Investigation of the binding mode of (−)-meptazinol and bis-meptazinol derivatives on acetylcholinesterase using a molecular docking method

J Mol model(2006)12:390-397 DOI10.1007/s008940050058y ORIGINAL PAPER Qiong Xie. Yun Tang. Wei Li Xing-Hai Wang Zhui-Bai Qiu Investigation of the binding mode of (-)-meptazinol and bis-meptazinol derivatives on acetylcholinesterase using a molecular docking method Received: 8 December 2004/ Accepted: 9 August 2005/ Published online: 11 January 2006 c) Springer-Verlag 2006 Abstract Molecular docking has been performed to Introduction investigate the binding mode of (-)-meptazinol (MEP) with acetylcholinesterase(AChE) and to screen bis- Alzheimers disease(AD)is a neurodegenerative disorder meptazinol(bis- MEP)derivatives for preferable synthetic that seriously threatens the health of elderly people around candidates virtually. A reliable and practical docking the world. Although with an unclear pathogenesis, AD is method for investigation of AChE ligands was established commonly believed to be associated with the dysfunction by the comparison of two widely used docking programs, of the central cholinergic system [1, 2]. Acetylcholinester- FlexX and GOLD. In our hands, we had more luck using ase (AChE) plays a key role in the regulation of the GOLd than FlexX in reproducing the experimental poses cholinergic system, and hence, inhibition of AChE has of known ligands(RMSD<1.5 A). GOLD fitness values of emerged as one of the most promising strategies for the known ligands were also in good agreement with their treatment of AD. So far, four AChE inhibitors have been activities In the present GOLd docking protocol, (-)-MEP approved by the FDa in the US for treatment of AD, seemed to bind with the enzyme catalytic site in an open- namely tacrine (THA)[3], donepezil (E2020)[4],rivas- gate conformation through strong hydrophobic interactions tigmine [5] and galanthamine(GNT)[6]. In China(-)- and a hydrogen bond. Virtual screening of a potential Huperzine A(HUPA)[7] was also approved for the same candidate compound library suggested that the most purpose. Due to drawbacks such as hepatotoxicity and low deni rising 15 bis-mEP derivatives on the list were mainly selectivity as in the case of tacrine [8], the search for new derived from(-)-MEP with conformations of(,S)and AChE inhibitors with high selectivity and low toxicity SR, RS)and with a 2-to 7-carbon linkage. Although there remains an urgent task are still no biological results to confirm the predictive Meptazinol (MEP, Fig. 1)is a potent opioid analgesic but power of this method, the current study could provide an with less respiratory depression and low addiction liability altermate tool for structural optimization of()-MEP as new [9], possibly due to a component of cholinergic activation AChE inhibitors presented in its pharmacological profile [10]. Early in 1986 Ennis et al. reported the inhibition of AChe by meptazinol Keywords Meptazinol (MEP) Molecular docking [ll], which not only confirmed MEP's involvement in the GOLD Acetylcholinesterase(AChE) cholinergic system, but also opened a door in the search for new AChE inhibitors. Ennis'experiments demonstrated that (+)-MEP acted on AChE in vitro with an ICso of 6. 4 uM, while the (r-enantiomer (IC50=3. 3 HM) was far more potent than its antipole (+)-enantiomer(ICso>1.0 mM). The Q.Xie.Y.Tang(凶)·W.Li·X.H.Wang:Z.B.Qu() inhibitory potency of(-)-MEp was almost equivalent to that School of Pharmacy, Fudan University of ()-galanthamine and tacrine, about 100 times less potent hanghai, 200032, People's Republic of china than physostigmine [11]. Recently, we determined the 54237419 absolute configurations of ()and (+)-MEP as S and R, respectively, by X-ray crystal structures(Fig. 1)[12] Tel:+86-21-54237595 Therefore, it was of considerable interest to investigate the mechanism of action of (-)-MEP on AChE for the search of Y Tang() hool of Pharmacy new mep derivatives as achE inhibitors East China University of Science and Technology, Molecular docking is an efficient tool for investigating Shanghai, 200237, People's Republic of China receptor-ligand interactions and for virtual screening Tel:+86-21-64251052 which plays a key role in rational drug design [13

