6S6

Enzyme Inhibitors. XVII. Kinetic Studies on the Irreversible Inhibition of Adenosine Deaminase

A kitietic method h:ih beeii developed for t,he evaluation uf the reversible etiz~ino~-iiihil)itt,l. tiinsttciatioii citit>stiiii I (Ki) and of t8he first-order rate constant for the alkylation of an enzyme by ati active-site-directed irreversible inhibitor. This method should be applicable to all irreversible enzyme inhibitors that, proceed t,hrorigh an initial reversible enzyme-inhibitor complex. It has been shown that irreversible inactivat,ion of adenosine deaminase caused by 9-(p-bromoacetamidobenzy1)adenirie (1) and by '3-(o-bromoac'etamidobeiizyl)adenine (3) is t tot a rail- dom bimolecular procesrj but proceeds by a first,-order react,ion in an initial reversible enzyme-inhibitor complex. ITsing this new method, the K i of 1 and 3 were found to be 1.4 X 10-j .TI and 43 X 10-5 M, respectively, in cs- cellerit agreement with the values obtained previously by t he rtaual doiible reciprocal plot rnet,hod. Further- more, i t was forirrd that, 3 alkylates the enzyme about aeveit time.? more rapidly than does 1. The first-order ratr roltrt,aiit ( k 2 ) for the alkylation iii the reversible E-I complex for 1 = 1.1 X niiiir' and for 3 = 7 .7 X IO-' 1 i i i i i - I .

During the past several years, a number of lahorx- torics have been interested i l l the design arid syrithe of irreversible inhibitors of certain enzymes.2-4 studies on esterases arid irreversible inhibitors, equa- tioris have been derived which allow one to evaluate K , (the affinity constant) and k , (the phosphorylation rate const'ant) or kzc (the carbamoylation rate con-

In this paper we wish to pre- sent M new mat,hematic*al description of the irreversible itihibitioii o f an enzyme which allows une to det'erminc Ki and k , (the first-order rat,e cwnstarit for the alkyh- tioti of t.he enzyme by the inhibitor).

Several recent papers have desrribed our research o i l

the irreversible inactivatiori of calf intestinal mucosi adenosine deaminase by 9-(p- , ut- arid o-bromoaret- ILniidobetizyl)adeiiines.6 It was found that' the para and oi'tho derivatives (1 and 3) were much more effective itwversihlc iiihibitors of adenosine deaminase thaii \vas the ~ c t a isomer (2). I t tvas also possible to mew- iii'c the itiit'ial revcrsihlc inhihitioil of adenosine deami- 1 1 : ~ ~ by these compounds, and it n-vas found that the effectiveness of reversible inhibition decreases i t i the followitig order: para isomer (1) > Illeta istrnier (2) > ovthuj isomer (3). Consequently the potency of thew ronipouiids its irreversiblc inhihitors i;i riot merely :I

rcflcc*tioti of their ability to form :t reversible romplcx

r wliicd must' tx coilsidered ivheri coni- c inhibitors of enzymes is the chemiral

reactivity of the alkylating o r acylating group of the itihibitor. -l-(p-l\jitrohe~izyI)pyridiiie ha.: been used previously to vompare the chemic*al reactivity of al- kylati t~g Using 4-(p-tiit'roberizyl)pyridirle ive csonipared the reaction rates of iodoacetamide with 1, 2, uric1 3. If the rate with iodotretnmide was normal-

tnt of the enzyme).j

! 11 This investigation \'ab supported by Grant T-337A from the American ('ancer Society, by a Public Health Service research grant (5-ROl-GM- 09775-051, by a research career program award (5-K3-CA-lST18-05) from the National Cancer Institute, and a training grant (5-T1-GM-555-05) from the Division of Medical Sciences, U. S. Public Health Services, Betheada, hld.

(2) 13. It. Baker, J . I'hurm. Sci., 53, 347 11964). (31 G. Sclioelmann and E. Sha\v, Biochemistry, 2 , 252 (1963).

( 5 ) (a) .I. R. Main, Science, 144, 992 (1961); (b) .4. R. hlain and E'. Iverson, Biochem. ,l,, 100, 625 (1966): i c ) h. R. 11ain and I'. I,. Hasting-.

