Design, synthesis, in silico and biological evaluations of novel polysubstituted pyrroles as selective acetylcholinesterase inhibitors against Alzheimer’s disease

The objective of this study was to design new polysubstituted pyrrole derivatives as selective acetylcholinesterase (AChE) inhibitors to target Alzheimer's disease. In this context, a highly efficient, one-pot, sequential, multi-component synthesis of a diverse range of polysubstituted pyrroles was developed through a sequential domino strategy by the condensation of amines with 1,1-bis(methylthio)-2-nitroethene (BMTNE), Knovenagle reaction of arylglyoxals with malono derivatives and subsequent Michael addition and intramolecular cyclization reaction in EtOH at reflux. Thirty-nine synthesized compounds were evaluated as AChE and butyrylcholinesterase (BChE) inhibitors. Among the synthesized compounds, compound 4ad (IC50 = 2.95 ± 1.31 µM) was the most potent and selective AChE inhibitor with no significant inhibition against butyrylcholinesterase BChE. A kinetic study of 4ad revealed that this compound inhibited AChE in an uncompetitive mode. Based on a molecular modeling study, compound 4ad due to its small size properly fitted into the active site of AChE compared to BChE and stabilized by H-bond and hydrophobic interactions with the critical residues of the AChE binding pocket. Consequently, it was proposed that the 4ad derivative can be an ideal lead candidate against AD with a simple and practical operation of synthetic procedures.

Multicomponent reactions (MCRs) have emerged as a powerful tool for the purpose that a series of chemical processes can be controlled in a one-pot operation 17,18 . These reactions involve forming multiple bonds in a single conversion in a one-pot process without separating intermediates [19][20][21][22] .
Nitro ketene dithioacetals are known as dicarbon synthons in organic synthesis due to their push-pull electronic nature. Reactions of nitro ketane dithioacetals with a variety of nucleophilic groups have so far been applied to construct various heterocycles 23,24 with diverse biological activities, including antitumor 25,26 , antianxiety 27 , antileishmanial 28 and antibacterial 29 .
Compounds containing N-heterocyclic skeletons are known to exhibit various biological and medicinal activities [30][31][32][33][34] . Among these compounds, pyrrole derivatives are generally shown to have several biological activities as antioxidant 35 , antifungal 36 , antituberculosis 37 , anti-inflammatory 38 , analgesic 39 , antidiabetic 40 and anticancer 41 agents. Some examples of pharmaceutical compounds with pyrrole ring skeletons are shown in Fig. 1 42,43 . In addition, pyrrole is an important core moiety in many biological compounds such as chlorophyll, heme, vitamin B 12 , bile pigments, and alkaloids.
A number of synthetic strategies for the construction of substituted pyrrole derivatives have been reported such as the Knorr reaction 44 , the Paal-Knorr reaction 45 , the Hantzsch synthesis 46 and different cycloaddition methods 47 . Some of these methods have disadvantages such as long reaction time, low yields, expensive reagents or catalysts, harsh reaction conditions, the use of toxic solvents, and methods that do not comply with the principles of green chemistry.
Owing to the attractive medicinal properties of pyrrole derivatives, the development of a new and simple synthetic strategy for the efficient synthesis of novel polysubstituted pyrroles will be a useful and attractive challenge. In continuation of our laboratory efforts on the development of multicomponent reactions for the synthesis of biologically important heterocyclic compounds [48][49][50][51] herein, an efficient and simple method was developed to synthesize a library of polysubstituted pyrroles as AChE and BChE inhibitors. Furthermore, the kinetic studies of the most potent derivative were performed. The most potent compound was then subjected to molecular docking and molecular dynamic (MD) studies to evaluate its binding affinity and mode of action within the binding site of the enzyme. Finally, drug-likeness predictions were performed to address the pharmacokinetic properties of the most interesting compounds.

