PH DEPENDENCE OF CARBOXYPEPTIDASE A
pH Dependence of the Hydrolysis of Hippuric Acid Esters by Carboxypeptidase At John W . Bunting* and Samuel S.-T. Chu
ABSTRACT: The pH dependence (pH 4.5-10.5) of the hydrolysis of seven hippuric acid esters (C6H5CONHCH2C02CRlR2C02H: la: R 1 = R2 = H ; lb: R l = R2 = CH3; IC: R l = H, R2 = p-C1C6H4;Id: R 1 = H, R2 = C2H5; le: R I = H, R2 = (CH3)2CHCH2;If: R1 = H, R2 = C6H5; lg: Ri = H, R2 = CsHsCH2) by bovine carboxypeptidase A has been investigated, and the pH dependence of the substrate activation of la-c and the substrate inhibition of Id-g have been compared. For all seven esters the catalytically productive binding of the first substrate molecule depends on enzymatic p K ,
In
recent investigations (Bunting and Murphy, 1974; Murphy and Bunting, 1974, 1975) we have been attempting to quantitatively describe the substrate activation and inhibition effects that complicate the analysis of the steady-state kinetics of the hydrolysis of many esters by bovine carboxypeptidase A. Esters (1) of hippuric acid are especially prone to these phenomena. Thus, for la-c, substrate activation produces sigmoidal curves for the dependence of initial velocity of enzymatic hydrolysis on substrate concentration (Murphy and Bunting, 1974). On the other hand, if Rl is H and R2 is C2H5 or a larger hydrocarbon unit, substrate inhibition is observed for the L enantiomers of the ester substrates (1) (Murphy and Bunting, 1975).
la,R1 = R2 = H b, R1 = R2 = CH3 C, R I = H ; R2 = ,D-CIC& d, R I = H ; R2 = C2H5 e, R1 = H ; R2 = (CH3)2CHCH2 f, R1 = H ; Rz = C6H5 g, Rl = H; R2 = C6HsCH2 We have shown (Murphy and Bunting, 1974, 1975) that both substrate activation and substrate inhibition' with these hippuric acid esters is consistent with the formation of a 1:2 enzyme-substrate complex ( E & ) according to Scheme I. Scheme 1 +S
E e ES K;
+S
A
E + PI + P ,
k
+S+
SE e ES? A E K,
PI + PL
f From the Department of Chemistry, University of Toronto, Toronto, Ontario M5S I A l , Canada. Receiced December 18, 1975. This work was supported by an operating grant to J.W.B. from the National Research Council of Canada. 1 I n addition to substrate activation, the esters l b and IC also display substrate inhibition at high concentrztions of substrate (Murphy and Bunting, 1974). This phenomenon has not been quantitatively analyzed and is not under consideration in the present paper.
values of 6.0 and 9.1. For Id, le, and l g the rate of hydrolysis (k2,PP) of this complex is p H independent, whereas for If k2aPP depends on a pK, of 5.9. The rate of hydrolysis (k3aPP)of the 1:2 enzyme-substrate complex (E&) is pH independent for Id-g, but for la-c k3,PP depends on the following pK, values: la, 6.1 and 9.1; lb, 5.4; IC,6.6. The pH dependences of k2,PP for If and k3aPP for IC are rationalized by the presence of c,stalytically nonproductive species. Equivalent ES2 species are believed to be productive for lc-g; however, the productive E& species for l b must be quite different.
Assuming that the equilibria are established rapidly relative to k2 and k3, this scheme generates eq 1 for the dependence of initial velocity ( u ) of enzymatic hydrolysis on the concentrations of enzyme ( E ) and substrate ( S ) provided that S >> E.
Equation 2 is a modified form of eq 1, and in general only the four parameters k2aPP,k3, KsaPP,and KssaPPcan be evaluated from the experimental rate data. It is clear that substrate activation will arise from Scheme I if k3 > k2,PP, whereas substrate inhibition will be observed for k3 < kZaPP.Provided that K I / K is~ considerably less than 1, these two criteria are simply satisfied by k3 > k2. and k3 < kz, respectively. However, substrate activation can also be generated even if k3 < k2. provided that K'/Kz is sufficiently large that k3 > k2aPP (= k 2 / ( 1 K1/K2)). This latter condition suggests that the observation of substrate activation or inhibition is then related to the relative binding strengths of substrate to two different sites on the enzyme, rather than to a change in the relative values of k2 and k3 as the alcohol unit of the ester, 1, is varied. In an earlier paper (Murphy and Bunting, 1974), considerations of relative k3 values for a number of hippurate esters led us to postulate that both of the above substrate activation mechanisms may occur independently for individual hippurate ester substrates. T o further explore these relationships between substrate activation and inhibition, we have now measured the p H dependence of enzymatic hydrolysis for a series of seven hippurate esters: three (la-c) that display substrate activation and four (Id-g) that display substrate inhibition. In initiating these studies, we felt that a knowledge of the pH dependence of the four parameters in eq 2 would provide further insight into the relationship between substrate activation and substrate inhibition, and would possibly allow a distinction to be drawn be-
+
BIOCHEMISTRY, VOL.
