Kinetic Model of Ethopropazine Interaction with Horse Serum Butyrylcholinesterase and Its Docking into the Active Site

Archives of Biochemistry and Biophysics Vol. 398, No. 1, February 1, pp. 23–31, 2002 doi:10.1006/abbi.2001.2697, available online at http://www.idealibrary.com on

Kinetic Model of Ethopropazine Interaction with Horse Serum Butyrylcholinesterase and Its Docking into the Active Site Marko Golicˇnik,* Goran Sˇinko,† Vera Simeon-Rudolf,† Zoran Grubicˇ,‡ and Jure Stojan* ,1 *Institute of Biochemistry and ‡Institute of Pathological Physiology, Medical Faculty, University of Ljubljana, 1000 Ljubljana, Slovenia; and †Institute for Medical Research and Occupational Health, Ksaverska c. 2, P.O. Box 291, 10001 Zagreb, Croatia

Received September 17, 2001, and in revised form October 24, 2001; published online January 2, 2002

The action of a potent tricyclic cholinesterase inhibitor ethopropazine on the hydrolysis of acetylthiocholine and butyrylthiocholine by purified horse serum butyrylcholinesterase (EC 3.1.1.8) was investigated at 25 and 37°C. The enzyme activities were measured on a stopped-flow apparatus and the analysis of experimental data was done by applying a six-parameter model for substrate hydrolysis. The model, which was introduced to explain the kinetics of Drosophila melanogaster acetylcholinesterase [Stojan et al. (1998) FEBS Lett. 440, 85– 88], is defined with two dissociation constants and four rate constants and can describe both cooperative phenomena, apparent activation at low substrate concentrations and substrate inhibition by excess of substrate. For the analysis of the data in the presence of ethopropazine at two temperatures, we have enlarged the reaction scheme to allow primarily its competition with the substrate at the peripheral site, but the competition at the acylation site was not excluded. The proposed reaction scheme revealed, upon analysis, competitive effects of ethopropazine at both sites; at 25°C, three enzyme–inhibitor dissociation constants could be evaluated; at 37°C, only two constants could be evaluated. Although the model considers both cooperative phenomena, it appears that decreased enzyme sensitivity at higher temperature, predominantly for the ligands at the peripheral binding site, makes the determination of some expected enzyme substrate and/or inhibitor complexes technically impossible. The same reason might also account for one of the paradoxes in cholinesterases: activ-

1 To whom correspondence and reprint requests should be addressed at Institute of Biochemistry, Medical Faculty, Vrazov trg 2, 1000 Ljubljana, Slovenia. Fax: ⫹386-(0)-1-5437641. E-mail: stojan@ ibmi.mf.uni-lj.si.

0003-9861/02 $35.00 © 2002 Elsevier Science All rights reserved.

ities at 25°C at low substrate concentrations are higher than at 37°C. Positioning of ethopropazine in the activesite gorge by molecular dynamics simulations shows that A328, W82, D70, and Y332 amino acid residues stabilize binding of the inhibitor. © 2002 Elsevier Science Key Words: butyrylcholinesterese; kinetic models; ethopropazine; progress curves; stopped flow.

Butyrylcholinesterases (BChEs), 2 closely related to acetylcholinesterases (AChEs), are found in blood of all vertebrates (cf. Ref. 1). From the resolved 3D structures of various AChEs (EC 3.1.1.7) it is known that the active sites residues of these enzymes lay at the bottom of a 20-Å-deep hydrophobic gorge (2– 4). Due to a larger cavity of this gorge BChEs accept, in comparison to AChEs, broader variety of substrates and inhibitors (5–7). For instance they metabolize butyrylcholine, the choline ester with large acyl moiety whose hydrolysis by vertebrate AChEs is negligible. Cholinesterases are non-Michaelian enzymes showing two deviations from classical hyperbolic kinetics. The first deviation, which is almost always present, is negative homotropic cooperativity at very high substrate concentrations. The second one was first recognized in BuChEs and has been described as activation at intermediate substrate concentrations, but carefully performed experiments, reveal that these enzymes too, are inhibited by excess of substrate. Recently, both phenomena have been described in wild-type enzymes from nematodes and insects (8) and also in many side 2 Abbreviations used: AChE, acetylcholinesterase; BChE, butyrylcholinesterase; ATCh, acetylthiocholine; BTCh, butyrylthiocholine; Etho, ethopropazine.

