Inhibition of Drosophila acetylcholinesterase by 7-(methylethoxyphosphinyloxy)1-methyl-quinolinium iodide

Chemico-Biological Interactions 119–120 (1999) 147–157

Inhibition of Drosophila acetylcholinesterase by7-(methylethoxyphosphinyloxy)1-methyl-

quinolinium iodide

Jure Stojan a,*, Veronique Marcel b, Didier Fournier b

a Institute of Biochemistry, Medical faculty, Uni6ersity of Ljubljana, Vrazo6 trg 2,1000 Ljubljana, Slo6enia

b Laboratoire d’Entomologie et Groupe de Chimie Biologie, Uni6ersite Paul Sabatier,31062 Toulouse, France

Abstract

The kinetic behaviour of Drosophila melanogaster acetylcholinesterase toward its substrateshows, in comparison with classic Michaelis–Menten kinetics, an apparent homotropiccooperative double activation-inhibition pattern. In order to construct an appropriate kineticmodel and obtain further information on the mechanism of the catalytic action of thisenzyme, the hydrolysis of acetylthiocholine in the absence and presence of different concen-trations of synthetic quaternary methylphosphonate, 7-(methylethoxyphosphinyloxy)1-methyl-quinolinium iodide (MEPQ), was followed on a stopped-flow apparatus. The reactionat low substrate concentrations was followed until the change of the absorbance becamenegligible and at high concentrations only initial parts were recorded. A simultaneousanalysis of the progress curves using numerical integration showed that the powerfulorganophosphonate inhibitor binds and compete with the substrate for the same bindingsites. The results are also in accordance with the hypothesis that virtually every substrate orquasi-substrate molecule that enters into the gorge of active site is hydrolysed. © 1999Elsevier Science Ireland Ltd. All rights reserved.

Keywords: Acetylcholinesterase; Kinetic model; Organophosphonate; Progress curves;Stopped-flow

* Corresponding author. Tel.: +386-61-1320019; fax: +386-61-1320016.E-mail address: [email protected] (J. Stojan)

0009-2797/99/$ - see front matter © 1999 Elsevier Science Ireland Ltd. All rights reserved.

PII: S0009 -2797 (99 )00023 -X

J. Stojan et al. / Chemico-Biological Interactions 119–120 (1999) 147–157148

1. Introduction

Cooperative effects in cholinesterases from different sources are known asinhibition or activation by the excess of the substrate [1]. Kinetic models proposedso far are only tied up with either of the two phenomena [2–4]. The existence ofperipheral binding surface, confirmed by resolving crystal structures [5,6] and by thebehaviour of mutated enzyme forms [4] seems to be related with this interestingcharacteristic. In the case of Drosophila melanogaster acetylcholinesterase (ACHE,EC. 3.1.1.7), however, apparent activation at low substrate concentrations isfollowed by almost complete inhibition at very high concentrations [7]. Thepossibility of the existence of a mixture of two enzyme species extracted from insectbrain tissue [7] was rejected by cloning and expressing it in baculovirus system [8].

In order to get additional kinetic information on the mechanism of doubleactivation-inhibition pattern in the catalytic hydrolysis of acetylthiocholine byAChE from the brain of Drosophila melanogaster, a detailed kinetic study of theeffect of relatively fast acting synthetic organophosphonate compound, MEPQ, onthis process was undertaken. It is believed that actions of various organophospho-rous substrate analogues proceed in the same way as the reaction with substrates[1]. The only difference seems to be in the different rates of individual steps. Ofthem all, the dephosphorylation step is often so slow that it is usually treated asnon-existent [1]. We assumed this when we constructed the kinetic model for thesimultaneous interaction of acetylthiocholine and MEPQ with D. melanogasterAChE.

2. Methods and materials

The hydrolysis of acetylthiocholine catalyzed by D. melanogaster AChE wasrecorded in the absence and presence of MEPQ, on a stopped flow apparatus. Theexperimental curves were obtained by following the product formation photometri-cally [9] after mixing together the aliquots of two buffer solutions, one containingthe enzyme and the other the substrate, MEPQ and Ellman’s reagent in the mixingchamber of the stopped-flow apparatus. The time course of the product formationwas followed at various concentrations of all reactants.