J Mol Model (2006) 12: 390–397 DOI 10.1007/s00894-005-0058-y OR IG INAL PAPER Qiong Xie . Yun Tang . Wei Li . Xing-Hai Wang . Zhui-Bai Qiu Investigation of the binding mode of (−)-meptazinol and bis-meptazinol derivatives on acetylcholinesterase using a molecular docking method Received: 8 December 2004 / Accepted: 9 August 2005 / Published online: 11 January 2006 # Springer-Verlag 2006 Abstract Molecular docking has been performed to investigate the binding mode of (−)-meptazinol (MEP) with acetylcholinesterase (AChE) and to screen bis￾meptazinol (bis-MEP) derivatives for preferable synthetic candidates virtually. A reliable and practical docking method for investigation of AChE ligands was established by the comparison of two widely used docking programs, FlexX and GOLD. In our hands, we had more luck using GOLD than FlexX in reproducing the experimental poses of known ligands (RMSD<1.5 Å). GOLD fitness values of known ligands were also in good agreement with their activities. In the present GOLD docking protocol, (−)-MEP seemed to bind with the enzyme catalytic site in an open￾gate conformation through strong hydrophobic interactions and a hydrogen bond. Virtual screening of a potential candidate compound library suggested that the most promising 15 bis-MEP derivatives on the list were mainly derived from (−)-MEP with conformations of (S,S) and (SR,RS) and with a 2- to 7-carbon linkage. Although there are still no biological results to confirm the predictive power of this method, the current study could provide an alternate tool for structural optimization of (−)-MEP as new AChE inhibitors. Keywords Meptazinol (MEP) . Molecular docking . GOLD . Acetylcholinesterase (AChE) Introduction Alzheimer’s disease (AD) is a neurodegenerative disorder that seriously threatens the health of elderly people around the world. Although with an unclear pathogenesis, AD is commonly believed to be associated with the dysfunction of the central cholinergic system [1, 2]. Acetylcholinester￾ase (AChE) plays a key role in the regulation of the cholinergic system, and hence, inhibition of AChE has emerged as one of the most promising strategies for the treatment of AD. So far, four AChE inhibitors have been approved by the FDA in the US for treatment of AD, namely tacrine (THA) [3], donepezil (E2020) [4], rivas￾tigmine [5] and galanthamine (GNT) [6]. In China (−)- Huperzine A (HUPA) [7] was also approved for the same purpose. Due to drawbacks such as hepatotoxicity and low selectivity as in the case of tacrine [8], the search for new AChE inhibitors with high selectivity and low toxicity remains an urgent task. Meptazinol (MEP, Fig. 1) is a potent opioid analgesic but with less respiratory depression and low addiction liability [9], possibly due to a component of cholinergic activation presented in its pharmacological profile [10]. Early in 1986 Ennis et al. reported the inhibition of AChE by meptazinol [11], which not only confirmed MEP’s involvement in the cholinergic system, but also opened a door in the search for new AChE inhibitors. Ennis’ experiments demonstrated that (±)-MEP acted on AChE in vitro with an IC50 of 6.4 μM, while the (−)-enantiomer (IC50=3.3 μM) was far more potent than its antipole (+)-enantiomer (IC50>1.0 mM). The inhibitory potency of (−)-MEP was almost equivalent to that of (−)-galanthamine and tacrine, about 100 times less potent than physostigmine [11]. Recently, we determined the absolute configurations of (−) and (+)-MEP as S and R, respectively, by X-ray crystal structures (Fig. 1) [12]. Therefore, it was of considerable interest to investigate the mechanism of action of (−)-MEP on AChE for the search of new MEP derivatives as AChE inhibitors. Molecular docking is an efficient tool for investigating receptor-ligand interactions and for virtual screening, which plays a key role in rational drug design [13, 14], Q. Xie . Y. Tang (*) . W. Li . X.-H. Wang . Z.-B. Qiu (*) School of Pharmacy, Fudan University, Shanghai, 200032, People’s Republic of China e-mail: ytang234@yahoo.com.cn Tel.: +86-21-54237419 e-mail: zbqiu@shmu.edu.cn Tel.: +86-21-54237595 Y. Tang (*) School of Pharmacy, East China University of Science and Technology, Shanghai, 200237, People’s Republic of China Tel.: +86-21-64251052