23

ˇ NIK ET AL. GOLIC

24

EXPERIMENTAL PROCEDURES

FIG. 1. Structure of ethopropazine.

specific mutants of vertebrate AChEs (cf. 7). It is expectable therefore, that considerable differences in kinetic behavior of these enzymes occur in spite of an essentially identical reaction mechanism of the catalytic process (9). Several investigators assume, on the basis of buried active site, that every substrate molecule that enters the active gorge is converted (10, 11). It seems that in cholinesterases graduate increase in the substrate affinity down the gorge guides it toward covalent acyl intermediate. Thus, when introducing a kinetic model valid for all cholinesterases (12), we have hypothesized that the only difference that discriminates one cholinesterase from another is the magnitude of various kinetic parameters, i.e., the rate of individual steps. Often certain rates are such that some important intermediates, like the Michaelis– Menten complex, do not accumulate. Such intermediates cannot be detected and could be omitted from the reaction scheme. Kinetic representation of these consecutive reactions can be summarized by rolling of all putative intermediate steps into a single irreversible one (13). On the other hand, several independent approaches identified various amino acid residues at the entrance of the active site gorge, in different cholinesterases, as primary contact sites, denoting them as the constituents of a peripheral anionic site (PAS) (14 –16). The existence of peripheral binding sites on BChEs has been revealed in reactions with substrates and inhibitors (17, 18). Interaction of ligands, especially with the peripheral sites, results in either enzyme activity inhibition or activation or both (19, 20). Inhibition of human BChE by ethopropazine (Fig. 1) has been characterized by noncompetitive and competitive enzyme–inhibitor dissociation constants, and two potential orientations of the ligand in the gorge of human BChE have been modeled (5–7, 21). To elucidate these relations in BChE from horse serum, we investigated kinetically and by molecular modeling the action of ethopropazine on this enzyme. In particular, the effect of higher temperature on various steps in the kinetic mechanism should be proposed.