(6) (it) 11. , J . Schaeffer arid E. Odin. .1. -IIet/. C h e m . , 9, 376 ilLl136~; ( 0 ) I f . ScliaeBer and R. N. .Johnson, J . I'hurm. Sci., 66, 929 (1966); ( c i 11. . J . IiaefTer and E. Odin. J . X e d . Cliem., 10, 181 (1987).

Gchramm, ibid. , 4, 377 (1965).

& C l F n C E . 164, -100 (1066).

NHL I "2

I

I

NHCOCH,Br

NH2COCH2Br 2

1

" 2

ized to 1, theti the rnte5 with 1, 2, a i d 3 n e i ~ ;3.0, 1.1. and 5.6, re\pertively. Since this ordei, of mtc*ti\4ty does not reprewit the effectivetie+ of 1, 2, a t i d 3 :i\

irreversible inhibitori of :idenositir deitminaic.. i t I \

:qiparcnt that ti morc hubtle effert or :L wnititi:ttioii of wveral efferti determine the effeetivcric~* of :ti1 11'-

rever4ble itihibitor of the erizyrric. 111 older t o gatit inore iiizight into the mecharii~m of thii re:wtion. :I

kinetic* itudy i d the irrcverhihle inactivxtioti ( i f :iclciio- sine deaminase by 1, 2, and 3 was undert akcn.

When ail enzyme iz irrever.ibly iiihibitetl by , t i i

alkylating or acylatitig agent, it would appear that t he two moqt probable mechanisms for the iri:wtiv:itioti are (A) a bimcrlec3ular attack of the inhibitor o i l Ihc enzyme or (B) the initial formation of a reversible (.om- plex through I\ hich cwvalent bond formation ( J W U Y ~ .

These two niech:mi-m\ may be distinguished liiiirt i - cdly qince. it1 the ( e of the bimoledar mechnriim

17) la) ,J . Eiistein, 11. \ \ . Ru~eiitlial, and I<. J. Esr. . l n i i l , Ciicnt., 27, 1 I G tIU53): (b) T. J. Ilardos, N. Datts-Gupta. P. Hebhurn, and I ) . .1 . l'riwlt!. J . .\led. Chem,. 8 , 1137 ( lUS5) ; f c ) 13, R. 13aker and J. 11. ,JurdaaIi, . I . I I c I t r o - r u d i c Cliem., 2, 21 (1!J65).

July 1967 ENZYME INHIBITORS. XVII 687

I I E I

I R E + R + \ / + X -

x I E . . . I

(A), an increase in the concentration of the inhibitor would rezult in a corresponding increase in the rate of inactivation at all concentrations of inhib tor. How- ever, in the second case where an initial reversible en- zyme-inhibitor complex is formed, increases in the con- centration of inhibitor would result in an increase in the rate of irreveriible inactivation only until the en- zyme was saturated with inhibitor. Further increases in the concentration of inhibitor would not affect the rate of inactivation.8 This phenomenon has been discusied previously by Baker, et U Z . , ~ who have de- scribed it a5 a "rate saturation effect." It must be emphasized that the observance of apparent first- order irreversible inactivation of an enzyme does not usually differentiate mechanism A from B since in the experimental design of most irreversible inactivations, the concentration of inhibitor is significantly greater than the concentration of enzyme; thus, both mech- anisms would exhibit apparent first-order kinetics. In order to differentiate these mechanisms, it is neces- sary to determine if a rate saturation effect is present or absent.

Baker, et U Z . , ~ have derived eq 3 to evaluate this rate zaturation effect. Briefly, the derivation is outlined below for an irreversible inhibitor that forms an initial reversible E-I complex.

K , = lEI[II/[EIl (1)

[Ed = [El + tEII (2)

Substituting (2) into (1) gives (3)

(3)

where [E] is the free enzyme concentration, [I] the con- centration of the inhibitor, [EI ] the concentration of the reverhible enzyme-inhibitor complex, K i the dis- sociation coristant of this complex, and [E,] the total enzyme concentration.

Beclause, as we have described earlier, an irreversible inactivation of an enzyme depends on both K, and on li2 (see eq B), we have derived an equation which sepa- rates these two kinetic parameters in order to gain a more complete understanding of the reaction mech- anism. For an irreversible inactivation of an enzyme that proceeds through an initial reversible complex and follows apparent first-order kinetics, the rate of inactivation (R) is described by eq 4.

