wi;gjj yJd'', --*

v ITtrFT rm ^?. 4 ; ! 4 ! ! ! ! ! l | u J

rhase artd. show:he three test d.e-r the range 5-15rhen it i-s noted{ 9 . 0 ; t h i s s u g -;y comparable tor i n h i c

' l i k c l 1 r i . .

i - e . -

l!i!s .

macromolecr.il-arf r l . r i c m a t . a r i o l

Ises tested. anoret systematicall i-rler d.erivativesr Cnambers et eL:eactor stuCieslblems of co-

rm, NationalInst i tutes of

ioRD, G. C. , HACIGRT,, r. E. , AT,LISON ,S . S . P t o c . I a t l .

I. & ROSSl.,{AltN, M.G.t l J .

7. 236:27L6, L96L.iT, L973.?. Bi.ophys. Aeta

\Rrcos, w.H. &

EIZIS,IATIC REGEIIEPATfON 0F ATP IBOM Al'G, AI'ID ADP :

?A-qT I. TIffPMODTNAI'trCS, KINETICS, AD FROCESS DE"/EL0Fl'gl'lT

c .R. Gardner , c .K . co l ton , R.S. La lger , B .K. Han i l ton '

M.C. Archer* s:C G-l{ . \^ Ih i+-esid-es**

Deparbnents of Cher1ical Engineering, Nutri'"ion and.

Food. Science*, d lc. Chenistry**Massachusetts Inst i tute of TechnologfCambri d.ge, Ifas s ach'Js et*u s

Ad.enosine triphosphate (iyp) pieys a prominent role in narry

biosynthetic pathways vherein chenical bond.s are maoe whi-ch other-

rrise worrld. not for:r in s:gnificant quan+.ity in d.i iute aquecus solu-

tion. In such reactions the brealieown of ATP is coupled via a

coxrmon interrned.iate vith synthesis of the new chenical bono. Tle

vork reported here is par, of a coord.j-nated. effort io d.enonstrate

large-sca-le, cell-free enzyoa+-ic synthesis of usel\: l- prod'ucts w:-+-h

sirmr-ltaneous reEeneration of the ATF consuned. in the biosynthe+.lcreact ion.

.A"lternative method5 available ior con'lterting AI'IP or ADF to

AIF includ.e: a) d.irect chenica-l synthesis, b) in uiuo raicrobialconversion using yeast or bacteria-l ferrnentations, c ) photosynthet:-c

phosphorylation using vhole cell s or i,colated. bacterial chrorn'ato-phores, and. d.) cell-free enzlme catalysis, The latter tvo methoCs

offer the ad.vantage of carryi-ng out ATP regeneration and. biosynthesisln the salne reactor or in read.i-ly compatible enrrironments. Photo-

slmthetic phosphoryIation is und.er stu(y by other menbers of our'

$roup, eJrd. we focus here on cell-free enz1nne catalysis.

We are concerned. with three interelated. components (fig. l).gtrnple raw materials are converbed. to more complex prod'ucts ln the

btosynthetic reactor with consunption of ATP and. prod.uction of A-lt'F

of ADP artd. inorgani.c phosphates. kre present moCel systern is syn-thesis of the cycl ic d.ecapept id.e ant ib iot ic, gra 'n i c id. in S, f rom i tsflve constituent amino acid's (f ). The loss of tvo te:mina-l phos-pboryl groups to prod.uce inorganic pyrophosphate and' AI@ is conmontn polypeptid.e, polysacchariCe, l ipid. and. nucleic acid. syntheses.Al'{P is iireo recycled. to an ATP regeneration system which also is

M

209

E N Z Y M A T I CR E G E N E R A T I O N

OF ATP

A T P + A M P = z A D P

B P + A D P : B + A T P

C H E M I C A LOR

E N Z Y M A T I C

S Y N T H E S I S

OFPHOSPHATE

DONOR( B P )

z].)I N O R G A N I '

- - pn-oscari flI?EF ilgI

C.R. GARDNER ET At

S U B S T R A T E S

ENZYMATIC

BIOSYT. I THES IS

WASTE PRODUCT(INORGANICPHOSPHATE )

III

tW A S T E

II

L - -

WASTE

I

I

J

Fig. 1: Enzlnoatic Synthesis with Enzymatic Regeneration of ATP-

fed. rith a phosphate d.onor (gp) synthesized. or regenerated' from

inerpensive raw materials.

Phosnhnr-v lat ion of A-DP is cataiyzed by a var iety of phoslhc-5 l V g V a l v l J * e v 4 v . . v _

t rarrsferases ( f ig . 2) . Conversion of AMP to ADP is achieved. by : : : 'e

reacti.on ca+.alyzeC by aCenylate kinase. ATP regeneration frorn A-I'LP

thus requires two ccupleC enzp.at ic reac+- ionsr wherease regenerat- :c l '