Materials. BChE from horse serum was a commercial preparation from Sigma Chemical Co. (St. Louis, MO). Butyrylthiocholine (BTCh), acetylthiocholine (ATCh), 5,5⬘-dithio-bis-nitrobenzoic acid (DTNB) and ethopropazine hydrochloride (Etho) were also purchased from Sigma. All other substances were also analytical grade. All experiments were done at 25 and 37°C in 0.1 M sodium phosphate buffer, pH 7.4. Substrate hydrolysis measurements. The hydrolysis of BTCh and ATCh catalyzed by horse serum BChE was recorded in the absence and presence of Etho, on a stopped-flow apparatus. Aliquots of two solutions, one containing only the enzyme and the other the substrate, Etho and DTNB reagent were mixed together in the mixing chamber of the apparatus. The absorbance of the reaction mixture was recorded spectrophotometrically according to Ellman et al. (22) at different substrate concentrations (5 ␮M–75 mM). At each substrate concentration, four different inhibitor concentrations (0.33– 3.3 ␮M) were tested. The final DTNB concentration was 1 mM. At low substrate concentrations the reaction was followed until its completion while at higher concentrations only the initial portions were measured. To avoid possible product modulation, we stopped the measurement when 60 ␮M product was formed. Concentration of enzyme active sites was determined by a stoichiometric titration with various concentrations of a high-affinity phosphorylating agent DEPQ [7-(O,O⬘-diethylphosphinyloxy)-1-methyl-quinolinium methylsulfate] (23). Concentrations of enzyme active sites during the assay were between 1.50 and 2.64 nM. Activities at 25 and 37°C with acetylthiocholine were also measured by the Ellman method on a spectrophotometer without stopped-flow device; the time of assay was several minutes. The observed rates were expressed in micromoles per second by using the molar coefficient 13,600 L mol ⫺1 cm ⫺1 for the anion of 5-thio-2nitrobenzoic acid. Concentration of active sites during this assay was 0.131 nM. Ethopropazine positioning in the active site. Modeling and docking was performed with WHATIF computer program (24). Further refinement and the molecular dynamics were carried out using the macromolecular simulation program CHARMM (25). The homologybuilt model of human BChE (CODE P06276) from SwissProt Model Repository was mutated and optimized to obtain a good atomic model of horse BChE. X-ray structure of ethopropazine hydrochloride was obtained from the Cambridge Crystal Structure Database and an ab initio geometry minimization using the GAUSSIAN program and 6-31g basis set at the Hartree–Fock level of ethopropazine protonated structure (Fig. 1) was performed. For molecular mechanics energy calculations we assigned atomic types for Etho molecule using types in the CHARMM distribution C27n1 topology file and charges calculated by Mullikan’s approximation. Under these conditions a satisfactory fit to the ab initio energy potentials and geometry was obtained. In the next step, we manually docked Etho molecule into the active site according to the position of tetrahydro-phenylmethyl-acridinamide in the active gorge of Drosophila melanogaster DmAChE (26, PDB entry 1DX4). Finally, constant temperature and pressure (CPT) dynamics calculations invoking the Ewald summation for calculating the electrostatic interactions were performed as follows: the protein molecule with docked Etho was put into a box of 14005 water molecules and the dynamics simulations were run for 370 ps (300 K, 1 bar) without any constraints, after a few picoseconds equilibration with fixed protein and the inhibitor. In all optimizations and simulations we used universal cut-off distance of 10 Å, except during EWALD dynamic where the electrostatic force is treated without cutoffs. Dissociable groups were treated as they appear at physiological pH: protonated arginines, lysines, and Nterminal; aspartates, glutamates, and C-terminal were deprotonated; all histidines were neutral with protonated N␦1. For dynamic simulations we used a cluster of four IBM compatible PCs with

25

INTERACTION OF ETHOPROPAZINE WITH HORSE SERUM BUTYRYLCHOLINESTERASE

SCHEME 1

450-Mz Celeron processors running under LINUX (10-ps simulation run, approximately 25 h). Data analysis. For the analysis of substrate hydrolysis progress curves in the absence of Etho we used the six-parameter model (12), which is defined by two dissociation constants (K 1, K 2) and four rate constants (k i, k 3, bk i, and ak 3) (Scheme 1). In this scheme, E is the free enzyme and EA the acylated enzyme. SE is a kinetically detectable species where the enzyme is in the high affinity complex with the substrate which may still undergo reversible dissociation and SEA represents the acylated enzyme with the substrate molecule bound to the nonproductive, modulation site. The

products P 1 and P 2 are thiocholine and acetate or butyrate, respectively. A system of stiff differential equations which describes the model under combined steady-state and equilibrium assumptions (cf. 27) was fitted simultaneously to the data of all experimental progress curves, using an appropriate computer program (28). For the analysis of the experiments in the presence of Etho we made an extension of the model to allow binding of Etho also with E, EA, SE, and SEA complexes. Additionally, we included in the model the binding of two Etho molecules to the free enzyme (Scheme 2). In this scheme, Etho, I, can bind either to the active site or to a nonproductive site on the enzyme. If d and f are zero the species IE and ISE represent enzyme–inhibitor complexes at the active site, while Etho in the complexes ISEA, IEA, and the second inhibitor molecule in I 2E must interact at the nonproductive site. The simultaneous analysis of substrate hydrolysis data in the presence of Etho, i.e., the evaluation of existing steps in Scheme 2 and the determination of kinetic parameters was done in two steps: (i) for the first estimation, a steady-state rate equation for the model in Scheme 2 was derived under the combined steady-state and equilibrium assumptions (27):