oh derivatives of both enantiomers connected by spacers length two to 14 carbons at substituents in NI-position(D) or C4-amino group (II)of the azepine ring. Bis-MEP derivatives were believed to make dual action, allowing interactions with both the ester-cleavage peripheral anionic site(PAS)of AChE Fig. 1 The structures of (=)-MEP, ()-MEP, and(+)-MEP Materials and methods especially when the crystal structure of a receptor or such as Auto Dock [15], DOCK [16], FlexX [17], Glide workstation. Molecular performed on a R14000 SGI Fuel [18] and GOLD [19], treat the ligand with full flexibility carried out with the software package SYBYL version 6.9 but assume the receptor to be rigid or apply very limited [32]. The programs GOLD [33] and CORINA [34] were flexibility to side chains. Various searching algorithms used for gOld docking and 2 d to 3 D structural conver such as fast shape matching(DOCK), simulated annealing sion, respectively. Standard parameters were used unless AutoDock), incremental construction(FlexX), genetic otherwise indicated algorithms(GOLD), Monte Carlo(Ligand-Fit, and Tabu earch, have been employed in docking studies [20] Scoring functions are usually categorized as force-field Preparation of proteins based methods(such as DOCK and GOLD), empirical free-energy scoring functions(such as FlexX), and knowl- Three crystal structures of Torpedo californica AChE edge-based scoring functions [20](such as PMF [21]).(TcAChE)complexed with E2020 [25], HUPA [29]and Among various docking algorithms, FlexX and GOLd are THA [31] were retrieved from PDB [24] with correspond the two most reliable methods in complex validation tests ing entry code IEVE, IVOT and lACJ. The proteins were [22] prepared by removing heteroatoms and water molecules Thusly, in the present study we first explored the and adding all hydrogen atoms. The active site of lEVE feasibility of FlexX and GOLD on AChE with seven was defined as residues with at least one atom within a known AChE inhibitors whose complex structures with radius of 10 A from any atom of E2020. For IVOT and AChE [23]are available from the Protein Data Bank(PDB) lAC], a 12 A radius around the corresponding ligand was 24], namely donepezil(E2020)[25], galanthamine(Gnt) defined. The active sites were saved as PDb files for FlexX [26], BW284c51[27], edrophonium(EDR)[28], huperzine docking and MOL2 files for GOLd docking A(HUPA)[29], huprine X(HUPX)[30], tacrine(THA) [ 31](Fig. 2). The better of the two programs was then used as the docking protocol to investigate the binding mode of Preparation of known ligands for pose reproduction ()-MEP on AChE. In addition, a small focused library with 91 bis-MEP derivatives(Fig. 3)was built and The seven known AChE inhibitors were extracted from the careened virtually using the specified docking method to corresponding complex structures with AChE, whose PDB give preferable candidates for synthesis. The compounds codes are lEVE, IQTl, lE3Q, 2ACK, IVoT, 1E66, and among the screening library included the bis-ligand lACJ, respectively. Atom types and bond types were OH BW284c51 E2020 EDR HUPX THA Fig 2 Structures of seven known AChE inhibitors

especially when the crystal structure of a receptor or enzyme is available. Almost all current docking programs, such as AutoDock [15], DOCK [16], FlexX [17], Glide [18] and GOLD [19], treat the ligand with full flexibility but assume the receptor to be rigid or apply very limited flexibility to side chains. Various searching algorithms such as fast shape matching (DOCK), simulated annealing (AutoDock), incremental construction (FlexX), genetic algorithms (GOLD), Monte Carlo (Ligand-Fit), and Tabu search, have been employed in docking studies [20]. Scoring functions are usually categorized as force-field based methods (such as DOCK and GOLD), empirical free-energy scoring functions (such as FlexX), and knowl￾edge-based scoring functions [20] (such as PMF [21]). Among various docking algorithms, FlexX and GOLD are the two most reliable methods in complex validation tests [22]. Thusly, in the present study we first explored the feasibility of FlexX and GOLD on AChE with seven known AChE inhibitors whose complex structures with AChE [23] are available from the Protein Data Bank (PDB) [24], namely donepezil (E2020) [25], galanthamine (GNT) [26], BW284c51 [27], edrophonium (EDR) [28], huperzine A (HUPA) [29], huprine X (HUPX) [30], tacrine (THA) [31] (Fig. 2). The better of the two programs was then used as the docking protocol to investigate the binding mode of (−)-MEP on AChE. In addition, a small focused library with 91 bis-MEP derivatives (Fig. 3) was built and screened virtually using the specified docking method to give preferable candidates for synthesis. The compounds among the screening library included the bis-ligand derivatives of both enantiomers connected by spacers of length two to 14 carbons at substituents in N1- position (I) or C4- amino group (II) of the azepine ring. Bis-MEP derivatives were believed to make dual action, allowing interactions with both the ester-cleavage site and the peripheral anionic site (PAS) of AChE. Materials and methods The study was mainly performed on a R14000 SGI Fuel workstation. Molecular modeling and FlexX docking were carried out with the software package SYBYL version 6.9 [32]. The programs GOLD [33] and CORINA [34] were used for GOLD docking and 2 D to 3 D structural conver￾sion, respectively. Standard parameters were used unless otherwise indicated. Preparation of proteins Three crystal structures of Torpedo californica AChE (TcAChE) complexed with E2020 [25], HUPA [29] and THA [31] were retrieved from PDB [24] with correspond￾ing entry code 1EVE, 1VOT and 1ACJ. The proteins were prepared by removing heteroatoms and water molecules and adding all hydrogen atoms. The active site of 1EVE was defined as residues with at least one atom within a radius of 10 Å from any atom of E2020. For 1VOT and 1ACJ, a 12 Å radius around the corresponding ligand was defined. The active sites were saved as PDB files for FlexX docking and MOL2 files for GOLD docking. Preparation of known ligands for pose reproduction The seven known AChE inhibitors were extracted from the corresponding complex structures with AChE, whose PDB codes are 1EVE, 1QTI, 1E3Q, 2ACK, 1VOT, 1E66, and 1ACJ, respectively. Atom types and bond types were OH N CH3 * OH N CH3 OH N CH3 (±)-MEP (-)-S-MEP (+)-R-MEP Fig. 1 The structures of (±)-MEP, (−)-MEP, and (+)-MEP O N N O N O O N OH O O N OH NH H O 2N Cl N NH2 N NH2 BW284c51 E2020 GNT EDR HUPA THA HUPX Fig. 2 Structures of seven known AChE inhibitors 391

Fig. 3 Structures of designed bis-meP derivatives ( n=2-14 2-14 n=2-14 corrected manually. Hydrogen atoms and Gasteiger-Marsili GOLd docking [35] atomic charges were then added. The structures were minimized in SYBYL for 100 steps with the Tripos force The default parameter settings for three times speed-up field and saved to a multi-MOL2 file were used, and no flipping was allowed In docking runs for pose reproduction, all ten solutions of each ligand were saved for further analysis. However, for virtual screening, Preparation of focused library for virtual screening only the best solution was kept for each molecule. Bis-MEP derivatives with spacer length of two to 14 arbons. a total of 91 molecules were drawn in ISIS/Base Results and discussion with ISIS/Draw from MDL [36] and exported into a 2 D structure data file(SDF)to form a small focused library. Comparison of prediction accuracy of Flexx The 2 D structures of the library were subsequently con- and GOLD verted into 3 D-structures with CORINA [37, still in SDF format. The 3 D structures were then read into SYBYL The accurate prediction of protein-ligand interaction geo- olecular SpreadSheet table for further treatment, such as metries is essential for the success of virtual-screening energy minimization for 100 steps with Tripos force field approaches in structure-based drug design. It requires and Gasteiger-Marsili [35 charges for each molecule. docking tools that are able to generate suitable conforma These structures were put into databases and then written to tions of a ligand within a protein binding site and reliable MOL2 files as input energetic evaluation indicating the quality of the interaction. Flexx [17] uses an incremental construction algorithm where ligands are docked starting with a base fragment Flexx docking The complete ligand is subsequently constructed by adding e remainin omponents. The best solution