Assessing the binding of cholinesterase inhibitors by docking and molecular dynamics studies

Accepted Manuscript Title: Assessing the Binding of Cholinesterase Inhibitors by Docking and Molecular Dynamics Studies Authors: M.Rejwan Ali, Mostafa Sadoqi, SGeir Moller, Allal Boutajangout, Mihaly Mezei PII: DOI: Reference:

S1093-3263(16)30403-X http://dx.doi.org/doi:10.1016/j.jmgm.2017.06.027 JMG 6958

To appear in:

Journal of Molecular Graphics and Modelling

Received date: Revised date: Accepted date:

15-11-2016 24-6-2017 26-6-2017

Please cite this article as: M.Rejwan Ali, Mostafa Sadoqi, SGeir Moller, Allal Boutajangout, Mihaly Mezei, Assessing the Binding of Cholinesterase Inhibitors by Docking and Molecular Dynamics Studies, Journal of Molecular Graphics and Modellinghttp://dx.doi.org/10.1016/j.jmgm.2017.06.027 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Assessing the Binding of Cholinesterase Inhibitors by Docking and Molecular Dynamics Studies M. Rejwan Ali (1), Mostafa Sadoqi (1), S Geir Moller (2), Allal Boutajangout (3) Mihaly Mezei (4) AUTHOR ADDRESS (1) Department of Physics, St John’s University (2) Department of Biological Sciences, St John’s University (3) Department of Neurology and Neuroscience & Physiology and Psychiatry, New York University Langone Medical Center (4) Department of Pharmacological Sciences, Icahn School of Medicine at Mount Sinai, New York Corresponding Author: M. Rejwan Ali, Email Telephone <718-990-6437> Fax <718-990-5812>

HIGHLIGHT In this report we assessed by docking and molecular dynamics the binding mechanisms of three FDA-approved Alzheimer drugs, inhibitors of the enzyme acetylcholinesterase (AChE): donepezil, galantamine and rivastigmine. Dockings by the softwares AutodockVina, PatchDock and Plant reproduced the docked conformations of the inhibitor-enzyme complexes within 2 Å of RMSD of the X-ray structure. Free-energy scores show strong affinity of the inhibitors for the enzyme binding pocket. Three independent Molecular

Dynamics simulation runs indicated general stability of donepezil, galantamine and rivastigmine in their respective enzyme binding pocket (also referred to as gorge) as well as the tendency to form hydrogen bonds with the water molecules. The binding of rivastigmine in the Torpedo California AChE binding pocket is interesting as it eventually undergoes carbamylation and breaks apart according to the X-ray structure of the complex. Similarity search in the ZINC database and targeted docking on the gorge region of the AChE enzyme gave new putative inhibitor molecules with high predicted binding affinity, suitable for potential biophysical and biological assessments.

ABSTRACT In this report we assessed by docking and molecular dynamics the binding mechanisms of three FDA-approved Alzheimer drugs, inhibitors of the enzyme acetylcholinesterase (AChE): donepezil, galantamine and rivastigmine. Dockings by the softwares Autodock-Vina, PatchDock and Plant reproduced the docked conformations of the inhibitor-enzyme complexes within 2 Å of RMSD of the X-ray structure. Freeenergy scores show strong affinity of the inhibitors for the enzyme binding pocket. Three independent Molecular Dynamics simulation runs indicated general stability of donepezil, galantamine and rivastigmine in their respective enzyme binding pocket (also referred to as gorge) as well as the tendency to form hydrogen bonds with the water molecules. The binding of rivastigmine in the Torpedo California AChE binding pocket is interesting as it eventually undergoes carbamylation and breaks apart according to the X-ray structure of the complex. Similarity search in the ZINC database and targeted docking on the gorge region of the AChE enzyme gave new putative inhibitor molecules with high predicted binding affinity, suitable for potential biophysical and biological assessments.

KEYWORDS: Acetylcholinesterase (AChE) inhibitors, molecular docking, molecular dynamics, donepezil, galantamine, rivastigmine, AutoDock-4, Autodock-Vina, AMBER, ZINC database.