f rom ADP requires a s ingle phosphotral lsferase react ion. For the

latter, casd.id.ate systerns are tabr.[ared. in Fig. 2 for which the

stand.ard. Gibbs free energ:y of nycroJ-ysls of t,he phcsphate oonor is

greater thas that of ATP, So that t,he coupled reacf-ions are d'riven

nearly to completion. The bottom four alternatives suffer the dis-

ad.vantage of a3 expensive phosphate Conor. Creatine phosphate re-

generati-on from creatine is feasible, but the enz].me is available

only from mansralian sources and. will inevitably be more e:pensive

than the remaining alternatives. Phospi:ora:nid.ate phosphotransferase

is labi le and. has low cel lu lar speei f ic ac+" iv i ty (Z), Both acetate

and carba:nate kinases are viable alternati-ves. The carbamate re-

act ion has been used. previously for AfP regenerat ion ( : ) r and. the

by-products are volati le and. thus easil-y separated. from the nucl-eo-

t id.es. However, acetate k inase has been selected. for in i t ia l stucY

because of the grealer stabi l i ty in solut ion of acetyl phosphate

( l+r5), the s igni f icant ly higher equi l ibr iuro constant for ADP phos-

pno*iat i .on (5,7), and the potent ia l ly lower cost of the phosphate

d.onor synthesized. by acylation of phosphoric acid. with ketene pro-

d.uced. by theraal cracking of acetone (B). B'esearch r:nd'enray is

Ez'!

P \

A F

A S

' a ' . i -

' , . i l a

iete

o - r i

' : i n a

;eeeI r -r' l

- : J i l

:CY

a n a

i i - a

aoavq i n n

T!,e

C C'ARDNER ET AI

WASTE PRODUCT(r r$oRGANICPHOSPHATE )

IJ

t a ? o ? ' l n r , l / 1 f - a ' l ' PI 9 l q v 4 v a a v r l t - 4

:egenerated. from

r e t r r n f nhOSnhO-\ I g J

I is achieveo by t . le

Ieneration frora AIvPtherease regenerat l -o l :l&c+. ion. For the2 for which thephosphate d.onor is

lac+.ions are d.rivenLves suffer the d.is-r t ine phosphate re-' t zvmc i s ava i i ab iea s J l r v

be nore eryensive;e phosphotransferaser ( 2 ) . B o t h a c e t a t eThe carba.nate re-

l t i o n ( : ) , a n d t h e:ed. from the nucleo-rd. for init ial sruqyacetyl pLtosphate

;tant for ADP Phos-: t of the PhosPhateid. wi th ketene Pro-erch und.enraY is

-{iznfr.c REG;m{ERATTON 0F AT? 211

C O M M E N T S

B P R E G E N E R A T I O N C O S T S

M O O E R A T E : M A M M A L I A N

E N Z Y M E S O U R C E O N L Y

B Y - P R O D U C T S V O L A T I L EE A S Y S E P A R A T I O N

L A E I L E E N Z Y M E , L O W

C E L L U L A R A C T I V I T Y

A O E N Y L A T E X I N A S E

A T P + A M P z A o P

M 9 -

' g ' K T N A S E

IgP + }AOP }ATP + ?B

M g 2 t

Z B p + A M p A T P + 2 8I E T R E A C T I O N :

' g ' K I N A S E 8 P C O S T

C R E A T I N E H I G H

A C E T A T E L o w

C A R B A M A T E M O O E R A T E

P H O S P H O R A M I O A T E L O W

P Y R U V A T E H I G H

A R G I N I N E H I G H

P H O S P H O G L Y C E R A T E H I G H

A S P A R T A T E H I G H

M A X . E O U I L .

8 P S T A E I L I T Y C O N S T A N T

G O O O

F A I R

P O O R

G O O 0

POO R

l o 5

4 0 0

) a

6 6 0 0

5 0 0

3 4 0 0

2 8 0 0

?!e. 2z .A,lternative Erzlncratic Routes for ATP Regeneration fron' Al'fP

:tncerned. w'ith the thernodynamics of the regenerat: 'on reaction,'. i .-r,h

-DTOCeSSeS fOr SeDaratLng nuCieot:-aes frCm acetate, vi- '"h the

ii letics and. stabil ity of ' ,,he two enzlmes in free solui ' ion and im-

=bilized. on various supports, anct lr-ith reactor d'evelopment for ATrP

-generation.- q

Over a range of free magnesir:n ion concentration of 10 ' lo

-.C l '1, the observed. equil ibrir.ur constar: 'ts vary from about l to 9

d from about jO to )rOO for the reactj.ons catalyzed. by acenylate

i : i&se and. acetate k inase, respect ively ( f ) . These resul ts have

icen correlated. by a thermo(yn"mic analysis which includ-es the mrrl-' - lp le equi l ibr ia Lxis*, ing between al l species in their completeiy

: issociated, chelated. and protonated. forns (9rfO). Tr is analysis

: :e perrni ts pred. ict ion of the equi l ibr iun composi t ion ( f ig ' : ) for

= arbitrary set of operating conaitions and. inputs to the regenera-- . loa reactor. Fig. 3a shows the conlrersion of ADP to ATP by acetate

l':hase. The ord.inate is the icole fraction of ad.enosine as ATP a'nd''--:e abscissa is the ratio of ace+,yt phosphate to AlP init ially ad'd'ed"' -c the reactor. With reac+"a-nts in stoict iornetr ic proport ion, %f ':::: 'rersion is pred.icted. and. observed.; essentially complete conver-:icn of ADP to ATP is obtained. l.;- ith a 5O/, excess of acetyl pirosphate'

}e equil ibriun constant for this reaction is sufficiently hish that

{.i' II!

