . Y- ] E , ( J € ) i t J
ootr-
E
!
.J)
o o )a
q d
< . e
-
-
=
l-. 1; l7 l2 l
I
Ifl
^-.?
U
I" la l- - . I
2lIIl l -
I
U,J
-
s-\H -
tI
<^)'x-\ -X
( )H. -
-N
3'' )
(
( )
It\ /t\\/
hl- )ol\
( )
( \
) (
U
ca
U
I- ^ l
7 lz l
-j7
aa
-
\q)
\
Q
\U
d \
6
q
q)
a-
\N
-
\l*.
a
N
t\
a
r-
H
a
fr
F
er
V)
CJa
L
c)
-()
60)
N
r r )
ak
h
i.)c)
a
d
Ia
iN
€
9 1 2
tr-\f,
clc.l
oov
co
t--$
t/)(r)
N
c{
N--1
v \\(,d-
I
III-^ |
J ) |
2 l^ l- l
III
I
ad
v
f , x
3 ' :U _
. e )' . a
> j^t r xq ) o
! oJ ^
€ Y .d = ' J_ L ^ )!
i u -
u b o 1 3z I = yF n 6 . -'J
? E U A- . I' J i
( n ' A O - J- a
d o l- ^-
\/ Y ). . . .g . . ' \ /
N 1 t \ - - l
\, .{= I"l+
IIl El c lr -
U
9 i 3
c{-, ) .
l-= cOv v
tr-.d-
a l
I^ lr l
i l d
T , Z
)c /
ac{
d v
=Z ^
III,/
II
IT
ZI EI A
l \ J o
^?-tE, ;\-' 9 -/al - ----\-
t v\ J F
d
5
h ia
+I
aq
c t \ J
c{
Z-r
=--\,,oI
/)I
: E Io l +
_ + i
- 4 -It -
T . I ' '
., )-== :-\
t^ -)<: - 1 " * ( tJ l r : . -
r v
*A - r \ d
. o -H
r )v
a{-U
Z E
qJ
t:\i)
\iJ
iJiJa
4
ar
N
.rl
-N
\a\
V)
N
a \
a
Cfi
a
q)
ti
F
.)
(n
a
L
'13
-
6
EN
: r l
a-)
.)
i)
V)
I!-
N
9 1 4
t//)
rn
ii
tr-s
v
;
o,w
s
c)
-5a-
E
X 9
A E
ic{
PIL
-I
+Il c
c I T^ l *J I Z
l r
I
T
o-Q
'{v +
Y fE- j -2
td
9 1 6 L A R G E S C A L E E N Z Y M I C S Y N T H E S I S
synthesis. Many of these activi t ies are for impure enzymes. and activi t ies forpure enzymes may vary widely depending on the source. Moreover, several ofthese transformations may involve more than one step. These data should thusbe considered only as lower l imits. Nonetheless, the total range of specif icactivi t ies shown covers almost lAe. Enzymes are known that are appreciablys lower than the s lowest l is ted. l t i s d i f f icu l t to draw genera l izat ions concern ingwhich types of transformations are l ikely to be fast and which slow. Protontransfer, hydration and dehydration, and phosphate transfer are often fast;individual steps in the biosynthesis of complex molecules may be very slow.Clear ly , the pract ica l i ty o f any scheme for synthes iz ing apprec iab le quant i t ies o fa substance using enzymatic catalysis rrust take into consideration the activi tyof the enzyme'. enzyme-catalyzed reactions are. of course. faster than un-catalyzed reactions, but they may st i l l be very slow.
Co fac to r Requ i r emen ts
The requi rernent o f rnanv b iosynthet ic react ions for cofactors p laces i rnpor ' -
tant econor l ic rest r ic t ions on the use of these react ions fur la rge-sca le s l n theses(Chapter I o f Par t I o f th is book out l ines the funct ions o f most o f the comrnon
cofactors) . A l though cer ta in cofactors (CoA. th iarn ine pyrophosphate. b io t in .
te t rahydrofo l ic ac id , 812) act as t rue cata lys ts- that is , they are regenerated
unchanged at the conc lus ion of the react ions in which they are invo lved others(NAD*, NADP.. nucleoside tr iphosphates, f lavins, pyridoxal phosphate) are in-
volved as stoichiometric reagents. Al l of these cofactors are expensivr ' . For
small-scale syntheses in whicl i the costs of reagents is not a major col.)ccnt.
cofactors can be added in quant i t ies s to ich iometr ica l ly equiva lent to those o l ' the
star t ing mater ia ls , and cons idered to be consumable reagents . For larse-sca le
synthes is . i t w i l l be necessary to recyc le the cofactors to ach ieve an eco l ' lo l t r ica l lv
acceptab le cost .
Processes for cata ly t ic regenerat ion o f two c lasses of co lactors . nuc leos idet r iphosphates ar td NAD(P)H der ivat ives, have been developed to the Stuge ot '
successfu l laboratory demonst ra t ions. The most impor tant o f the f l rs t o t ' thcsec lasses is ATP. which is in some sense the b iochemica l equ iva lent o f DCC' t r rtosy l ch lor ide, that is . phosphate or adeny la te t ransf -er f iom ATP to ox l 'ser rcen te rs se rves t o ac t i va t c oxygen as a l eav ing g roup [ - 51 I Fo r exan tp le . ca t -boxy l g roup ac t i va t i on can p roceed by e i t he r pa th [ 5 1 .52 ] . and a ve ry , ' w idc 'range of impor tant substances-prote ins, carbohydrates. nuc leot ides. r ruc le icac ids, terpenes, and others- requi re ATP at one or severa l s tages in the i r b iosvn-thesis. NAD./NADH and NADP./NADPH are similarly widely used as hycir idedonors and hydr ide acceptors .