15,
NO.
1 5 , 1976
3237
BLYTIYG
25
t
I
I
/ /* 5'5
-
c
0
20
60
40
S
I
1-
4.5 I
80
mM
I ' Dependence of initial velocity of enzymatic hydrolysis of l a on substate concentration at pH values as indicated (25 'C, ionic strength 0.5). Data for pH 5.0,6.0, 7.0,8.0.9.0, 10.0 not shown. Curvescalculated as described i n text FIGLRE
tween the two modes of substrate activation previously postulated. At the very least, the data collected in such a study would be a valuable supplement to the relatively meager data that is currently available (Riordan and Vallee, 1963; Carson and Kaiser, 1966; Hall et al., 1969; Auld and Holmquist, 1972; Bunting et a]., 1974) on the p H dependence of the esterase activity of carboxypeptidase A. Materials and Methods Synthetic routes to the substrates used in this study have been described previously (Murphy and Bunting, 1974, 1975). The sodium salt of O-hippuryl-~-3-phenyllacticacid was purchased from Bachem Inc. Fine Chemicals, Marina Del Rey, Calif., and was found to be only 75% L isomer based on complete enzymatic hydrolysis. Stock solutions of each substrate were prepared as previously described (Bunting and Murphy, 1974). The initial velocities of enzyme-catalyzed ester hydrolysis were determined on a Radiometer pH-stat a t 25 O C , ionic strength 0.5 (substrate NaC1) by automatic titration with standard potassium hydroxide solutions. All reactions a t p H I6.0 were carried out in the presence of M zinc nitrate to counteract dissociation of the zinc ion from the enzyme (Auld and Vallee, 1971). Corrections to the apparent velocities were necessary at both low and high p H to give the true initial velocities of enzyme-catalyzed hydrolysis. At p H I5.5, incomplete dissociation of the carboxylic acid products results in less than a stoichiometric liberation of hydrogen ions per ester molecule hydrolyzed. From the total base uptake a t complete hydrolysis at each pH, the percentage (p) of protons released upon complete ester hydrolysis can be calculated. The apparent initial velocity is then multiplied by the factor 1OO/p
+
3238
BIOCHEMISTRY. VOL.
15, NO. 15, 1976
A h D 9. Values of k2,PP/ KsaPP, k3, and KsaPPKssaPPwere evaluated as described above and are plotted in Figure 4.The curve through the data for ~ z ~ P P I K s ~isPbased P on pK, values of 6.0 and 9.1, while k3,PP depends on a pK, value of 6.6. The pH dependences of substrate inhibition in the hydrolysis of L-2-hippuroxybutanoic acid (Id), L-2-hippuroxyisocaproic acid (le), U-hippuryl-L-mandelic acid (If), and O-hippurylL-3-phenyllactic acid (lg) are similar, and the data for Id in Figure 5 are typical. The data on each curve in Figure 5, and all related sets of data were fitted to eq 1 by the iteration procedure that has been previously described (Bunting and Murphy, 1972). In the case of the mandelic acid derivative (If), unique sets of parameters could not be obtained a t p H > 7.5 by fitting the experimental data to eq 2. This problem has also been encountered previously, and the reasons for multiple solutions have been discussed (Murphy and Bunting, 1974, 1975). I n such cases only values for k2aPP/K~aPPcan be confidently estimated. Values of k2aPP/K~aPPfor each of Id-lg are plotted as a function of p H in Figure 6. It is clear that the same p H dependence is observed for each of these esters and for Ib and c (Figure 4), with two enzymatic acid dissociation constants controlling the binding of each substrate. Each curve in Figure 6 is based on pK, values of 6.0 and 9.1, which give satisfactory agreement between the calculated curve and experimental data for each ester. The p H dependences of kZaPP, k3,PP, and KssaPP for Id-g
3240
U 8
BIOCHEMISTRY, VOL.
15, NO. 15, 1976
i
51
-B
3L 2
I
are compared in Figure 7. For each of these esters k3dPP appears to be p H independent over the entire p H range investigated. For three of these esters, kzaPP is also p H independent; however, the mandelic acid derivative (If) is unusual, and displays a dependence on the conjugate base form of an acidic group of pK, 5.9. Since it is not possible to unequivocally evaluate k2aPP a t p H > 7.5 for this ester, it is not possible to determine whether a group of pK2 9 may also be important in controlling kZaPP for this ester. The various pK, vaiues derived in this work for the esters la-g are collected in Table
-
1.
In Table 11, the pH independent k3 values obtained from curve-fitting in Figures 2, 4, and 7 are compared with the corresponding second-order rate constants for the nonenzymatic base-catalyzed hydrolysis of the same esters. Comparison of the k 3 / k o ~ratios gives a direct estimate of the catalytic efficiencies of the various ES2 species in ester hydrolysis.
