Biochimica et Biophysica Acta 912 (1987) 203-210 203 Elsevier

BBA 32801

The pH dependence of pre-steady-state and steady-state kinetics for the papain-catalyzed hydrolysis of N-c~-carbobenzoxyglycine p-nitrophenyi

ester

Paolo Ascenzi a, Patrizia Aducci a, Antonio Torroni a, Gino Amiconi a, Alessandro Ballio a, Enea Menegatti b and Mario

Guarneri b

a C.N.R., Center for Molecular Biology, Department of Biochemical Sciences, University of Rome La Sapienza, Rome,

and b Department of Pharmaceutical Sciences, University of Ferrara, Ferrara (Italy)

(Received 16 June 1986) (Revised manuscript received 8 December 1986)

Key words: Papain; Cysteine proteinase; Proteinase kinetics; pH effect; (C. papaya L.)

Pre-steady-state and steady-state kinetics of the papain (EC 3.4.22.2)-catalyzed hydrolysis of N-a-carbo- benzoxyglycine p-nitrophenyl ester (ZGlyONp) have been determined between pH 3.0 and 9.5 (I = 0.1 M) at 21 5:0.5 o C. The results are consistent with the minimum three-step mechanism involving the acyl • enzyme intermediate E • P:

k+l k+2 k+3 E + S ~ E . S ~ E . P ~ E + P 2

k_t k_ 2 1~i k-3

The formation of the E- S complex may be regarded as a rapid pseudoequilibrium process; the minimum values for k+i are 8.0 pM - l s - i (pH ~< 3.5) and 0.40 pM -i s-1 (pH > 6.0), and that for k_ 1 is 600 s -1 (pH independent). The pH profile of k+ 2 ~Ks (= k~t/Km; Ks = k_ 1/k+l) reflects the ionization of two groups with pK' values of 4.5 5:0.1 and 8.80-t-0.15 in the free enzyme. The pH dependence of k+2 and k + 3 (measured only at pH values below neutrality) implicates one ionizing group in the acylation and deacylation step with pK" values of 5.80 :t: 0.15 and 3.10 5: 0.15, respectively. As expected from the pH dependences of k+2 /K S ( = k ~ t / K m) and k+2, the value of K s changes with pH from 7.5.101 /~M (pH ~< 3.5) to 1.5" 10 3 pM (pH > 6 . 0 ) . Values of k-2 and k-3 are close to zero over the whole pH range explored (3.0 to 9.5). The pH dependence of kinetic parameters indicates that at acid pH values (~< 3.5), the k +2 step is rate limiting in catalysis, whereas for pH values higher than 3.5, k +3 becomes rate limiting. The observed ionizations probably reflect the acid-base equilibria of residues involved in the catalytic diad of papain, HislS9-Cys as. Comparison with catalytic properties of ficins and bromelains suggests that the results reported here may be of general significance for cysteine proteinase catalyzed reactions.

Abbreviations: ZGIyONp, N-a-carbobenzoxyglycine p-nitro- phenyl ester; ZGly, N-a-carbobenzoxyglycine.

Correspondence: P. Ascenzi, C.N.R., Center for Molecular Biology, Department of Biochemical Sciences, University of Rome "La Sapienza", Piazzale Aldo Moro 5, 00185 Rome, Italy.

Introduction

Over the past fifty years, a number of cysteine and serine proteinases have been found and/or isolated from fruits, leaves and latex of several plants [1]. Among cysteine proteinases, papain

0167-4838/87/$03.50 © 1987 Elsevier Science Publishers B.V. (Biomedical Division)

204

(EC 3.4.22.2), isolated from the latex of the unripe melon-like fruit of the tropical pawpaw tree (Carica papaya L.), has been extensively studied from both dynamic and static viewpoints (see Refs. 1-6 for reviews), and therefore is generally taken as the molecular model for this class of enzymes. However, in spite of the general interest in structure-function relationships in papain, only limited pre-steady-state measurements are availa- ble (see Refs. 3-8). Since these kinds of data are relevant for a detailed characterization of the cata- lytic mechanism of papain, pre-steady-state and steady-state kinetics for the hydrolysis of ZGlyONp, one of the most investigated substrates for this cysteine proteinase [3,7,9,10], have been determined between pH 3.0 and 9.5 ( I = 0.1 M), and T = 2 1 +0 .5°C. Catalytic and molecular properties of papain are compared with those of ficins and bromelains to support structure-func- tion relationships in these homologous cysteine proteinases.