Introduction Alzheimer’s Disease (AD) is a chronic neurodegenerative disorder that is characterized by the destruction of nerve cells, and rapid deterioration of memory and other vital cognitive functions (Burns and Iliffe, 2009). In the United States, about 5.3 millions of Americans are suffering from AD (Weuve et al., 2014). AD is ranked the sixth leading cause of death among the general U.S. population and is the third leading cause of death among older adults, after heart disease and cancer (Hebert et al., 2013). For the past thirty years, much of the AD treatments revolved around the cholinergic hypothesis, which proposes that dysfunctionalities of acetylcholine-containing neurons in the brain contribute to the cognitive decline observed in those with advanced age and AD (Francis et al., 1999). Eventually, AD causes lethal destruction of neuron cells in the brain from initial onset of severe interference in the neurotransmitters at synapses actuated by Acetylcholinesterase (AChE), an enzyme that hydrolyzes the acetylcholine neurotransmitter that is associated with memory and learning. It is now accepted that AD destroys neurons and disrupts the communication pattern of neurotransmitters at synapses (Francis et al., 1999). Another hypothesis is that the disease progression is associated with plaques and tangles in the brain (Serrano-Pozo et al., 2011).

A number of marketed AD drugs work by inhibiting acetylcholinesterase that breaks down the key acetylcholine neurotransmitter. The first FDA-approved AD drug, Donepezil (Aricept®), is a cholinesterase inhibitor that works by slowing down the process of neurotransmitter breakup. Subsequently, two more cholinesterase inhibitors, Galantamine (Razadyne®) and Rivastigmine (Exelon®), received FDA approval. Although all cholinesterase inhibitors target and bind to AChE to slow down the breaking down of the neurotransmitter, each marketed inhibitor binds to different residues of the AChE binding pocket as revealed by the inhibitor-bound X-ray structures of Torpedo California acetylcholinesterase (Tc-AChE) (Kryger et al., 1999) (Bar-On et al., 2002a) (Greenblatt et al., 1999). The enzyme binding site is identified as a narrow 20 Å deep gorge that penetrates AChE halfway into the enzyme and widens close to the bottom (Sussman et al., 1991) (Koellner et al., 2000). Donepezil has a unique bound orientation with the lower part of it near a tryptophan residue (Trp-84) and the top part near another, Trp-279 (Kryger et al., 1999). Galantamine makes interaction with Trp-84 of the activesite gorge and it forms a strong hydrogen bond with Glu-199 as found in Tc-AChE (Greenblattet al., 1999). A detailed X-ray and molecular simulation study with ligands on gorge dynamics in Tc-AChE was performed previously (Axelsen et al., 1994). However, they did not report any inhibitor-enzyme complex studies for the FDAapproved drugs. It is interesting that another marketed inhibitor for AD, rivastigmine, is reported to actuate carbamylation on Tc-AChE (Spencer and Noble, 1998). X-ray structural studies done on the complex of rivastigmine with Tc-AChE showed that rivastigmine undergoes molecular fragmentation in forming the bound complex (Bar On et al., 2002). Furthermore, simultaneous kinetic studies, carried out also on Tc-AChE,

show rivastigmine as a slow-binding reversible inhibitor (Bar-On et al., 2002a). Also, X-ray structures of donepezil and galantamine complexes with human AChE have been reported recently (Cheung et al., 2012a). The aim of this work is to further study the binding of these three known inhibitors of AChE using molecular docking and molecular dynamics and to examine structural analogs of these inhibitors as putative hits. Although biophysical assays will be essential for further assessment of the efficacy of the new molecules, such efforts can serve as reasonable initial stages to explore novel potent hit molecules. Molecular docking involves significant approximations to be able to achieve the speed required for the screening of large libraries. Several algorithms and softwares have been developed for this purpose, involving different algorithms for sampling ligand positions and conformations and for scoring the likely affinity of a binding conformation (pose). In this work using four docking software we were able to benchmark their accuracy by docking donepezil and galantamine to the crystal structure of their receptor. Molecular dynamics is a computationally involved process, providing an ensemble of ligand-target conformations in their aqueous environment. This allows the detailed analysis of the various interactions the ligand and the target protein make, as well as obtaining an indication of the stability of the complex. More precise estimates could also be obtained from a series of molecular dynamics simulations to calculate the free energy of binding, but that has not been attempted in the present work. We also focused on studying the kinetics of hydrogen bond formation and break-up processes of the AD inhibitors in binding gorge region.

The case of rivastigmine presents an additional question. First, there is no X-ray structure of rivastigmine with human AChE, only with Tc-AChE. Next, in the Tc-AChE system, due to carbamylation, rivastigmine is broken up into two chemical moieties in the bound complex. We will show, however that unbroken rivastigmine is also able to bind into the binding gorge of Tc-AChE, and shows identical binding pattern as for donepezil and galantamine. While the carbamylation hypothesis may need more studies, our current docking and molecular simulation can provide binding and dynamics information of rivastigmine based on non-bonded interactions only. It is not well understood if carbamylation is essential for hydrolysis of the neurotransmitter by rivastigmine as it does not happen for donepezil or galantamine. We carried out molecular dynamics simulations mainly to explore if extreme structural instability is involved in the carbamylation and rivastigmine break-up hypothesis.