Discussion It is clear from Table I that two enzymatic acid ionizations control the binding of the first substrate molecule to car-
PH DEPENDENCE OF CARBOXYPEPTIDASE A
TABLE I:
pH Dependence of the Hydrolysis of Hippuric Acid Esters (1) by Carboxypeptidase A.a
la lb IC Id le If 1g
6.0b 6.0 6.0 6.0 6.0 6.0 6.0
9.1 9.1 9.1 9.1 9.1 9.1 9.1
C
C
C
C
c
6. I 5.4 6.6
C
C
pH independent pH independent pH independent pH independent
pH independent pH independent 5.9
9.1
C
pH independent
From k 3 a P P / K ~ a P P K ~ ~ a Psee P ; text. Not experimentally observable.
At 25 OC, ionic strength 0.5.
boxypeptidase A for each hippurate ester. This suggests that Scheme I should be modified to include three ionization states of each enzyme species. Scheme I1 is the most general case of Scheme I1
TABLE 11: Comparison of k3 and koH for Hydrolysis of Hippuric Acid Esters (1).
Ester
k3
(min-I)
koH M-I min-')
k d k o ~ (MI
~~
la lb
2.6 x 104 7.0 x 103
IC
800 220 1IO 300 1800
Id le If lP
a
T EHS EHS,
k
h
A
200 5000
7.8 8.8
7.9 4.2 60
From Murphy and Bunting (1974). This work.
+S
+ P, + P, EH + S + P , + P , EH
aEH2S2
such a situation, where H represents a proton and the relative charges of the various species are omitted for the sake of clarity. This scheme is based on the assumption that only one of each of the 1:l and 1:2 enzyme-substrate complexes can react to produce hydrolysis products. On the assumption that all equilibria (both substrate binding and ionization) are established rapidly relative to k2 and k3, Scheme I1 generates the rate eq 6, where C Y E H=~ 1 4- ([H+]/KEH,) 4- (KEH/[H+]), etc. k2S
129a 1.4a 1020 25 a 14b 71a 30b
k3S2
At constant pH, eq 6 has the same form as eq 2, with the four parameters being described by eq 7- 10. (7)
Of the four parameters, only k3,PP depends on the ionization constants of only one of the four types of enzyme-substrate species in Scheme 11. Thus, k3aPP is the only single parameter whose pH dependence can in general be used to define catalytically important enzymic ionization constants ( K E H ~and s~ KEHS2). There are, however, two combinations of parameters that may be used to define ( Y E H ~(Le., K E H and ~ KEH). Thus, k2aQP/K~aPP= ~ ~ / ( K s c Y E H and ~ ) k3app/(KSaPPKSSaPP) = ~ ~ / ( K s K s s ( Y E HNow, ~ ) . k2aPP/K~aPPis of course equivalent to k,t/K, for the binding of the first substrate molecule to the enzyme, and the pH dependence of this ratio reflects the ionization constants of enzymatic functional groups that control the binding of the substrate molecule in the catalytically productive k2 binding site. The pH dependence of this ratio (Table I) gives a clear indication that the same two enzymatic functional groups ( P K E H ~= 6.0 and PKEH = 9.1) control the binding of the first molecule of each ester in its catalytically productive mode. The pH dependence of k3aPP/ KsaPPKssaPP is indicated in Figure 8 for three substrates and is typical of all seven esters. This function allows both an alternative evaluation of P K E Hand ~ PKEHfor these esters and also gives values for these pKa values for l a as substrate, for which kZaPP/KSaPP is not directly accessible (see Results). Figure 8 confirms that for l a productive binding of the first substrate molecule is controlled by the same pK, values that are found for the other esters. The agreement between P K E Hand ~ PKEHevaluated in BIOCHEMISTRY,
VOL. 15, NO.
15, 1976
3241
BUNTING A N D CHU
In the light of the experimentally observed influence of p H on the parameters for the enzymatic hydrolysis of the esters Id-f, Scheme I1 may be modified as in Scheme 111. The obScheme 111
SEH-
-
PKhIH‘
7
SEH
0
5-
,L
+S
I
4
8
6
10
F I G U R E 8: Dependence of log (k3aPP/KsappKssapp) for la ( 0 ) .l b (0), and Id (A).Each curve calculated using P K E H *= 6.0 and PKEH = 9.1.
these two different ways is satisfying, and gives further confidence in the interpretation of the current data in terms of Scheme 11. The pH independence of k p P for Id, le, and l g (Figure 7) S KSaSEH2/KS‘ are not p H dependent suggests that C Y E H ~and in the p H range investigated. For CYEH~S’.this requires that K E H ~and S KEHSshould lie outside the range p H 5-10, and C Y E H ~ S= 1 at all p H values in this range. The value of KSaSEH2/KS’ may also be p H independent because K S E Hand ~ KSEHlie outside this same p H range, or alternatively if K s