A usefu l scheme fbr ATP regenerat ion must combine two features: i t must beable to accept e i ther ADP or AMP as the s tar t ing rnater ia l , s ince e i ther or bothmay be generated in a par t icu lar b iosynthes is . and i t must u t i l i ze an inexpens ive
RCZ\o-
3 N E W P R O C E S S E S 9 1 7
ool r r l
RC-O-P -O_ + ADPIo-
ot l
RC-OAMP +
ool l l l-oPoPo-t lo- o-
substance as phospliorylating agent. The most highly developed schemes for ATP
regeneration are based on the coupled activity of adenylate kinase [53] and
another kinase uti l izing the ultimate phosphate donor as substrate. In one regen-
erat ion scheme. X-P is acetyl phosphate. generated by react ion of ketene with
AMP + arp {*q IADP
2ADP + 2X-P I kI"*^
2ATP + 2X
phosphor ic ac id [54 ] , and X k inase is ace ta te k inase [55 ] ; in the second X-P is
carbamyl phosphate. produced by react ion between potassium cyanate and
phosphate. here. carbamyl phosphokinase is used as catalyst [56] .The est i rnated
cost of regenerat ing ATP ( in very large cluant i t ies) by a routc based on acetylphosphate is less than S1.00 per pound [57] ;detai led economic est imatesbased
o
\
on carbarny l phosphate have not been made. S imi lar processes could be used t t l
regenerate nuc leot ide t r iphosphates o ther than ATP. or ATP could be used.
together wi th appropr ia te phosphot ransferases. to e f tec t these phosphory la t ions.
H.C
nCH - :C :O
- C H ,H.
of l ) H ,PO, l l
f f ic l t3coPo'r-
ol l ,Po,t-
/ - - ' /I t2N \- /
-N:C:O + HPOa2- J
9 1 8 L A R G E S C A L E E N Z Y M I C S Y N T H E S I S
X D P + A T P + X T P + A D P
Schemes for regeneration of cofactors in the NADPH (NADH) series rely on analcohol dehydrogenase to catalyze hydrogen transfer [58]
o
-/\'-\ * /
|,,o.tJ- o
XOH OH
Many prob lems, not the least o f which is that o f coupl ing the cofactor regener-at ion step to the enzymic synthesis step(s), rernain to be solved before processesfor regenerating ATP and NAD(P)H derivatives can be used direct ly for large-scale synthesis. Nonetheless, many of the fundarnental uncertaint ies in t l-reseschemes have been reso lved, and i t seenls c lear that these cofactors can beeconomica l ly regenerated in quant i ty , i f requ i red for a large-sca le synthes is . Noat tent ion has been devoted to the prob lems of regenerat ing o ther cofactors . andthe practical i ty of these regenerations remains to be demonstrated.
R e a c t o r D e s i g n [ 5 9 ]
Enzymes to be used as catalysts in biosynthetic reactors wil l be expensive. Abiosynthetic reactor using cell-free enzymes as catalysts should be designed toconserve enzymrc activity: specifically, a continuous process should retainenzyrnic activity in the synthetic reactor with high efficiency, and a batcfiprocess should permit easy separat ion of products f rom catalysts. In ei ther type.the reactor should be designed to permit maximum stabi l izat ion and operat inglifetime of the enzymes. Two strategies are available for design of biosyntheticreactors: either the enzymes may be irnmobll lzed on an insoluble support, or t lrel 'may be used in solution, and reisolated from products (by ultrafi ltration, nricro-encapsulat ion, or possibly by precipi tat ion). Of the two, i rnmobi l izat ion seeptsthe more ef fect ive approach when appl icable:not only are problems of isolat ionminimized by immobilization, but imrnobil ized enzyrnes automatically enjoy rhigh degree of protection against hydrolysis by proteases and are subjecteclneither to the potential for shear and surface degradation present during ultra-fi l tration, nor to the rigors of precipitation or microcapsule formation. Thesubject of enzyme immobilization is discussed in Chapter IX of this volume; hereit is worthwhile to make only a few general points about the characteristics
H 2
JJ . C H 3 C H 2 O H ---
H H O
r#rn,\ n /
- + c H 3 c H o
[,o{t'l\ /
/1OII OH
3 N E W P R O C E S S E S 9 1 9
requi red of a suppor t for immobi l iza t ions for la rge-sca le synthet ic reactors . The
suppor t should present a ba lance of hydrophi l ic and l ipophi l ic character that
perrn i ts maximum stab i l iza t ion o f the act ive confbrmat ion of the enzyme: i ts
funct iona l izat ion s l iou ld be s t ra ight forward. and the coupl ing o f the enzyme to
the suppor t should idea l ly preserve the charge type of the amino ac id invo lved
(usuai ly the 7-arn ino group of lys ine) and produce a permanent l ink between
enzyme and suppor t ; the suppor t should have good mechanica l s t rength and
t low character is t ics for use in large co lumns and s t i r red or f lu id ized reactors : i t
shguld perrn i t rap id access of s tar t ing mater ia ls to the enzy l re . and rap id d i f fu-
s ion of the product f ront the enzyrne: i t should not be subject to tn ic rob io log ica l
degradat ion, and, f or cer ta in appl icat ions. should to lerate s ter i l i za t ion: i t should
be inexpens ive.