Materials and Methods

Papain was isolated according to Baines and Brocklehurst [11] from commercial preparations (Sigma Chemical Co., St. Louis, U.S.A.). The pro- teinase purity and homogeneity were checked and corresponded to criteria proposed by Baines and Brocklehurst [11], and Zucker et al. [12].

Papain concentration was determined (i) by titration with L-3-carboxy-2,3-trans-epoxypropion- yl-leucylamido-(4-guanidino)butane [12,13] and (ii) from the absorbance at 280 nm (e = 58.5 mM -1 cm -1 [14]). Papain was approximately 60% active in purified enzyme preparations; according to Zucker et al. [12], the inactive proteinase present in papain preparations consists of unactivable (irreversible oxidized) enzyme, as observed previ- ously [4]. Therefore, all molar concentrations of papain stated in the text refer to active enzyme determination by titration with L-3-carboxy-2,3- trans-epoxypropionyl-leucylamido-(4-guanidino)- butane.

The following compounds: ZGlyONp, ZGly, L-3-carboxy-2,3-trans-epoxypropionyl-leucylami- do-(4-guanidino)butane, acetonitrile, p-nitrophe- nol, L-cysteine and ethylenediaminotetraacetic acid (tetrasodium salt) were purchased from Sigma Chemical Co. (St. Louis, U.S.A.). All other prod-

ucts were from Merck AG (Wuppertal, F.R.G.). All chemicals were of analytical grade and used without further purification.

The papain-catalyzed hydrolysis of ZGlyONp was monitored spectrophotometrically between 290 and 410 nm [7,9,10,15] with a VARIAN dou- ble-beam spectrophotometer (Cary 118 or 219) or by a DURRUM-Gibson rapid-mixing stopped- flow apparatus.

Under all the experimental conditions, the ini- tial velocity for the papain-catalyzed hydrolysis of ZGIyONp was corrected for the acid or alkaline hydrolysis of the substrate [16].

Following the literature [7,9,10,17], papain was preactivated in the assay buffer containing 2.0. 10 -3 M ethylenediaminotetraacetic acid (tetra- sodium salt) and 5.0.10 -3 M g-cysteine, before each experiment.

ZGlyONp was dissolved in acetonitrile and then diluted in the buffered reaction mixtures; the final acetonitrile concentration was 1.6% (v/v). As ex- pected [7,9,10], control experiments indicated that acetonitrile, up to 3.0% (v/v), never affected sig- nificantly (less than 5%) the papain catalyzed hy- drolysis of ZGIyONp.

The pH profile was explored using the follow- ing buffers: formate (pH 3.0-3.5); acetate (pH 3.5-6.0); phosphate (pH 6.0-8.5); borate (pH 8.5-9.5), all at 0.1 M ionic strength (sodium salts). As expected [10], control experiments with differ- ent buffers overlapping in pH showed no specific anion effects.

All measurements were performed at 21 + 0.5°C.

Determination of kinetic parameters The pre-steady-state and steady-state parame-

ters for the papain-catalyzed hydrolysis of ZGlyONp were obtained from the experimental data according to the standard treatment of the minimum three-step mechanism (Scheme I) of cys- teine proteinases [3-5,15,18]:

k+l k+2 k+3 E+S~ E.S~ E-P~ E+P z (I)

k 1 k-2 k-3 +

P1

where E is the enzyme (i.e., papain), S is the substrate (i.e., ZGlyONp), E. S is the reversible

rapidly formed enzyme-substrate complex, E - P is the acyl. intermediate, P1 and P2 are the hydro- lysis products (i.e., p-nitrophenol and ZGly, re- spectively).