is selected All Flexx runs used the version 1. ll.I embedded within according to the docking score, using a pure empirical SYBYL. Formal charges were assigned and the Flexx scoring function by Bohm or a knowledge-based scorin scoring function was chosen to evaluate the docking poses. function Drug Score. The GolD [19]( Genetic Optimiza- Table 1. a Fitness scores, RMSD from crystal structures and ICso of known ligands by gold IEVE IVOT IACJ ICso(nM) MOL NAME Fitness RMSD(A) Fitness RMSD(A) Fitness RMSD (A) E2020 66.15 0.5437 48.64 3.3821 46.45 3.7311 13.6 GNT 5753 0.7819 4.8748 4.2323 19954 Half-open BW284c51 80.28 3.5743 85.36 4.1937 63423.6457 0.0036 EDR 44.09 4.2878 43.93 46.02 4.1636 46.35 3.6063 1.21 51550.528 58.4 HUPX 5994 5.3004 59.79 5.0281 0.9428 0.026(K1) THA 45.13 5.8934 49.265.638 50.47 5.6056 From Reference [47] bFrom Reference [48] From Reference [49] dRank 2 scored 50.19 with an rMsD of 0.5404

corrected manually. Hydrogen atoms and Gasteiger-Marsili [35] atomic charges were then added. The structures were minimized in SYBYL for 100 steps with the Tripos force field and saved to a multi-MOL2 file. Preparation of focused library for virtual screening Bis-MEP derivatives with spacer length of two to 14 carbons, a total of 91 molecules, were drawn in ISIS/Base with ISIS/Draw from MDL [36] and exported into a 2 D structure data file (SDF) to form a small focused library. The 2 D structures of the library were subsequently con￾verted into 3 D-structures with CORINA [37], still in SDF format. The 3 D structures were then read into SYBYL Molecular SpreadSheet table for further treatment, such as energy minimization for 100 steps with Tripos force field and Gasteiger-Marsili [35] charges for each molecule. These structures were put into databases and then written to MOL2 files as input. FlexX docking All FlexX runs used the version 1.11.1 embedded within SYBYL. Formal charges were assigned and the FlexX scoring function was chosen to evaluate the docking poses. GOLD docking The default parameter settings for three times speed-up were used, and no flipping was allowed. In docking runs for pose reproduction, all ten solutions of each ligand were saved for further analysis. However, for virtual screening, only the best solution was kept for each molecule. Results and discussion Comparison of prediction accuracy of FlexX and GOLD The accurate prediction of protein-ligand interaction geo￾metries is essential for the success of virtual-screening approaches in structure-based drug design. It requires docking tools that are able to generate suitable conforma￾tions of a ligand within a protein binding site and reliable energetic evaluation indicating the quality of the interaction. FlexX [17] uses an incremental construction algorithm where ligands are docked starting with a base fragment. The complete ligand is subsequently constructed by adding the remaining components. The best solution is selected according to the docking score, using a pure empirical scoring function by Böhm or a knowledge-based scoring function DrugScore. The GOLD [19] (Genetic Optimiza￾OH N * OH N * H2 C n (3',3'') n (S,S) n=2~14 (R,S) n=2~14 (R,R) n=2~14 OH N * OH N * * * H N H N H2 C n (3'4',4''3'') n (SR,RS) n=2~14 (SS,SS) n=2~14 (RR,RR) n=2~14 (RS,SR) n=2~14 3' 3'' 3' 4' 3'' 4'' · · · · Fig. 3 Structures of designed bis-MEP derivatives Table 1.a Fitness scores, RMSD from crystal structures and IC50 of known ligands by GOLD 1EVE 1VOT 1ACJ IC50 (nM) MOL_NAME Fitness RMSD (Å) Fitness RMSD (Å) Fitness RMSD (Å) Open E2020 66.15 0.5437 48.64 3.3821 46.45 3.7311 13.6a GNT 57.53 0.7819 54.82 4.8748 53.29 4.2323 1995a Half- open BW284c51 80.28 3.5743 85.36 4.1937 63.42 3.6457 0.0036b EDR 44.09 4.2878 43.93 1.2034 46.02 4.1636 240b HUPA 46.35 3.6063 50.78 1.2146 51.55 0.528 58.4a Closed HUPX 59.94 5.3004 59.79 5.0281 70.48 0.9428 0.026 (Ki)c THA 45.13 5.8934 49.26 5.638 50.47 5.6056d 93a a From Reference [47], b From Reference [48], c From Reference [49], d Rank 2 scored 50.19 with an RMSD of 0.5404 392