In the first step, the time course of the reaction was recorded in the absence ofMEPQ at substrate concentrations from 2 mM to 0.05 M. Additionally, at lowconcentrations the reaction was followed until its completion while at high concen-trations only the initial portions were measured. The simplest possible kinetic modelwas constructed on the basis of various models for other cholinesterases [2–4,7,10]and in accordance with the biphasic activation–inhibition behaviour of theDrosophila enzyme.

In the second step, the model was enlarged in accordance with the obtainedexperimental data in the presence of MEPQ (see Section 3). It was tested withessentially the same kinetic approach as introduced previously [11] and by usinggeneral criteria for the goodness of fit proposed by Cleland [12].

J. Stojan et al. / Chemico-Biological Interactions 119–120 (1999) 147–157 149

Experiments were done at 25°C in 10 mM phosphate buffer, pH 7.0, with a total ionicstrength of 0.2 M, obtained by addition of NaCl. The enzyme was AChE from D.melanogaster, produced in baculovirus infected cells [8], purified by affinity chromatog-raphy and titrated with MEPQ which was synthesized according to Levy and Ashani[13]. Acetylthiocholine iodide and 5,5%-dithio-bisnitrobenzoic acid (Ellman’s reagent)were obtained from BDH Biochemicals. All substances were reagent grade. Themeasurements were carried out on a stopped-flow apparatus PQ-SF 53 with atheoretical dead time of 0.7 ms, manufactured by Hi-Tech, Salisbury, UK (Fig. 1).

Fig. 1. Ball and stick models of acetylcholine and MEPQ.


3. Results and discussion

Progress curves for the hydrolysis of acetylthiocholine catalyzed by DrosophilaAChE are presented in Fig. 2. The inset shows the dependence of the initial rate ofeach progress curve on its substrate concentration and the diagram is in accordancewith the double activation-inhibition pattern [7]. In order to explain the observedpattern, the following kinetic model was the simplest that was able to phenomeno-logically describe the obtained data: where E is AChE, S is the substrate acetylthio-choline with an acetyl group A (P2) and the leaving group thiocholine (P1). SE isa reversible addition complex and S on the left may denote the binding of thesubstrate at the site other than acylation site. K3 and K4 are the equilibriumconstants, while ki and k3 are the rate constants. a and b are proportional factors.

Among numerous tested models, this one with six kinetic constants and only twosubstrate binding sites fitted the best. In the case of more complicated threesubstrate binding sites and cubic formed schemes with as many as 14 constants,considerable instability among several parameters was observed during the analysisalthough most of them could account for the obtained experimental data as well.According to Gibson [14] such models are possible but unstable steps are kinetically

Fig. 2. Progress curves for the hydrolysis of acetylthiocholine catalyzed by Drosophila acetyl-cholinesterase. The concentrations of the substrate are: 2 mM, 5 mM, 10 mM, 20 mM, 50 mM, 0.1 mM,0.2 mM, 0.5 mM, represented by dots and 1, 2, 5, 10, 20, 40 mM represented by triangles; solid curvesare theoretical, obtained by fitting differential Eqs. (1)–(5) to all experimental curves simultaneously.(For details see text). The inset represents the dependence of initial rates of the progress curves on thelogarithm of the corresponding substrate concentrations.


invisible and should be omitted. It must be pointed out that the model in Scheme1 is only the other variant of the model introduced by Cauet et al. [2] for the actionof butyrylcholinesterase; instead of combining acylation and deacylation, we com-bined the other two steps: complex formation at the acylation site and the acylation(including thiocholine release), thus denoting ki as a bimolecular rate constant.Rolling these two steps together only means that under these conditions theykinetically cannot be separated (the same is true for the binding of MEPQ, seebelow). So, the differential equations for the model presented in Scheme 1 are asfollows:

d[(E)+ (SE)]dt

= −ki � (S) �[K3+b(S)]