Values of kcat, K m and kcat/K m were de- termined from the intercepts on the ordinate and the abscissa and the slope, respectively, of plots of 1/ ini t ial velocity versus 1/[So] with [So] >/5. [Eo]. Values of k+2, K s ( = k _ J k + l ) and k+2//K s were determined from the intercepts on the ordinate and abscissa and the slope, respectively, of plots of 1 / k app versus 1/[Eo] with [Eo] >/5. [So]. At pH values higher than 6.0, values of K s become much greater than the solubility of papain [7], and thus only the ratio k+2 /K s can be de- termined; therefore, the single values of k+2 and K s have been obtained only between pH 3.0 and 6.0. Values of k+3, calculated by substitution of the experimentally determined values of kcat, Kin, k+2 and K s into Eqns. 1 and 2:

k+3 = (kca t .k+2)/(k+2 - kcat) (1)

k + s = ( k + 2 " K m ) / ( K s - K m ) (2)

were the same within the errors (see Table I). At pH values higher than 6.0, values of k+3 for the papain-catalyzed hydrolysis of ZGlyONp cannot be evaluated, since single values of k+ 2 and K s are not available (see above).

Under all the experimental conditions, the pre- steady-state and steady-state parameters for the papain-catalyzed hydrolysis of ZGlyONp are not affected by the addition of p-nitrophenol and /o r ZGly (i.e., P1 and P2, respectively, of Scheme I), up to 500 /~M, in the reaction mixtures. Thus, it can be deduced that values of k_ 2 and k_ 3 are always close to zero.

When [So]>[Eo], the differential equations arising from Scheme I may be solved to describe the time course of p-nitrophenol release (P~) [15,18]:

[Pll = {(k¢~t'[Eol't)/(l+ Km/[So] )} + a [ E o l ' ( 1 - e -k')

(3)

where

205

corresponds to the amplitude of the burst phase of p-nitrophenol release, and

k = ( k+2/(l + KJISol) } + k+3 (5)

is the first-order rate constant for the burst phase of p-nitrophenol release. The low solubility of papain [7] prevented the possibility to evaluate the amplitude and the first-order rate constant of the burst phase of p-nitrophenol release from ZGlyONp at pH values higher than 6.0 (see above).

An average error value of + 8% was evaluated for kinetic parameters according to Atkins and Nimmo [19].

Analysis of the p H dependence of catalytic parame- ters

The pH dependence of k+ 2 /K s (--kcat/Km) has been analyzed according to Cleland [20] and Tipton and Dixon [21], with Eqn. 6:

log y = log( C/(1 + [H + ]/K( + K:~/[H + 1) } (6)

where y is the relevant catalytic parameter, C is the pH-independent value of y, and K 1' and K 2 are the acidic dissociation constants referring to acid-base equilibria of amino acid residues, occur- ring in the acidic and alkaline pH limbs, respec- tively, for the ZGly(ONp)-free papain.

The pH dependence of k+2, k+3 , kcat, k, 'n'[So] (= a[Eo]) has been analyzed according to Cleland [20] and Tipton and Dixon [21], with Eqn. 7:

log y = log{ C/(1 + [H + l/K;" + K~'/[H + ])} (7)

where y is the relevant catalytic parameter, C is the pH-independent value of y, and K(' and K~' are the acidic dissociation constants referring to acid-base equilibria of amino acid residues, occur- ring in the acidic and alkaline pH limbs, respec- tively, for the ZGly(ONp)-bound papain.

The pH dependence of K s and K m has been analyzed according to Eqn. 8, assuming the sim- plest model (Tipton and Dixon [21]), which in- volves the pK shift of one (for K s values) or two (for K m values) ionizing group(s) on substrate binding:

( (III+]+ l¢;)'(tn+]+ r~) log y = log c + log ([H+ ]----~+--K;,--~. ([H------W ] ~ K~') a= {k+2/(k+2 + k+3)/(l+ Km/[So])} 2 (4) (8)

206

where y is the relevant catalytic parameter, C is the pH-independent value of y, K( and K~ refer to the functionally relevant residues in ZGly- (ONp)-free papain, and K( ' and K~' to the same residues in ZGly(ONp)-bound papain (see above). Eqn. 8 shows that if a change in pH shifts the apparent value of the constant ( K S or g m ) , then a change of substrate concentration in a given range should modify the apparent p K of the ionizing group(s) linked to substrate binding. It is, there- fore, possible to calculate the protonated fraction of papain at each pH and substrate concentration as well as the difference in protein-bound protons with increasing the substrate concentration from zero to any [ZGly(ONp)]. The proton uptake is zero for [H +] very different from p K values. Therefore a maximum proton uptake has to be observed at some intermediate pH value, i.e., pHm~ × = ( p K ' + p K " ) / 2 , prima x is usually indi- cated as mid-point of the titration. Once the mid- point and the distance between the end points of the curve are known, K ' and K " may easily be calculated by Eqn. 8.