Result and discussion
Design of new scaffold as AChE inhibitor. According to crystallographic studies, the binding pocket of AChE comprises a narrow gorge approximately 20 Å deep, consisting of two binding sites: catalytic activity site (CAS) at the bottom of the gorge and peripheral anion site (PAS) near the entrance of the gorge. According to several studies, dual binding site inhibitors of AChE occupy CAS and PAS pockets and exert their effects more efficiently. Generally, dual binding site AChE inhibitors may not only improve cognition by inhibiting AChE but also slow down the aggregation of Aβ to generate toxic plaques 52 .
The traditional AChE inhibitors comprise three parts including a core ring that interacts with amino acid residues of PAS, a basic moiety to interact with the aromatic residues of CAS, and a linker to link these two parts, which lies in a narrow part of the active site of AChE. Compounds A-F (Fig. 2) are some AChE examples [53][54][55][56][57][58][59] . However, those inhibitors that possess a positive charge (quaternary amines) are mostly unable to cross the blood-brain barrier (BBB) and show poor efficiency in CNS.
There are limited data available about the ChE inhibitory activities of polysubstituted pyrroles. Amongst compounds G 60 , H 61 , and I 62 with moderate to good activity against AChE were presented in Fig. 3. Recently, compound bearing pyrrole-sulfonyl amine was reported as an example of pyrrole derivatives with IC 50 values ranging from 6.50 to 37.46 nm against AChE 63 .
It was assumed that the pyrrole ring can play a role as core moiety to interact with the aromatic residues of the AChE binding pocket while different substitutions on the ring might improve its potencies. In this context, a library of 39 molecules was designed and synthesized to properly evaluate the structure-activity relationships (SARs) of pyrrole derivatives against AChE and BChE. (1 mmol) and amine (2 eq) or diamine (1 eq) were reacted in EtOH at reflux for 6 h to form the corresponding intermediate A. Then arylglyoxal 1 (1 mmol) and malono derivative 2 (1 mmol) were added to obtain the desired product 4. It should be mentioned that some other solvents such as H 2 O, CH 3 CN, CHCl 3 , THF, and DMF were tested and the results showed that EtOH was the best solvent. Having optimized the reaction conditions a series of polysubstituted pyrroles were successfully synthesized under the given reaction conditions and the results are summarized in Table 1.
As shown in Table 1, a diverse range of different arylglyoxals, amines and malono derivatives were successfully employed. In this study, methoxy, methyl and chloro substituents on the phenylglyoxal ring showed similar productivity in comparison with phenylglyoxal. In addition, different malono derivatives such as malononitrile and alkyl cyanoacetates were used in these reactions. However, the use of malono nitrile leads to a shorter reaction time and higher yields compared to ethyl or methyl cyanoacetate. Diamines such as 1,3-propyl diamine, 2,2-dimethyl-1,3-propane-diamine and 1,8-diaminonaphthalene were used for the effectient synthesis of fully substituted pyrroles in good to excellent yields (compounds 4a-4 m). Also, two equivalents of alkylamines such as benzylamine and furfuralamine were used and the corresponding products were synthesized in excellent yield (compounds 4n-4t). In this study, aqueous ammonia solution (25%) was used as a source of amine to obtain N-substituted-free pyrroles and these type of pyrroles were prepared efficiently (compounds 4u-4y www.nature.com/scientificreports/ noted that in this case, ammonia was used in excess amount (1.5 mL for 1 mmol of BMTNE). Aromatic amines were also used and studies shown that only one equivalent of amine is presented in the final product even when we used two equivalents of the amine. On the other hand, this method has successfully led to the formation of polysubstituted pyrroles haveing thiomethyl substituent in their structure, which can be potentially biologically important compounds (4z-4am). When we used 1,2-Phenylenediamine derivatives as the source of amine, the reaction proceeded in another direction that is currently being investigated in our laboratory. The most probable reaction mechanism for the formation of compounds 4 depicted in Fig. 5). The initial reaction between diamine and BMTNE 3 lead to intermediate A. In  In vitro ChE inhibitory activities. First derivatives 4a-l were synthesized to evaluate their potency as AChE and BChE inhibitors. Results are summarized in Table 2 in terms of % inhibition at 50 µM and IC 50 s. According to IC 50 values of pyrrolo[1,2-a]pyrimidine analogs, this set of compounds recorded, mostly, better inhibitory activity toward AChE compared to BChE. The exception in this trend came back to 4a and 4j.
Assessment of SARs indicated interesting results so that type of substitution at R 1 and R 2 positions played the most dominant role in AChE inhibitory activity compared to R 3 substitution. As can be seen compounds 4b (IC 50 = 37.15 µM), 4e (IC 50 = 36.30 µM), and 4h (IC 50 = 67.60 µM) bearing 3,3-dimethyl at the R 1 with CN moiety at R 2 position significantly improved the inhibitory potencies and resulted in the most potent derivatives in this set compared to the other derivatives.
The substituted groups at the R 3 position did not disclose straightforward inhibition and significant differences. It seems that this position had the least effect on the AChE inhibitory activity.
Next to better extract the SARs of pyrrole-acetamide derivatives 4n-ak were also synthesized. The activity of 4n-t bearing aromatic rings at R 1 and R 4 positions showed different results so these analogs were more potent against BChE compared to AChE (Table 3). It seems that the increase bulkiness in this position makes the derivatives more potent as BChE inhibitors. The exception in this trend came back to 4p. The activity of BChE changes in the following order so that 4q (R 1 : -aminomethylfuran; R 2 : COOEt; R 3 : 4-CH 3 and R 4 : -methylfuran) showed IC 50 values of 18.62 µM followed by 4o (R 1 : -aminomethylfuran; R 2 : CN; R 3 : 4-CH 3 ; R 4 : -methyl-furan; IC 50 = 69.18 µM).
Changing the type of R 3 position while R 1 and R 4 are the same exhibited that the presence of para-methyl as an electron-donating group is more favorable compared to para chlorine as a halogen-substituted group as BChE inhibitor. In this case, 4o and 4q recorded better activities compared to 4n and 4s as well as 4r and 4t versus 4p. www.nature.com/scientificreports/ Based on the obtained biological results related to 4u-y (R 1 = NH 2, R 2 = CN), three derivatives were almost inactive against both enzyme. 4w (R 3 = 4-Cl) as the most active compound in this set showed an inhibition percent of 48.61 against AChE and 33.50 against BChE at 50 M followed by 4X (R 3 = 4-MeO) with 47.73% and 19.43% inhibition percent against AChE and BChE, respectively.