COMPUTATIONAL METHODOLOGIES 1. Molecular Docking via Patch-Dock Web-server, PLANTS, AutoDock and Vina Molecular Docking is the method of molecular modeling that assesses the atomic level interactions between a small molecule and a receptor like an enzyme or any other protein (Lengauer and Rarey, 1996). Molecular Docking is a cost effective, fast technique that has been an essential tool in the process of drug discovery as well as a complementary tool to many experimental biophysical techniques (Kitchen et al., 2004)(Meng et al., 2011). In this work molecular dockings on the AChE enzyme have been carried out by four tools: Autodock-4 (Garrett M. Morris et al., 2009), AutodockVina, (Trott and Olson, 2010), PLANTS (Korb et al., 2006) and the web-based

PatchDock Server (Schneidman-Duhovny et al., 2005). While both Autodock-4 and Autodock-Vina use genetic algorithms to repeatedly dock each ligand to the target, PLANTS is based on a class of stochastic optimization algorithms called ant colony optimization (ACO). ACO is inspired by the behavior of real ants finding a shortest path between their nest and a food source and the algorithm PatchDock uses is based on object recognition and image segmentation techniques usually used in computer vision, as docking can be compared the assembly of a jigsaw puzzle (Schneidman-Duhovny et al., 2003). For the docking calculations we obtained the initial AChE complex crystal structures from the Protein Data Bank (www.pdb.org) – (PDB ids: 4EY6, 4EY7, 1GQR). Prior to docking all the crystal-catalyzers as well as water molecules have been removed from the human AChE enzyme target as none of them plays any role in inhibitor binding (Wong and Lightstone, 2011). AutoDockTools have been used to prepare the enzyme prior to the docking. Gasteiger (Gasteiger and Marsili, 1980) partial charges have been assigned to both the inhibitor and enzyme atoms. The docking sampled the ligands in a 126 × 126 × 126 grid with 0.375 Å resolution that was positioned to encompass the AChE binding gorge. The quality of the re-docking has been quantified by the root meansquare deviation (RMSD) between the top-scoring docked pose from each docking method and the corresponding X-ray crystal structure. Prior to docking all X-ray ligand structures were compared with PubChem (www.pubchem.com) structures for any missing atoms and then optimized using the program LigPrep from the Schrodinger (Schrödinger Release 2016-2: LigPrep, version 3.8, Schrödinger, LLC, New York, NY, 2016). The optimization is based on molecular

mechanics and used the OPLS force field (Jorgensen et al., 1996). The docking results were analyzed with the programs AutoDockTools (Garrett M Morris et al., 2009), DOCKRES (Mezei and Zhou, 2010), as well as with VMD (Humphrey et al., 1996).

2. Molecular Dynamics Simulations of Inhibitor-AChE Complexes. To assess the strength of binding of the inhibitors in the enzyme pocket as revealed under X-ray studies, the role of water molecules as well as of conformational dynamics, we performed 50 ns full-scale atomistic molecular dynamics simulation using the AMBER general atom force field (Wang et al., 2004) for the ligands and the AMBER 12SB, v12 force field (Case et al., 2012) for the protein. For donepezil and galantamine, the X-ray structure of the inhibitor-enzyme complexes with water molecules have been used as the initial conformation to prepare the topologies (Cheung et al., 2012b). Since as of today no human rivastigmine-AChE complex structure is available, we took the Autodock-Vina docked pose for the un-fragmented rivastigmine to build the topologies of the complex with the crystal water molecules (Bar-On et al., 2002b). All atomic charges (Jakalian et al., 2002) and the force field parameters of the inhibitor molecules interacting via non-bonded interaction with the enzyme were obtained with the Antechamber tools (Wang et al., 2004) using the AM1-BCC charge scheme. In analyzing the structure of AChE enzyme, we identified three pairs of Sulphur atoms in cysteine

residues that form disulfide bonds, and incorporated them into the topologies accordingly (Case et al., 2005). The topologies and the initial coordinates of the inhibitorenzyme complexes for the molecular simulation run were built with the Leap module (tleap) (Wang et al., 2004) of AMBER. Molecular dynamics of each complex was carried out in a rectangular cell of size 65x65x75 Å3 under periodic boundary conditions (PBC) with 10 Å non-bonded cut-off and using Particle Mesh Ewald (PME) for the long-range electrostatics. Each initial inhibitor-enzyme complex structure has been first energy minimized to 0K using a total of 1000 minimization steps (500 steepest descent followed by 500 steps with the conjugate gradient method). This minimization was followed by two heating cycles consisting of 5000 MD steps each. The first cycle raised the system temperature to 100K. Finally the system was equilibrated to the simulation temperature of 300 K via the second cycles. The production MD simulations of the complexes at constant pressure and temperature were carried out for 50 ns each with a 2 fs integration step size using SHAKE (Ryckaert et al., 1977). The trajectories were analyzed with SIMULAID (Mezei, 2010) and visualized with VMD (Humphrey et al., 1996).