Solid lines (see Figs. 1 and 2), calculated with parameters given in the text and the appropriate equation (i.e., Eqns. 6, 7 or 8), were obtained with an iterative non-linear least-squares curve fitting procedure, according to Cleland [20]; an average error value of _+ 15% was ascribed to the values of the acidic dissociation constants. Within the ex- perimental error, the agreement between the ex- perimental points and the calculated solid lines is satisfactory (see Figs. 1 and 2).

Results

Under steady-state conditions ([So] >i 5-[Eo]), the papain-catalyzed hydrolysis of ZGlyONp con- forms to simple Michaelis-Menten kinetics. In fact, the dependence of the initial velocity on the sub- strate concentration follows simple saturating kinetics, under all the experimental conditions. Moreover, over the range explored (1.0-10 -8 to 1.0- 10 5 M), the initial velocity is strictly linear with the papain concentration.

Over the pH range explored (3.0-9.5), no lag phase is detectable in p-nitrophenol release after mixing E with S ([Eo] >i 5. [So] ), indicating that the process E + S ~ ( k + l / k _ l = K s 1) E - S is

complete within 3 ms, the dead-time of the rapid- mixing stopped-flow apparatus, and therefore k ~< 600 s-~. Such a finding is fully consistent with the classical assumption k 1 >> k+2 , as predicted from Eqns. 1-5 [15,18]. Taking k ~ = 600 s -1, the minimum values of the second-order rate constant for the formation of the E . S (i.e., papain . ZGlyONp) complex (k+l = 6 0 0 / K s ; K S values are. shown in Fig. 1) are 8.0 g M - 1. s - 1 (pH ~< 3.5) and 0.40 g M -1 s -1 (pH > 6.0).

The pH dependences of catalytic parameters kzat, k + 2, k + 3, Kin, Ks, k ~ a t / K m and k + 2 / K S for the papain-catalyzed hydrolysis of ZGlyONp are shown in Fig. 1; data obtained at 25°C from literature [7,9,10], and included for comparison, agree very well with present results.

Over the pH range explored (3.0-9.5): (i) val- ues of k+z, k+3 and kca t for the papain-catalyzed hydrolysis of ZGIyONp (see Fig. 1) are affected

+ 3 . . . . ,_ : ,

÷1

3 4 5 6 7 8 9 10

p H

Fig. 1. pH dependence of log kca ~ (0), log k+2 (zx), log k+3 (12), log K m (O), log K s (E]), log k c a t / K m (O) and log k + 2 / K s

(m) for the papain-catalyzed hydrolysis of ZGIyONp. Values of k+3 are averages of those reported in Table I. Data obtained from literature [7,9,10] for the papain-catalyzed hydrolysis of ZGlyONp have been also included for comparison (filled symbols; T = 25 o C). Solid lines, calculated with parameters given in the text and according to Eqns. 6, 7 or 8, are theoretical curves for one or two ionizing group(s). For all the other experimental details, see the text.

by one ionizing group with pK(' values of 5.80 + 0.15, 3.10 + 0.15 and 3.8 +__ 0.1, respectively; (ii) values of K s and K m for the papain/ZGlyONp system (see Fig. 1) depend on titrations with one or two apparent mid-point value(s) at pH 5.15 + 0.10 (= (4.5 + 5.8)/2), and 3.65 + 0.10 (= (3.25 + 4.05)/2) and 8.6 (= (8.4 + 8.8)/2), respec- tively; and (iii) values of k+2/K s (= k c a t / g m ) for ZGlyONp hydrolysis (see Fig. 1) show a good fit to two ionizing groups with pK( and pK 2 values of 4.5 + 0.1 and 8.80 + 0.15. The values of the Hill coefficient [21], related to the pH dependence of catalytic parameters, are always equal to 1 (0.98 to 1.02), within the experimental error. The pH vari- ation of kca t is that predicted by the separate pH profiles for k÷2 and k÷3 (see Fig. 1).