I

S U B S T R A T E S

ENZYMATIC

BI OSYN THES I S

2r2 C.R. GARDNER ET ft

predicted. conversion is relat ively insensi t ive to errors in themod.el . wi th the coupred react ions (r is. 3b), 9o/ , conversion ofAl"lP to ATP occurs vith reectants in stoichiometric proportions, a:i lthe remnining adenosi-ne is essentialiy a-l l in the fona of ADp. A502 excess of acetyl phosphate gives greater than 99/, conversionto AfP.

/ c o u P L E D R E A c r t o N s

a - _ _ . ^ - - _ A d l n y l a r ?

a r P + a I P f f i

z A O e

A c t t o t ?

2 A c P + 2 A o PK r n O t ?

J A I P] A D PI A[aP

- Th ro ry

a ) b )

Fig. 3: a) Equil ibrium Con'rersion of A.DP to ATP; b) Equil ibriumProduct Distribution witir Coupled Reac'bions

Removal of acetate ion from the recirculating nucleotid.estrea. la is necessaryl other* ise a bui ld-up in i ts concentrat : -on ' r - i : 'reduce ATP eonversion. Se'reral- processes have been exanined (Taote1) for this purpose. Ad.sorption on ac-,ivated. carbon, foJ-' lowed. byethanol elution and. d.isti i l-ation, leads to essentia-l ly conpleteseparat i .on and recovery of ATP. Upon ac:-d. i f icat ion to pi{ 3, acetaieis converted. to voiaii le acetic acid. which rnay be removeci by vacuulevaporat ion or by srr i tabl-e gas/t io.uia conlact ing. .LLternat ively,the aqueous solution folloving acio.if ication nay be contac+,ed. withan appropriate innnisc'ble organ:-c sol' 'rent and the acetic acid. re-rnoved.. For exampie, wi-rh t r ibuty l phosphate, the distr ibut ion ra-. :c:of aeetate and. ATP d. i f fer by a factor of 5000 ( f f ) . T}r is systernunfortunately produces a stable emufsion r-hich is d.iff icult tobreak. fon exchange chrornatography, follorred. by pil grad.ient elut:.c::,is used. to fractionare the nucleot j-d.es, acerate, anrd. acetyl phoscira-.e(fe). Membrane separat ion processes such as rr l t raf i l t rat ion havethe ad.vantage of simplicrty, but the comnierci.ally available nenbrs,e:ve have tested are not retentive enough to prevent significant nuc-leo t id .e 1oss . Means fo r inc reas ing ATP re jec t ion are be i .ng exami :e : ,and more retentive merrbran:,es wil l- be testea as they become availeb-e.l lhile it is presentJ-y uncertain which approach lri lJ. be mosr econon:-c"1, it is clear that se' 'reraL practical aLternatives erist for nuc-leotid.e separation from acetate.

tl

a o P

j

t--, O fI

o.el-

IO E F

Ii

o" f

o.rl-

l .

- t ;a l :

6

9t1 | 4 2t'

( x g ) r ' O O l 5 M

lotol Adanorraa . O0O75 M

R . ( A c P '

( l n t r , o l A r r P l

.

c 9. CARDNER ET AL

to errors in theOfr conversion ofric proportions r g::j,he forrr of ADP. Aan 99/ 'conversion

2 1 ?

fl:r,Jflf REGENERAfTON 0F ATP

TASLE 1

SEFANATION OF ADM{INE NUCLEOTIDES

I I E T H O D O P E R A T I O N

CARBON AOSORPI ION IOY" ETHANOL ELUTION

FOLLOWED BY OISI ILLATION

rROM ACETATE

R ECOVERY

> 9 9 % A T P

COMPLETE SEPARAI ION

C O U P L E D R € A C T I O I { S VACUUM EVAPORATION OR

GAS/L IOUID CONTACI ING

ACETAIE P H3 '

ACETIC

A C I D ( v o l o t r l e )

9 7 % A T P

COMPLETE SEPARAI IONo A6en t l o t?- A T P r A | l | P - z r C a

f , l 6 o i e

A c ! t o t a

2 A c P + 2 A O P : ? A c r 2 l r r

I O N E X C H A N G E

CH RO MATOGRA PH Y

8 I O - R A D A G I - X 2 D O W E X

RESIN, pH GRADIENT ELUII0N

I O O % A M P , 9 6 % A D P ,

9 2 % A T P , 9 0 % A c P

COMPLETE SEPARATION

O F A L L S P E C I E S

9 , ( A c P )

( l n r t ' o l A L P )

t . h l E n r r i i i h r i g ] n) w I ! Y 4 4 4 I V I

; ions

rt i nucleotid.e.ts concentrat ion - 'r i l -

been examined. (Tabie

:arbon, followed. bY

lntially complete, * r ' a * * ^ n l r

" a c e t - a : gL U I \ . , i l I U V l / r r J ,

be removeo. bY 'racutl

lB. .LlternativelY,ry be contac+"ed r i--h

;he acet ic ac id . re-

;he d-istr ibution r&-uLl j

f : - ) . f i r is systen.is d. i f f icul-t to

tY pH grad.ient el-ut ic:.