K3+ (S)� [(E)+ (SE)]

+k3 �[K4+a(S)]

K4+ (S)� [(EA)+ (SEA)] (1)

d(S)dt

= −ki � (S) �[K3+b(S)]

K3+ (S)� [(E)+ (SE)] (2)

d(EA)+ (SEA)dt

=ki � (S) �[K3+b(S)]

K3+ (S)� [(E)+ (SE)]

−k3 �[K4+a(S)]

K4+ (S)� [(EA)+ (SEA)] (3)

d(P1)dt

=ki � (S) �[K3+b(S)]

K3+ (S)� [(E)+ (SE)] (4)

d(P2)dt

=k3 �[K4+a(S)]

K4+ (S)� [(EA)+ (SEA)] (5)

Although it was shown that the rate of detection reaction might change the shapeof progress curves [15], we can assume that under our experimental conditions therate of color production is proportional to the product formation.

To check the adequateness of the model, the evaluation of kinetic parametersfrom Eqs. (1)–(5) was done in two steps: (i) for the rough estimation, a steady-state

Scheme 1.


Table 1Characteristic constants for the action of MEPQ on Drosophila acetylcholinesterase in the presence ofacetylthiocholine according to Schemes 1 and 2a

7.33×107980111 M−l s−1ki

1544.391.1 s−lk3

K3 3.59590.014 mM22.690.1 mMK4

a 0.058690.00470.136590.0002b

Kl 25.290.08 nM47.190.71 mMK2

090c0.126090.0004d

4124189430 M−l s−1kl

0.002390.0002ef 2.01690.002

a The values in the table were obtained by direct fitting of differential Eqs. (1)–(5) and Eqs. (7)–(13)to the progress curves in Figs. 1 and 2, respectively. (For details see Section 2 in the text).

rate equation was derived under combined steady-state and equilibrium assump-tions [16,17]:

60=E0k3(S)

�1+a

(S)K4

nS�

1+(S)K4

�+

k3�

1+(S)K4

��1+

(S)K3

�ki�

1+b(S)K3

�(6)

and the equation was fitted to the initial rates obtained from the curves for the timecourse of product formation using the non-linear regression program developed byDuggleby [18]; (ii) the obtained values of kinetic parameters were used in a muchmore reliable analytical approach [19]: as first estimates in fitting numerically solvedsystem of differential equations (Eqs. (1)–(5)) simultaneously, to all experimentalprogress curves using a previously introduced computer program [11]. This wasnecessary because at low substrate concentrations complete conversion of thesubstrate into the products violated the assumptions St=S0 and St�E. The resultsof the fitting are shown in Table 1 and in the form of solid curves throughexperimental points in Fig. 2. It should be remembered, that the time course ofproduct P1 is a measured quantity represented by Eq. (4).

The progress curves for the hydrolysis of acetylthiocholine at four differentconcentrations in the presence of various concentrations of MEPQ are shown inFig. 3. All curves in the presence of MEPQ are downwardly concave. Thecurvature, however, has two origins: the first is a consequence of an irreversibleinhibitory action of MEPQ, but the second is expected only with low substrateconcentrations in the case that its conversion is faster than the action of the


inhibitor. Therefore the depletion of the substrate during the reaction also had tobe taken into account.

It is clear from the comparison between the uppermost line-curve in each panelof Fig. 3 without MEPQ, with other curves that the initial rates are MEPQconcentration independent. This means that MEPQ acts on the insect AChE as aslow, irreversible inhibitor, and that at first glance, there is no observable instanta-neous phase in its action. As in Fig. 2, the curve in the panel A with the lowestsubstrate concentration obtained without MEPQ reaches the plateau which corre-sponds to the starting concentration of 0.02 mM. Again, this means that completeconversion of the substrate took place and should be considered in the analysis.

On the other hand, some curves obtained in the presence of the inhibitor reachthe plateau within the linear portion of the line obtained in the absence of it. Thissuggests that although the inhibitory action is slow, it is in fact very effective sinceit takes place within several minutes despite of relatively high substrateconcentrations.