Table 1. b. Flexx scores and rMSd from crystal structures and ligands by Flexx MOL NAME EVE F Score RMSD(A) F Score RMSD(A) RMSD(A) E2020 17545 -114 9.6344 -9.7 13.3023 GNT 0.5628 113 154656 0.7694 Half-Oper BW284c51 -8.9 4.6601 3.764 EDR -16.5 1.2315 HUPA -12.2 134537 -18.3 3.5703 -117 184646 Closed HUPX -14 11.8 15.3927 11.3687 THA -16.3 174666 1.854 13.5 14.7315 tion for Ligand Docking) program uses a genetic algorithm plexes, namely the complexes of E2020(IEVE)[25], GNT to explore the full range of ligand conformational flexibil-(1QTD)[26], BW284c51(1E3Q),[27] EDR(2ACK),[28] ity and the rotational flexibility of selected receptor HUPA(IVOT), [29] HUPX (1E66), [30] and THA (IACd hydrogen atoms. The mechanism for ligand placement is [31]. Among them, the first two complexes show open-gate based on fitting points, which are added to hydrogen- conformations, whereas the last two form closed-gateway onding or hydrophobic groups on both protein and ligand. binding styles. And half-open gate structures are found for And then the points between acceptors and donors map the middle three complexes. In the present study, we each other. The docking poses are ranked based on a performed both FlexX and goLd docking protocols on nolecular mechanics-like scoring function, named fitness seven known AChE inhibitors. Taking three dominant function. These two protocols are verified as two of the orientations of the Phe330 side chain(open, half-open, and most reliable methods in complex validation tests among closed gate)into account, we have considered all three various docking algorithms [22] possibilities in separate docking runs for each of the seven From a comparison of the published X-ray crystal ligands structures of various TcAChE complexes, it is obvious that Scoring calculations showed that six of the seven the side-chain orientation of Phe330 is the only site with ligands'optimal poses determined by GOLD were very flexibility, which is responsible for substrate trafficking close to the original orientation found in the crystal. The down the gorge. Three major orientations of Phe330 have root mean square deviation(RMSD) for all heavy atoms been observed. The TCcAChE-E2020 complex (IEVE)is between the docked and crystal ligand coordinates was characterized by the open-gate conformation, while the below 1.5 A with the exception of BW284c51, when the complex with THA (lACD) displays a closed one, and the actual gate conformations of AChE were considered half-open conformation is observed in the TcAChE-HUPA(highlighted in Table 1. a). And the fitness scores concluded cOmplex(IVOT)[38]. Since is still difficult to deal with from the actual gate conformations conformed to the docking methods, the open, half-open and closed gate of inhibitors was, the higher their fitness score showed For conformations of AChe were investigated separately in the case of BW284c51, molecular complexity and flexi this study to determine the importance of the side-chain bility might play a part in the failure of binding pose flexibility of Phe330 for trafficking prediction. It is worth noting that Flexx performs better To explore the feasibility of the two methods Flexx and than GOLD in many cases. However, in the present OLD, we compared their predictive power of reproduc- exercise, it seemed that Flexx performed less well than ing the binding poses of the known seven AChE com- GOLD (Table 1. b) Table 2 Fitness score and evaluation of the interactions of(-)MEP and(+)-MEP with TcAChE EVE IVOT lACE (-)MEP (+)-MEP ()-MEP (+)-MEP ()-MEP (+)-MEP Fitness score 47.37450244.95 45.7348.34 H-bond interactions Donor Ph-OH Trp84-NH Ph-OH, Ser122-OH Acceptor Glu199-0 None Ph-o Trp84-CO, Ph-0 None Asp72-CO0 Distance 2.57 2.681,2.470 2.054 Hydrophobic interactions(distance/A)Ar-Trp84.320 Ar - Phe330 4.086 3.958 Az-Phe3304.003 Ar-aromatic cycle of MEP bAz--azepine cycle of MEP