3. Targeted Docking of Analogs from Zinc Library Simulation of the inhibitor-enzyme complex provides rich information about the nature of ligand binding in the enzyme site. Molecular dynamics trajectory analysis often suggests region of high ligand affinity that results in identifying new hit molecules with higher biological efficacy (Durrant and McCammon, 2011). For AChE, both molecular docking and molecular dynamics studies confirmed that the binding gorge region of

AChE is the active binding pocket for strong inhibition of AChE from hydrolyzing and disabling the neuro-transmitter. Fig. 5A shows the grid box spanning the space of the targeted region of ligand conformations for donepezil analogs. Similar targeted dockings for analogs of galantamine and rivastigmine have been carried out that were obtained with similarity search in ZINC database. Prior to docking the initial analog searches were performed on the ZINC database using the inhibitors’ SMILES strings, finding 16, 19 and 10 analogs of donepezil, galantamine and rivastigmine, respectively. The ZINC database has the 3D structure of the molecules dock-ready form. However, prior to docking, we have still performed the same geometry optimization on these analogs with LigPrep as we did it with the known inhibitors. For the targeted docking, we used Autodock-Vina and AutoDock-4 and analyzed results for the top scoring ligands from both of these methods.

Results & Discussions (1) Molecular Docking of the Inhibitors on AChE Analysis of the molecular docking of donepezil with human AChE indicates a strong interaction with the gorge-like binding site associated with high free energy score and also method-specific high scores for PatchDock and PLANTS as presented in Table 1. While AutoDock-4 and Autodock-Vina have standard free energy units, exact interpretations of the scoring functions for other two methods have been detailed in the reported works (Korb et al., 2006). As shown in Fig. 1A and Table 1, overall inhibitor docking conformations produced by Autodock-Vina, PLANTS and the Patch-Dock server are within 2.5 Å of RMSD of the X-ray conformation. However, the binding site

predicted by AutoDock-4 (green) has noticeable deviation from the X-ray structure with 18 Å RMSD. In case of galantamine as shown in Fig 1B, we observe that the binding sites predicted by Autodock-Vina (1.34 Å) and PatchDock (2.37 Å) server the are very close to the X-ray site. However, docking conformations from both Autodock-4 and PLANTS have detectable deviations. Molecular docking of rivastigmine on AChE has been interesting, given the fact that so far no X-ray structural studies on human rivastigmine-AChE complex have been reported. X-ray studies of rivastigmine with TcAChE show that rivastigmine undergoes carbamylation upon binding to AChE and the two fragments of 3-[1-(dimethylamino)ethyl]phenol (NAP) and carbamyl moiety are in close proximity of 2.5 Å. Our molecular docking studies of rivastigmine on Tc-AChE (Fig.1C) are likely to suggest that before carbamylation, the inhibitor binds to the same gorge binding sites as NAP and the carbamyl moiety with high affinity (Table 1). The RMSD of docked poses were calculated only for NAP moiety; values for different docking methodologies have been shown in Table 2. Molecular dynamics of 50 ns duration, discussed in next section, does not indicate any instability of the docked complex. It is an open question if rivastigmine undergoes similar reaction in human AChE as in Tc-AChE; however, the exploration of the mechanism of carbamylation would require tools that are different from those used in the present study.