! , and. acetl l PhosPi:e::. traf i l t rat ion have

Jy available membrue"rent signif icant nuc-

-on a re be ing exam iae : '

they becone avaiLab-e'lril]- be nost econon:-

rt ives exist for nuc-

ULTRAFILTRATION A M I C O N U M - 0 5

M ILL IPORE PSAL

%REJECTION ATP> 9 0

> 9 6M E M B R A N E

M EM BRAN E

N O C I

U

W A I E R , O R G A N I C D I S T R I B U T I O N R A T I O

LIOUID / LIOUIDEXTRACI ION

A C I D I F Y , C O N I A C I W I T H

TRIBUTYL PHOSPHATT

A T P 5 0 0 : tA C E T A T E | . i 0

Kinetic stud.ies \.rith acetate kinase from E' eoLi have shown a"

:rcad. pH opt iu,r :mr over t i re pi- t* t tgu 6 ' ' +t 'o B' which is insensi t ive

. .o al l other operat ing conO.i t ions. Large scale operat ions are

iavored. by high concentrations so as to rninimj-ze the 'rolume of

i idd. which must be hand.led and. to cut Com the size of necessary

;rccess1ng equipment. EXperinents therefore i rave been focused on

: lEb concentrat ions of substrates ( f ig. )+), prod'ucts, amd o*uher

species. Acetyl phosphate concentrat ions as high as 300 urM show

l. \ inhihi fnrrr ef fect . wi th total magnesir :n, l held. constarr t , in i t ia l- y r u r r u r s v r J

: : - " : . - ; : ^ _ ^ ^ ^ ^ - ^ - ; f h o n , , l e o r e a . S e s - . . - 5 L i n a r o a . i n f l

:eect iOn velOci ty f j - rst increaseS anC then d'ecreaSes wrurr J-rr \ - r=*o*

.JP concentrat ion. From the plot of act iv i ty vs ' concentrat ion of

: '4ADP complex (r ' ig. )+), i t is c lear that there is no inhibi t ion by

: :e act ive substrate. The foregoing reSults therefore suggest that

' - tere is inhibi t ion by large .oi"u* i tat ions of totat ly d ' issoci-ated' ADP

r-rt MoH nnh?r species , such aS nonovalent ions , inorganiC plrOsphate o' r . s r ' I t 5 . vU^ i l

sid, acetate have l i t t le or no inhibi tory ef fect l whereas ATP' and-.c a greater extent AI,IP, have substanti.al inhibitory effec+-s at high

:cncentrations. Both ATP and. Allp aTe cornpetit ive inhibitors with

:esleet to A-DP, while ATP is a noncompetii ive inhibitor with respect

to acetyl phosphate. fanri f isalions of these observations with

K r n o r !

O A I Pt r AOP: AMP

- t h l o r y

A O P

2fl+ C.n. GARDNER ET n

o too ?@ 300M9 AOP CONCENTRATION (mM)

I O O m M A O P2 O O m M M g C 1 2

ffi,

o.9

o.8

f-r

o.7

^ t .o'=ae

trJoE

o-F

a6 0.5

=O

U

F

z ^

o.7

o.6

23' oH 7.4

O . 2 M T r i s 8 u f f c r

2 m M O f T

5 0 m M A c P

m M M g C I Z

o t 5. 5 0

A 0 P C O N C E N T R A T I O N ( m M )

r 5m M M q C 1 2

u.o -

o too zco 300AcP CONCENTRATION (mM )

Fig. )+: Kinet ic StuCies rr i th Acetate Kinase

regard. to optimr:n operation of an ATF regeneration reactor are

und.ergoing further stud.y, aiong with the kinetics of ad.erlyl-ate kin-

ase from yeast.

T:nz.rrmc stabil ity is perhals the single most important factor4 a g J s v

wb,rch d.etermines the economi c feaslbi l i * . i - of the process. In i i ia i

stud. ies of the storage stabi l i ty of acer-ate k inase in f ree solui icn

indicateC vir tual ly complete d.eact ivat ion af ter ihree d.ays ( f ig. l ) .

Experiments with various protective agen+-s d.enonstrated. that oxi-

d.ation of thiol grouls was the principal factor in l-oss of enzjnoe

act iv i ty. Upon ad.d. i t ion of d. i th iothrei to l (mf ) ( f : -naf ccnc. 2 n!1),

reduction of d.isrrtf id.e bond.s substantia].ly increased. init ial- enzyme

actirrity over the cou.rse of one d.ay. Activity remained neariy con-

stant for about ten d.ays, after which there folloved a precj-pitous

decline. DTT had. no effec+, if addeC after the enzlme had. remainea

at a low level- of actiuity for severel d'ays; but there vas again

an increase in activity, although not to its pre'rious maxim:m va.lue,

if UfI vas ad.d.ed. during the period. of rapid. d.eeline. These obser-

vations are consistent with re.rersible conversion of sr:-lfhyd.rylgroups to d.isul-f id.e bond.s, followed. by irreversible oxid-ation to

sulfonic acid'. Ferric or ferrous ion (f aiV) caused rapid' d'eacti-

vat ion, even l . r i th DTT presentr but ' rh is ef fect vas prevented by

ad.d.it ion of substrates. By period.ic ad.d.it ion of DTT, acetate. kin-

ase has been maintained. active for long period.s of t ine z 80% of

na,1gimg actirity arter 7 weeks and 30% of naxi.mum actirrity after

13 veeks. These results are encouraging and- suggest that an ATP

regeneration reactor conposed. of enzime in free sclution may be

feasible. Hovever, preliminary resuJ-ts vith an rrltrafi l tration

reactor indicate a sensi t iv i ty of acetate k inase to f lu id ' shear ( f f )"