As shown in Fig. 1, k+2 (the acylation step) is rate limiting for the papain-catalyzed hydrolysis of ZGlyONp at pH ~< 3.5, whereas, for pH values higher than 3.5, k + 3 (the deacylation step) be- comes rate limiting. Therefore, according to Eqns. 3-5 (when [So] >~ lEo] and k+2 >_ k + 3 ) , only at pH >~ 3.2, a burst of p-nitrophenol release of am- plitude a[Eo], corresponding to ~r[So], with a first-order rate constant k is experimentally detec- table; as expected (see Fig. 2), such a burst phase decreases by lowering pH from 6.0 (where k÷2 > k + 3 ) t o 3.0 (where k+z < k + 3 ). As shown in Fig. 2, the experimental and calculated (according to Eqns. 3-5) values of the amplitude and of the first-order rate constant of the burst phase of p-nitrophenol release from ZGlyONp catalyzed by papain are in good agreement.

The pH dependence of the amplitude (i.e., or[So] ) and of the first-order rate constant (i.e., k) of the burst phase of p-nitrophenol release for the papain-catalyzed hydrolysis of ZGlyONp (see Fig. 2) may be fitted with simple titrations with ap- parent pKl" values of 3.7 + 0.1 and 3.80 + 0.15, respectively.

Discussion

The pre-steady-state and steady-state results in- dicate that kinetics for the papain-catalyzed hy- drolysis of ZGIyONp may be consistently fitted to the minimum three-step mechanism (Scheme I) of cysteine proteinases (see Refs. 2-6 for reviews), over the whole pH range explored (3.0-9.5). In

207

fact, the treatment of kinetics according to Scheme I is consistent with: (i) the excellent agreement between values of k+2/K s (obtained where [Eo] >/ 5. [So] ) with those of k c a t / K m (determined where [So] ~> 5-[Eo]) (see Fig. 1); (ii) the coincidence between values of k + 3 calculated according to Eqns. 1 and 2 (see Table I); and (iii) the good agreement between the experimental and calcu- lated (according to Eqns. 3-5) values of the ampli- tude and of the first-order rate constant of the burst phase of p-nitrophenol release from ZGlyONp catalyzed by papain (see Fig. 2). How- ever, it should be emphasized that this reaction mechanism is only minimal (i.e., phenomenologi- cal) and other more complex models may apply as long as certain conditions are satisfied (see Refs. 8 and 22). Thus, a conformational change of the enzyme, preceding both the acylation and deacyl- ation steps, has been taken into account as the rate-limiting process in the hydrolysis of lysine and arginine derivatives catalyzed by papain as well as ficins (see Ref. 8). Moreover, the possibil- ity that the departure of the leaving group is kinetically significant has been considered [22] as

TABLE I

VALUES OF THE DEACYLATION RATE CONSTANT (k÷ 3) FOR THE PAPAIN-CATALYZED HYDROLYSIS OF ZGIyONp CALCULATED ACCORDING TO EQNS. 1 AND 2 BETWEEN pH 3.0 AND 6.0 ( I = 0,1 M; T = 21 +0 .5°C)

For all the other experimental details, see the text as well as Fig. 2. At pH values higher than 6.0, values of k+3 cannot be evaluated, since single values of k+2 and K s are not available, the solubility of papain [7] being lower than K~ (see the text).

pH k + 3 (s - l ) calculated from:

Eqn. 1 Eqn. 2

3.0 2.3 2.2 3.2 2.5 2.3 3.4 3.1 3.0 3.6 3.0 2.8 3.8 3.4 3.3 4.0 3.6 3.5 4.3 4.0 4.0 4.5 4.8 4.8 4.75 4.7 4.6 5.0 4.5 4.4 5.3 4.5 4.5 5.5 4.4 4.5 5.75 4.5 4.5 6.0 4.0 4.1

2O8

7o- r,,,e~,~ _/~" ~ o [] ~ 14o

[ . J li 20

I0 hme (s) ]