(2) Molecular Dynamics of AChE-Inhibitor Complexes For the simulated time span of 50 ns, both donepezil and galantamine have shown general stability and confinement in the gorge-shaped binding pocket of AChE, in

conformity with the conformations from X-ray as well as the docking results, indicating that the inhibitor-enzyme complexes are thermodynamically favored conformations. We have not observed any significant conformational change in the AChE surface during this simulation time window. Analyzing the MD trajectory, we observe that both inhibitors have shown strong tendency to be localized in the binding gorge region of AChE. In case of donepezil, two localized water molecules have been observed in its proximity to form hydrogen bonds with the oxygens of the fused aromatic ring. On the other hand, in case of the galantamine-AChE complex, we observed one hydrogen-bond forming water molecule with galantamine’s oxygen at the location of the fused aromatic ring. Figs. 2 and 3 show the conformations of the bound water molecules with donepezil and galantamine at 25 ns and 50 ns showing stable affinity and proximity of the water molecules in forming the hydrogen bonds. The molecular dynamics simulation of rivastigmine does not indicate any strong abrupt local force that could potentially break up the inhibitor or can even delocalize it from the binding site. As reported in the X-ray studies, rivastigmine undergoes carbamylation and is bound to Tc-AChE in fragmented molecular parts of NAP and carbamyl moiety (Spencer and Noble, 1998). For our simulation duration, rivastigmine has been observed to be also confined in the gorge-like pocket by non-bonded interactions. As shown in Fig. 4, a significant number of water molecules have been observed to stay close to the rivastigmine binding site. To capture events prior to fragmentation of rivastigmine, one might need to carry out longer time scale simulation in the range of micro or even millisecond order. Longer simulation is likely to capture events prior to carbamylation as it is speculated to occur at slower time scale than fast

kinetics. Abundance of water molecules in the vicinity of rivastigmine indicates water medium likely to have role in rivastigmine break-up process. Ultimately, however, the simulations would need to include the possibility of bond breaking and bond forming, as provided, e.g., in simulations where the reactive part of the system is treated quantum mechanically and the rest with molecular mechanics.

(3) Docking of the Analogs from the Zinc Library: Putative Hits for Biophysical Assessments For all three inhibitors we performed similarity searches in the ZINC database (http://zinc.docking.org/), followed by targeted docking of each set of the analog molecules found to their respective receptors to assess their theoretical affinity to the gorge region of the AChE. Figs. 5A and 5B show the grid box for the targeted docking as well as the conformational clustering from docking results for donepezil using Autodock4 and Autodock-Vina. We also carried out similar studies on galantamine as well as on rivastigmine; the free-energy estimates of the top-scoring poses for each target are presented in Table 3. Although the top scoring putative hit molecules in Tables 3(A) and 3(B) have structural similarity with the marketed drug donepenzil, our literature search in Sci-Finder, PubChem, ChemSpider and ChemDB does not report any of the compound found reported as AChE inhibitor. One of the rivastigmine analogs in Table 3(C), has been also identified via virtual screening and pre-ADMET software studies as an AChE binder (Singh et al., 2014). The predicted scores for hit molecules are promisingly high

and thus these molecules are of further interest for both biophysical and biological assessments.

Conclusion We performed molecular docking, molecular dynamics and targeted docking studies on three marketed drugs for Alzheimer disease that inhibit the AChE enzyme from hydrolyzing and destroying the neurotransmitter acetylcholine. Molecular docking conformations show reproducibility of the X-ray complex structure of human AChE with donepezil and galantamine within small degree of uncertainty. Molecular dynamics simulations show general stability of both donepezil and galantamine in the binding gorge and the tendency of the inhibitor to form hydrogen bonds with water, indicating that the gorge binding site is indeed the active site of the enzyme. Docking studies on rivastigmine indicate that rivastigmine binds to the same gorge region as other two inhibitors. The observed stability of rivastigmine docked on Tc-AChE may indicate that the carbamylation process as revealed in the X-ray structure is a slow irreversible process that is likely to be water assisted. One may speculate at what stage of neurotransmitter inhibition, carbamylation does occur as gorge binding region showed stability for rivastigmine docked conformation as non-bonded interaction. A longer simulation in water environment can potentially give more information about this particular inhibitor’s action on the AChE enzyme. Our definitive identification of the region for inhibitor binding opens up for testing new analogs for further efficacy assessments. We have narrowed down the region for targeted docking from molecular dynamics trajectory. Docking scores of the analogs

of the three FDA-approved drugs suggest them to be putative hits and thus good candidates for testing as AChE inhibitors. Experimental tests, in particular surface plasmon resonance binding kinetic experiments, of the new molecules on the AChE target in comparison to the marketed drugs will be of immense interest to understand the inhibitory functions of the molecules and possible lead toward discovery of new AChE inhibitors.

ACKNOWLEDGEMENTS We would like to acknowledge St John’s University’s Summer Faculty Research Support (2016-17) and the NSF-supported XSEDE Super Computer Resources (Allocation # TG-MCB140084) as well as the Computational Resources and staff expertise provided by the Department of Scientific Computing at Icahn School of Medicine at Mount Sinai.