It is therefore l ikely that a successf\: l regeneration reactor r{i l l

r t -

t t

l O m M F r e c M g 2 '

5 0 m M A c P

o.8

0.6 F

o roo 200 300AcP CONCENTRATION (mM)

te Kinase

uion reactor areics of ad.enylate kj.n-

cst inportant factorre --ocess. In i t : -a i3a in f ree solut ionr * , l t ree d.ays (Fig. 5 ; .rnstrated. that oxi-r in Loss of enzymeg) ( f inaf conc. 2 xoM) ,reased. init ia-t enz1meremaineo. nearly con-

Lloved. a precipitousenzJme had remained.

rut there was agai-n:evious mocimum vaiue,: l ine. fhese obser-Lon of sr.il-f'hyd.ry1rible oxid.ation torused. rapid. d.eacti-vas prevented by

lf D'IT, acetate kin-i of t ine z 80% ofnr.rm activity afterrggest that an ATPl solution rnay ber rrltrafiltratlonie to f lu id . shear ( f : ; .

:ration reactor wil]

U . R. GARDI{"ER ET At

l l

?"1-

*zyt"fr.c REGETIER.{IION 0F ATP 2r5

o r o ? o 3 0 4 0 5 0 6 0T I M E ( 0 o y s )

Fig. 5: Acetate Kinase Stabi l i ty in Free Solut ion

-quire that the enzjrme function, if not imobil ized., in a statici l 'dd environment. Development of imrnobil ized. enzfne reactors iselso undeffey (B).

A CISI O I,fLED GE IVIETI T S

sr:nnnrrod. in part by grants from The llational Sciencev 5 ! / ! , v 5 v v

icr:ad.ation, GI-31+28)+, the f,emrllfs and. IIenry Dreyf\rs Found.ation,ud' by a Fi l lsbury Fel lowship ( fnst i tute of Food Technologis+,s) aner Sigraa Xi grar:,t-in-aid.-of-research to BIC{. Contributions of .f .! 'cyle, B. Bunow, C. Cooney, M. Kopels, and M. Lapid.us are gratefullyxcknovled.ged., as is the participation of M. GreenwalC, P. Hernleyri . Johnsonr M. Levenstein, S. Schul+uz, E. Siegal , and B. Yorgeylder auspices of the lf iT Und.ergrad.uate Research 0pportr:nit ies. - :0grFm.

REEERXNCES

5I

{,el ' .ot!ore

F

oo6

- n F

=tr(J

I

;:

l .o t

34 ' 9H 7 .4

O.2 M Tr is Euf fer

O.O2 M Free Mgz'

wirh o.oo2 M 0TTWrthout DTT

A c l d i t r o n o f 0 T T

I O O m M A D P2 O O m M M g C l 2

-. t lAl, t ILTON, B.K., MONTGolnuRY, J.P., wArG, D.I.c. f tr is vorr:m.e, p.153.

2t6 C. R. GARDNER ET er,

2. MARSIIALL, D., wAr,TER, J. & FALB, R. rnterin Report on Develop_ment of Techniques for the fnsolubil ization of Enzynes t;convert Glucose to Starch, Battelle Coh:mbus Laboratories,submitted' to Nationa-t Aeronautics and. Space AdministrationAmes Research Center, June 29, LgTz.

J. MARSIIALL, D. BioteehnpL. Bioeng, LS:\ \ | , L973,1. DiSABATO, G. & Jrr\IcKS, i,tr.p. J, ArT7. chen. soe. gg;l+hoo, 1951.q QDIANMAD T) t . ' f&uru! l r r , . , JONES, M.E. & LTPMANN, F. Meth. EnzymoL. 3:653,

L95T .6 p n e F T ^\ r o AvD . r ! 1 J - .A . r MANAGO, M .G . , COREY, S .R . & ocHoA, S . J . B ioL . ( r 7em"

ZLL:737 , L95 ' \ .\. I-,ANGER , R. S . Ph . D. fhes is , WT , Cambri dge ,8. lrHrrESrDES , G. M. , ci{MItRNy, A. , GARFETT , p. ,

COLTON, C.K. Th is vo l_une, p . 2LT,A-LBERTY, R.A. J , ELIL , Chem. 245:L337, 196g.AIBERTY, R. A. J , BioL . (rten. 244 : 32gO , L96g .DAlffS, J. Personal commur:,ication.JACI(SON, F. M.S. Thesis, l ,4fT, Canrbr id.ge, Lg73.QIIINTERO, R. M. S. Thesis , I l fT, Cambridge, L9T\.

L > l + .

LAIvIOTTE , A. &

9 .10 .11.! 2 .t <

AIII

ttj

It

t J .' 21 ,

l ' t .

filfrir%*..

( GARDNER ET ii

r Report on Devel-oc-rtion of Enzyrres to.rlnbus Laboratories,ipace Adrni nislration,

' 3 .r. ag.. )+l+oo, 1951.