tion for Ligand Docking) program uses a genetic algorithm to explore the full range of ligand conformational flexibil￾ity and the rotational flexibility of selected receptor hydrogen atoms. The mechanism for ligand placement is based on fitting points, which are added to hydrogen￾bonding or hydrophobic groups on both protein and ligand. And then the points between acceptors and donors map each other. The docking poses are ranked based on a molecular mechanics-like scoring function, named fitness function. These two protocols are verified as two of the most reliable methods in complex validation tests among various docking algorithms [22]. From a comparison of the published X-ray crystal structures of various TcAChE complexes, it is obvious that the side-chain orientation of Phe330 is the only site with flexibility, which is responsible for substrate trafficking down the gorge. Three major orientations of Phe330 have been observed. The TcAChE-E2020 complex (1EVE) is characterized by the open-gate conformation, while the complex with THA (1ACJ) displays a closed one, and the half-open conformation is observed in the TcAChE-HUPA complex (1VOT) [38]. Since is still difficult to deal with flexibility of the protein thoroughly in current molecular docking methods, the open, half-open and closed gate conformations of AChE were investigated separately in this study to determine the importance of the side-chain flexibility of Phe330 for trafficking. To explore the feasibility of the two methods FlexX and GOLD, we compared their predictive power of reproduc￾ing the binding poses of the known seven AChE com￾plexes, namely the complexes of E2020 (1EVE) [25], GNT (1QTI) [26], BW284c51 (1E3Q), [27] EDR (2ACK), [28] HUPA (1VOT), [29] HUPX (1E66), [30] and THA (1ACJ) [31]. Among them, the first two complexes show open-gate conformations, whereas the last two form closed-gateway binding styles. And half-open gate structures are found for the middle three complexes. In the present study, we performed both FlexX and GOLD docking protocols on seven known AChE inhibitors. Taking three dominant orientations of the Phe330 side chain (open, half-open, and closed gate) into account, we have considered all three possibilities in separate docking runs for each of the seven ligands. Scoring calculations showed that six of the seven ligands’ optimal poses determined by GOLD were very close to the original orientation found in the crystal. The root mean square deviation (RMSD) for all heavy atoms between the docked and crystal ligand coordinates was below 1.5 Å with the exception of BW284c51, when the actual gate conformations of AChE were considered (highlighted in Table 1.a). And the fitness scores concluded from the actual gate conformations conformed to the inhibition activities as well (Table 1.a). The lower the IC50 of inhibitors was, the higher their fitness score showed. For the case of BW284c51, molecular complexity and flexi￾bility might play a part in the failure of binding pose prediction. It is worth noting that FlexX performs better than GOLD in many cases. However, in the present exercise, it seemed that FlexX performed less well than GOLD (Table 1.b). Table 1.b. FlexX scores and RMSD from crystal structures and ligands by FlexX MOL_NAME 1EVE 1VOT 1ACJ F_Score RMSD (Å) F_Score RMSD (Å) F_Score RMSD (Å) Open E2020 −12 17.545 −11.4 9.6344 −9.7 13.3023 GNT −27 0.5628 −11.3 15.4656 −24.7 0.7694 Half-Open BW284c51 −8.9 7.049 −14.3 4.6601 −8.7 3.764 EDR −16.5 3.1461 −18 1.2315 −15.2 0.7693 HUPA −12.2 13.4537 −18.3 3.5703 −11.7 18.4646 Closed HUPX −14 16.688 −11.8 15.3927 −13.9 11.3687 THA −16.3 17.4666 −12.8 11.854 −13.5 14.7315 Table 2 Fitness score and evaluation of the interactions of (−)-MEP and (+)-MEP with TcAChE Index 1EVE 1VOT 1ACJ (−)-MEP (+)-MEP (−)-MEP (+)-MEP (−)-MEP (+)-MEP Fitness score 47.37 45.02 44.95 44.34 45.73 48.34 H-bond interactions Donor Ph-OH – Trp84-NH Ph-OH, Ser122-OH – Ph-OH Acceptor Glu199-O None Ph-O Trp84-CO, Ph-O None Asp72-COO Distance/Å 1.953 – 2.557 2.681, 2.470 – 2.054 Hydrophobic interactions (distance/Å) Ara - Trp84 4.320 – 4.045 – – 4.720 Ar - Phe330 – 3.519 4.118 – – 4.107 Azb -Trp84 – 4.086 – 3.958 3.777 – Az - Phe330 4.003 –– – 4.281 – a Ar—aromatic cycle of MEP b Az—azepine cycle of MEP 393

