TREVOR SPENCER AND JULIAN M. STURTEVANT
1874
the phototube and by reducing the aperture anterior to the analyzing Polaroid. It is probable that in the dispersion curves obtained earlier with the Rudolph instrument the anomaly may also have been caused by stray light, partially polarized by reflection.
Vol. 81
Experimental p-(p-Dimethy1aminobenzeneazo)-benzoylaminoacetic acid
~ ifigkpW& ~~
”
toria.
(2) F. Karush,
~
~ ~ Cutter ~ ; Labors~‘ r
~
J. Phys. Chem., 66, 7 0 (1952).
BUFFALO,NEWYORK
NO. 1504 FROM THE STERLING CHEMISTRY LABORATORY, YALE UNIVERSllY]
[CONTRIBUTION
The Mechanism of Chymotrypsin-catalyzed Reactions. I11 BY TREVOR SPENCER AND JULIAN M. STURTEVANT RECEIVED JUNE 23, 1958 Kinetic experiments, employing fast reaction techniques, on the chymotrypsin-catalyzed hydrolysis of p-nitrophenyl acetate are reported. These experiments demonstrate t h a t the acetylchymotrypsin isolated by Balls and Wood is a true intermediate in the reaction and give further indication t h a t the position of acylation is at a serine hydroxyl rather than a t R histidine imidazolyl group. The rate of the final hydrolysis of the acetyl enzyme is controlled by a group a t the catalytic site having a n apparent pK of 7.4. It is shown on the basis of experiments involving competition between 9-nitrophenyl acetate and a specific substrate, acetyl-L-tyrosine ethyl ester, t h a t hydrolysis of the latter probably involves the same three-step mechanism as the hydrolysis of p-nitrophenyl acetate. This conclusion emphasizes t h a t the specificity of the enzyme can be even more important in later stages of the catalysis than i t is in the initial formation of a Michaelis-Menten complex. A new stopped-flow fast reaction apparatus is briefly described.
Introduction It has been shown in previous publications1s2 that the kinetics of the chymotrypsin (CT)catalyzed hydrolysis of p-nitrophenyl acetate (NPA) is consistent with a mechanism involving three distinct steps E
ki
+S
(ES’)t k-
kf
(ES’)Jr (ES”)t k-
(1)
1
+ P’
(2)
ka
2
ki
(ES”)Jr E k-
The differential equations of this kinetic scheme are readily solved by the method of Laplace transf o r m , if attention is confined to the early stages of the reaction so that the substrate concentration can be assumed to remain essentially constant a t its initial value, [SIo,and the reversal of the second and third steps can be neglected. I t can be shown that the concentration of the product P’ is given by the expression
f
P”
nm
(3)
a
In the first step there is a very rapid formation of a Michaelis-Menten complex, ES’, which in the second step is decomposed to give the product I” (p-nitrophenol (NP) in the case of NPA) and a n acylated derivative, ES”, of the enzyme. Finally, the acyl enzyme is hydrolyzed to regenerate the enzyme and forni the product P” (acetate in the case of NPA). The rates of the second and third steps are dependent on the state of ionization of a group in the catalytic site, only the unprotonated form being active. In equations 1 to 3, (ES’)t and (ES”)t represent the sum of both the ionized and unionized forms of the intermediates, the distribution between these forms following, a t least approximately, the equation for a single group ionization
where the quantities in square brackets represent the molar concentrations of the corresponding species and a H is the hydrogen ion activity (in this paper operationally defined as the quantity measured by a glass electrode). (1) H. Gutfreund and J. M. Sturtevant, Biockcm. J . , 63, 656 (1956). ( 2 ) H. Cutfreund and J. M. Sturtevant. Proc. Nal. Acad. Sci.. 42, 719 (1956).
