Neurochemical Research, VoL 17, No. 12, 1992, pp. 1207-1210

Velocity of Ellman's Reaction and Its Implication for Kinetic Studies in the Millisecond Time Range

Jure Stojan I and Milo~ R. Pavli~ 1

(Accepted April 8, 1992)

A detailed study of the velocity of the reaction between Ellman's reagent and thiocholine was undertaken, in order to test the possibilities of this reaction as a detection method for the earlier stages of cholinesterases reactions. Experiments were carried out on a stopped-flow apparatus with a built-in spectrophotometer. The obtained experimental data were analyzed by fitting the data to theoretical kinetic equations derived for the reaction. In this way, a complete kinetic characteri- zation of the reaction was obtained. An important practical result derived from our investigations is the finding that, under most experimental conditions, the Ellman's reaction is more than suffi- ciently rapid as a detection method. However, in the case of reactions in the time scale of 200 milliseconds or less, this being 5 times the half life of Ellman's reaction at standard conditions, one has to consider the interference of this reaction with the enzyme reaction itself.

KEY WORDS: Cholinesterases; Ellman; Stopped-flow.

I N T R O D U C T I O N

In the year 1959 Ellman introduced 5,5-dithiobis (2-nitrobenzoic acid) as a new effective reagent for the determination of the concentration of thiol groups (1). Later, in 1961, a new and rapid colorimetric determi- nation of acetylcholinesterase activity based on the re- action between this reagent and enzymatically released thiocholine was proposed by Ellman and co-workers (2). Since then the method has become generally used in most of kinetic studies of acetylcholinesterase and bu- tyrylcholinesterase.

It is also the experience in our laboratory that the Ellman's method represents a convenient tool for the determination of enzyme activity when the time period between mixing reactants and the beginning of the mea- surement is relatively long. Recently, however, when studying some earlier stages of butyrylcholinesterase re- actions using stopped flow technique on the millisecond

1 Institute of Biochemistry, Medical Faculty, 61000 Ljubljana, Slov- enia.

1207

time scale (4), some unexpected experimental results were obtained with no reasonable explanation in terms of en- zyme reaction mechanism. It came into mind that the reason for this finding might lay in the detection method, more precisely, in the velocity of El lman's reaction. Therefore, a detailed study of the velocity of the reaction between El lman's reagent and thiocholine was under- taken.

E X P E R I M E N T A L P R O C E D U R E

The velocity of the reaction between Ellman's reagent, 5,5-di- thiobis(2-nitrobenzoic acid), and thiocholine (2)

Ellman's reagent + thiocholine ~ yellow color (Scheme 1)

was recorded on a stopped flow apparatus which made it possible to follow the product formation in a millisecond time scale (5), from a few milliseconds after the mixing of reactants until reaching the "'fi- nal" state. The apparatus was manufactured by Hi-Tech Scientific Ltd., Brunel Road, Salisbury SP2 7PU, U.K. The apparatus consists of a preparative quench and stopped flow sample handling unit (PO-

0364-3190/92/1200-1207506.50/0 �9 1992 Plenum Publishing Corporation

1208 Stojan and Pavli~

SP 53), a spectrophotometer unit (SU 40), and a MSC-2 data acqui- sition system. In a stopped flow mode at a pneumatic pressure of 5 bar, using Berger MklI ball mixer, a dead time period is nominally 0.7 millisecond.

Experiments were done in the following way: aliquotes of two solutions, one containing Ellman's reagent and the other one contain- ing thiocholine were mixed together in the mixing chamber of the stopped-flow apparatus. The absorbance of the reaction mixuture at 412 nm was recorded till the maximum absorbance was reached. Data were captured by using the orginal Hi-Tech data acquisition system. This time course of product formation was followed at various fixed surplus concentrations of Ellman's reagent and at varied concentrations of thiocholine. The final concentrations of Ellman's reagent were be- tween 0.33mM (concentration in the original EUman's procedure for the determination of butyrylcholinesterase activity) and 4 raM, and those of thiocholine between 8.3 IxM and 33 IsM.

Under the conditions of the original Ellman's procedure (pH = 8 and surplus of Ellman's reagent, cf.2), the bimolecular reaction rep- resented by Scheme 1 follows a first order pattern with an apparent first order rate constant (k~pp) being equal to the second order rate constant (k from Scheme 1) multiplied by the concentration of Ellman's reagent.