REFERENCES Axelsen, P.H., Harel, M., Silman, I., Sussman, J.L., 1994. Structure and dynamics of the active site gorge of acetylcholinesterase: synergistic use of molecular dynamics simulation and X-ray crystallography. Protein Sci. Publ. Protein Soc. 3, 188–197. Bar-On, P., Millard, C.B., Harel, M., Dvir, H., Enz, A., Sussman, J.L., Silman, I., 2002a. Kinetic and Structural Studies on the Interaction of Cholinesterases with the AntiAlzheimer Drug Rivastigmine,. Biochemistry (Mosc.) 41, 3555–3564. doi:10.1021/bi020016x Bar-On, P., Millard, C.B., Harel, M., Dvir, H., Enz, A., Sussman, J.L., Silman, I., 2002b. Kinetic and structural studies on the interaction of cholinesterases with the antiAlzheimer drug rivastigmine. Biochemistry (Mosc.) 41, 3555–3564.

Burns, A., Iliffe, S., 2009. Alzheimer’s disease. BMJ 338. doi:10.1136/bmj.b158 Case, D.A., Cheatham, T.E., Darden, T., Gohlke, H., Luo, R., Merz, K.M., Onufriev, A., Simmering, C., Wang, B., Woods, R.J., 2005. The Amber Biomolecular Simulation Programs. J. Comput. Chem. 26, 1668–1688. doi:10.1002/jcc.20290 Case, D., Darden, T., Cheatham, T., Simmerling, C., Wang, J., Duke, R., Luo, R., Walker, R., Zhang, W., Merz, K., Roberts, B., Hayik, S., Roitberg, A., Seabra, G., Swails, J., Goetz, A., Kolossváry, I., Wong, K., Paesani, F., Vanicek, J., Wolf, R., Liu, J., Wu, X., Brozell, S., Steinbrecher, T., Gohlke, H., Cai, Q., Ye, X., Wang, J., Hsieh, M., Cui, G., Roe, D., Mathews, D., Seetin, M., Salomon-Ferrer, R., Sagui, C., Babin, V., Luchko, T., Gusarov, S., Kovalenko, A., Kollman, P., 2012. AMBER 12. Cheung, J., Rudolph, M.J., Burshteyn, F., Cassidy, M.S., Gary, E.N., Love, J., Franklin, M.C., Height, J.J., 2012a. Structures of human acetylcholinesterase in complex with pharmacologically important ligands. J. Med. Chem. 55, 10282–10286. doi:10.1021/jm300871x Cheung, J., Rudolph, M.J., Burshteyn, F., Cassidy, M.S., Gary, E.N., Love, J., Franklin, M.C., Height, J.J., 2012b. Structures of Human Acetylcholinesterase in Complex with Pharmacologically Important Ligands. J. Med. Chem. 55, 10282–10286. doi:10.1021/jm300871x Durrant, J.D., McCammon, J.A., 2011. Molecular dynamics simulations and drug discovery. BMC Biol. 9, 71. doi:10.1186/1741-7007-9-71 Francis, P.T., Palmer, A.M., Snape, M., Wilcock, G.K., 1999. The cholinergic hypothesis of Alzheimer’s disease: a review of progress. J. Neurol. Neurosurg. Psychiatry 66, 137–147. Gasteiger, J., Marsili, M., 1980. Iterative partial equalization of orbital electronegativity—a rapid access to atomic charges. Tetrahedron 36, 3219–3228. Greenblatt, H.M., Kryger, G., Lewis, T., Silman, I., Sussman, J.L., 1999. Structure of acetylcholinesterase complexed with (-)-galanthamine at 2.3 A resolution. FEBS Lett. 463, 321–326. Hebert, L.E., Weuve, J., Scherr, P.A., Evans, D.A., 2013. Alzheimer disease in the United States (2010-2050) estimated using the 2010 census. Neurology 80, 1778– 1783. doi:10.1212/WNL.0b013e31828726f5 Humphrey, W., Dalke, A., Schulten, K., 1996. VMD: visual molecular dynamics. J. Mol. Graph. 14, 33–8, 27–28. Jakalian, A., Jack, D.B., Bayly, C.I., 2002. Fast, efficient generation of high-quality atomic charges. AM1-BCC model: II. Parameterization and validation. J. Comput. Chem. 23, 1623–1641. doi:10.1002/jcc.10128 Jorgensen, W.L., Maxwell, D.S., Tirado-Rives, J., 1996. Development and Testing of the OPLS All-Atom Force Field on Conformational Energetics and Properties of Organic Liquids. J. Am. Chem. Soc. 118, 11225–11236. doi:10.1021/ja9621760 Kitchen, D.B., Decornez, H., Furr, J.R., Bajorath, J., 2004. Docking and scoring in virtual screening for drug discovery: methods and applications. Nat Rev Drug Discov 3, 935–949. doi:10.1038/nrd1549 Koellner, G., Kryger, G., Millard, C.B., Silman, I., Sussman, J.L., Steiner, T., 2000. Active-site gorge and buried water molecules in crystal structures of