冉

vi ⫽

冉

关S兴 1 ⫹

关S兴 关S兴关I兴 关I兴 ⫹ ⫹ K2 K 2K 6 K4

冊

k3 ⫹

冉

v i is initial steady-state rate of substrate hydrolysis in the absence or presence of the inhibitor, E o is the total concentration of the enzyme, S and I are the initial substrate and Etho concentrations, respectively. By setting [I] ⫽ 0, Eq. [1] reduces to the steady-state rate equation valid for Scheme 1. Equation [1] was fitted to the initial rates determined from the progress curves in substrate hydrolysis measurements; (ii) the obtained parameters were used in the second step as initial estimates in fitting the numerically solved system of differential equations (cf. 28) simultaneously to all progress curves. We collected four sets of progress curves, at two temperatures (25, 37°C) using ATCh or BTCh as substrates. In these sets, there were between 56 and 77 progress curves, each consisting of approximately 20,000 data points.

SCHEME 2

冊 冊 冉

关S兴 关S兴关I兴 关I兴 ⫹e ⫹c K2 K 2K 6 K4 关S兴 关S兴关I兴 关I兴 关S兴 关S兴关I兴 关I兴 关I兴 2 1⫹a ⫹e ⫹c 1⫹ ⫹ ⫹ ⫹ K2 K 2K 6 K4 K1 K 1K 5 K 3 K 3K 7 关S兴 关S兴关I兴 关I兴 ki 1 ⫹ b ⫹f ⫹d K1 K 1K 5 K3

E 0 k 3 关S兴 1 ⫹ a

冉

冊

冊

[1]

RESULTS

The progress curves for the hydrolysis of BTCh by horse serum BChE at 25°C in the absence and presence of Etho are shown in Fig. 2. Analogous results were obtained in similar experiments performed at 37°C and with ATCh as the substrate at the same temperatures (diagrams not shown). The curves at low substrate concentrations reach plateaus which correspond to each starting substrate concentration, and with rising Etho concentrations these plateaus are achieved later. Thus, at low substrate concentrations, Etho inhibits BTCh hydrolysis. The inhibition can also be established by comparing initial slopes of progress curves obtained at high substrate concentrations. More instructive representation of the same data is given in the form of theoretical pS curves (v vs log[S]) in Fig. 3. The points in Fig. 3a are numerical derivatives at time 0 of each theoretical progress curve in Fig. 2b–2d display the analogous initial rates at 37°C and with ATCh as the substrate at 25 and 37°C. Theoretical progress curves, again, were obtained by fitting differential equations for the model in Scheme 2 to all progress curves in each figure simultaneously (see Data analysis). The results of this fitting are presented in Table I and the corresponding parameters were put into Eq. [1] to calculate the curves in Fig. 3.

26

ˇ NIK ET AL. GOLIC

FIG. 2. Progress curves for the hydrolysis of butyrylthiocholine catalyzed by the wild-type horse serum BChE in the absence and presence of ethopropazine at 25°C. In a and b the enzyme concentrations were 1.5 nM, in c and d 1.31 nM, and in e and f 1.4 nM. Concentrations of ethopropazine in a–f were 0, 0.066, 0.33, 0.66, 1.66, and 3.3 ␮M, respectively. Substrate concentrations were from 5.4 ␮M to 75 mM (77 curves, 20167 data points).

Theoretical pS curves in all figures show that Etho inhibits BTCh and ATCh hydrolysis by horse serum BChE at all tested substrate concentrations at 25°C as well as at 37°C. By careful inspection, however, one can notice special characteristics of these pS curves: (i) inhibition by excess of substrate is seen in the experiments with BTCh at 25°C, and is only perceivable at 37°C; (ii) in the case of ATCh hydrolysis, substrate inhibition is not visible and at 37°C even the plateau is not reached; (iii) the course of pS curves in the absence of inhibitor at 25°C is almost linear in the range of intermediate substrate concentrations indicating higher enzyme activity as expected by simple Michaelis–Menten reaction mechanism; (iv) it is a common tendency in all panels of Fig. 3 that with rising Etho concentrations this apparent activation at intermediate substrate concentrations is less and less significant.