Ph
In the last set (4z-4ak), NH 2 of R 1 position was replaced with SMe. The most potent compounds with high selectivity against AChE were presented in this set. The most active compounds among all tested derivatives were 4ad (R 2 = CN; R 3 = 4-Cl; R 4 : 4-hydroxyphenyl) with an IC 50 value of 2.95 ± 1.31 µM followed by 4ah (R 2 = CN; R 3 = 4-Cl; R 4 = 4-methylphenyl) with IC 50 value of 4.78 ± 1.63 µM. In this group COOEt or COOMe substitutions at R 2 inferior the activity (% inhibition at 50 M ranging from not active to 38.86) compared to the CN (% inhibition at 50 M ranging from 26.72 to 89.84) substituted counterparts.
The summary of SAR was presented in Fig. 6. The highest potency to inhibit AChE was observed in compounds bearing SMe group at R 1 , CN moiety at R 2 and 4-Cl at R 3 position. Compounds 4ad and 4ah with IC 50 values of 2.95 ± 1.31 µM and 4.78 ± 1.6 were the most potent AChE inhibitors, respectively.

Kinetic studies of AChE inhibition.
To determine the mechanism of inhibition of AChE, a kinetic study of 4ad was done against AChE. The reciprocal Lineweaver-Burk plots (Fig. 7) illustrate that K m and V max reduced with the increasing concentration of 4ad which indicates that 4ad is an uncompetitive type inhibitor.
Furthermore, the plot of the K m versus different concentrations of 4ad gave an estimate of the inhibition constant, K i of 2.84 µM which is in accordance with the IC 50 value of 2.95 ± 1.31 µM (Fig. 8).