acetylcholinesterase from Torpedo californica. J. Mol. Biol. 296, 713–735. doi:10.1006/jmbi.1999.3468 Korb, O., Stützle, T., Exner, T.E., 2006. PLANTS: application of ant colony optimization to structure-based drug design, in: Proceedings of the 5th International Conference on Ant Colony Optimization and Swarm Intelligence. Springer-Verlag, Brussels, Belgium, pp. 247–258. Kryger, G., Silman, I., Sussman, J.L., 1999. Structure of acetylcholinesterase complexed with E2020 (Aricept®): implications for the design of new anti-Alzheimer drugs. Structure 7, 297–307. doi:10.1016/S0969-2126(99)80040-9 Lengauer, T., Rarey, M., 1996. Computational methods for biomolecular docking. Curr. Opin. Struct. Biol. 6, 402–406. doi:10.1016/S0959-440X(96)80061-3 Meng, X.-Y., Zhang, H.-X., Mezei, M., Cui, M., 2011. Molecular docking: a powerful approach for structure-based drug discovery. Curr. Comput. Aided Drug Des. 7, 146–157. Mezei, M., Zhou, M.-M., 2010. Dockres: a computer program that analyzes the output of virtual screening of small molecules. Source Code Biol. Med. 5, 2–2. doi:10.1186/1751-0473-5-2 Morris, G.M., Huey, R., Lindstrom, W., Sanner, M.F., Belew, R.K., Goodsell, D.S., Olson, A.J., 2009. AutoDock4 and AutoDockTools4: Automated Docking with Selective Receptor Flexibility. J. Comput. Chem. 30, 2785–2791. doi:10.1002/jcc.21256 Morris, G.M., Huey, R., Lindstrom, W., Sanner, M.F., Belew, R.K., Goodsell, D.S., Olson, A.J., 2009. AutoDock4 and AutoDockTools4: Automated Docking with Selective Receptor Flexibility. J. Comput. Chem. 30, 2785–2791. doi:10.1002/jcc.21256 Ryckaert, J.-P., Ciccotti, G., Berendsen, H.J.., 1977. Numerical integration of the cartesian equations of motion of a system with constraints: molecular dynamics of n-alkanes. J. Comput. Phys. 23, 327–341. doi:10.1016/0021-9991(77)90098-5 Schneidman-Duhovny, D., Inbar, Y., Nussinov, R., Wolfson, H.J., 2005. PatchDock and SymmDock: servers for rigid and symmetric docking. Nucleic Acids Res. 33, W363-367. doi:10.1093/nar/gki481 Schneidman-Duhovny, D., Inbar, Y., Polak, V., Shatsky, M., Halperin, I., Benyamini, H., Barzilai, A., Dror, O., Haspel, N., Nussinov, R., Wolfson, H.J., 2003. Taking geometry to its edge: fast unbound rigid (and hinge-bent) docking. Proteins 52, 107–112. doi:10.1002/prot.10397 Serrano-Pozo, A., Frosch, M.P., Masliah, E., Hyman, B.T., 2011. Neuropathological Alterations in Alzheimer Disease. Cold Spring Harb. Perspect. Med. 1, a006189. doi:10.1101/cshperspect.a006189 Singh, A., Jatav, V.K., Sharma, S., 2014. VIRTUAL SCREENING AND ADMET ANALYSIS FOR IDENTIFICATION OF INHIBITORS AGAINST ACETYLCHOLINESTERASE ASSOCIATED WITH ALZHEIMER’S DISEASE. Int. J. Pharm. Pharm. Sci. 5. Spencer, C.M., Noble, S., 1998. Rivastigmine. A review of its use in Alzheimer’s disease. Drugs Aging 13, 391–411.