These character is t ics are. o f course, a lso des i rab le for suppor ts to be used for
i rnrnobi l iza t ions on a snta l l sca le . l t seems rnost un l ike ly that a s ing le suppor t
wi l l ever serve every need adequate ly , and i t is d i f f lcu l t to judge a pr ior i the best
suppor t lb r a new appl icat ion. Nonethe less. cer ta in d i f terences between large-
and smal l -sca le processes are common. For one. the t ra ture o f la rge-sca le pr t l -
cesses p laces a greater prenr ium on the mechanica l and f low charaoter is t ics o f the
suppor t . These cons iderat ions a lone are probably suf f lc ient to exc lude the cross-
l inked agarose and dext ran ge ls commonly used in snta l l -sca le work f rom wide
appl icat ion in large-sca le processes. Second. s ince the enzyme preparat ions used
in large-sca le b iosynthes is wi l l p robably be less pure than those in srna l ler syn-
theses. the capac i ty o f the suppor t is o f greater inrpor tance in the former than
the ia t ter . S ince the act iv i ty per un i t o f ' reactor vo lu tne t l - ra t can be ach ieved wi th
any enzytne preparat ion wi l l depend upon the suppor t capac i ty , and s ince both
the expense o1- the reactor and i ts operat ing character is t ics (par t icu lar ly pressure
drop across large co lur r rns) depends upon the s ize requi red. there is a s t rong
econor l ic incent ive to work wi th h igh-capac i ty suppor ts . F ina l ly . s ince the l i fe -
t ime of the enzyme uuder operat ing condi t ions is cent ra l to large-sca le pro-
cesses. se lect ion o f a suppor t and coupl ing procedure that rnax in t izes th is para-
r n e t e r i s i r t t p o r t a n t .
There is present ly no co lnmon agreement concern ing the best suppor ts fbr
large-scale inrrnobi l izat ions. In the relat ively rare appl icat ions in which pl iysical
adsorpt ion on an inexpens ive. h igh-sur face area inorganic (d ia tomaceous ear th)
or organic (DEAE ce l lu lose) suppor t is successfu l . th is procedure wi l l p robably
be d i f f icu l t to beat . In the more f requent appl icat ions in which the enzyme
must be rnanipulated more careful ly. organic polyrner supports derived frorn
vinyl ntonomers seem l ikely to prove rnost useful. Supports derived frotn vinyl
po lymer izat ion have the combined advantages of re la t ive ly low cost and great
l lexibi l i ty in propert ies. Thus. polyacrylamide crossl inked with,A/,rv'-methylene-
bisacrylamide provides a hydrophil ic polymer with fair but not outstanding
rlechanical propert ies. Inclusion of rnore hydrophobic rnonomers (2-hydroxy-
O\ f--\- O
>-d v-J
\ \>*i l_J >r-6
O)-
CH)*HtcH2)6N"<:
ocH3
J
l ink ing with nrore hydrophobic mater ia ls (hexanrethylene ,V,M-bisacrylarnide)may provide i rnproved structural and nrechanical propert ies, a l though possiblywith some decrease in sui tabi l i ty of the resul t ing polyrr ier matr ix as a support .Coupling of the enzyme to the gel can be accornplished either by inclLrdingappropriate rnonomers (glycidyl methacrylate. rV-acryloxysuccinirnide) in theini t ia l polymerizat ion or by funct ional izat ion af ter polyrner izat ion. Sinceacrylic acid, rnethacrylic acid. and acrylarnide are all readily available andsubject to faci le der ivat izat ion of a wide range of types, i t should be possible totailor the resulting polymers to have those properties most appropriate lbr apart icular enzyme.
Of the other n-raterials commonly considered as potential supports lbr largc'-scale reactors, only the control led-pore ceramics recent ly introduced by Corningseem likely to be useful [60] . Controlled-pore glasses from the same sourccseem too expensive to be useful [60. 61].
A L T E R N A T I V E S A N D P R O S P E C T I V E S
Expe r imen ta l p rocedu res f b r enzyn re i so la t i on . s tab i l i za t i on , and i n rn rob i l i zu -
t ion are suf f ic ient ly advanced that i t is pract ica l to cons ider cata lys is by ce l l - t ' reeenzymes as an in tegra l , i f re la t ive ly unexplo i ted. technique in in termediate- andlarge-scale organic synthesis. Given the avai labi l i ty of the technique, i t is inter-esting to ask in what types of reactions enzymic catalysis rnight be useful. Thisarea requi res a sophis t icated b lending of techniques f ror r i organic chemist ry ,rr icrobiology, enzymology, and polymer chemistry. I ts technical cornplexity issuc i r that i t w i l l on ly be usefu l when cornpet i t ive "s tandard" synthet ic tech-
9 2 O L A B G E . S C A L E E N Z Y M I C S Y N T H E S I S
ethyl methacrylate, glycidyl methacrylate,methyl methacrylate), or cross
o1\H{H.,
_/
o)- o .oH
4L/C H .