kar + mz(n -
117 )
k3‘
7t2(?l
-n
- WZ)
e-,,t]
(5)
+
where t is the time, kz’ = kzK,’/(K,’ U I I ) and k3’ = U H ) are apparent first-order rate constants, and m, n are given by
kaKi”/(Ki”
2m = klA‘
+
+
kl‘
-d(kId’ m
+n
+ kS’)* - 4k,(kz‘[S]o+ ks‘A’,l
= klA’
+
+ k3’
A ’ = [SI@ K,’
(6) (7) (8)
The second exponential in equation 5 reflects the build up of the steady-state concentration of the Michaelis-Menten complex ES’, while the first exponential describes the formation of the steadystate concentration of the acyl enzyme, ES”. I n the present case, available techniques are inadequate to detect the phase of the reaction involving the second exponential so that we may conclude that n>>m. This implies that klA’>> kZ‘and klA’ >>-k3’. I t then follows with adequate accuracy that
Equation 5 becomes, after decay of the second exponential
o
~
m
April 30, 1959
MECHANISM OF CIWMOTRYPSIN-CATALYZED REACTIONS
Measurements made during the period when the exponential term in equation 11 is comparable in magnitude to unity may be employed in favorable cases t o evaluate the quantity kz'[S],/A' k3'. Measurements innde by classical methods usually involve values of t large enough so that the exponential, term is negligible compared to unity. The rate of formation of P' is then given by
+
1875
also performed with a sample of ACT kindly supplied by Dr. George Hess of Cornell University. Substrates.-ATEE was purchased from Mann Research Laboratories, New York, N. Y., and was used without further purification; m.p. 79.1' cor., lit.6 79-80'. NPA was prepared by the acetylation'" of N P by acetic anhydride and was purified by recrystallization from aqueous ethanol; m a p .77.9" cor., lit.'b77.&78'. Kinetic Methods Titration Method.-Hydrolyses of ATEE were followed by the titration method of Schwert, et al.,*using 1 M NaOH delivered from a 0.1-ml. microburet with manual control. Stopped-flow Apparatus.-Most of the kinetic observations were made with a modification of the spectrophotometric stopped-flow apparatus described by Gibson .$ A schematic diagram of the apparatus is shown in Fig. 1.
It is important to note t h a t application of Eadies plots to data giving r as a function of [SI0 can then only lead to the evaluation of the apparent kinetic constants
and 1
1
1
G)Tf+iE;i LlGtjT The approximations involved in arriving at these equations are certainly valid in connection with experiments where initial rates are determined with substrate concentrations much larger than enzyme concentrations. In experiments such as those illustrated in Fig. 3, in which the substrate concentrations were only two to five times the enzyme concentrations, the adequacy of equation 11 is indicated by the fact t h a t the expected linear variation of [P'] with time was observed for a considerable period after the decay of the exponential term was complete. The present communication is concerned primarily with three questions relating to this mech- Fig. l.-%hematic diagram of stopped-flow apparatus. See anism : (1) is the acetylchymotrypsin (ACT) text for details. isolated by Balls and Wood4 a true intermediate in The solutions which are to be mixed are contained in two the reaction; (2) a t which position is the enzyme 2411. Luer-Lok glass syringes, A. The syringes are conacylated; and (3) does the same three-step mech- nected by 1 mm. capillary channels in a Lucite block to a "mixing chamber" which in this design is a simple T-joint. anism apply to other substrates of CT?
Materials Buffer.-In mast of the experiments where buffering was necessary, appropriate mixtures of acetic acid and tris(hydroxymethy1)-aminomethane( T H A M ) of 0.02-0.04 M ionic strength were employed. All solutions were also 0.10 M in NaCl. PH values were determined with reference to Beckman p H 7.0 standard buffer by means of a glass electrode and Beckman Model G pH-meter Enzyme.-Salt-free crystalline a-chymotrypsin was purchased from Worthington Biochemical Sales, Freehold, N. J. Nitrogen determinations by the micro-Kjeldahl method indicated 14.2% N. Where molar enzyme concentrations are given, they were computed on the assumption that CT was the only nitrogenous material present, taking the enzyme t o contain 16.2% N and t o have a molecular weight6 of 23,000. Acetylchymotrypsin.-ACT was prepared according to the method of Balls and Wood.' The product was assayed before and after being held a t PH 9, using acetyl-L-tyrosine ethyl ester (ATEE) as substrate a t PH 6 and 25'. A typical preparation was found to contain 9% active CT, and 96.5% of the original activitywwas recovered after treatment a t PH 9 for 30 minutes. Some experiments were
.
(3) G. S. Eadie, J . B i d . Chcm., 146, 85 (1942). (4) A. K. Balls and H.N . Wood, ibid.. 319, 245 (1956). (5) M. S. N. Rao and G. Kegeles. T-a JOURNAL, 80, 5724 (1958).