Molecular modeling studies of 4ad in AChE and BChE active site.
Given the uncompetitive-type inhibitory nature of 4ad, its interaction with the active site of AChE was subsequently investigated. First, the docking procedures were validated by accurately redocking the co-crystallized ligands into the AChE and BChE models. CAS located at the bottom of the gorge consist of two sub-units namely, the catalytic triad (esteratic subsite) and the anionic subsite and the second is PAS. Glutamic acid (E202), serine (S203) and histidine (H447) are the main residues of the catalytic triad while the anionic subsite consists of tryptophan (W86). PAS consists Table 1. Multicomponent synthesis of polysubstituted pyrrole derivatives a,b . a Isolated yield. b Reaction conditions: BMTNE 3 (1 mmol) and amine (1 or 2 mmol) were added to EtOH (5 mL) at reflux and after 6 h arylglyoxal 1 (1 mmol) and malono derivative 2 (1 mmol), were added to obtain the desired product 4. Computational docking studies were performed to model the interaction of 4ad with AChE (PDB ID: 4BDS) at the target-binding site (Fig. 9). The nitro and acetamide moieties of 4ad are accommodated in CAS. In detail, the acetamide group showed three hydrogen bound interactions with Trp86 (anionic subsite) and Tyr124, and also the nito group recorded interaction with Trp 86. On the other side of the molecule, 4OH-benzyl is oriented towards PAS and displayed two hydrogens bound interactions with Phe295 and Tyr341 of PAS. This ring also exhibited interaction with Tyr341. Para chlorophenyl also demonstrated hydrophobic interaction with Thr75. These interactions were in line with the results of biological assessments and confirmed the high potency of this derivative.

Scientific Reports
Also, the molecular docking study of 4ad as one of the inactive compounds was performed against BChE. 4ad did not tightly interact with BChE active site and just showed interactions with His438 and His438. A salt bridge interaction of the nitro group with Glu197 was observed (Fig. 10).
These results justify the high potencies of this derivative against AChE compared to BChE. It should be noted that as the active site of AChE is smaller than BChE, the smaller molecules, in this case, 4ad, properly fitted in the binding site of the enzyme and exhibited interaction with PAS and CAS pocket. However 4ad derivative could not effectively occupy all pockets of BChE where this compound lacks hydrogen bonds or pi-pi interactions with active site residue.

Molecular dynamics simulation. MD simulation was performed to model the time-dependent motions
and stabilization of the 4ad-AChE complex. The root mean square deviation (RMSD) showed the average change in displacement of a selection of atoms for a particular frame. The RMSD value of less than 3 Å is perfectly acceptable for small, globular proteins. As can be seen, the RMSD stabilizes around 2.1 Å indicating the simulation is in equilibration. Also, it can be seen that after 15 s the RMSD of the ligand was slightly fewer than the RMSD protein which confirms that the ligand is fixed in the binding site (Fig. 11).
The root mean square fluctuation (RMSF) is useful for characterizing local changes along the protein chain. As can be seen, residues between 270 and 285 (PAS pocket) showed the most fluctuation which showed the highest movement. Also, the residues that more effectively interact with the ligand are marked with green-colored vertical bars (Fig. 12).
The protein interactions with the ligand can be monitored throughout the simulation. The interactions were categorized by type are summarized in Figs. 13   www.nature.com/scientificreports/ (100% of all time) and seems to be close to the entrance of the active site gorge. Also, interaction with the Trp86 of the anionic site was recorded (75% of all time). The molecular interactions of AChE with 4ad over the binding site of AChE at the equilibrated phase of MD simulation were represented in Fig. 14. As can be seen, 4ad formed H-bound interaction with Ser125, and Tyr341, through both of its NH 2 and OH groups during the equilibrated phase. Pi-cation interaction between Trp86 and NO 2 plus two pi-pi stacked interactions between phenyl ring and Tyr341 were also recorded.
Physicochemical properties. In order to assess the drug-like characteristics of the most active compounds (4ad and 4ah), various physicochemical parameters of these compounds were evaluated using the pkCSM and swissADME server.
Both the compounds showed good drug-like properties and follow Lipinski's rule of five (see Table 4). The logP value of both the compounds is less than five and hydrogen bond donor and acceptor atoms are in the optimum range. Both the compounds pass the drug-likeness criteria of Pfizer and Amgen. Thus, it can be concluded that 4ab and 4ah compounds have an optimum chemical skeleton that can be developed as potential drug molecules.