Enzymol. S:613,

tA, S. J. BioL. ChqT.

LgT\.I4l40TrE, A. &

TTZYMATIC BEGN{ENATION OF ATP FROM AMP AND ADP

?ART II : ENZYME TIVMOBILIZATICN AI{D ruACTOR DEVELOPI{N'IT

G.M. Whi tes ioes , A . Chmrny , P . Gar re t t tA. Lamotte and C.K. Col ton

Depat'-inents of cheni stry aJ1d. chenical legineenng

Massachusetts Inst i . tute of Technolog;Cambrid.ge, l4as s achusei-+,s

f6e biosynthesis of many natr:ral prcducts consumes ATP (l).

fre involvement of ATP in biosyn*,hetic reactions can be i l lus*"rated

by tvo well established. pathvays fcr ',,he activation of an alkyl

carborglate ion ( f ig. l ) . in one, t ransfer of the tenr inal phosphate

o--

O-P=O'J l--t

, ' O Jt ' l

--.' o-?=o-...?, fH"-*[:'.r"4410

c>

o''il ?

o,P-O-6-

+ ADP

o#

b-n O 0

R-i,- + - o-F-o-F-o-b-anp - d 6-

PPiHO OH

ATP

Fig. 1: Pathvays for Activation of Alkyl CarboryIate

group of ATP to the carbo4llate grouping (phosphoryl transfer) resultsta the prod.uction of an acyl phosphate anc ADP. In a second pathwaytaucleophil ic attack of carbo>qllate i.on on the a-phosphate groupingof ATP (ad.enylyl transfer) generates arl acyl ad.enylate and inorganicgyrophosphate ion. Trese tvo types of acyl d.erj-vatives are bothrctlve esters a3d. cal take part in further reactions at the carbonylFoup.

As part of a proJect to test the pract ical i ty of large scale

2 1 75 _ l

218 G.M. IIItrTESTDES ET AI

synthesis of the cyclic d.ecapeptid.e antibiotic Gra-nicid.in S usingcell-free enz]nxes (Z) , we have erylored. nethod.s for the econorricllregeneration of AT? fron A-l@ or A-Dp. we have settl_ed. on a coup:-ienzlane system consisting of adenylate kinase and acetate kinase(f ig, 2) . Acetyl phosphate is chosen as the ul t imate phosphate

AMP + ATp :d""'""- ? Aopk inosc

? 9--cHacoPr

- + ADP- 'o -

AcP

oca to td

k inosc

K = 2

+ ATP K.t40O,o

cH3c\o-

Fig. 2: Coupled. Syste:n for ATp Regeneration

source. Reacti.on of acetyl phosphate w-ith sp, cataryzec by ace-tate k inase, generates acetate icn and ATF ( : r t*) . AT? read. i ly d is_proportionates with AltP i-n the presence of ad.enylate kinase to gener-ate tvo molecul_es of ADp (>-f ) . Thus, reaction of Al@ with aceiy:-phosphate in the presence of sc.al_l nmounts of ATF a-nd. ad.enylatekinase and. acetate kinase resul-ts in consunption of acetyl phosphatea.Ird. the generation of acetate ion and ATp.

fn searchj-ng for a synthesis of acetyl phosphate that is bot,hinexpensive and' amenable to process d.evelopment on an ind.ustrialscnler w€ have sett led on the acylat ion of phosphor ic acid. wi t i : .keteae (8). Ketene is reaci ly generated. on a large sca_le by the

o 700cl l

xac/\cxa CHa:g=g + CH+

so{z-c--o + Hspo4 (Hzo) ---+ .rrErE lX + .r.Eo8.r.

I ?.oucHscoPloui

(5O % lso lo ted)

Fig. 3: Synthesis of Acetyl phosphate

thermal cracking of acet ic acid. or acetone (r ig. 3) . React ion ofketene with phosphoric acid. generates, ini.t ialiy, monoacetyr phos-phate' Furbher reaction of ketene with the prod.uct nirtqre con-taiaing the monoacetyl phosphate a-lso generates di- and. triacetylphosphates. In ad.d.it ion, water ori.ginatly present in the phosphor:.cacid' reacts lrith ketene to produce acetic anhyd.rid.e. Frocedr:resfor the conversion of phosphoric acid. to monoacetyl phosphate are

,?

&b l&C(

trq

TTt

ai

3

t

2 ) E thono l

t{. InfiTESfDES ET AL

Q1'nrn'i sfflin S using: for the economicaltet t led on a coupledrd acetate k inaseLtimate phosphate

t<= ?

P K3400

rneration

catalyzed by ace-.) . ATP readi iy d. is-ylate kinase to gene:-, of AI@ with acerylTP anr i nr ienwl gtrg5 4 * s * v r r J 4

n of acetyl phosphate

osphate that is bothon an ind.ustrraJ-

ph c acid. with&Fgv scale by the

sphate

, 3 ) . R e a c t i o n o f't monoacetyl phos-luct nixtrrre con-di- and. triacetylnt in the phosphoric' id.e. Procedr:resrtyl phosphate are

fr|.wc REGENERATToN oF Arp 2 1 4

:ct yet opt imi zed.; nonetheless, using exist ing proceCures ' phos-

:r;;" ""ia can be read.iry converted. into the dil i thir:n salt of

,.rtyl phosphate in approxirnnteiy 507' yteLd, basea on starting phos-

;rri" acid.. It shoutd. u-lt ir1atel-y be possible to make large quantl-

.. i., .r acetyl phosphate, very ine:pensively, using this proced'ure.