o
4 A L T E R N A T I V E S A N D P R O S P E C T I V E S 9 2 1
niques, conventional organic Synthesis and fermentation, are unsatisfactory'
Organic synthetic procedure has the virtue of great f lexibi l i ty. I t is, however,
notab ly unsuccessfu l when appl ied to prob lems requi r ing the se lect ive manip-
Lrlat io' of sint i lar functional i t ies, and part icularly for materials that are soluble
in water and related solvents. Thus, the modif icat ion and synthesis of carbohy-
drates. nucleotides, and certain types of polypeptides is seldom easi ly accom-
pl ished by organic synthetic techniques. Further, the procedures developed for
low-molecular-weight substances are not adequate for easi ly synthesizing and
characterizing proteins, nucleic acids, and other biological macromolecules'
Femrentation often offers a very successful approach to these types of problems'
It is dif f icult to improve on a successful fermentation for simplici ty ' Fermenta-
t iol suff-ers. however, from disadvantages of i ts own. I t is intr insical ly a slow and
inf lexible technique: the start ing materials that are acceptable for a f-ermenta-
t ion. anc l t l ' re products o f the fermentat ion, are re la t ive ly f ixed by the tnetab-
o l is r r - r o f the microorganisrn . l t i s somet imes d i f f icu l t to iso la te a par t icu lar
transformation of interest from the rnyriad other reactions usually taking place in
the cel l . and. in consequence, improving a f 'ennentation yield nray be a very
dif f icult task. Fermentation is also intr insical ly ineff icient. in the setrsc that the
part icular material of interest in a fermentation is usually only a small fract ion
of the cel l 's synthetic output. Part icularly when fermentation yields are low.
the disposal of fermentation resiclues may present a signif icant problertr. and
other aspects of fermentation processes present environmental problerns.
Enzymic synthesis offers a colnpromise between fermentation and chenlical
synthesis having dist inct potential advantages. I t affords higher selectivi ty than
chemical synthesis, but is lnore amenable to control and modif icat ion than
fermentat ion. I t permi ts s ing le react ions or react ion sequences in cot l tp lex n te ta-
bo l ic pathways to be iso la ted and ut i l i zed. ar - rd pronr ises h igher y ie lds . h igher
pur i ty . and greater ease of iso la t ion than e i ther '1-er t t tenta t io l l o r c l ie r t r ica l syn-
thes is . E lzyrn ic react ions are run at roo l l t teu lperature in water : e t lzynt ic
cata lys is thus combines the at t rac t ive process character is t ics o f fer t t te t t ta t io t t
w i th the e f f ic iency of chern ica l synthes is . a l though at the cost o f h igh technrca l
complex i ty .
Large-sca le enzyrn ic synthes is ho lds prorn ise in a number o f areas o1- t inc
c6e r r r i ca l s s -Vp thes i s , pa r t i cu la r l y i n a reas re l e l an t t o b i c l l og i ca l l y ac t i v c ' c t r t l t -
pou lds. Spec i t ic areas in which researc l t cot r ld produce usefu l sv l l thet ic
inethods inc lude these :
l . Synthes is o f nuc leot ides and nuc leos ides f ro tn sugars and pur ines. pyr i r t r i -
t l ines. or analogs lr ight be carr ied out eff iciently using enzynles. The avai labi l i ty '
of enzymes frorn the so-cal led "salvage patl-rways" that accept the prelortrtcd
l i t roge l heterocyc les bypasses many of t l te s teps in the main b iosynthet ic
pathways [62] , and would capital ize on the abi l i ty of cel l- free enzyrnic synthesis
to se lect one, minor . enzyrn ic pathway f iom atnong the large nu lnber o f paths
9 2 2 L A R G E S C A L E E N Z Y M I C S Y N T H E S I S
competing in vivo fbr the same substrate. Unlike chemical syntheses. l ' lo pro-tect ive groups would be involved in the enzyrnic syntheses.
2. Stero id nrod i f lca t ion, par t icu lar ly ox idat ive funct iona l izat ion, w i l l p rob-ab ly present unusual and d i f f icu l t prob lerns in iso la t ion and s tab i l iza t ion o f theenzy t l t es i nvo l vcd [ 63 l .T l t e a rea i s none the less one o f g rea t p rac t i ca l i r npo r tance ,and fermentat ion methods. a l though extens ive ly developed [6a l . do not a lwaysproduce the desired products in high yield or puri ty.
3 . Asynmetr ic and s tereoselect ive syntheses cxp lo i t the ab i l i ty o f e nzyrncsto catalyze the formation of optical ly act ive products. This area is ful lydescribed in Chapter IV of Part I of this book.
4. The synthes is o f nonr ibosomal pept ides appears poss ib le [65] An exarnp leof an impor tant member o f th is c lass o f compounds that is w i th in reach ofpresent tec l tn iqLres is the an inra l f i rod ant ib io t ic bac i t rac in [66] The b iosyn-theses of pen ic i l l in and cephalospor in are not present ly unders tood, and thelarge-scale cel l- free enzymic synthesis of these types of nraterials l ies in thefu ture. Stud ies d i rected toward the s impler but re la ted ant ib io t ic Grarn ic id i r rS have been descr ibed [65]
5. In termediates usefu l in the synthes is o f semisynthet ic ant ib io t ics arepresent ly obta ined by cherr r ica l or enzyrn ic degradat ion o f f 'e rn tentat io l t prod-uc t s . I n p r i nc i p l e , i t shc lu l d a l so be poss ib l e t o p roduce t hesc o r con tp len ten ta rymaterials by cel l- free enzyrnic synthesis. Such reactions would require a f lrrnknowledge of the b iosynthes is o f the target ant ib io t ic . and would in ter rupt th isb iosynthes is by us ing on ly those enzymes requi red to synthes ize in ternrec l ia tesa long the path to the product obta ined by fer rnentat ion 1671. The ava i lab i l i tyo f in termediates to penic i l l ins , cephalospor ins , or the aminoglycos ide ant ib io t icswould be par t icu lar ly usefu l , and the enzyrn ic synthes is or subsequent modi f ica-t ion o f these in ter rnedia tes might present an eas ier prob lern than the to ta l syn-thes is o f the complete ant ib io t ic .