Conclusion
In summary, a novel series of polysubstituted pyrroles derivatives 4a-4ak were synthesized via simple multicomponent chemical reactions, furnishing excellent yields and producing diverse libraries of polysubstituted pyrroles. Four sets of compounds (39 derivatives) were synthesized and the inhibitory activity of all derivatives was evaluated against AChE and BChE. SARs studies showed that SMe group at R 1 , and CN moiety at R 2 significantly improve the AChE inhibition compared to BChE. Among them, compound 4ad exhibited an IC 50 value of 2.95 ± 1.31 μM against AChE.
In addition, compound 4ad possessed an uncompetitive type of inhibition in the enzymatic assay. Furthermore, the IFD docking study of the 4ad in the AChE active site showed key interacting residues with the PAS and CAS pocket of the enzyme. Also, MD simulations revealed compound 4ad interacted with Tyr341 of PAS as well as Trp86 which are of the key residues anionic site. In summary, 4ad was presented as a suitable anti-AChE agent with simple multi-components synthetic procedures that are ideal for lead optimization studies. www.nature.com/scientificreports/                                   6, 142.6, 129.5, 129.3, 129.0, 125.7, 116.8, 115.0, 114.7, 97.9, 62.4, 55.9, 55.8, 53.4, 36.8, 16.5, 14.2.

Molecular docking.
To understand structural aspects of ligand binding using induced-fit molecular docking (IFD), we used. The SMILE format of 4ad was converted to a three-dimensional structure within the Maestro software package. The X-ray structures of AChE (PDB code: 4EY7). and BChE (PDB code: 4BDS) were selected as targets for dockings and were prepared with the Protein Preparation Wizard interface of Maestro via removing the ligand and water molecules, adding hydrogen atoms, optimizing their position, and assigning the ionization states of acid and basic residues according to PROPKA prediction at pH 7.0. The molecular docking was performed using IFD mode with the ligands as flexible, the force field was set as OPLS-2005, and all other parameters were set to default. The binding site was used to generate the grid for IFD calculation. The maximum 20 poses with receptor and ligand van der Waals radii of 0.7 and 0.5, respectively considered. Residues within 8 Å of the crystallographic ligands at the active site were refined followed by side-chain optimization. Structures in which primeenergy is more than 30 kcal/mol are eliminated. The re-docking experiment for validation of the used docking protocol was done and recorded the RMSD value of 0.79, indicating the docking experiment is reliable 70,71 .
Molecular dynamic simulations. Molecular simulations of this study were performed using the Desmond v5.3 using the Maestro interface (from Schrödinger 2018-4 suite). The appropriate pose for the MD simulation procedure of the compound was achieved by the IFD method. To build the system for MD simulation, the protein-ligand complex was solvated with SPC explicit water molecules and placed in the center of an orthorhombic box of appropriate size in the periodic boundary condition. Sufficient counter-ions and a 0.15 M solution of NaCl were also utilized to neutralize the system and to simulate the real cellular ionic concentrations, respectively. The MD protocol involved minimization, pre-production, and finally production MD simulation steps. In the minimization procedure, the entire system was allowed to relax for 2500 steps by the steepest descent approach. Then the temperature of the system was raised from 0 to 300 K with a small force constant on the enzyme to restrict any drastic changes. MD simulations were performed via NPT (constant number of atoms, constant pressure i.e. 1.01325 bar, and constant temperature i.e. 300 K) ensemble. The optimum system was finally subjected to produce MD simulations for 30 ns for the protein-ligand complex. During the simulation, every 1000 ps of the actual frame was stored. The dynamic behavior and structural changes of the systems were analyzed by the calculation of the RMSD and RMSF. Subsequently, the representative frames of the simulation were extracted based on the clustering method from the equilibrated trajectory system for investigating of ligand-protein complex interaction. The geometric criteria for H-bond interaction are the distance of 2.5 Å between the donor and acceptor atoms, a donor angle of ≥ 120° between the donor-hydrogen-acceptor atoms, and an acceptor angle of ≥ 90°. The π-cation happened between aromatic and charged groups within a 4.5 Å distance. The π-π and other non-specific hydrophobic side chains were seen within 3.6 Å of a ligand's aromatic or aliphatic carbons. Ionic interactions or polar interactions happened between two oppositely charged atoms within 3.7 Å of each other which did not involve in the H-bond. Water Bridge is hydrogen-bonded protein-ligand interaction mediated by water with a distance of 2.8 Å between the donor and acceptor atoms, a donor angle of ≥ 110° between the donor-hydrogen with acceptor atoms, and an acceptor angle of ≥ 90° between the hydrogen-acceptor-atoms. www.nature.com/scientificreports/