The time course of the yellow color production in Scheme 1 may be represented by the following equation:

A(t ) -- A . ( 1 - e -kappt) --t- C Equat ion (1)

where A(t) is the absorbance of yellow color solution at time t, A, the absorbance at time 00, t time and C the value of A ~,=o).

Each individual experimental progress curve for the time course of the yellow color production at a particular concentration of Ellman's reagent and of thiocholine was fitted to Eqn.1 by means of the Gauss- Newton iterative nonlineai least square method (6). Each fitting re- sulted in a pair of values for the parameters A= and k,~, each pair characterizing the phenomenology of the reaction under the particular concentrations of Ellman's reagent and thiocholine. From the depend- ency of k,pp on the concentration of Ellman's reagent, k (Scheme 1) was determined as the slope of the line in the corresponding diagram.

However, it must be recognized that the reaction in Scheme 1 is a simplification of the following process (cf. 1,2):

K 1

thiocholine ~ th iochol ine- + H + th iochol ine- + RSSR ~ th iochol ine-SR + R S -

R S - + H + .-~ RSH

Scheme 2

In this scheme RSSR represents Ellman's reagent and RS- the yellow anion of 5-thio-2-nitro-benzoie acid whose concentration is actually being photometrically detected. Kx and Kz are the corresponding dis- sociation constants, and k' is the second order rate constant.

In excess of Ellman's reagent, the solution of this system (Scheme 2), according to the method of Cha (7), yields again Eqn.1, where, however,

A= = (AoK2)/(K2 + (H+)), Equat ion (2)

kapp = (k'KI(RSSR))/(K1 + (H+)) Equation (3) and Ao is the absorbance corresponding to the total concentration of thiocholine.

In order to evaluate the relevant constants in Scheme 2, experi- ments for the determination of A. and k~pp were carried out at three pH values (7, 8 and 9). Then, the constants k' and K~ were determined.

The determination was carried out by a direct fitting of the obtained values of k~pp to Equation 3. It should be emphasized at this point that Eqn.3 includes two independent variables, (RSSR) and (H+); for this reason a computer program developed by Duggleby (8) for such cases was applied to our case.

Experiments were performed on a Hi-Tech stopped-flow appa- ratus, consisting of the units PQ/SF 53, SU-40 and MSC-1. Computer programs were run on an Apple II/e and PC-AT compatible computer.

The buffer solution used in all experiments was 23.5mM borax- HCI (9), with a total ionic strength of 0.2M, obtained by the addition of NaC1. The temperature was 25~ Ellman's reagent (5,5'-di- thiobis(2-nitrobenzoic acid) and thiocholine iodide were purchased from BDH Biochemicals Ltd. All substances were reagent grade.

R E S U L T S A N D D I S C U S S I O N

Measurements of the time course of the yel low color product ion resulted, for each part icular pH value , in four sets of progress curves, each set conta in ing data for a

fixed concentra t ion of E l l m a n ' s reagent and four differ- ent concentra t ions of thiocholine. The corresponding diagrams for pH = 8 are shown in Figure 1.

Two things can be seen easi ly f rom these diagrams. First , under all exper imental condi t ions a f inal absorp- t ion (A=) is attained and it depends on the concentra t ion of thiocholine but is independent on the concentra t ion

of E l l m a n ' s reagent. Second, the curvature of progress curves, (characterized by k~pp), depends on the concen- tration of E l l m a n ' s reagent and is independent on the concentrat ion of thiocholine. A closer inspect ion of the diagrams shows that individual A= values are directly

1 2

3 4

, " r

�9 p ' , , , A

Time (msec)

Fig. 1. Time course of product formation in the reaction between Ellman's reagent and thiocholine (pH = 8). x-axis in each diagram represents the time from zero to 200 milliseconds and y-axis the con- centration of product from zero to 50 ~M. The drawings on the figure are reproductions from the original hardcopies performed by the stopped flow apparatus shifted to give a common origin at A~t=o) = 0. Con- centrations of Ellman's reagent: 1(5 mM), 2(10 raM), 3(20 mM), 4(40 mM). Concentrations of thiocholine: A(8.3 IxM), B(16.7 IxM), C(25 txM), D(33.3 I~M).