R = kz[E-I] (4 )

(8) This argument implies that the initial concentration of inhibitor was insufficient to saturate the enzyme. If the initial concentration of in- hibitor saturated the enzyme, then further increases would not increase the rate. In this case one could decrease the initial concentration of in- hibitor to observe the variation in inactivation rate.

(9) B. R . Baker, \I-. \V. Lee, and E. Tong, J . Theoret. B i d . , 8, 459 (1962).

Substituting (3) into (4) gives ( 5 ) ,

1 El Since

R/[Etl kobsd (6)

where kobRd equals the observed first-order rate con- stant for enzyme inactivation, then

Taking the reciprocal of (7) gives (8). Thus, if a series

of irreversible enzyme inactivations are performed at various inhibitor concentrations, a plot of l /kobsd

against 1/[I] should give a straight line whose inter- cept would be l /kz and whose slope would be K i / k z .

A similar equation has been derived for the reaction of isopropyl methylphosphonofluoridate (farin) with a variety of subst,ituted catechols.'O

Experimental Section Adenosine deaminase (Type I, calf intestinal mucosa) was

purchased from the Sigma Chemical Co. The adenosine deaminase inactivation procedure 11s a modifica-

tion of a method described in the literature.6a.11 A solution of the enzyme was prepared such that a 50-~1 aliquot when diluted in phosphate buffer to 3.1 ml which was 0.081 mM in adenosine gave the desired initial velocity of enzymic reaction. Equal volumes of this enzyme solution were then mixed with equal volumes of phosphate buffer containing 10% dimethyl sulfoxide (DMSO) or solutions of the irreversible inhibitor in phosphate buffer containing 10% DMSO. These solutions were incubated a t 37", and a t various intervals 0.5-ml samples were removed and immediately cooled to 0". The amount of enzyme remaining in each aliquot was determined a t 25" in triplicate experiments by using a 100-~1 sample of each aliquot diluted with phosphate buffer which was 0.081 mM in adenosine. I t Wafs shown that the irreversible inhibitor did not inactivate the enzyme during the few minutes it was kept at 0". Each point on the plots is an average of three determinations; each inactivatioi experiment was repeated a t least tnTice.

Results and Discussion In Figures 1 and 2 are shown some of thle apparent

first-order rates of irreversible inhibition of adenosine deaminase by the para and ortho compounds (1 and 3). The meta derivative (2) was also capable of causing irreversible inact>ivation of adenosine deaminase, but the rates were t'oo low to allow an accurate kinetic evaluation due to considerable t'hermal inactivation of the enzyme over a 2448-hr period. In the case of 1 and 3, the irreversible inactivation of adenosine deami- nase was relatively rapid; therefore, the 1;hermal in- activatjon of %he enzyme was ignored. The apparent first-order rate constants for the irreversible inhibition of adenosine deaminase were calculated from the slopes of the lines shown in Figures 1 and 2 and a complete

(10) (a) M. A. Schmarta, Ph.D. Thesis, University of Wisconsin. 1959; (b) T. Higuchi. Progress Report submitted to Armed Forces Chemical Cen- ter, Edgewood, Md. (May 20, 1959); (e) T . C. Bruice and S. J. Benkovic, "Bioorganic Mechanism," W. A . Benjamin, Inc., New York, N. Y. , 1966, p 143.

(11) B. R. Baker, Biochem. Pharrnacol., 11, 1165 (1962).

688 H. J. SCHAEFFER, 11. A. SCHWARTZ, AND E. ODIs 1-01. 10

\ \ \

listing of the data i5 given i n Table I. A plot of 1/ kcobid 1's. 1/ [I] is shown in Figure 3. An examination of Figure 3 reveals that for the para derivative (I) the slope of the line is 1.24 arid the intercept is 91. E'roni these data it can he calculated that for 1 the K , = 1.4 x 10-j JZ arid k , = 1.1 x lo-' niiii-l. I n the

V n - IO 20 30 40 50 60 70 80 90

I (I)