A I cm. length of 1 mm. channel connects the mixing chamber to the glass-lined observation chamber, which is of 2 mm. diameter with quartz end plates 1.00 cm. apart. From the observation chamber the fluid passes to the stopping syringe, B. The apparatus is kept full of liquid a t all times. At the start of a reaction, the two solution syringes are pushed by a common pushing bar, C, to expel about l/, ml. from each syringe. This amount of liquid is adequate to flush the old liquid from the observation chamber. Time is counted from the instant the flow of liquid is stopped by the stopping syringe hitting an adjustable stop. Side-arms (not shown) together with appropriate 3-way stopcocks are provided for removing air bubbles trapped when the syringes are screwed onto the plastic block and for expressing old fluid from the stopping syringe. Copper blocks through which water from a thermostat is circulated enclose the plastic block and the solution syringes. Provision is included for inserting a small low-lag thermocouple to a (6) S. Kaufman, H. Neurath and G.W. Schwert, J . B i d . C h e m . , 177, 793 (1949). (7) (a) F . D. Chattaway, J . Chcm. Sac., 2495 (1931); (b) M. Bender and B . W. Turnquest, THISJOURNAL, 79, 1652 (1957). (8) G. W. Schwert, H. Neurath, S. Kaufman and J . E. Snoke, J. B i d . Chcm.. 179, 221 (1948). (9) Q. H. Gibson, J . PhysMI., 117,49P (1962). See also E. Chance, in "Investigations of Rates and Mechanisms of Reactions," edited b y S. L. F r i e s and A. Weissberger, Interscience Publishing Co., New York. N. Y.,1953, p. 690.
187G
TREVOR SPENCER AND JULIAN 11.STURTEVANT
position just beyond the observation chamber for checking the adequacy of temperature control. The rate of flow of fluid during the initiation of the reaction averages 5 ml. set.?. T h u s the average age of the mixed solution which comes t o rest in the obseivation chamber is approximately 0.005 sec. T h e efficiency of mixing in the present design is not particularly good, so t h a t reactions having half-times below about 0.02 sec. cannot be properly observed. T h u s the flow rate is high enough t o avoid significant timing errors, A Unicamlo SP.500 spectrophotometer is employed as a monochromator, the light source being either a hydrogen lamp and associated power supply’o or a tungsten lamp and regulated power supply.11 The stopped-flow apparatus fits in place of the usual cell holder of the SP.500. T h e photocell compartment of the SP.500 is replaced by a similar compartment containing a 1P28 photomultiplier and associated voltage divider network for supplying appropriate voltages t o the dynodes. The high voltage for the photomultiplier is delivered by a very stable electronic power supply.12 The load resistor of the 1P28 is the 1 megohm input resistor of a stable linear D-C amplifier‘3; a variable low-pass resistor-capacitor filter is inserted between the phototube and the amplifier so t h a t noise voltages can be reduced when high speeds of response are not necessary. T h e amplifier drives a direct-inking o~cillograph,’~ the combination having a flat response up to 100 cycles per sec., which is adequate for recording any processes for which the present speed of mixing is sufficient. A very useful feature of the amplifier is t h a t the zero can be offset 5 or more chart widths by a very stable and accurately reproducible adjustment, so t h a t high amplifier gain can be employed t o give a full scale deflection with as little as 570 change in transmission. The oscillograph is supplied with 6 readily interchanged chart speeds ranging from 2.5 mm. per min. t o 25 mm. per sec. Ovel.-all sensitivity is adequate for observations down t o 220 mp. The stability of the light sources has been found t o be excellent, so t h a t experiments extending over many minutes can be performed even though there is no opportunity to check light intensity during a n experiment. I n fact, the chief source of instrumental instability appears t o be phototube fatigue. The detection and recording system is linear in per cent. transmission, so t h a t conversion t o optical density is always necessary. In usual practice, the system is adjusted, by means of the monochromator slit width, t o give full scale deflection with water in the observation tube and the amplifier set at a low gain. T h e amplifier gain and zero offset are then adjusted as necessary to give a satisfactory record of the reaction under study. Optical densities are thus obtained by direct calculation, calibrating solutions of known optical density being unnecessary. A very convenient feature of the apparatus is t h a t no more than 5 min. are needed t o replace the stopped-flow components by the usual SP.500 cell holder, either with the normal photocell detector and circuitry of the SP.500 (which is similar to t h a t of the Beckman DU spectrophotometer) or with the 1P28 detector and associated recording equipment described above.
The Value of pKif for the Hydrolysis of NPA.In an earlier publication,’ the variation of the over-all rate of hydrolysis of KPA by CT was investigated by ordinary spectrophotometry over the range of pH (3.45 to 7.75 a t 27”. I n these measurements the solvent contained 20yo (v./v.) isopropyl alcohol and 0.05 M phosphate, and the )If. initial substrate concentration was 2.5 X These results, when interpreted in terms of a single group ionization, indicated a value of 7.3 for pKi” (cf. equation 4). However, since interference from non-enzymic hydroxide ion catalyzed hydroly(10) Unicam I n s t r u m e n t s L t d . , Cambridge, England.
Thestopped-
flow a p p a r a t u s has been briefly described a n d illustrated b y photo-
g r a p h s in Spectrovision, No. 5 , p . 5 (1957), published by Unicam. (11) Nobatron MAG501, Sorensen and Co., Inc., S t a m f o r d , Conn. (12) hIodel HV-RA, Technical Measurements Corp., New H a v e n , Conn. ( 1 3 ) Model BL-530, Brush Electronics Co., Cleveland, Ohio. (11) Model 131,-201, Brush Electronics Co.. Cleveland. Ohio.