Sussman, J.L., Harel, M., Frolow, F., Oefner, C., Goldman, A., Toker, L., Silman, I., 1991. Atomic structure of acetylcholinesterase from Torpedo californica: a prototypic acetylcholine-binding protein. Science 253, 872–879. Trott, O., Olson, A.J., 2010. AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. J. Comput. Chem. 31, 455–461. doi:10.1002/jcc.21334 Wang, J., Wolf, R.M., Caldwell, J.W., Kollman, P.A., Case, D.A., 2004. Development and testing of a general amber force field. J. Comput. Chem. 25, 1157–1174. doi:10.1002/jcc.20035 Weuve, J., Hebert, L.E., Scherr, P.A., Evans, D.A., 2014. Deaths in the United States among persons with Alzheimer’s disease (2010-2050). Alzheimers Dement. J. Alzheimers Assoc. 10, e40-46. doi:10.1016/j.jalz.2014.01.004 Wong, S.E., Lightstone, F.C., 2011. Accounting for water molecules in drug design. Expert Opin. Drug Discov. 6, 65–74. doi:10.1517/17460441.2011.534452

Figures:

Figure 1A: X-ray structure of AChE complexes with the top-scoring docked

inhibitor conformations from different docking methods. The inhibitor conformations are color coded: X-ray-Red, Vina-Blue, AutoDock 4.0– Green, PatchDock Server–Yellow and Plant–Orange with Donepezil docked on human AChE (PDB ID: 4EY7).

Figure 1B: X-ray structure of AChE complexes with the top-scoring docked

inhibitor conformations from different docking methods. The inhibitor conformations are color-coded: X-ray-Red, Vina-Blue, AutoDock 4.0– Green, PatchDock Server–Yellow and Plant–Orange with Galantamine also docked on human AChE (PDB ID: 4EY6).

Figure 1C: X-ray structure of AChE complexes with the top-scoring docked

inhibitor conformations from different docking methods. The inhibitor conformations are color-coed: X-ray-Red, Vina-Blue, AutoDock 4.0–Green, PatchDock Server–Yellow and Plant–Orange with Rivastigmine on Torpedo California AChE (PDB ID: 1GQR).

Figure 2A: Stable conformation of donepezil with hydrogen-bonded water molecules (blue) in AChE complex at 25 ns.

Figure 2B: Stable conformations of donepezil with hydrogen-bonded water molecules (blue) in AChE complex at 50 ns.

Figure 3A: Stable conformations of galantamine with a hydrogen-bonded water molecule (blue) in AChE complex at 25 ns.

Figure 3B: Stable conformations of galantamine with a hydrogen-bonded water molecule (blue) in AChE complex at 50 ns.

Figure 4: Abundance of water molecules in rivastigmine proximity in gorge region of Tc AChE. 50 ns of MD simulation studies show rivastigmine binds and stays stable before potential water assisted break-up into NAP and other chemical moiety.

Figure 5A: Targeted docking of analogs on donepezil. The target grid box (40x20x40) is centered at (-15,-43, 26).

Figure 5B: The binding conformations of Donepezil analogs from ZINC database in gorge binding region of human AChE.

TABLE. 1. Binding free energy of AChE complexes.

Donepezil Galantamine Rivastigmine

AutoDock-4 (Kcal/mole)

AutodockVina (Kcal/mole)

Plant Scoring PatchDock function (AU) Atomic Contact Energy (AU)

-9.3 -6.5 -7.3

-11.7 -9.2 -7.7

-94.8 -81.2 -77.4

-233 -175 -157

Table 2: RMSD deviation of docked poses from X-ray conformation (RMSD in Å) AutoDock-4 AutodockPlant PatchDock Vina Donepezil Galantamine Rivastigmine

18.06 7.22 5.19

1.18 1.34 3.64

2.47 2.37 3.27

1.93 9.16 4.16

Table 3A: Free energy of the top-scoring analogs of Donepezil[(O=C1[C@H](CC2CCN(CC2)Cc2ccccc2)Cc2c1cc(OC)c(OC)c2] from targeted docking on gorge region of human AChE by Autodock-Vina and Autodock-4. Autodock-Vina

Autodock-4

-12.10 kcal/mole ZINC22058276

-9.56 kcal/mole ZINC34964948

Table 3B: Free energy of the top-scoring analogs of Galantamine [O1[C@@H]2[C@]3(c4c1c(OC)ccc4CN(CC3)C)C=C[C@H](O)C2] From targeted docking on gorge region of human AChE using Autodock-Vina and Autodock-4. Autodock-Vina

Autodock-4

-9.42 kcal/mole ZINC13449462

-8.65 kcal/mole ZINC22058276

Table 3C: Free energy of the top-scoring analogs of Rivastagmine O(c1cc([C@@H](N(C)C)C)ccc1)C(=O)N(CC)C from targeted docking on gorge region of human AChE by Autodock-4 and AutodockVina. Autodock-Vina

Autodock-4

-7.13 kcal/mole ZINC90411665

-6.28 kcal/mole ZINC90411521