Figure 3--I'lot of l/k,,,,,l for irreversible iiihibitioti of : & t i c ) -

rille deamiiiase cs. 1/ [I] (mnl ) : 0, !~-(p-bromoac.eta~nidol,er~~~.I)- adenine (1): 1. !)-(o-bromoacetarnidober~z~l)ade~~i~le 13).

case of the oltho derivative (3), the A p e of thcl liiic I C

3.Z7 and the intercept i4 13. Therefore, K , = 4 4 X 10-5 J/ arid k , = 7.7 x 10-2 niiii-I. We liavcl pro- viously determined that both 1 arid 3 are (*ompetit ivc inhibitors of adeno4ne deaminase by the Liiicn c a v c ~ ~

Burl< method arid found that the K , of 1 = 1.3 X 10V' A l l a i d the I<, of 3 = 14 X 10-j JI.fiajC The detcwiii-

nation of the inhibitor conitant ( K , ) by the irrevcrkible inhibition experinierit ii therefore iii excelleiit agree- merit with our previous data. We believe that these data offer itroiig experimental support to the mech- aniim a> outlined in eq B, especially in view of the fac t that iodoaretamide did not irreversibly inhibit aderio- aine deaminase even at concentratior~s of 1 . A 11~11.

I t is also clear that the alkylation reaction ( k J betweeii 3 arid the enzyme i i approximately seven times faster than in the c:\+c of 1 arid the enzyme, even though 3 is not even twice a> reactive a5 1 when measured with 4-(p-nitrobeiizy1)pyridine as the riucleophilic reagent.

July 1967 ENZYME INHIBITORS. XVIII 689

Finally, the results of the irreversible inactivation of adenosine deaminase by 1 and 3 represent an in- teresting example which must be considered when a comparison is made between two irreversible inhibitors of an enzyme. If the irreversible inactivation of an enzyme is performed only with a single concentration of each inhibitor, it is possible to draw an erroneous conclusion concerning the relative effectiveness of the inhibitors. For example, if the irreversible inhibition of adenosine deaminase were performed with 0.1 m3l concentratioris of 1 or 3, it would be found that 3 in- activates the enzyme more rapidly than 1. If, however, a similar experiment were performed at 0.03 mill con- centration of inhibitor, it would be found that 1 is more effective than 3. This apparent reversal of potency of 1 and 3 as irreversible inhibitors of adenosine deaminase occurs because the observed first-order loss of enzyme activity is a function of both K , and k2. When the ir- reversible inactivations are carried out at concentrations

of inhibitor which do not saturate the enzyme, the amount of the total enzyme, [Et], in the reversible enzyme-inhibitor complex is dependent on K,. The amounts of E, in the reversible E-I complex at 0.10 mdl and 0.03 m J I concentrations of 1 are O.SS[E,] and 0.70[Et], respectively, whereah in the case of 3, the amounts of E, in the reversible E-I complex at the same concentrations are 0.1s [ E t ] arid 0.067 [E,]. Thus, for 3, a much larger percentage change in the conceritration of the reversible E-I complex will occur than in the case of 1 which, in turn, would result in an apparent reversal of effectiveness of these compounds as irreversible inhibitors of adenosine deaminase. From these data it is clear that in comparing irreversible inhibitors of an enzyme, the observed first-order in- activation rates should not be employed unlesy the K,'s of the compounds are equal. Rather, the comparison should be made by the procedure outlined in this paper so that both K , and k 2 are evaluated.

Enzyme Inhibitors. XVIII. Studies on the Stereoselectivity of Inhibition of Adenosine Deaminase by DL-, D-, and ~-9-(2-Hydroxypro-pyl)adeninel"

HOWARD J. SCHAEFFER AND ROBERT VISCE'~

Department of Medicinal Chemistry, School of Pharmacy, State L'niversity of ,Yew Y o r k at Buffalo, Buffalo, Sew York 14214