To stress another very interesting characteristic of measured initial rates at two temperatures we present in Fig. 4 the pS-curves for BTCh and ATCh hydrolysis in the absence of Etho at 25 and 37°C. It should be emphasized that in this figure too, the initial rates were calculated from Eq. [1], but we applied only the enzyme–substrate interactions constants listed in Table I; all terms in the equation referring to enzyme– inhibitor interactions were zero, due to the absence of the inhibitor. The pS curves show the inversion of the initial rates at low substrate concentrations: unexpectedly, the course of pS curves at low and intermediate substrate concentrations is equal or slightly higher at 25 than at 37°C. On the other hand, the initial rates at 37°C exceeded those measured at 25°C with BTCh (above 0.25 mM) and ATCh (above 4 mM). To confirm the temperature dependence, we repeated the measurements with ATCh on a classical spectrophotometer

INTERACTION OF ETHOPROPAZINE WITH HORSE SERUM BUTYRYLCHOLINESTERASE

27

FIG. 3. The dependence of initial rates of horse BChE on the concentration of ethopropazine at various butyrylthiocholine (a, b) or acetylthiocholine (c, d) concentrations. Data in a and c were obtained at 25°C and those in b and d at 37°C. The concentrations of ethopropazine in a and b were 0, 0.066, 0.33, 0.66, 1.66, and 3.3 ␮M, while in c and d were 0, 0.33, 0.66, 1.66, and 3.3 ␮M. Initial rates are normalized to the same enzyme concentration and the points are theoretical, obtained as numerical derivatives of each fitted progress curve at time 0.

without rapid stopped-flow equipment in the absence of Etho; the obtained activities fitted well into the calculated pS curves (Fig. 4b). Molecular dynamics calculations, using the homology-built 3D model of horse BChE, show during 370 ps a slight netto shift of Etho from the starting position toward free space around Ala328. The position seems to be stable especially because of three groups of interactions: (i) two lateral rings of Etho make ␲–␲ interactions with Trp82 and ␲–␴ interactions with methyl group of Ala328, respectively, with the distances around 4 Å; (ii) an electrostatic interaction with an average distance of 4.4 Å between protonated nitrogen of the aliphatic side chain of ethopropazine and Asp70; and (iii) ␲–␴ interactions between the phenol ring of Tyr332 and the terminal ethyl groups of ethopropazine.

DISCUSSION

Kinetic behavior of cholinesterases originates from the specific architecture of the active site, which has been revealed by resolved crystal structures (2– 4). Unfortunately, even crystal structures themselves cannot provide all answers concerning diversity in catalytic function of these enzymes. Although, various kinetic models have been proposed for these purposes, a universal one is still urged. Two types of approach have so far proven to introduce some clarification in this field. The first starts with the assumptions predicted from chemical reaction theory and subsequently trying to integrate kinetic evidence with nonkinetic data to confirm and quantitatively evaluate all steps in the reaction mechanism (15). The second one is doing basically the same, but suggests at the beginning the simplest reaction

ˇ NIK ET AL. GOLIC

28

TABLE I

Characteristic Constants for the Interactions of Horse Serum Butyrylcholinesterase with Various Substrates and Ethopropazine at Two Temperatures, According to Scheme 2 Substrate and temperature

Butyrylthiocholine (25°C)

Butyrylthiocholine (37°C)

Acetylthiocholine (25°C)

Acetylthiocholine (37°C)

k i ⫻ 10 ⫺6 (M ⫺1 s ⫺1) k 3 (s ⫺1) K 1 (␮M) K 2 (M) a b K 3 (␮M) K 4 (M) c d K 5 (␮M) K 6 (M) e f K 7 (␮M)

7.21 2030 166 0.011 0.55 0.079 0.21 — 0 0 0.16 ⬎1 a 0 0.02 1.1

7.46 2190 43 0.13 a 0.71 a 0.32 0.71 — 0 0 0.37 — 0 0 —

2.81 910 155 0.14 a 0.74 a 0.141 0.16 — 0 0 0.45 — 0 0 4.4

1.68 1190 223 ⬎1 a 0.72 a 0.143 a 0.23 — 0 0 1.32 a — 0 0 —

Note. Due to a large number of experimental data points (approximately 20,000 in each set), the standard deviations of fitted parameters were less than 1%. a See Discussion.