~ 0 3 4 5 6

oH

Fig. 2. pH dependence of the amplitude ( . , ©; left scale) and of the first-order rate constant (11, []; right scale) of the burst phase of p-nitrophenol release during the papain-catalyzed hydrolysis of ZGlyONp (the left scale represents the per- centage of p-nitrophenol released from ZGIyONp at different pH values; 100% is equivalent to 100 #M of ZGIyONp, see below). Filled and open symbols refer to the experimentally measured and calculated (according to Eqns. 3-5) values of the amplitude and of the first-order rate constant of the burst phase of p-nitrophenol release. Solid lines, calculated with parameters given in the text and according to Eqn. 7, are theoretical curves for one ionizing group. At pH = 3.0 (inset A), the burst phase of p-nitrophenol release is experimentally undetectable (see the text). At pH = 6.0 (inset B), the ampli- tude of the burst phase of p-nitrophenol release (rr[So]) corre- sponds to 0.155[So] (i.e., 0.775[Eo] ), and the related first-order rate constant is 31.0 s -1. At all pH values, papaln was 20 btM and ZGlyONp was 100 #M. For all the other experimental details, see the text as well as Fig. 1.

the rate-limiting step in the papain-catalyzed hy- drolysis of N-a-benzoylglycine p-nitrophenyl es- ter, at least in the presence of high concentrations of organic solvents.

The pH dependence of the catalytic parameters for the papain-catalyzed hydrolysis of ZGlyONp can be described in terms of only two apparent ionizing groups, one dissociating in the acidic pH region and the other in the alkaline limb.

It is generally accepted that the pH dependence of k + 2 / K s (= k c a t / K m ) , k+2 and k + 3 follows the ionization(s) in the free enzyme (E), in the E-S and E. P complexes (see Scheme I), respectively [23]. Thus, according to one theory [23], the pH variation of k + z / K s ( = k c a t / K m ) , n o t only for papain but also for ficins- and bromelains-cata- lyzed reactions, is independent of the nature of the substrate being in each case bell-shaped and re-

flecting ionization(s) with p K ' values ranging be- tween pH 4.0 and 4.6, and 8.4 and 9.0 (see Refs. 2-6 for reviews). Unfortunately, due to technical difficulties (e.g., low papain solubility at pH val- ues higher than 6.0 [7]), data in the alkaline limb are not extensive enough to allow a reliable mech- anistic interpretation. On the other hand, an abun- dance of accurate and precise results in acidic media allows one to assume that just one func- tionally relevant acidic group is associated with the active site. The correctness of this assumption, which is simply a device for describing the data, is supported by the value of the slope (equal to 1) of the straight-line sections in the pH profile plot, such as k + 2 / K s (= k ~ , t / K ~ ) versus pH. Since this is a phenomenological point of view, it does not exclude a priori that two amino acid residues may participate in modulating k + z / K s (= kcat/ Kin) below pH 6.0, as suggested in the literature [4,5,24].

Data shown in Fig. 1 imply a pK shift of the ionizing group in the enzyme on substrate bind- ing, between pH 3.0 and 6.0. In fact, K s changes over the same pH range, and k+2 versus pH profile implicates one ionizing group with a pK~" value of 5.80 + 0.15 in the E. S complex, 1.3 pH units higher than that of the residue in the free enzyme (pK[ = 4.5 + 0.1). Assuming that these two pK a values are associated with the same titrable group, the value of K s should increase by 20-fold on going from pH 3.0 to 6.0. According to linkage relations [21], the value of K s for the papain-cata- lyzed hydrolysis of ZGlyONp changes from 7.5 • 101/~M, at pH ~< 3.5, to 1.5 • 103/~M, at pH > 6.0. Moreover, the pH dependence of K s indicates a preferential binding of the substrate to the enzyme form in which the acidic group is protonated, as also reported for the ficin-catalyzed hydrolysis of N-a-benzoylglycine p-nitrophenyl ester [15].