Fig. 3. Progress curves for the hydrolysis of acetylthiocholine catalyzed by Drosophila acetyl-cholinesterase in the absence and presence of MEPQ. In the panel with acetylthiocholine concentrationof 0.02 mM, the enzyme concentration is 0.49 nM and the concentrations of MEPQ are: 0, 8.8, 19, 40,110, 175, 200, 250, 320 and 530 nM; for 0.1 mM substrate, the enzyme is 0.46 nM and MEPQconcentrations are: 0, 92, 170, 260, 490 nM, 1 mM, 1.35 mM, 2 mM, 2.3 mM, 3.5 mM and 4.6 mM; for1 mM substrate, the enzyme is 0.23 nM and the concentrations of MEPQ are: 0, 120, 160, 430, 740 nM,l mM, 1.25 mM, 1.5 mM, 2.1 mM, 5.4 mM and 8.8 mM; for 5 mM substrate, the enzyme is 0.3 nM andMEPQ is 460 nM, 740 nM, 1.4 mM, 3.7 mM, 7.3 mM and 9.2 mM. Solid curves are theoretical, obtainedby fitting differential Eqs. (7)–(13) to all experimental curves simultaneously. (For details see text).


Scheme 2.

Scheme 3.

The analysis of curves in Fig. 3 was done essentially in the same way as those ofFig. 2. Scheme 1, however, was enlarged to allow competition between the substrateand MEPQ at both substrate binding sites (see Section 1).

In Scheme 2, I on the left of E may represent the complex between the enzymeand MEPQ at the site other than acylation site, while on the right side representsthe phosphorylated enzyme.

According to Cha [17], Scheme 2 can be written in a form where only slow stepsare included:

Here X and X % contain the instantaneously formed enzyme species E, SE, IE andEA, SEA, IEA, respectively. Y contains the relatively slowly formed irreversiblyinactivated enzyme forms EI, SKI, IEI.

Differential equations for the system described by Schemes 2 and 3 are:

d(X)dt

= − [ki(S)a+k1(I)h ](X)+k3b(X %) (7)

d(S)dt

= −ki(S)a(X) (8)

d(X %)dt

= [ki(S)a+k1(I)h ](X)−k3b(X %) (9)

d(P1)dt

=ki(S)a(X) (10)

d(P2)dt

=k3b(X %) (11)

d(I)dt

= −k1(I)h(X) (12)

d(Y)dt

=k1(I)h(X) (13)

where


a=[K1K3+b(S)K1+d(I)K3]

K1K3+ (S)K1+ (I)K3

(14)

b=[K2K4+a(S)K2+c(I)K4]

K2K4+ (S)K2+ (I)K4

(15)

h=[K1K3+e(S)K1+ f(I)K3]

K1K3+ (S)K1+ (I)K3

(16)

For Scheme 2, the initial rate equation is not the same as in the absence of MEPQ(Eq. (6)). This seems to be in contradiction with the pattern of the curves in Fig. 3unless the proportional factors b, d and a, c, respectively, are approximately equal.On the other hand, the rates at infinitive reaction time are zero, due to theirreversible action of MEPQ. Therefore, the evaluation of all rate and equilibriumconstants for the reaction according to Scheme 2 was carried out by fitting thedifferential Eqs. (7)–(13) (using the expressions in Eqs. (14)–(16)) to all curves fromFig. 3 simultaneously. Since this is a next to impossible task, the evaluation wasbroken down into several steps. Besides, the first six parameters out of 13 in Scheme2, were those already determined for the reaction without the inhibitor. For thisreason, they were set as constants in the preliminary evaluation of the other sevenparameters. When good estimates corresponding well to each of the four sets ofprogress curves in Fig. 3 were found, all curves [20] were used for the finalevaluation (Table 1).