Received December g4, 1966

Previous studies on the inhibit,ion of adenosine deaminase with compounds t'hat contain an asymmetric center utilized the corresponding racemic compounds. In order t,o det,ermiiie if adellosine deaminase exhibits a stereo- selectivit,y in complexation with one enantiomer of a DL mixture, the syiit,heses of D-( - ) and L-( +) isomers of 9-(2- hydroxypropy1)adenine ( 6 ~ and 6 ~ ) were undertaken because the DL racemate of this compound has previously been shown to be a good reversible inhibitor of this enzyme. Enzymic evaluation of L( + )-9-(2-hydroxypropyl)- adenine (6~) and D-( - )-9-(Zhydroxypropyl)adenine ( 6 ~ ) revealed that 6 - ~ is a much better inhibitor of adeiio- sine deaminase than is 6 ~ ; the ( [ I ] / [S ] )o .6 of 6 - ~ = 0.148, whereas the ( [ I ] / [S] ) , . j of 6 - ~ = 1.48. .4 rationaliza- tion for this difference in inhibitory properties is presented. Calculations based on the ([I]/[S])o,j of 6 - ~ com- pared to the ([I]/[S])O.: of 9-(2-hydroxyethyl)adenine reveal t,hat the free energy of binding of t'he methyl group of 6~ is - 1.15 kcal, a value which cannot be accounted for on t,he basis of hydrophobic forces alone. Thus, the positive involvement of van der Waals forces is invoked. Based on these and other dat,a, it is concluded that there is a specific methyl binding region on adenosine deaminase which forms a unique "tight fit'' or "lock and key" type of fit with the methyl group of ~-(+)-9-(2-hydroxypropyl)adeniiie (6~).

It is well-known t,hat many enzymes exhibit stereo- selectivity when complexes are formed between t'he enzyme and a molecule that contains one or more asym- metric centers. Recently, it has been suggested t'hat calf intestinal mucosal adenosine deaminase has both polar and nonpolar areas which are important for binding the substituents at the 9 posit'ion of a 6-sub- stit'uted purine derivative.2

Since many of the compounds which have been found to inhibit adenosine deaminase contain an asymmet'ric center, the possibility existed that the enzyme was only combining with one of t'he enantiomers of a DL mixture. It became desirable, t'herefore, to det'ermine if adeno- since deaminase exhibits either a specificity or selec- tivit'y3 when combining with an inhibitor molecule.

(1) (a) This investigation was supported by Grant T-337A from the American Cancer Society, by a Public Health Service research grant (5- R01-GM-09775-05). by a research career program award (5-K3-CA4-18718-05) from the National Cancer Institute, and a training grant (5-T1-GM-555-05) from the Division of Medical Sciences. U. S. Public Health Services, Beth- esda, Md. (b) Recipient of 1966 Lusford Richardson Pharmacy Award.

(2) H . J. Schaeffer and R. Vince, J . M e d . Chem., 8, 507 (1965). (3) The term selectivity implies tha t the inhibition of a n enzyme occurs

mainly with one enantiomer of a DL pair, whereas specificity implies that the activity is exclusively in one enantiomer of a DL pair.

I n order to study this problem, me decided to prepare t'he optically active forms of 9-(%hydroxypropyl)ade- nine, a good reversible inhibitor of adenosine deaminase, and evaluate these compounds as reversible inhibitors of this enzyme.

Chemistry.-The general method of synthesis of D- and L-9-(2-hydroxypropyl) adenines was patterned after the method utilized for the preparation of the racemic c o m p o ~ n d . ~ Optically pure D-( -)-l-amino-2- propanol (2-D) was obtained from D-( -)-lactic acid ( l - ~ ) in three steps by a modification of a previously reported procedure; (see Chart I). Treatment of 2-D wit'h 5-amino-4,6-dichloropyriniidine (3) resulted in the formation of D-( -)-5-amino-4-chloro-6-(2- hydroxypropy1aniino)pyrimidine ( 4 - ~ ) . Cyclization of 4-D wit'h a mixture of triethyl orthoformate and con- centrat'ed HCl gave the required 6-chloropurine deriva- tive (5-D) , which upon t,reat,ment with liquid ammonia gave D-( -)-9-(2-hydroxypropyl)adenine (6-D). The series of L isomers was obt,ained by a sequence of reac-

(4) H. J. Schaeffer, D. Vogel, and R . Vince, J . J f e d . Chem., 8, 502 (1965). ( 5 ) D. E. Wolf, 11.. H. Jonea, J. Valiant, and K. Folkers, J . A m . Chem.

SOC., 72, 2820 (1950).