6 . Carbohydrates are par t icu lar ly d i f f icu l t to modi fy chern ica l ly . A largenumber o f enzyrn ic t rans lbrmat ions have recent ly been cata logued [68] . butmost have not been exp lo i ted for synthes is .
7 . A less spec i l ic app l icat ion o f enzymic synthes is L ies in the area of drugmetabolism. Tlte usual approach to the isolat ion of drug metaboli tes is to f 'eeclthe drug to an appropriate animal, and to isolate metaboli tes from blood orur ine. These iso la t ions o f ten present severe exper imenta l d i f f icu l t ies . A s i r r ip ler(a l though not ent i re ly equiva lent ) pre l iminary approach is to examine t rans-format ion of the drug of in terest by enzynes or enzyrne mix tures be l ievec l tobe invo lved in metabol ism (e.g . , c rude ox idase preparat ions f rom l iver ) . Themater ia ls generated by such a procedure would probably , a t least , be re la ted tothose formed in vivo, and their isolat ion would be less dif f icult .
8. I t may prove possible to use enzymes ordinari ly involved in hydrolysis tcrform peptide, phosphodiester, or related bonds by changing solvent or other
4 A L T E R N A T I V E S A N D P R O S P E C T I V E S 9 2 3
environmental pararneters. These hydrolyses are reversible, and the practical
problems to be solved in using enzymes Lo catalyze the reverse reactions are
those of stabi l izing the enzyme under the condit ions necessary to run the
reaction, deal ing with the substrate specif ici ty of the enzymes, and preventing
mult iple addit ions through suitable blocking groups. Several examples of
react ions o f these types have been repor ted [9 , 69] -g. Ribosomal peptide synthesis of proteins represents a possibi l i ty for the
distant future. Elegant studies by Schechter have pointed to a general procedure
tor isolat ing specif ic in-RNA's from eukaryotic cel ls, by precipitat ing the ir ' t -
RNA, together with associated r ibosomes and growing polypeptide ohains, using
ant ibod ies against these po lypept ides [70] .These m-RNA's can then. in pr in-
c ip le . be used in combinat ion wi th mix tures o f r ibosomes, / -RNA's , a tn ino ac ids.
and nucleoside tr iphosphates to generate proteins. This system would clearly be
a dif f lcult one to develop for practical synthesis, but for suff iciently valuable
prote ins (e .g . , human growth hormone) , i t might prove very usefu l '
' These and similar subjects al l are described in terms of cel l- free enzymes.
Progress in large-sca le enzymic synthes is wi l l a lso depend on two re la ted and
complernentary sub jects . The f l rs t is the developrnent o f n tore genera l tec l tn iq t res
for synthes is us ing in tmobi l ized whole ce l ls . As present ly pract iced. th is tcch-
l ique wi l l p robably be l i rn i ted in i ts appl icat ion to re la t ive ly s i rnp le react io t ts .
Combined wi th the procedures developed to protect . immobi l ize . and s tab i l ize
cel l- free enzylnes, i t might have very great usefulness by providing amethod of
circumventing the labor required to isolate and (possibly) stabi l ize cel l- free
enzymes. Alternatively, a compromise between the use of immobil ized enzymes
and immobil ized whole cel ls-synthesis using, S?y, irnmobil ized intact mito-
chondria or membrane fragments-rnight offer important advantages. The second
is the ref inement of microbiological techniques for isolat ing andior mutating
rrew strains of organisms capable of carrying out new types of transfbrmations.
One important advantage of synthesis using immobil ized enzymes relat ive to
synthesis by ferrnentation is the abi l i ty of the former to take advantage of
relat ively unimportant metabolic pathways. Thus, i f microorganisms can be
developed that show detectable activi ty of a part icular type, isolat ion and
stabi l izat ion of the enzymes responsible oft 'ers a possible synthetic route even i f
the enzymes are present a t such low concent ra t ions that fermentat ion would not
be p rac t i ca l .
Acknowledgment
Many of the ideas incorporated in to th is chapter were generated in d iscuss ions
wi th rny co l leagues in the M. l .T . Enzyme Technology Group, par t icu lar ly
Professors D. I . C. Wang, C. K. Col ton, A. Demain, and A. S insky, and wi th
Drs. A. Ramel and A. H. Nishikawa, of the Biopolymers Laboratory of
Hoffmann-LaRoche.
9 2 4 L A R G E S C A L E E N Z Y M I C S Y N T H E S I S
REFERENCES
l . Rev iews: E. K. Pye and L. B. Wingard, J r . , Enzyme Engineer ing, YoL2,Plenum Press, New York, l9 l4 ; B. Wolnak , Present and Future Technolog-ical and Commercial Status of- Enzymes, Bernard Wolnak and Associates,407 South Dearborn St . , ch icago, I l l . , 1972; A. wiseman, E,d. , I lancrbookof Enzyme Biotechnology, Wi ley- In tersc ience, New York, 1975; K. J .Sk inner , Chent . Eng. l t lews, (August l8) 23 (1975) .
2 . K. L . Agarwal , A. Yamazaki , P. J . cash ion, and H. G. Khorana, Angew.Chem. I n t . Ed . Ens . , 11 , 45 | ( 1912 ) .