Vol. 81
sis prevented observations at p H values above 7.75, the distinction between the p H dependence to be expected from equation 4 and t h a t corresponding to a direct hydroxide ion catalysis of the last step of the enzymic process was not as definite as desirable. Dixon and Neurath15 measured the rate of recovery of activity toward A T E E by ACT as a function of pH and found the variation with p H to be in accord with a single group ionization with pK = 7.0. If ACT is identical with ES” in equation 3, the pK for this process should be expected to be equal to pK,“ for the NPA hydrolysis. In view of this discrepancy and the restricted p H range of the earlier data, i t seemed important to redetermine pK,”. The stopped-flow apparatus makes i t possible to determine rates with relatively large enzyme concentrations and thus to minimize the importance of the contribution to the rate resulting from non-enzymic catalysis. Experiments were performed a t 25’ in the pH range 7.0 to 8.8 with the results summarized in Fig. 2 . In each experiment, a solution containing CT, NPA and 0.008 AII acetic acid was mixed in the stoppedflow machine with an equal volume of THAMacetate buffer, and the reactions were followed a t 400 mp. The initial “burst” of N P liberation was essentially complete in the low @ Hsolution before mixing. The observed rates, in optical density units per sec., were converted to mole 1.-’ sec.-l using the measured absorption of N P as a function of p H . The substrate concentration used in these experiments was sufficient nearly to saturate the enzyme, so t h a t it is very unlikely that the results were significantly affected by variation of Km’(dppl within the p H range covered. The observed rates include the contribution due to hydroxide ion catalysis. As illustrated in Fig. 2, i t was found, by successive approximations, t h a t a plot of yo v ’ s . ~ ~ u H where , ro is the corrected initial enzymic rate, adhered reasonably well to a straight line if the hydroxide ion catalysis was assumed to be governed by the equation
with k’ = 43 X mole 1.-’ see.--’. The slope of the line gave a value of 7.41 for $Kif’, in fair agreement with the earlier value obtained in a different solvent. A series of measurements of the initial rate of the hydroxide ion catalyzed reaction was performed with the results listed in Table I. T h e mean value of k‘ is identical with that which had to be assumed to have the enzymic rates fall as well as possible on a single group ionization curve. Holleck and Melkonian16 studied the hydroxide ion catalyzed reaction in 0.1 M borate buifers by a polarographic method, obtaining a rate constant over twice as large as t h a t reported here. This discrepancy is due a t least in part to the fact t h a t they employed 10% ethanol as solvent. From the enzymic rate extrapolated to zero hydrogen ion activity and the enzyme concentration, we obtain a first-order rate constant of 0.021 sec.-’, (15) G . H. Dixon and H . K e u r a t h , J. B i d . Chem., 326,1049 (1957). (16) I.. Hitlleck and G . A . hlelkonian, Z. E l e k f r o c i w n . , 68, 867 (198.1).
I o
2 4 G Initial rate X aH X 10'5. Fig. 2.-The initial rate of the steady-state CT-catalyzed hydrolysis of NPA at 25" in THAM-acetate buffers plotted against the product of the initial rate times the hydrogen ion activity. Total ionic strength 0.12 M; [SI0 = 5.0 X lo-' X; [E]o = 8.8 X 10-6 d l ; 2.57, ( v . / v . ) ethanol. Open circles, rates including OH- ion catalysis; filled circles, rates corrected for O H - ion catalysis. 0
in satisfactory agreement with the previously published' estimate. I n the experiments of Dixon and Neurath,15 A T E E was presumably specifically bound during the hydrolysis of the ACT. It is possible t h a t this might be the cause of the lower p K observed by these authors. TIiE