Two points are to be clarified in connection with this analysis. (i) It emerged thatthe proportional factor c was unstable and approached zero. This was expectedsince the highest substrate concentration was still relatively low for the final part ofthe substrate inhibition. But higher substrate concentrations would call also forhigher MEPQ concentrations which were already enormous. And exactly this is anindication for the competitive effect of MEPQ, proposed by Scheme 2. This alsomeans that in fact c$a. This is one of the conditions for virtual in-existence of aninstantaneous phase in the action of MEPQ. (ii) In order to measure the progresscurves under very different reactant concentrations, the concentrations of enzymeactive centers had to be chosen properly. In the final fitting of all curves, theconcentrations of the enzyme active sites were set as determined titrimetically andthe set substrate of varying concentrations as added (therefore, the panels in Fig. 3are not absolutely comparable).

In this way, all parameters with very low standard deviations and calculatedcurves which fit well the experimental data were obtained. Again it must beemphasized that in spite of 12 parameters, Scheme 2 represents a very simple kineticmodel, which is able to reproduce the time dependent kinetic behavior of theinteraction between acetylcholinesterase and an irreversible organophosphonateinhibitor in the presence of the substrate. A wide range of experimental concentra-tions of the substrate and the inhibitor are also a very good test of the adequatenessof the model. All peculiarities of the insect enzyme, including substrate activation-inhibition pattern and the effect of MEPQ, on these phenomena are wellreproduced.


At this stage the question on the applicability of the model to other systemsarises: It has been used to explain not only the action of reversible anti-cholinesterase agents such as tetrametylammonium, edrophonium or propidium(submitted for publication) but also the kinetic behavior of various mutated enzymeforms and of horse serum butyrylcholinesterase (in preparation).

But in spite of all this, one could argue that this is only a phenomenologicalexplanation which does not tell much of the molecular mechanism during thesubstrate turnover and the effect of MEPQ on it. There is an element of truth inthis as much as there is for every kinetic model. However, a reasonable explanationfor each step in the model is possible together with known facts from literature: (i)the basic concept are two binding sites [2–4]; (ii) it is known that the action ofcholinesterases is fast but not diffusion rate limited [21] and the value of ki supportsthis hypothesis; (iii) on the other hand, the binding of small molecules onto thesurface of proteins is usually only limited by diffusion [22] and K3, K4, Kl and K2

could represent instantaneous binding to the site different from the acylation site;(iv) the same argument can also answer how can another molecule reach theacylation site while the first one is still bound in the vicinity. In another way: at thesaturating concentrations for the binding to the site other that acylation site thereis always a next molecule at this site while the first one is accommodating at theacylation site; (v) the steady-state flux into the cleft is in accordance with thehypothesis that virtually every substrate or quasi-substrate molecule which entersthe active site is converted [21]. It is, however, the particular orientation (thespecificity in binding) of various ligands at the peripheral site which decide whetherthe entrance is enhanced or slowed. This is reflected by the corresponding values ofthe proportional factors b, d, e and f in Scheme 2 and also by the fact that theW359L mutant of Drosophila enzyme is activated rather inhibited by D-tubocurarine (unpublished data).

Acknowledgements

This work was supported by the Ministry of Science and Technology of theRepublic of Slovenia, grant No.: P3-8720-0381, and by grants from DRET (94-084), IFREMER and CNRS (GDR 1105 and ACS-SV3).

References

[1] A.R. Main, in: A.M. Goldberg, I. Hanin (Eds.), Biology of Cholinergic Function, Raven Press,New York, 1976, pp. 312–315.

[2] G. Cauet, A. Friboulet, D. Thomas, Horse serum butyrylcholinesterase kinetics: a molecularmechanism based on inhibition studies with dansylaminoethyl-trimethylammonium, Biochem. Cell.Biol. 65 (1987) 529–535.

[3] J. Eastman, E.J. Wilson, C. Cervenansky, T.L. Rosenberry, Fasciculin 2 binds to the peripheral siteon acetylcholinesterase and inhibits substrate hydrolysis by slowing a step involving proton transferduring enzyme acylation, J. Biol. Chem. 270 (34) (1995) 19694–19701.