3 . Rev iews : N . F . o l son and r . R i cha rdson , , l . F -oo t l s c i . , 39 ,653 (1974 ' ) ,w .L . S tan ley and A . c . o l son , i b i d . , 39 ,660 (19 j4 ) ; I . ch iba ta , T . Tosa , T .Sato, T . Mor i , and Y. Matsuo, tn Proc. Four th . In tent . Ferrnentot io t tSymp. , Soc. Fermentat ion Technol . , Osaka, Japan, 197 2; B. Wolnak.Repo r t PB-219636 , Na t i ona l Techn i ca l I n fo . Se rv i ce , U .S . Dep t . Commerce .Sp r i ng f i e l d ,Ya . 22151 , 197 ) . : A . c . o l son and c . c . cooney , lmmob i l i zec lEnzyvlgs in Food ortd l l l icrobial Processes, plenum, New york, 1974.
4 . T . T o s a , T . M o r i , N . F u s e , a n d I . C h i b a t a , L ' n z y m o l o g y , 3 l , l l 4 . l l 5( 1 9 6 6 ) . 3 2 . 1 5 3 ( 1 9 6 1 ) , B i o t e c h . B i o e n g . , 9 , 6 0 3 ( 1 9 6 1 ) . A g r . B u t l . C l t t : n t . .3 3 , 1 0 4 7 ( 1 9 6 9 ) : T . T o s a . T . M o r i , a n d I . C l h i b a t a . i b i d . . 3 3 , 1 0 5 3 ( 1 , ) 6 e ) :E n z y m o l o g y , 4 0 . 4 9 ( 1 9 7 1 ) : T . l ' o s a . T . M o r i , a n d I . C h i b a r a . l l u k k oKogoku Zass l t i , 49 ,522 , ( l 9 l l ) . T . Sa to . T . N {o r i . T . ' f osa . an t l I . ( ' h i l r a r r .Arch. B i r . tche nt . B iopht ,s . , | 47 . 7 88
5. H. H. Weetal. Foorl Protl . DeveloSt.,and M. D. L i l ly , B io tec l t . B ioeng. ,
) .( 1 9 7 3 ) : D . W a r b u r t o n . P . l ) u n t r r l l
1 3 ( 1 9 1 3 ) . f r . A c e v e d o a n d ( ' . L .
t97 |, 4 65 .
C o o n e y , i b i d . . 1 5 , 4 9 3 ( 1 9 7 3 ) .6 . I . ch iba ta , T . Tosa , and r . Sa to , . 4pp t . , r l i c rob io l . , 27 , g7g (19 i4 ) .7 . T . T o s a . T . s a t o , a n d I . c h i b a t a , A p p l . r l i c r t b i o l . , 2 7 . g g 6 ( 1 9 1 4 1 .8. M. Friedrnan, Chemistry und Biochemistr l , of the Sulptty,r)rS,l ( ;) .oltp,
Pergamon. New York, 1973. P. C. Joce lyn,B iochern is t r l ; c . tJ the SI I Orot tp .A c a d e m i c P r e s s , N e w Y o r k , 1 9 1 2 .
9 . C . J . S u c k l i n g a n d K . E . S u c k l i n g , C l t e m . S o c . R e v . , 3 . 3 g 7 . l 9 7 4 ) .
10. H. F. . St reet , F .d . , P lant T issue ond Cet t Cul ture , B lackwel l . Oxfor r l .Eng land , 197 3 .
I l . C . Mossback and P . D . La rson , B io tech . B ioeng . , l Z , 19 (1970 ) .12. K. Venkatasubramanian, R. Sain i , and w. R. V ie th .J . Fermentat ior t T ' t , t ,h . .
5 2 , 2 6 8 ( 1 9 7 4 ) .
1 3 . S . J . U p d i k e a n d D . R . H a r r i s . t \ o t u r e , 2 2 4 . l l l l 0 9 6 9 ) .14. W. B. Jacoby. Ed., Xlet l tods in L'nz1,rt tolog1,; Enz1,me puriJicatiort urtt l
Re la ted rechn iques , Vo l . 2 l , Acaden r i c p ress . New yo rk . l 97 l .15. M. Dixon and E. C. Webb, I : - r ' tz rn tes, Acadern ic Press. New York. 196.+.
Chap te r I I I .16 . Rev iew : A . w i seman , P rocess . B iochen r . , 8 , 14 (Augus t l 913 ) .17. E. Mihalyi, Application o. l- Proteol l ,slc Enzymes in protein Struc,turt,
Stud ies, CRC Press, C leve land, Ohio, 1972.
R E F E R E N C E S 5 2 5
18. G. E. Per lman and L. Lorand. Eds. , Methods in Enzymology: Proteo ly t ic
Enzymes , Vo l . X IX , Academic P ress , New Yo rk ,1970 .
19. P. D. Boyer , Ed. , The Enzymes: Pept ide Bond Hl tdro lys is , Vo l . I I I .
Academic Press. New York . 197 l .
10. N. R. Lazarus, A. H. Ramel , Y. M. Rustum, and E. A. Barnard, B iochent -
is t ry , 5 , 4003 (1966) .
I 1 . H. Gor ish and F. L ingens, B io che mis t ry , I 3 , 3190 ( 197 4) .
l l . J . R . P r i ng le , B iochem. B iophys . Res . Comrnun . , 39 ,46 (1970 ) .