The principle of microscopic reversibility re- quires that the k+3 versus pH profile should implicate in the acyl .enzyme adduct the ioniza- tion of the same functional residue observed in the E- S complex, unless a rate-limiting change in the protein conformation precedes the deacylation step. The last hypothesis is suggestive, in that the apparent p K [ ' (= 3.10 + 0.15) of the more acidic ionization, relevant in the deacylation step, is lower than that in the free papain (1.4 pH units) and

that in the acyl.enzyme intermediate (2.7 pH units). Such a finding, also observed in the ficin- catalyzed hydrolysis of N-a-benzoylglycine p- nitrophenyl ester [15], may be interpreted in terms of change(s) in the microenvironment of the func- tionally relevant ionizing group which becomes more positive (or less negative) in the E-P com- plex than in the free enzyme and in the E. S adduct. This explanation, based on the assump- tion that the same titrable group affects E, E. S and E. P (see Scheme I), is consistent with the view, expressed for both cysteine and serine pro- teinases, that conformational changes precede or occur during the deacylation step (see Refs. 2-6, 25-27 for reviews).

As previously reported for the ficin-catalyzed hydrolysis of N-a-benzoylglycine p-nitrophenyl ester [15], the results of the present study (see Figs. 1 and 2) demonstrate that, at pH values higher than 3.5, the k+3 step (deacylation process or the related conformational change(s)) is rate limiting in papain catalysis, whereas, at pH values lower than 3.5, the k+2 step (acylation process) becomes rate limiting. It is relevant that the pK(' values of the pH dependence of the amplitude and of the first-order rate constant of the burst phase of p-nitrophenol release for the papain-catalyzed hydrolysis of ZGlyONp (see Fig. 2) agree with the pH value (--3.5) at which the change in the rate-limiting step occurs (see Fig. 1). This il- lustrates the potential danger in interpreting: (i) kca t versus pH profile for cysteine proteinase ac- tion on p-nitrophenyl esters as substrates on the basis that the deacylation step is rate limiting throughout the whole pH range; and (ii) K m as

the true affinity constant for enzyme, substrate association.

Although transient structural changes leading to differences in pK and/or mid-point values of ionizing group(s) involved in the free enzyme and in the catalytic intermediates (i.e., E. S and E. P) cannot be directly interpreted on the basis of the stereochemical models (see Refs. 5, 28 for reviews), the catalytic properties of papain are likely to be related to different enzyme, substrate interactions occurring in the different steps of catalysis. According to literature (see Refs. 1-6, 24 for reviews), the observed pH effects probably reflect

209

the acid-base equilibria of the components of the catalytic diad of papain, His159-Cys 25 [28]. In other words, increase in k+2/Ks (= kcat/K,,,) as the pH is increased to pH 6.0 appears to be essentially synchronous with the generation of the ImH÷/S - ion pair. Thus, the pK' values in the ZGly- (ONp)-free papain are 4.5 + 0.1 for the protona- tion of the Cys- residue, and 8.80 + 0.15 for the deprotonation of the HisH ÷ side chain, as sug- gested from the pH profile of k+2/K s (= kca t /Km) . Accordingly, when His-159 is deproto- nated, ZGlyONp affinity for papain should de- crease drastically, since kca t values are pH inde- pendent at high pH values. In the acid-pH limb, ZGlyONp binding raises the pK 1 of Cys-25 from 4.5 ___ 0.1 to 5.80 + 0.15 (as suggested from the pH titrations of k+2/K ~ (= kcat/Krn ) and k+2, re- spectively), and conversely, the substrate then binds papain more tightly below pH 4.0 than above pH 6.0 (as shown by the pH dependence of Ks). Once ZGIyONp has reacted to form the acyl.enzyme intermediate (i.e., E. P adduct in Scheme I), the pK(' value of Cys-25 drops to 3.1 (as shown from the pH profile of k+3 ), indicating the difficulty of protonating this group when the active site is filled with the E. P thioester, and is possibly neutral.

The nature of further modulation(s) of catalytic events in papain, by another positively cooper- ative protonic dissociation with pK~ ranging be- tween 3.0 and 4.0 [29,30], has never been estab- lished, although it has usually been attributed to an electrostatic-field effect contributed by the carboxy group of Asp-158.

As a whole, the reported data show that the catalytic diad is a sufficient condition for func- tional activity of papain and therefore, contrary to other cysteine proteinases (e.g., cathepsin B [31]), further modulation(s) by ionic or polar groups in the catalytic site of strategic regions of the protein does not appear to be a strict requirement.

Acknowledgements

This work was supported by grants from the Italian Ministry of Education (Ministero della Pubblica Istruzione) and the Italian Research Council (Consiglio Nazionale delle Ricerche).