With an assp,red. supply of acetyl phosphate as starting mater-

lt l , attention has been focused. on important problems deaU'ng with

::e tvo enz1m.es required. for the conversion of AMP to ATP' A matter

ci centraf concern in consid.ering potential conmercial applica*uions

,l uorr.ru, is the stabil ity of the enz'ne. An enz1rne that can be

.:sed for extend.ed. period.s of t ime is a practical subject for d'evelop-

-qtal work, even if i ts ini+,ial cost is high. An enzylne with a

lbort l i fetirne r:nd.er operating conditions j-s less attractive, even

i , f appreciably less expensive. The pr incipal contr ibutor to the

,.u.i i.r"t ion of rabbit muscle r4lokinase is autoxid-ation of mercaplan

Sroupings in the prorein. The conversion of the sullyd4lf groups

rf .yt tu ine to c isul- f id.e groupings, a:ro subsequentJ-y *"o cysteine

sutf tn ic aci .d. , is ' , -e l l kno'rn (9). In the part icular case cf nqlo-

i: iase, this autoxid.ation can be part,ially re'rersed by add':-+"ion or-

ippropr iate red.ucing agents: a i th iothrei to l , d. l th ioerythrytol and

:-rercaptoethanol . I f the solut ion containing the enzlme is kept

:igorously free of oq;gen, and. the oxid.ation potenria-I of the solu-. , ion is stabi i izeo by aCoit i -on of d. i th iothrei to i , tne enzjr lne re 'uain 'sj , ts act iv i ty for weeks. Surpr is ingly, the stabi l i ty of myokinase

isobil ized. on Sepharose is mucLr higher than that of myokinase in

t'ree solution. T|e inrnobil ized. enzyme has retained. its acti 'r i+'y

lcr four to six weeks, even in suspensions from whicir o>qlgen has not

been exclud.ed.. Tl3.e origin of t,he increaseo s*,abil-ity of the inmobi-

i ized. enz1me relative to that of the seme en3]1ae in free solution

1s not c lear at present. I f general , the iar ter phenomenon shoulC

prove to be of enornous practical importance in the coro:nercial ap-

plications of im'nobil ized. enz;ruxes.

We have successfull-y inmobil izeC both acetate kinase and. ad-

eaylate kinase on cyaJrogen bromid.e Sepnarose and' operated' a small

denonstration reactor for the conversion of AI4P and. ADP to ATP (Fig"

L). TLre circulating solution flovs througn the colu:ut containing

acetate kinase ag{ ad.enylate kinase, and subsequently through a

col:on containing potato apyrase inmobil izeC on Sepharose' The po-

tato apyrase column converts ATP to ADP and. AIvlP. T[e product strea:n

euerging from the apyrase column is mj-xed. with a solution eontaining

acetyl phosphate and. fed. into the acetate kinase and. ad-enylate kin-

ase containing co}:mn. Ttris colr.mn has been operated' for a nr:m'ber

cf hoqrs, converting all of the input Al'F to ATP. The total actrvity

tf the reactor was sufficient to generate about one gra& of ATP perh ^ r r - A ] * l a a r r a h i h a ' l a n c * o m c * a h i " l i * ' r n f t h i s r e a c t o f h a S n O t b e e n. .es . .L r - u r ]v , .gh

t rhe lOng tg fm S LaUi t I t , J \ . / r vu+J r ese v

:tudied. in continuous operation, 'uhe ATP regeneration colulrn retained'

activity for greater than si.x veeks.

0 0+ cx.8o8cx.

:itL

I

I

tI*{I

j

3 d /tin-'-

BrCN- Sephorose ?45O mq)

ketota Kinosa F250 units)

Adenylata Kinose ?5AO unifs)

BrCN- Sephorose

Poloto Apyrose

220

AMP+ AOP + AcP.-

ATP+ 21' '

G.M. I'NIITESIDES ET AL ,n;

rl,c';at[ ! v

vltr?4.'.t'o

3ect , E g

r f ?

:lre'.tg

iesqEiniESifor3fr

. ;AMP + p.

';z AOP + t

. . A T P

Rasidencc Timc in Regeneroi ion Column < I min

( A T P I * I / ( A T P ) . n = g - 9

P r o d u c t i v i t y - l g A T P / h r

Fig. \: Al,F to ATP Regeneratj.on Reactor

As part of arr efforb to d.esign enzlrunatic reactors for ATP re:generation and. other problens tha*,, provid.e alternatives to the com-mon-ly used. f ixec bec, we have begun work on a reactor based. on astirred. bed. of polyacrylamid.e beads eontaining both entrapped. enzim,esa.nd ' sme11 nagnet ic par t i c res (F ig . 5 ) . fhese bead.s , 1 -10 mlc rons ind'ianeter' are sufficiently small that d.iffusional- l imitations on therates of reaction of substrate solution with the enz;mes are relat-ively r:nimportant. Conventional- f i l tration of these particies is e

qJ r.

pbe

! ^ v

tvotheatt3C.!,- h {

:gx,t +

mognetic polyocrylomide beods

contoining entropped enzyme

lMognetic beods withenzyme A

Non-mognetic beods wrthenzyme B

I .