scheme which is generally applicable (12). In the present investigation we are using the second approach and try to expand the basic reaction scheme (Scheme 1) by identifying intermediates in the reaction mechanism, which accumulate during the cata-

lytic process in the presence of Etho. The applied six-parameter model foresees modified Michaelis– Menten mechanism which takes into account well defined kinetic characteristics observed in cholinesterases: (i) undetectable Michaelis–Menten complex

FIG. 4. Activities calculated from the constants in Table I according to Eq. [1] for butyrylthiocholine (a) and acetylthiocholine (b) at 25 (- - -) and 37°C (—). Activities were measured with acetylthiocholine on a spectrophotometer (without a stopped-flow device) at 25°C (E) and 37°C (F).

INTERACTION OF ETHOPROPAZINE WITH HORSE SERUM BUTYRYLCHOLINESTERASE

29

FIG. 5. Stereoview of the active site of the horse BChE homology-built 3D model. The structures of tacrine (yellow) and ethopropazine (red, starting; blue, final position) are superimposed.

is omitted in the traditional reaction scheme, thus, a single acyl enzyme intermediate is proposed in the main reaction pathway, (ii) an intermediate is included which represents the formation and accumulation of an enzyme–substrate complex at the site or position, distinct from acylation site and finally, and (iii) the affinity of the enzyme to form reversible substrate complexes may change substantially upon cooperative effects or different orientations, forced by the enzyme acylation and vice versa. It has been proposed, that Etho noncompetitively inhibits mouse AChE by wedging into its active site and that there are two potential orientations of this ligand in the gorge of human BChE (6). We use horse BChE which is more than 90% identical with human enzyme and can therefore be seen, in terms of peripheral site differences, as a human A277V/G283D/P285L triple mutant (W279, D285, I287, homologous in Torpedo). To analyze our kinetic measurements in the presence of Etho at two temperatures, we have therefore enlarged the reaction scheme to allow competition of Etho with the substrate at the peripheral site, but our representation is not in contradiction with possible interaction of Etho at the acylation site (Scheme 2). During the analysis, it turned out that some parameters in the scheme appeared unstable, suggesting that the fluxes through the corresponding steps are unimportant. According to the principles of kinetic theory those steps can be omitted (20, 29, 30), so they are missing in Table I. Additionally, some steps and intermediates, like those responsible for substrate inhibition are obvious, but they cannot be determined reliably because limitations in solubility prevented the collection of full information (parameters marked with a in Table I). What is still more interesting are the consequences of omitted steps: the absence of the acylation of IE complex (d ⫽ 0) and the nonexistence of

IEA complex (missing K 4) suddenly converts the competition between the substrate and Etho at the peripheral site into the model where the two compete at the active site. Visualization and docking of Etho showed that when inhibitor is in interaction with Trp82 substrate could not interact with choline binding site and vice versa (Fig. 5). The comparison of the corresponding constants in Table I show, that with rising temperature, the affinity for the substrates and Etho decreases. If one assumes realistically that all rates are enhanced at higher temperature than, in our case, it influences dissociation more effectively. However, we can see significant differences in this effect on two types of constants: while K 1 , K 3 , and K 5 are little changed, K 2 and K 7 change drastically. As mentioned, some even cannot be determined reliably any more. It seems that at higher temperature the binding represented by these constants, even if it occurs, does not affect the enzyme activity. Since this is true for the inhibitor as well as for the substrate, the following speculation might be drawn: less affected affinities represent the binding into the active site and those which are significantly affected reflect interactions with the peripheral site. Although the measurements in the presence of ATCh could technically not be performed at sufficiently high concentrations and thus not provide direct information for several steps in the reaction (values denoted with a in Table I), we can perceive an interesting feature: Etho affinity is better for free (K 3) and butyrylated (K 6) enzyme than for acetylated enzyme (absence of K 6). As suggested, K 3 might reflect high affinity binding of Etho into the active site. The existence of Etho binding step, characterized by K 6, might be explained as the interaction with the nonproductive site allowed only by better stabilization of larger butyryl moiety which