TABLE I HYDROXIDE I O N CATALYZED IIVDROLYSIS 25 O PIT
Initial rate, mole l . - ' s e c . - ' X 107
s.21 8.52 8.63 8.78
OF
k',
m o l e l . - I s w - ' X 1011
0.41 . DX
62
70 1.13
12
,
N P L AT ~
42
3s~Mean 13 Icig. 2 with enzyme oiriittetl. ~
Solution composition
:IS ill
Experiments with Acetylchymotrypsin Acetylchymotrypsin as an Intermediate in NPA Hydrolysis.-It has seemed reasonable to assume that the ACT isolated by Balls and h:00d4 is indeed an intermediate in the hydrolysis of NPA. Dixon and Neuratlilj found t h a t ACT is converted to a form of the enzyme active in catalyzing the hydrolysis of A T E E , the conversion following first-order kinetics with a rate constant of 8.8 X l o p 3set.-' a t PH 7 and 25". Using iheir value for the pK, 7.0, of the group controlling the rate of this process, the limiting value of the rate constant a t high PH is calculated to be 1.8 X lo-* sec.-', which is fairly close to the limiting rate constant for the hydrolysis of NPA. The experiments illustrated in Fig. 3 give direct indication t h a t ACT is an intermediate, a t least so far as the lyophilized ACT redissolved a t low p H is concerned. I n the experiment represented by curve A, CT at pH 4.8 was mixed in the stopped-
0
5
10
16
Sec. z,-opticai
dellsity changes a t 400 ~
I acconipanying P
tile l,ydrolysis of sp.4 by C T anti ACT a t 28' in THAMacetate buffers. Total ionic streIlgt1l 0.12 'If; IE1o = 4.4
x 10-5 ' ~ f , curve :I, CT, and curve B, .iCT, mixed with Jf, 1,0% (F'./v.) ethanol; Kp.4 a t PI-I 8.2, [SIo= 2.0X the rnte constants refer to the approach to the steady state with respect to ACT. Curve C, C T p l ~ sXPA a t PH 4,8 llfi tnixctl \ v i l l i huffer to give pFI 8.2, /S]o = 2.5 x 2.5'1 ( v . / v j e t i ~ a l l o ~C. u r v e I), ACT plus KPA a t PH 4.8 mixed ivit1i tmffcr to give PH 8 . 2 , [SI" = I .O x .If,0.5% ( v . / v . ) ethanol.
flow apparatus with a solution of NPA in a buffer such that the final p H was 8.2. Rapid acylation of the enzyme with liberation of somewhat less than a mole of N P per mole of C T was followed by a zero-order steady-state liberation of N P and acetate. Curve B c:n-responds to a similar experiment performed with ,It,T. A small "burst" of N P liberation was observed because the enzyme was incoinplctely acylated in the sampleI7 of ACT used; as soon as this was coinpleted to give the steady-state concentration of ACT, the reaction proceeded with essentially the same zero-order rate as observed with the free enzyme. Thus, within a time interval no larger than 10 to 20 niillisec., the ACT was behaving precisely as expected for the true intermediate contaminated by soiiie unchanged enzyme, Curves C and D were obtained in similar experiments in which CT (curve C) or ACT (curve D) and substrate were present in the same solution a t pH 4.8 which was then mixed in the stopped-flow machine with an equal volume of buffer to give a final PH of 5.2. Under these circumstances, the acylation took place before the pH was raised, so t h a t no rapid liberation of N P was observed. In each case the initial rate of N P liberation was actually a little less than the steady-state value, presumably because the steady-state degree of (17) T h e incomplete acylation was p r o b a h l y in p a r t d u e t o t h e f a c t t h a t t h e sample was two months old. I t has heen observed (H. h-eurath, private communication) t h a t lyophilized ACT undergoes slow hydrolysis during storage.
15;s 0.03
1
1
TREVOR SPENCER AND
JULIAN
;
a
;
-
I
Seconds,
Vol. 81 I
1
0
Fig. 4.-Optical
M. STURTEVANT
M O
1
I IO00
density changes a t 245 NP following raising the $H of an ACT solution from 4.8 to 8.2 at 25', THAMacetate buffer, total ionic strength 0.14 Ai, 4.4 x 10-5 M.4CT.