Velocity of Ellman's Reaction and Its Implication for Kinetic Studies 1209

proportional to the concentrations of thiocholine. All these observations are in full agreement with Ellman's repre- sentation of the reaction between 5,5'-dithiobis (2-nitro- benzoid acid) and thiocholine (cf.2) described by Schemes 1 and 2.

The fitting of each individual progress curve (from Fig. 1) to Eqn. 1 yielded the values of A~ and the corre- sponding values of kapp. The dependencies of the values of A~ and kapp on the concentrations of thiocholine and

DH 8

Ithioc holine~ (pM)

0

0 0

Fig. 2. Dependence of final absorption (As) on the concentration of thiocholine at pH = 7, 8 and 9. x-axis in each diagram represents the concentration of thiocholine from zero to 35 ~M and y-axis the final absorption in terms of product concentration. The points are experi- mental and represent the levels of individual curves on Figure 1 (pH = 8) and of analogous ones for pH 7 and 9. The lines on the diagram are theoretical; they have been calculated by simultaneous fitting of all experimental points at all pH values to Equation 2.

' 0

pH c

0

PH7

OH9 oH8

~EHmans reagent~ imM, n

Fig. 3. Dependence of the apparent first order rate constant (k~pp) on the concentration of Ellman's reagent at various pH values. The points (several overlap!) represent the individual values of k~pp obtained by fitting each individual curve in Figure 1 and from analogous figures at pH 7 and 9 to Equation 1 (see Experimental Procedure). The lines on the diagram are theoretical; they have been calculated by simulta- neous fitting of all experimental points at all pH values to Equation 3.

Ellman's reagent, respectively, are presented in Figure 2 and Figure 3 (pH 8).

It can be seen that both relationships are linear, again in accordance with Scheme 1 and 2. The slope of the line in Fig. 3 represents the second order rate con- stant kpH= 8 from Scheme 1; the value is given in Table I.

The attempt to characterize the actual mechanism of Ellman's reaction by means of experiments at pH 7 and 9, in addition to those at pH 8, yielded the following results.

The values of A. for each concentration of thioch- oline and Ellman's reagent at pH 7 and 9, respectively, are the same as the values at pH 8 (see Figure 2). This is in full agreement with Scheme 2 and confirms the derived meaning of A. and the equality of A. with ,4o at all three tested pH values. This had to be expected, since, at pH 7 to 9 the product 5-thio-2-nitro-benzoic acid appears, at the end of the reaction, only in the form of the yellow anion (pK2 = 5.1, ref. 2).

On the other hand, the values of kapv at different pH values differ strongly from each other. This is re- flected in different slopes of the lines in Fig. 3. The values of the slopes, i.e. of the second order rate con- stants are given in Table I.

Finally, the values of the constants K1, I{2 and k' had to be established. Since Kz is known (2) but no reference was found for the value of K1, the values of K1 and k' remained to be determined. This was carried out by direct fitting of all obtained values of kapp at all three pH's to Eqn. 3. The fitting resulted in the following values: Ks= (149 ___ 7) x 10-1~ (pK1 = 7.83)which is reasonable for such a thioalcohol group, and k' = (105 _ 2) x 1031mol-ls-~ (Table I).

It seems important to correlate our findings pre- sented above with the applicability of Ellman's reaction as a kinetic detection method for the activity determi- nation of cholinesterases. On the basis of our results it is obvious that Ellman's reaction has a certain limitation. For the conditions of the original Ellman's procedure,

Table I. Characteristic Constants for the Reaction Between Ellman's Reagent and Thiocholine (for Details See Text)

kpn=7 = (13.7 __. 0.3) x 10almol-ls -1 kprt=8 =(58.1 + 0.7) x 1031mol-ls -1 kpn=9 =(102 4- 4.8) x 1031tool-Is -a k' =(105 ___ 2) x 1031mol-~s -~ K~ =(149 4- 7) x 10-1~ (pK = 7.83) K2 = 7.9 x 10-6M (pK2 = 5.1) ~

'~From reL 2.