30

ˇ NIK ET AL. GOLIC

rumble less in fully occupied active site. The similar reasoning might help us to understand why the values of k i and k 3 are lower with ATCh. Smaller ATCh has less chance to accommodate for successful acylation in relatively large BChE gorge and forms less water susceptible acyl– enzyme complex. This phenomenon can be even potentiated at higher temperatures, but only at low substrate concentrations. At higher substrate concentrations, however, two phenomena are effective: increased number of successful entrances into the gorge and lower affinity at the nonproductive inhibition site dramatically enhance the turnover of the enzyme. We see, therefore, in Fig. 4 the crossing of the corresponding pS curves measured at two temperatures. The pS diagrams in Fig. 3 obtained at 37°C show no detectable substrate inhibition. The simultaneous analysis of the data in the absence and presence of Etho still yields the “a” values slightly lower than unity, thus anticipating inhibition at such a high substrate concentration which cannot technically be achieved. Accordingly, it also yields low affinity constant for binding of the substrate to the modulation site of the acyl enzyme intermediate (K 2). The analysis, however, reveals unexpectedly an interesting feature: although our assumption was that Etho interacts with the modulation site, the absence of parameters c, d, e, f, K 4, and K 6 clearly points to active-site competition with the substrate. The competitive type of inhibition proposed by our kinetic experiments is visualized in Fig. 5. It shows the position of Etho in the active site of the homology-built 3D model of horse BChE. The starting position was obtained by superimposing Etho on tacrine derivative tetrahydro-phenylmethyl-acridinamide in the gorge of DmAChE (26). After 370 ps of EWALD dynamic simulations of the complex, the inhibitor moved with one of the rings to form ␲–␴ interaction with methyl group of Ala328. The other ring preserved its ␲–␲ interaction with Trp82 as well as the protonated nitrogen of Etho aliphatic side chain its electrostatic interaction with Asp70 (deprotonated during the simulation run), the major component of the PAS (5, 6, 17). The later residue stayed, during the complete simulation, hydrogen bonded with its peripheral site counterpart Tyr332. It seems that all these interactions can stabilize Etho in the gorge of horse BChE where no obvious steric obstructions neither of the shape nor of the size could prevent its entering. Finally, such a positioning of Etho in the active site of horse BChE is further sustained by the fact that it gradually diminishes apparent substrate activation and decreases substrate metabolization at all concentrations. In contrast, the analogous experiments on various D. melanogaster AChEs have

shown, that “omitting” the peripheral anionic site or covering it by peripheral ligand, leads to a disappearance of apparent activation but to much better overall enzyme effectiveness (20). If Etho were only a peripheral ligand a similar effect would be expected.

ACKNOWLEDGMENTS Our thanks are extended to Dr. E. Reiner for critical reading of the manuscript. The authors are grateful to Dr. H. Leader, Dr. Y. Ashani, and Dr. B. P. Doctor for the gift of DEPQ. This work is part of a Slovenian–Croatian joint project supported by the Ministries of Science and Technology of the Republic of Slovenia and Republic of Croatia.