acylation is somewhat larger a t low than at high
be partially protonated, and since the pK of acetylimidazolyl should be much lower than that of The Position of Acylation of the Enzyme.-It is imidazolyl, we would expect protons to be liberated now generally agreed that both the imidazolyl by the enzyme if the imidazolyl group became apgroup of a histidine residue and the hydroxyl group preciably acetylated. Although there is some of a serine residue are involved in the active site of doubt as to the actual value of fiKi', there is no CT. There has, however, been considerable dis- doubt that it is considerably smaller than fiKi". agreement as t o which of these groups is acylated This fact is not only consistent with the observed in the most stable intermediate ES" formed during uptake of protons during acetylation but is in the hydrolysis of NPX. I n a recent publication, itself clear indication that during the final rateDixon and Neurath'* reported that a slow de- controlling hydrolysis the imidazolyl group is crease in absorption a t 245 mfi occurs when the free and in equilibrium with respect to proton PH of a solution of ACT is raised from 4 to 8 or 9. binding. Since i t is known t h a t acetylimidazole absorbs There remains the problem of accounting for the a t this wave length, they postulated that the ob- spectral changes a t 245 mp observed by Dixon and served decrease in absorption is due to the hy- Neurath.I8 We have investigated the behavior of drolytic removal of the acetyl group from the -4CT in the stopped-flow apparatus with results imidazolyl group at the active site of the enzyme similar to those of Dixon and Neurath, as shown in and that the decrease in absorption is preceded by Fig. 4. Immediately after raising the pH from a very rapid increase as the acetyl group shifts 4.8 to 8.2, there is a rapid increase in absorption, from a serine residue to the histidine residue. followed by a slow decrease which proceeds acTheir proposed mechanism would thus place the cording to approximately first-order kinetics with set.-'. The deacyl group on imidazolyl for most of the time a rate constant of 1.7 X as large during which it is bound to the enzyme in the actual crease in absorption was approximately as observed by Dixon and Neurath. hydrolytic process a t high pH. It is immediately evident that the process shown There are several reasons for believing that in fact the acetyl group is bound most of the time at in Fig. 4, which took place at 25", cannot be dia serine residue during hydrolysis of NPA. The rectly associated with the rate-controlling step in experiments summarized in Fig. 3 show that if the hydrolysis of NPA since it proceeded a t a rate there is an acyl shift from oxygen to nitrogen as approximately one$jteenth as large as the rate of postulated by Dixon and Seurath, i t must be the final step in the hydrolysis of NPA. I t is extremely rapid. Furthermore, it has been shown,2 interesting that the half-time of the decay of abby experiments in XP buffer, that when CT is sorption at 245 mp a t 25" shown in Fig. 4 is 395 acetylated by NPA a t a fiH of 6.6, hydrogen ions are sec., as compared with the value of 415 sec. found taken up by the enzyme from the buffer. A t this by Dixon and Neurath a t 10". I n several experip € I , the imidazolyl group a t the active site should ments at 15" we have also observed half-times of about 400 sec. It thus appears that this reaction, (18) G. 13. Dixon and €1. Neurath. THISJ O U R N A L . 79, 4558 (1957).
kH.
April 20, 1959
TECHA AN ISM
OF
1879
CHYMOTRYPSIN-CATALYZED REACTIONS
-
Seconds.
Fig. 5.-Optical density changes a t 245 mji following raising the PH of a C T solution from 4.8 to 8.2 a t 25'. THAMHCl buffer, total ionic strength 0.13 M , 4.5 X 10-6 M CT. Age of C T solution a t start of experiment: open circles, 0.3 hr.; filled circles, 6.1 hr.; filled squares, 26.6 hr. The maximum optical density reached is indicated for each solution.
whatever i t may be, has a rate essentially independent of temperature. Quite coincidentally, the rate of the conversion of ACT to active C T was found by Dixon and Neurathls a t 10" and p H 7 to have a rate almost identical with the rate of decay of absorption a t 245 mp a t 10" and p H 9. The significance of the changes in absorption a t 245 mp, so far as the hydrolysis of NP.4 is concerned, is further made doubtful by the experiments shown in Fig. 5 . A solution of chymotrypsin a t p H 4.8 was raised after various time intervals to p H 8.2 in the stopped-flow apparatus, using a THAM-HCl buffer containing no acetate. Again, rapid increases in absorption a t 245 mp followed by slow decreases were observed, though the rates of both processes were considerably faster than in the case of ACT. The magnitudes of the changes decreased markedly with increasing age of the low PH solution. ( I t is not known whether a similar effect of age is observable with ACT.) A solution of C T which had been aged overnight a t p H 4 and which was shown to give a small change in absorption a t 245 mp when its p H was raised to 8.2 was held a t pH 8.2 for 30 min. and its p H was then again dropped to 4.5. When the PH of this solution was returned to 8.2 in the stcpped-flow apparatus, no significant changes in absorption a t 245 mp were detected. This experiment indicates t h a t the process observed a t 245 mp is not reversed by lowering the pH. It appears that some change in the enzyme is caused by the process of lyophilization ; this change is reversed when the enzyme is redissolved, the rate of reversal increasing with PH. The present experiments give no basis for an understanding of the changes involved or for the large differences in rates observed with C T and ACT.