[4] Z. Radic, R. Duran, D.C. Vellom, Y. Li, C. Cervenansky, P. Taylor, Site of fasciculin interactionwith acetylcholinesterase, J. Biol. Chem. 269 (15) (1994) 11233–11239.

[5] J.L. Sussman, M. Harel, F. Frolow, C. Oefner, A. Goldman, L. Toker, I. Silman, Atomic structureof acetylcholinesterase from Torpedo californica: A prototypic acetylcholine-binding protein,Science 253 (1991) 872–879.

[6] Y. Bourne, P. Taylor, P. Marchot, Acetylcholinesterase inhibition by fasciculin: Crystal structure ofthe complex, Cell 83 (1995) 503–512.

[7] V. Marcel, L.G. Palacios, C. Pertuy, P. Masson, D. Fournier, Two invertebrate acetyl-cholinesterases show activation followed by inhibition with substrate concentration, Biochem. J. 329(1998) 329–334.

[8] D. Fournier, J.M. Bride, F. Hoffman, F. Karch, Acetylcholinesterase: Two types of modificationconfer resistance to insecticides, J. Biol. Chem. 267 (20) (1992) 14270–14274.

[9] G.L. Ellman, K.D. Courtney, V. Andres, R.M. Featherstone, A new and rapid colorimetricdetermination of acetylcholinesterase activity, Biochem. Pharmacol. 7 (1961) 88–95.

[10] J. Stojan, M. Zorko, Kinetic characterization of all steps of the interaction between acetyl-cholinesterase and eserine, Biochim. Biophys. Acta 1337 (1997) 75–84.

[11] J. Stojan, Analysis of progress curves in an acetylcholinesterase reaction: a numerical integrationtreatment, J. Chem. Infor. Comp. Sci. 37 (1997) 1025–1027.

[12] W.W. Cleland, Statistical analysis of enzyme kinetic data, Methods Enzymol. 63 (1963) 103–138.[13] D. Levy, Y. Ashsni, Synthesis and in vitro properties of a powerful quaternary methylphosphonate

inhibitor of acetylcholinesterase., Biochem. Pharmacol. 35 (7) (1986) 1079–1085.[14] Q. Gibbson, Rapid reaction methods in biochemistry, in: A. Neuberger, L.L.M. Van Deenen (Eds.),

Modern Physical Methods in Biochemistry, Part B, Elsevier, Amsterdam, 1988, pp. 65–84.[15] J. Stojan, M.R. Pavlic, Velocity of Ellman’s reaction and its implication for kinetic studies in the

millisecond time range., Neurochem. Res. 17 (12) (1992) 1207–1210.[16] J. Stojan, Equations for progress curves of some kinetic models of enzyme-single substrate-single

slow binding modifier system, J. Enzyme Inhibition 13 (1998) 161–176.[17] S. Cha, A simple method for derivation of rate equations for enzyme-catalyzed reactions under the

rapid equilibrium assumption or combined assumptions of equilibrium and steady-state, J. Biol.Chem. 243 (4) (1968) 820–825.

[18] R.G. Duggleby, Regression analysis of nonlinear Arrhenius plot: an empirical model and acomputer program, Comput. Biol. Med. 14 (4) (1984) 447–455.

[19] M. Markus, T. Plesser, in: L. Endrenyi (Ed.), Kinetic Data Analysis, Plenum, New York, 1981, pp.317–339.

[20] D.F. Senear, D.W. Bolen, Simultaneous analysis for testing of models and parameter estimation,Methods Enzymol. 210 (1992) 463–485.

[21] J. Antosiewicz, M.K. Gilson, I.H. Lee, J.A. McCammon, Acetylcholinesterase: diffusional encoun-ter rate constants for dumbbell models of ligand, Biophys. J. 68 (1995) 62–68.

[22] J.F. Morrison, C.T. Walsh, The behavior and significance of slow-binding enzyme inhibitors, Adv.Enzymol. 61 (1988) 207.

.

Documents

Inhibition of Drosophila acetylcholinesterase by 7-(methylethoxyphosphinyloxy)1-methyl-quinolinium iodide