REFERENCES 1. Massoulie, J., Pezzementi, L., Bon, S., Krejci, E., and Vallette, F. M. (1993) Prog. Neurobiol. 41, 31–91. 2. Sussman, J. L., Harel, M., Frolow, F., Oefner, C., Goldman, A., Toker, L., and Silman, I. (1991) Science 253, 872– 879. 3. Bourne, Y., Taylor, P., and Marchot, P. (1995) Cell 83, 503– 512. 4. Harel, M., Sussman, J. L., Krejci, E., Bon, S., Chanal, P., Massoulie, J., and Silman, I. (1992) Proc. Natl. Acad. Sci. USA 89, 10827–10831. 5. Saxena, A., Redman, A. M. G., Jiang, X., Lockridge, O., and Doctor, B. P. (1997) Biochemistry 36, 14642–14651. 6. Saxena, A., Redman, A. M. G., Jiang, X., Lockridge, O., and Doctor, B. P. (1999) Chem.–Biol. Interact. 119–120, 61– 69. 7. Radic´, Z., Pickering, N. A, Vellom, D. C., Camp, S., and Taylor, P. (1993) Biochemistry 32, 12074 –12084. 8. Marcel, V., Palacois, L. G., Pertuy, C., Masson, P., and Fournier, F. (1998) Biochem. J. 329, 329 –334. 9. Usdin, E. (1963) in Anticholinesterase Agents (Karcsmar, A. G., Ed.), pp. 61, 63, 142, 162, Pergamon Press, Oxford. 10. Zhou, H. X., Wlodek, S. T., and McCammon, J. A. (1998) Proc. Natl. Acad. Sci. USA 95, 9280 –9283. 11. Antosiewicz, J., Gilson, M. K., Lee, I. H., and McCammon, J. A. (1995) Biophys. J. 68, 62– 68. 12. Stojan, J., Marcel, V., Estrada-Mondaca, S., Klaebe, A., Masson, P., and Fournier, D. (1998) FEBS Lett. 440, 85– 88. 13. Cleland, W. W. (1977) Adv. Enzymol. 45, 273–277. 14. Taylor, P., and Lappi, S. (1975) Biochemistry 14, 1989 –1997. 15. Szegletes, T., Mallender, W. D., and Rosenberry, T. L. (1998) Biochemistry 37, 4206 – 4216. 16. Botti, S. A., Felder, C. E., Lifson, S., Sussman, J. L., and Silman, I. (1999) Biophys. J. 77, 2430 –2450. 17. Masson, P., Froment, M. T., Bartels, C. F., and Lockridge, O. (1996) Eur. J. Biochem. 235, 36 – 48. 18. Reiner, E., Sˇkrinjaric´-Sˇpoljar, M., and Simeon-Rudolf, V. (1996) Period. Biol. 98, 325–329. 19. Reiner, E., and Simeon-Rudolf, V. (2000) Eur. J. Physiol. 440(Suppl.), 118 –120. 20. Golicˇnik, M., Fournier, D., and Stojan, J. (2001) Biochemistry 40, 1214 –1219. 21. Simeon-Rudolf, V., Sˇinko, G., Sˇtuglin A., and Reiner, E. (2001) Croat. Chem. Acta 74, 173–182.

INTERACTION OF ETHOPROPAZINE WITH HORSE SERUM BUTYRYLCHOLINESTERASE 22. Ellman, G. L., Courtney, K. D., Andres, V., and Feathersone, R. M. (1961) Biochem. Pharmacol. 7, 88 –95. 23. Luo, C., Saxena, A., Ashani, Y., Leader, H., Radic´, Z., Taylor, P., and Doctor, B. P. (1999) Chem.–Biol. Interact. 119–120, 129 –135. 24. Vriend, G. (1990) WHAT IF: A molecular modeling and drug design program. J. Mol. Graph. 8, 52–56. 25. Brooks, B. R., Bruccoleri, R. E., Olafson, B. D., States, D. J., Swaminathan, S., and Karplus, M. (1983) Comput. Chem. 4, 187–217.

31

26. Harel, M., Kryger, G., Rosenbery, T. L., Mallender, W. D., Lewis, T., Fletcher, R. J., Guss, J. M., Silman, I., and Sussman, J. L. (2000) Prot. Sci. 9, 1063–1072. 27. Cha, S. (1968) J. Biol. Chem. 243, 820 – 825. 28. Stojan, J. (1997) J. Chem. Inf. Comput. Sci. 37, 1025–1027. 29. Gibbson, Q. (1983) in Modern Physical Methods in Biochemistry, Part B (Neuberger, A., and Van Deenen, L. L. M., Eds.), pp. 65– 84, Elsevier Science, Amsterdam. 30. Cleland, W. W. (1963) Biochim. Biophys. Acta 67, 104 –137.