Inhibition Experiments There is a t present no direct experimental evidence t h a t the kinetic scheme embodied in equations 1-3 is applicable to the CT-catalyzed hydrolysis of specific substrates such as ATEE. Even if acylated enzyme were formed in considerable amount, the reaction rates in this case are so large as to preclude the possibility of detecting the acylation reaction. If the method of "initial acceleration" devised by Gutfreundlg is extended to the 3-step mechanism, it can be shown that the intercept r of the steady-state hydrolysis curve with the time axis must be less than 1/k3, under circumstances such that the approximations employed in the treatment of equation 6 are valid. I n the case of ATEE, k3 = 160 set.-' (see below), so that 7 < 0.006 sec. With a substrate such as acetyl-L-tyrosinamide, which is hydrolyzed approximately 1000 times slower than ATEE, rate-limitation undoubtedly occurs a t the acylation step, so t h a t the steadystate concentration of acylated enzyme is extremely small. If the hydrolyses of NPA and A T E E utilize the same active site on the enzyme, i t should be possible to observe mutual inhibitory effects of these two substrates. We have accordingly investigated?O the influence of NPA on the hydrolysis of A T E E and of ATEE on the hydrolysis of NPA. Inhibitipon of ATEE Hydrolysis by NPA.-The rate of hydrolysis of ATEE by C T a t p H 7.0 and 25' was determined in the presence of various (19) H. Gutfreund, Disc. Faraday Soc., 30, 167 (1955). ( 2 0 ) The inhibition experiments reported here were undertaken as a result of a conversation with Professor Carl Nlemann of the California Institute of Technology. We acknowledge our indebtedness to Professor Niemann for directing our attention to the problem of inhibition of NPA hydrolysis.
cciiiceiitr:ttiot~sof SP.1. The solutions containecl o.OO.>.1/ 'I'HALI and 0.10 M NaCI. Two series of iiieasurciiients were made, one at an initial substrate conceiitration of 7.1.5x -11 arid one a t 180 x lo-.' -I[. Suiliciently lon. enzyme concentrations w r e eiii1~10y~l so that reliable initial rates could lie obtaiiied in the nianunlly operated PH-state.S Ilytlrolysis o f t h t iiihihitor, including the amount iiccessary to acetylatc the enzyme, was negligible during the iiiter\xl required to evaluate the initial r:ite (if li>-tlrcil>.sis. The data are sunimtrized in t i i v ioriii o f rate : N T .rate X [SP-I] plots in Fig. O, ivlicrc tlic. rates ;ire csiiresse(1 in nrhitrary units. UOlIIC
x
2
0
I .~
1
14
'7 \ '
\
%
O.G
.'..=
L
2
1
rOi~10 x
3
0 o'2
105.
I:ig. ii.---Iiiliil,it ioti ! ~ y SP.4 of the CT-catalyzed liydroly-
.ITEI-gen of which is in\-olved in I:ucleophilic atleast srinic~r-hat1:irgcr than k? for ATEE. tack of the carbonyl. In 2111~7 case, it can be stated t h a t the inhibitory The hydrogen bond in -1is considered to be the action of .1TEE is cscrtctl a t either the adsorption of the change of the apparent pk' of tlie imicause site or the aq-lation site of the enzyme, or perhaps rat both sitas. This result and that of the pre- dazolyl group resulting from acetylation by SP.1. ceding section gil-e definite support to the assump- \{-hen this hydrogen bond is broken by acetylation, tion that thc hydrolysis of .\TEE also proceeds by the imidazolyl becomes available for (weaker) hydrogen bonding with a water molecule. the %step niechanisin. X s is to be expected, there are difficulties inDiscussion herent in this mechanism, which actually are enThe partial mechanism for chymotrypsin catal- countered in one form or another by the other yses presented schematically in Fig. 9 is in accord (2.;) I.. \Y C i , n i i i r l p l i n n i , .Siii)ict. 1 2 5 , 1 1 4 5 119;:) Sre :LIL.O I: 11, with the salient facts a t present available. This !Vrstlitiirier, I ' r o i .\ai . I r d . S < i ,4 3 , 969 ( I O J i ) .
t- -. \,
TREVOR SPENCER AND JULIAN M. STURTEVANT
1882
mechanisms so far proposed. For example, why does imidazolyl in the k2 step serve to transfer a proton from serine hydroxyl to H20, while in the final step it serves to transfer a proton from H20 to serine hydroxyl? One would suppose that H20 molecules are readily available at all times, and t h a t the intermediate acylation of serine would be unnecessary.26 It is possible t h a t there is some significant entropic gain for the final step resulting from the additional point of attachment of the substrate to the enzyme furnished by serine hydroxyl. A more specific problem arises from some of the very large differences in rates observed in C T reactions. Table I1 summarizes values for various kinetic constants; some of those listed are approximate values only, b u t this is of no concern in the present discussion. Since i t is not certain a t which step the rate of hydrolysis of A T E E is limited, cornparison of the rate constants for this reaction with those for the NPA hydrolysis can only be expressed in the form of inequalities MATEE) ki(NPA)
k3(
acetyltyrosyl) kO(acetyl)
2
GODO
(18)
Vol.