1210 Stojan and Pavli6

the value of kap p is about 19 s -1 with the corresponding half time of approximately 36 ms. Therefore, when in- vestigating reactions on the millisecond time scale or less, one has to consider the interference of Ellman's reaction with the reactions themselves. However, in the case of standard experiments where dead time periods exceed the half time by many times, Ellman's reaction is more than sufficiently rapid.

This can be illustrated by the following example. The hydrolysis of butyrylthiocholine catalyzed by cho- linesterase was studied by following the product for- mation according to Ellman et al. The process can be represented by the following scheme:

k--i k+2 k+3 E + S ~ ES --> E'S--> E + P2

k+l P1

+

Ellman's reagent

Sk Yellow Color

Scheme 3

Experiment was performed in the millisecond time range by using a stopped-flow apparatus. The obtained prog- ress curve is shown in Figure 4. Unexpectedly for this time range, the curve is concave upwards in the first portion, and becomes straight line only after a time pe- riod of approximately 200 milliseconds. Since it is gen- erally accepted that the formation of steady-state between

J

Time(msec) 1000

Fig. 4. Progress curve for the hydrolysis of butyrylthiocholine (0.5 raM) catalyzed by cholinesterase. The product formation was followed according to Ellman et al. on the millisecond time scale, using a stopped-flow apparatus. The points on the figure are reproduced from the original hardcopy performed by the stopped-flow apparatus. The curve is theoretical, obtained by fitting the points to Equation 4.

enzyme and substrate occurs in the microsecond time range the curvature seems to be a consequence of a rel- atively slow velocity of Ellman's reaction.

If this conclusion is correct, the curvature in Figure 4 is characterized by k in Scheme 3 and it's value must be around 19 s -x, which is the value of kapp for the conditions of original Ellman's procedure.

Under the conditions of our experiment, the explicit solution of the system presented in Scheme 3 is

Yel. Color(t) = A-t + B-[1 - e TM] (4)

By fitting the experimental data from Figure 4 to this equation (Eqn. 4), using Duggleby's program, the value of k was found to be (21.5 +_ 1.4) s -1. This agreement between the two values (kapp and k) demonstrates that the curvature, indeed, is a consequence of a relatively slow velocity of Ellman's reaction.

ACKNOWLEDGMENT

We would like to express our thanks to Dr. M. Brzin for his valuable comments on our manuscript. The financial support of Re- search Community of Slovenia is gratefully acknowledged.

REFERENCES

1. Ellman, G. L. 1958. A colorimetric method for determining low concentrations of mercaptans. Arch. Biochem. Biophys. 74:443- 450.

2. Ellman, G. L., Courtney, K. D., Andres, V., and Featherstone, R. M. 1961. A new and rapid colorimetric determination of ac- etylcholinesterase activity. Biochem. Pharmacol. Vol. 7:88-95.

3. Pavli~, M. R. 1987. The role of hydration in an enzyme reaction. Arch. Biochem. Biophys. 253(2):446--452.

4. Stojan, J., and Pavli~, M. R. 1991. On the inhibition of cholin- esterase by o-tubocurarine. Biochem. Biophys. Acta 1079:96- 102.

5. Gibbson, Q. 1988. Modem Physical Methods pages 65-84, in Neuberger, A., and Van Deenen, L. L. M. (eds.) Biochemistry, Elsevier, Amsterdam.

6. Stojan, J. 1987. Fitting the kinetic equation for slow-binding li- gand by microcomputer. Vest. Slov. Kern. Drust. 34(2):233-241.

7. Cha, S. 1968. A simple method for derivation of rate equations for enzyme-catalyzed reactions under the rapid equilibrium as- sumption or combined assumptions of equilibrium and steady- state. J. Biol. Chem. 243(4):820-825.

8. Duggleby, R. G. 1984. Regression analysis of nonlinear Arrhen- ius plots: an empirical model and a computer program. Comput. Biol. Med. 14(4):447--455.

9. Weast, R. C. 1978-79. CRC Handbook of Chemistry and Physics, 59 ~h Ed. Page D-186, CRC Press, Boca Raton.

10. Riddles, P. W., Blakeley, R. L., and Zemer, B. 1979. Ellman's reagent: 5,5'-Dithiobis(2-nitrobenzoid Acid)-a reexamination. Analyt. Biochem. 94:75-81.