13. D. E. Fahrney and A. M. Gold, J . Amer . Chem. Soc. , 85, 997 (1963) .
)4 . F . Markward t , M . R i ch te r , P . Wa lsmann , and H . Landmann , FEBS I ' e t t . , 8 ,
170 (1970 ) and re fe rences c i t ed t he re in .15 . H .
' I s chesche , Angew. ( . hem. I n t . Ed . Eng . , 13 , l 0 ( 1914 ) .
26. O. R. Zaborsky, lmrnobi l ized Enzynzes, CRC Ptess, C leve land, Ohio,1913.
)7 . J . M . R i f k i nd , B iochemis t r y , 13 , 2415 11914 ) .28 . W. W. C le l l and , B iochemis t r y , 3 ,480 (1964 ) .
29. L . l . Rothf ie ld and D. Romeo, in St ructure and Funct ion oJ B io log ico l
] l embranes , L . I . Ro th f i e l d , Ed . , Academic P ress , New Yo rk , l 9 l l , Chap te r
6 .30. A. Adamson. I 'hys ica l Chemist ry o f Sur foces,2nd ed. , Wi ley- ln tersc ience.
New Yo rk ,196 l ;B . W. Mor r i ssey and R . R . S t romberg , J . Co l l o i d . I n te r -
f a c e 9 c i . , 4 6 , 1 5 2 ( 1 9 1 4 ) ; D . J . L y m a n , A n g e w . C h e m . I n t . E d . [ ' r t g . , 1 3 .
r 0 8 ( t 9 7 4 ) .31 . N . E . Good , G . D . W inge t , W. W in te r , T . N . Conno l l y , S . I zawa , and R . t { .
H. S ingh, B iochemist ry , 5 , 47) (1966) .
32. W. P. Jencks, Coto lys is in Chemist ry and Enzymolc tg l ' , McGraw-Fl i l l . NewYork , 1969 .
33. C. Tantord. The Hydrophobic Ef fect , Wi ley- ln tersc ience, New York , 1913.34. W. B. Dandl iker and V. A. de Sanssure. rn The Cl tent is t r , t , t t l 'B iosur . foces,
Vol . I , M. L . Hai r , Ed. , lV larce l Dekker , New York, l9 l l , Chapter l .3 5 . P . H . v o n l { i p p e l a n d T . S c h l e i c h , A c c t . C h e m . R e s . , 2 , 2 5 7 ( 1 9 6 9 ) .36. P. H. von Hippel and T. Schle ich, tn St ructL t re und Stab i l i t . t , oJ 'B io log ico l
i l lacromolecu les, S. N. T imashef f and G. D. Fasman, Hds. , Marce l Dekker ,New Yo rk , 1969 , pp . 411 f f .
3 7 . J . J . O ' M a l l e y a n d R . W . U l m e r , B i o t e c h . B i o e n g . , 1 5 , 9 1 1 ( 1 9 7 3 ) ; J F .C l l a r k and W. B . Jakoby , J . B io l . Chem. , 245 . 6072 ( 1970 ) .
3 8 . N { . J . R i r v a r t a n d C . H . S u e t t e r , J . B i o l . C h e m . , 2 4 6 , 5 9 9 0 ( 1 9 7 1 ) .
39 . S . L . B radbu ry and W. B . Jakoby , P roc . Na t l . Acod . Sc i . U .5 . , 69 . ) 313( 1 9 1 J ^ ) .
4 0 . A . D . E l b e i n . A d v a n . E n z y m o l . , 4 0 , l 9 ( 1 9 1 4 ) .
41 . J . N . Umbre i t and J . L . S t rom inge r , P roc . Na t . Acad . Sc i . U .5 . , 70 ,2991
0 e 1 3 ) .4 2 . C . B l i n k h o r n a n d M . N . J o n e s , B i o c h e m . J . , 1 3 5 , 5 4 7 ( 1 9 7 3 ) ; F . I l .
K i rkpat r ick , S. E. Gordesky, and G. V. Mar inet t t , B ioch im. B iophys. Acta,
3 4 s . 1 5 4 ( t 9 7 4 ) .
4 3 . R . J . P . W i l l i a m s , Q u a r t . R e v . , 2 4 , 3 3 1 ( 1 9 7 0 ) ; C . H . S u e l t e r , S c i e n c e , 1 6 8 .
9 2 6 L A B G E - S C A L E E N Z Y M I C S Y N T H E S I S
7 8 9 ( 1 9 7 0 ) .44. A. Martel l and L. G. Si l len, Eds., Srabi l i t l t Constants, Special Publication
No . 17 , Chemica l Soc ie t y , London , 1964 .45. Repor t o f the Commiss ion on [ . -nz1tmes, l .U.B. Symposium Ser ies, Vo l . 20,
Pe rgamon P ress , New Yo rk , 1961 .46 . P . Ta la l ay and J . Boye r , B ioch im . B iophys . Ac ta , l 0S , 389 (1965 ) ; p .
Ta la lay e t . a l . , X le thctds in Enzymology, Vol . XV, R. B. Clayton. Ed. ,A c a d e m i c P r e s s , N e w Y o r k , 1 9 6 9 , p . 6 4 2 .