210

References

1 Ryan, C.A. and Walker-Simmons, M. (1981) in Plant Pro- teinases (Stumpf, P.K. and Corm, E.E., eds.), pp. 321-350, Academic Press, New York

2 Cooreman, W.M., Scharpe, S., Demeester, J. and Lauwers, A. (1976) Pharm. Acta Helv. 51, 73-97

3 Lowe, G. (1976) Tetrahedron 32, 291-302 4 Brocklehurst, K., Baines, B.S. and Kierstan, M.P.J. (1981)

Top. Enzyme Ferment. Biotechnol. 5, 262-335 5 Polgar, L. and Halasz, P. (1982) Biochem. J. 207, 1-10 6 Aducci, P., Ascenzi, P., Ballio, A. and Antonini, E. (1984)

in Proteinase Action (Symposia Biologica Hungarica 25) (E16di, P., ed.), pp. 453-466, Akad~miai Kiad6, Budapest

7 Hubbard, C.D. and Kirsch, J.F. (1968) Biochemistry 7, 2569-2573

8 Hollaway, M.R. and Hardman, M.J. (1973) Eur. J. Bio- chem. 32, 537-546

9 Kirsch, J.F. and llgestr6m, M. (1966) Biochemistry 5, 783-791

10 Williams, D.C. and Whitaker, J.R. (1967) Biochemistry 6, 3711-3717

11 Baines, B.S. and Brocklehurst, K. (1979) Biochem. J. 177, 541-548

12 Zucker, S., Buttle, D.J., Nicklin, M.J.H. and Barrett, A.J. (1985) Biochim. Biophys. Acta 828, 196-204

13 Barrett, A.J., Kembhavi, A.A., Brown, M.A., Kirschke, H., Knight, C.G., Tamai, M. and Hanada, K. (1982) Biochem. J. 201, 189-198

14 Glazer, A.N. and Smith, E.L. (1971) in The Enzymes, 3rd Edn., (Boyer, P.D., ed.), Vol. 3, pp. 501-546, Academic Press, New York

15 Hollaway, M.R., Antonini, E. and Brunori, M. (1971) Eur. J. Biochem. 24, 332-341

16 Ascenzi, P., Sleiter, G. and Antonini, E. (1983) Gazz. Chim. Ital. 113, 859-861

17 Arnon, R. (1970) Methods Enzymol. 19, 226-244 18 Gutfreund, H. (1972) Enzymes: Physical Principles, Wiley-

Interscience, London 19 Atkins, G.L. and Nimmo, I.A. (1973) Biochem. J. 135,

779-784 20 Cleland, W.W. (1979) Methods Enzymol. 63, 103-138 21 Tipton, K.F. and Dixon, H.B.F. (1979) Methods Enzymol.

63, 183-234 22 Henry, A.C. and Kitsch, J.F. (1967) Biochemistry 6,

3536-3544 23 Peller, L. and Alberty, R.A. (1959) J. Am. Chem. Soc. 81,

5907-5914 24 Lewis, S.D., Johnson, F.A., Ohno, A.K. and Sharer, J.A.

(1978) J. Biol. Chem. 253, 5080-5086 25 Neuman, R.C., Jr., Owen, D. and Lockyer, G.D. Jr. (1976)

J. Am. Chem. Soc. 98, 2982-2985 26 Ascenzi, P., Menegatti, E., Guarneri, M., Bolognesi, M. and

Amiconi, G. (1984) Biochim. Biophys. Acta 789, 99-103 27 Ascenzi, P., Pertollini, A., Bolognesi, M., Guarneri, M.,

Menegatti, E. and Antonini, E. (1984) in Proteinase Action (Symposia Biologica Hungarica 25) (El6di, P., ed.), pp. 339-353, Akad~miai Kiad6, Budapest

28 Drenth, J., Kalk, K.H. and Swen, H.M. (1976) Biochem- istry 15, 3731-3738

29 Sluyterman, L.A.E. and Wydenes, J. (1973) Biochim. Bio- phys. Acta 302, 95-101

30 Brocklehurst, K. and Malthouse, J.P.G. (1980) Biochem. J. 19l, 707-718

31 Willenbrock, F. and Brocklehurst, K. (1984) Biochem. J. 222, 805-814