Fig 4 Best docking solution of MEP into the open pocket (EVE): a(-MEP (leff); b(+)- MEP (right) Phe330 Phe330 The low RMSD and the consistency of calculated site consisting of some aromatic residues such as Trp84 and affinity with experimental activity indicated that the gold Phe330 method and parameter set were reasonable to reproduce the In order to explore the mechanism of action of MEP X-ray structure and could be extended to search and isomers on AChE and explain the reason of their differ- evaluate the binding poses of other AChE ds ences in activity, which is valuable for structure-based accordingly design of MEP derivatives with desired pharmacological On the other hand, comparison of the results of a specific properties, both S and R isomers of MEP were considered ligand from three conformations of AChe demonstrated for GOLD docking runs. ()-MEP and (+)-MEP were that the highest scoring pose was derived from the actual submitted to GOLd separately with the three gate gate conformation and indicated the lowest RMSD. Thus, conformations of AChE to consider the flexibility of the the experimental gate conformation of AChE complexed binding site of AChE with an unknown ligand could be predicted roughly by Fitness scores of the best solutions and detailed inter comparing the fitness score concluded from three gate action indices of both isomers are listed in Table 2.The conformations of achE hydrogen-bond (HB)interaction was evaluated by the dis tance between HB donor and acceptor atoms. The hydro phobic interaction was estimated according to the distance Predicted binding conformation of(-)-MEP between the centroid of aromatic or azepine cycle of MEP with achE and the centroid of the side train of the hydrophobic residue Trp84 or Phe330. Figs. 4, 5, 6 illustrated the binding maps The X-ray structure of AChE illustrated that the active site of (-)-MEP and (+)-MEP with the binding site of three of the enzyme was a deep and narrow gorge with such TcAChE conformations features [39]: a catalytic triad composed of residues The highest score of ()-MEP was found for the open Ser200, Glu327 and His440; a peripheral anionic site at pocket(IEVE), while that of (+)-MEP was found for the the entry centered by residue Trp279; and a hydrophobic closed one (lAC)(Table 2). Although the actual gate 5 Best docking solution of into the half-o T): a()-MEP (lefn); b(+)- Phes 30 Phe330

The low RMSD and the consistency of calculated affinity with experimental activity indicated that the GOLD method and parameter set were reasonable to reproduce the X-ray structure and could be extended to search and evaluate the binding poses of other AChE ligands accordingly. On the other hand, comparison of the results of a specific ligand from three conformations of AChE demonstrated that the highest scoring pose was derived from the actual gate conformation and indicated the lowest RMSD. Thus, the experimental gate conformation of AChE complexed with an unknown ligand could be predicted roughly by comparing the fitness score concluded from three gate conformations of AChE. Predicted binding conformation of (−)-MEP with AChE The X-ray structure of AChE illustrated that the active site of the enzyme was a deep and narrow gorge with such features [39]: a catalytic triad composed of residues Ser200, Glu327 and His440; a peripheral anionic site at the entry centered by residue Trp279; and a hydrophobic site consisting of some aromatic residues such as Trp84 and Phe330. In order to explore the mechanism of action of MEP isomers on AChE and explain the reason of their differ￾ences in activity, which is valuable for structure-based design of MEP derivatives with desired pharmacological properties, both S and R isomers of MEP were considered for GOLD docking runs. (−)-MEP and (+)-MEP were submitted to GOLD separately with the three gate conformations of AChE to consider the flexibility of the binding site of AChE. Fitness scores of the best solutions and detailed inter￾action indices of both isomers are listed in Table 2. The hydrogen-bond (HB) interaction was evaluated by the dis￾tance between HB donor and acceptor atoms. The hydro￾phobic interaction was estimated according to the distance between the centroid of aromatic or azepine cycle of MEP and the centroid of the side train of the hydrophobic residue Trp84 or Phe330. Figs. 4, 5, 6 illustrated the binding maps of (−)-MEP and (+)-MEP with the binding site of three TcAChE conformations. The highest score of (−)-MEP was found for the open pocket (1EVE), while that of (+)-MEP was found for the closed one (1ACJ) (Table 2). Although the actual gate Fig. 4 Best docking solution of MEP into the open pocket (1EVE): a (−)-MEP (left); b (+)- MEP (right) Fig. 5 Best docking solution of MEP into the half-open pocket (1VOT): a (−)-MEP (left); b (+)- MEP (right) 394