s1
of the carbonyl group for the hydrolytic attack, whereas in ACT configurational entropy is lost in forming the transition state for the last step. If we adopt the approximate procedure suggested by Laskowski and S ~ h e r a g awe , ~ ~can estimate in the case of ACT this disadvantage might contribute as much as 3500 cal. per mole to the free energy of activation. TABLE I1 SOMEKINETICCONSTANTS FOR CHYMOTRYPSIN-CATALYZED REACTIOSS AT 25" PKit
SIC.-'
3
P-Nitrophenyl acetate
kr,
'
kr, Substrate
sec.-l
6.6
0.025
~s(~PP),
pK~t
K-, M
7.4
0 005
PKi(wp)
Krn~w)
sec. Acetyl-L-tyrosine ethyl ester 160 6.8" 0.005 Methyl hippurateb 0.23 .. ,0038 Methyl hydrocinnamate" 0.028 .. ,004 a G. H. Dixon and €1. Neurath, ref. 15. * A t p H 7.9, 0.15 M NaC1. R. B . Martin and C. Niemann, THIS JOURNAL, 79, 4814 (1957). C I n 20y0 ( w . / w . ) methanol. K . J . Laidler and M. L. parnard, Trans. Fayaday SOL.,52, 407 (1956).
In each case the inequality becomes an equality if Very recently Rydon28 has suggested t h a t in the hydrolysis of A T E E is rate-limited a t the corplace of serine a t the active site there is a Azresponding step. An important general conclusion is obvious oxazoline group formed by a ring closure involving from a comparison of these two substrates: the the serine hydroxyl and the adjacent pcptide boric1 specificity of an enzyme may be primarily exerted to an aspartyl residue. This hypotliesis appears at a stage in its catalysis subsequent to the initial to be consistent with the information availablt~ binding of substrate. The true K, for X T E E is a t concerning the acylation step and perhaps to I i a \ ~ ~ least as large as that for XPA; the fact that it is scjiiic advantages over the mechanism rcpresciitc,tl a much better substrate arises from the much greater in Fig. 9. However, Kydoti further sugqcsts ease of cithcr the forination or the hydrolysis of t1i:it the hydrolysis of the :icy1 enzyme is :iitletl I N acetyltpros~lcli\.iiiotr)~~)sinas compared with the adjacent carboxylate of the aspartyl recii I i i c and that the stability of .ICT a t low p1I is diic to ncetylchpiiiot~psin. If we :iswine that k l > k2 for A T E E hydrolysis the fact that the aspartyl carboxylate is 1)rotoii:itc.d. (see the preceding scctioii), Ire must conclude that ?'his view seems untenable in T-ieu. of the fact tli:tt tlie rate of hydrolysis of :tcet\.ltyrosplchyrnotrJ-p- tlie hydrolysis of .\CT is s h o w ~ ihy o u r c~\rpcriiiic~!its to be coiitrollctl by :i group h:i\.iiig :I pIi' o f : t i ) sin esciwls that c l i :\CT by a factor of more th:in that 311 atltlitioii:tl (i000, xrliicli correspoiids to a diiference in free proxiliiatc,ly 7.4. I t tlius :t~)l)c:trs groiip, pri'siiiiiaL1y :in iiiii~l:i:/ol:cl, is ~)rcscc"t :it tlic, c.iii.r,yics c j i :icti\.:ition o f inore than 5200 cal. IIcr riicJc. I t sc'i'iiis iiiilikcly tliat verq' niuch of this c:it;ilytic sitc ; i n c 1 i i l i i q t bc i i i tlic utiprotoii:i t ( , i I (liffvreiic-c c i i i hi' :ittril)utetl to a tliffcrence i n fori11i(jr the fiiial li!.iIrolytic stc.1) t o t;tl;cB pl:icc~. Acknowledgments. 'l'lic :iiitliors i r . i i l i to : i ( ~ Iic.:its of :ictiv:itioii, i i i view of the f:tct th:it ncitlicr k r i o \ v I e c l q tlic IlV1lJ qi\.cri I i J . 1)r. [ I . (;utirc,lli;(l i i i l i i ~ ~ ~ i i ~ ~ l c l i ~ i nrjr i i ~ ~hytlrr~c'inn:uii~lcliyi~i~~tr~~~~si~i sc\.cr:il stiiiilil;itiliy c l i x - l l < > i i i i i s ; i i i < l t l i c , t t c * l i l i i c : t l trylisiil :ti)litb:irs to be li~drrilpzetliiiuch faster t1i:tii . \ C T st^, '1'~tI)leI 1 1, a i i ( 1 IY(> must therefore s~ippose :thsist:ti!L~c ( , i l i r , .I. I C . (;rt,:i In.. '1.Iiis rc,si',irc.Ii t!i:it tlic~rc~ is :Lsiyliifiwiit diffc'rmcc. in entrrij)ies o f ir:t:; aicit,c! I ) \ .
I
III,,
i v i