47. Th. E. Barman, Enzyme Handbook, Vols . I , I I , Spr inger-Ver lag, New york ,
1 9 6 9 .48. Y. Abiko, in l l ' [e thods in Enzymology, Yol . xv I I I , D. B. Mccormick and
L . D . Wr igh t , Eds . , Academic P ress , New Yo rk , l 970 , p . 350 .49. P. D. G. Dean, in Methods in Enzymology, Vol . XV, R. B. Clayton l Ed. ,
Academic Press, New York , 1969 , p . 495.50. D. G. Parkes and T. [ r . E l ing, B iochemist ry , 13,2598 (1974) .5 1 . E . R . S t a d t m a n , i n T h e E n z y m e s , 3 r d e d . , V o l . 8 , p . D . B o y e r , E d . ,
Academic Press, New York , 1972, Chapter l .52 . J . F . Mor r i son and E . l { eyde , Ann . Rev . B iochem. ,4 l , 29 (1912 ) .53 . L . Noda , r n The L -nz t ' n tes , Yo l . 8 , P . D . Boye r , Ed . , Academic p ress , New
York , | 97 2 , Chap te r 8 .54 . R . Ben t l ey , J . Amer . chem. sc t c . , 70 ,2183 (1948 ) ; see a l so R . w . po r te r .
M . O . Modeke , and G . R . S ta rk , J . B io l . Chem. , 244 , 1846 ( t 969 ) ; LHeyde, A. Nagabhushanlam, and J . F . Morr ison, B iochemist ry , 12,47r8(1913 ) ; G . M . Wh i tes ides , M . S iege l , and P . Ga r re t t , . / . O rg .Chem. , 40 ,2 s r 6 ( r 9 7 5 ) .
55. c . R. Gardner , c . K. co l ton, R. S. Langer , B. K. Hami l ton, M. c . Archer .and G. M. whitesides, in l : .nzym.e Engineering, vol. 2,1:. K. pye ani l L. B.W i n g a r d , J r . , E d s . , P l e n u m P r e s s , N e w y o r k , 1 9 7 4 , p . 2 0 9 ; G . M . W h i t e -s i d e s , A . c h m u r n y , P . G a r r e t t , A . L a m o t t e , a n d c . K . c o l t o n , i b i d . . t ) . 2 1 7 .
56. D. L . Marshal l , B io tech. B ioeng. , 15,447 (1973) .51 . C. C. Col ton, pr ivate communicat ion.58 . K . Mosbach and P . O . La rsson , B io tech . B ioeng . , 13 ,393 (19 l 1 ) :M . K
weibe i , H. t t . weeta l l , and H. J . Br ight ,B iochem. B iophys. Res. contmun. ,44 ,347 (1971 ) . See a l so t he sec t i on on coenzyme recyc l i ng , Pa r t I , Ohap te rI V .
59 . w . R . v i e th and K . Venka tasub raman ian , chem. Tech . , 677 (1973 ) ,47 .309 ,434 (1974 ) ; R .w . cough l l n and M . cha r l es , Enzyme Tech . D iges t , 3 .6 9 f i e 7 4 ) .
60 . R . A . Mess ing , Resea rch f Deve lopme t . t t , 25 ,32 (1914 ) .61. H. H. weeta l l and R. A. Mess ing, in The chemist ry o l 'B iosur faces, M. L .
Ha i r , Ed . , Ma rce l Dekke r , New Yo rk , 1972 , pp . 563 f f ; H . H . wee ta l l , F . r l . .Immobil ized Enzymes, Antigens, Antibodies, and Pepti t les; Preporatiort ontlCharacterization, Marcel Dekker, New York , l9j 5.
62. J . F . Henderson and A. R. P. Paterson, . ly 'uc leot ide Metabol ism, AcademicP r e s s , N e w Y o r k , 1 9 7 3 ; T . Y . S h e n , A n g e w . C h e m . I n t . E ' d . E n g . , 9 , 6 7 go970) ; R. J . Suhadolov lk , Nuc leos ic te Ant ib io t ics , Wi ley In tersc ience, NewYork , 1970 .
R E F E R F N C F S S 2 ]
63 . C . A . Tyson , J . D . L i pscomb , and I . C . Gunza lus . J . B io l . Chem. ,247 ,5177(1912) : V. U l l r ich , Angew. Clhem. In t . Ed. L .ng. , I l , 701 (1912) .
64. G. S. Fonken and R. A. Johnson, Chemica l Ox idat ions wi t l t Microorgon-is rns, Marce l Dekker . New York, 1912.
65 . B . K . Hami l t on , J . P . Mon tgomery , and D . I . C . Wang , tn I ' nzyme Eng inee r -i r t g , Yo I . 2 , b , .K . Pye and L . B . W inga rd , J r . , Eds . , P lenum, New Yo rk ,1 9 7 4 , p . 1 5 3 .
66 . K . Ku rahash i , , 4 nn . Rev . B ioche rn . , 43 , 853 (1914 ) .
67 . D . Pe r lman and M . Bodonszky , Ann . Rev . B ioc l t em. , 40 ,449 (197 l ) .
68. A. Herp, D. Hor ton, and W. P igman, in The Carbohyt l ra tes, Vo l . IL , \ .Academic P ress , New Yo rk ,1970 , pp . 836 f f .
6 9 . I I . J e r i n g a n d H . T s c h e s c h e , A n g e w . C h e m . I n t . E d . E n g . , 1 3 , 6 6 0 , 6 6 1 ,662 (1914 ) ; S . Baue r , R . Lamed , and Y . Lap ido t . B io tech . B ioe t t g . , 14 ,8 6 1 , ( 1 9 1 2 ) ; M . B r e n n e r a n d A . V e t t e r l r , H e l v . C h i n t . A c t u , 4 0 , 9 3 7 ( 1 9 5 7 ) .
70 . I . Schecht er , B io chemist ry , I 3 , I 81 5 (1914) .