2 .

?

I t .

5 .(

'll a

g .

stirred or f ixed bed reoctor f or

coupled enzymes hoving different

stob,, i t ies

Fig. 5: Magnet le Processing

f'' o3'' ol')

Ifi^^",,,J--ATP

IIIIT--SfDES ET Al

+ P .I

+t

S,ctOr

reaetors for ATP re-la 3s to the cors-r .ctor based on ar r h o n r r - q n n o r i o n 7 \ r 9 q

r d q ' 1 - ] n m 1 . r Y . ^ n q

i ' i

i r n i f q f i n r n < / 1 r r t i ^ o

enz]mes are reiat-rese part ic les is a

letic beods withrzyme A

rnognetic beods wtthrzyme B

I reoctor for

tving different

fl?..w]c REGENERATTON OF ATP 22l'

slo,;, and. inefficient process. I{ovever, they are very read.ily sepa-

rated. from solution in the presence of high magnetic f ield. grad'ientsl. 'n rr ) Morq importantly, rcagnetic beao.s offer a nethod- of d.ealingt l V t l r r

r

,rith enzyne systems in which tvo enzJrmes irave arasticaily d.ifferent

stabi l i t ies und.er the operat ing condi t ions of the reactor. I f these-,yo enz1mxes were coinrnobilized. on a colnllon support, it would. be

eecessary to d.iscard. the activity or- the longer l ived enzlncl.e when.,be shorter I ived. enzyme had. become inactive. In the sti-rred nag-

let ic bead. reactor, th is problem can be oeal t wi th by enclosing the

:?o enzlm,es in d.ifferent beao.s, one rnagnetic and one nonmagnetic.'.ten the activity of one of the tvo enzymes has d.ecreaseo to a use-

iess vaIue, i t is then easy to separate the act ive and. inact ive en-

rynes by raagr:etic f i l tration. The inactive enzyne can be d.iscarCeo

aad the active enz1nxe mixed rr.ith a new batch of its :eaction partner"

aad. ad.d.ed. back +,o the reac?-or. Tre potentla-i of magnetic separallons

for d.eaJ-ing with tnls ana other separation probiems in appiied en-

zynolory is high and is being actively e>ploi' ied-.

The central conclusi-on frour "his

wori< is that the en-

4;natic regeneration of ATF from A-!IP ar:d/or ADP using acetyr phos-

phate, acetate k inase and adenylate k inase is an ent i re iy pract ical

proposi t ion. Acetyl phospha+-e appears io be reaCi ly avai lable, the

tvo enzlmes have gooo stabil itY r-:naer '-he oP€ra-u1-I1$ coni:--:cns, ano

the thermodynamics of r,he coupleo enzyraatic react-i-cn system are

attract ive for possible use in large scale : 'eactors. Of the var ious

scbemes that have been proposed. for enzla-xatic regenera*,,i.on of ATPt' -b is scheme Seems +,o have rnost appeal t because i t is capabie of

lendling both A-DP and. Al4P, ani. because the phosphate source on rrnich

i'. ult ieately relies appears to be ihe nost economical.

ACIC{OruEDGII,M{T

This research was supported by the National Scienceloradation, Grant GI-31+28\.

f i n f f i f i n i ? d F d

l s r^nrM^T\T q .R. In t tThe Enz lm.es : , 3 rd . ed . . , Vo l . I (ga . P ,D. Boyer )-. UJ..t l l r t l 'Jf l l l , L

Acad.emic Press , New York , L972, Chapt . 1 .?. EAMTLToN, B.K., MONTGOI,ERY, ,J. ts. & ' ' I IANG, D.r .c. This volune,

n t \ .P ' t ) J '

3. GARDNER, c .R. , coLTON, c .K . , LA1 ' IGER, R.s . , t iA I ' t tLTgN, B .K. ,AnCIf iR, M. C. & I I IT{ITESIDES, G.M. fh is volume ' F. 209.

i . A l f l tHoNy, R.S. & SPECTOR, L .B . ,1 . B ioL . Chem. 247:2120, L972.

7, DerEER, P. & LOIrE, A.G. J. BioL. Crtem. 248:2829, L973.5. NODA, L. rn "T?re Enz1ruees" 3rd. ed. . vol . B (na. P.D. Boyer)

Acad.snic Press, New York , L972, Chapt. B.7. BLAIR, J.M. Eu.z. . ,J. Bioehem. L3: -?Bh, L970.8. BETllrLEy, R. J. Arn. chem. soe, 70:2L83, 191+8.

222

9, JOCELI1I, p.C. "BiocheroJ.stry of the SIINev york, L972,

10. KArsER, R. , C0LTON, C.K. , MrsKol,czy,- P rog , Samp,6g :115 , 1971 .r l . BIIRTON, C. Crit. . Reu , Bioengineering

G.M.I,EIITESfDES ET AL

Grouptt Acad.emic press,

G. & MIR, L. Chem, Erg.

L : 2 7 3 , L 9 T Z , .

4 . .

ar-

t c :r t i -.v.t--

ttt'isr

ge:be

5E:

by8)

n {

81

\J.l'

ce

f"

tn

Jcvte7