Structure-Based Scaffold Repurposing toward the Discovery of Novel Cholinesterase Inhibitors.

Cholinesterases (ChE) are well-known drug targets for the treatment of Alzheimer's disease (AD). In continuation of work to develop novel cholinesterase inhibitors, we utilized a structure-based scaffold repurposing approach and discovered six novel ChE inhibitors from our recently developed DNA gyrase inhibitor library. Among the identified hits, two compounds (denoted 3 and 18) were found to be the most potent inhibitor of acetylcholinesterase (AChE, IC50 = 6.10 ± 1.01 μM) and butyrylcholinesterase (BuChE, IC50 = 5.50 ± 0.007 μM), respectively. Compound 3 was responsible for the formation of H-bond and π-π stacking interactions within the active site of AChE. In contrast, compound 18 was well fitted in the choline-binding pocket and catalytic site of BuChE. Results obtained from in vitro cytotoxicity assays and in silico derived physicochemical and absorption, distribution, metabolism, and excretion (ADME) properties indicate that repurposed scaffold 3 and 18 could be potential drug candidates for further development as novel ChE inhibitors.

Introduction

Alzheimer’s disease (AD), the most common type of age-related neurodegenerative disorder, is characterized by a progressive loss of memory, cognitive function, and difficulties in performing daily activities.[1,2] According to the World Alzheimer’s Reports, approximately 50 million people worldwide were affected with dementia in 2019. Further, it is anticipated that this number will increase 3-fold by 2050.[3] In 2019, the annual economic and societal costs associated with dementia reached US$1 trillion, a figure set to double by 2030.[3] Various hypotheses previously reported have now been verified with regard to their influence on the pathogenesis of AD. These include β-amyloid aggregation,[4,5] tau protein hyperphosphorylation,[6] oxidative stress,[7] and dysfunctional cholinergic hypothesis.[8] Several treatments are currently based on the cholinergic hypothesis, which states that deficiency of the neurotransmitter acetylcholine (ACh) is the primary cause of AD. Moreover, this deficiency is linked with either the reduced production or degradation of ACh by cholinesterases.[9,10] Two different types of cholinesterase (ChE) enzymes, i.e., acetylcholinesterase (AChE) and butyrylcholinesterase (BuChE), are responsible for the hydrolysis of ACh within the brain.[11] Furthermore, recent reports suggest that AChE also plays an essential role in the aggregation of β-amyloid peptide and their deposition as senile plaques on the brain.[12] BuChE is a nonspecific cholinesterase as it also hydrolyzes other molecules such as adipoylcholine, benzoylcholine, and neurotoxic peptides.[10,13] In the healthy brain, AChE plays a significant role in the hydrolysis of ACh. However, in contrast, it is BuChE that hydrolyzes ACh in the AChE-deficient brain.[14] Further, in patients suffering from AD, AChE levels within the brain diminish over time, while BuChE activity remains either unchanged or increases to 165% to that of normal values.[14,15] Therefore, inhibition of both ChEs is considered a valid treatment of mild- to late-stage AD. Tacrine, the first ChE inhibitor approved for the treatment of AD, demonstrated potent activity against both AChE and BuChE. However, poor tolerability and hepatotoxic consequences resulted in its early market withdrawal (Figure ).[16] Rivastigmine, galantamine, and donepezil, currently approved for the symptomatic treatment of AD, are nonselective cholinesterase inhibitors.[17] Recent clinical studies have revealed that donepezil and rivastigmine can cause diarrhea and vomiting, respectively.[18−20] Due to these adverse effects and the devastating consequences of the disease, research to identify novel ChE inhibitors remains paramount.

Figure 1

Current and former ChE inhibitors used in the treatment of AD.

Drug discovery is a complex, time-consuming, and expensive process as newly developed drugs must pass many hurdles. As a result, only a limited number have been approved for use within the last few years.[21] Due to the importance of ChEs in AD drug advancement, diverse drug discovery approaches have been employed to achieve this goal. Among them, lead optimization and hybridization of drugs already approved are utilized by many researchers.[22,23] However, none of the designed compounds have been clinically approved. Within the last decade, the computer-aided drug design (CADD) approach has emerged as an important drug discovery tool. Such computational methods hasten the optimization and identification of lead candidates, thus saving time and money during the drug discovery process.[24,25] To date, several computational studies have successfully identified novel ChE inhibitors.[14,26−29] Therefore, utilization of computational methods to identify novel ChE lead compounds is a well-known approach to discover innovative treatments for AD.

Scaffold repurposing is a way of discovering new medicinal uses for compounds designed for other clinical indications. This concept is similar to drug repurposing, whereby clinically approved or investigational drugs are assessed for their efficacy against other medical ailments which were not within the original scope of the targeted conditions.[30] For example, aspirin, a well-known analgesic that has been in general use for over a century, was approved for the treatment of colorectal cancer in 2015.[31] Moreover, the availability of several hundred X-ray structures of AChE and BuChE co-crystallized with various ligands[22,32,33] provides further insights regarding essential features necessary for the development of novel ChEs inhibitors. Figure provides an overview of the significant structural differences between the active sites of AChE and BuChE. Human AChE (hAChE) and human BuChE (hBuChE) share 50% amino acid sequence homology,[34] with the active site of both enzymes consisting of a catalytic triad, and choline-binding and acyl-binding pockets. The active site cavity of hAChE consists of 14 aromatic amino acids,[27] although 6 are replaced by aliphatic amino acids compared to hBuChE.[35] In combination with scaffold repurposing, these structural features also need to be taken into account while designing novel ChE inhibitors.[36]

Figure 2

Major structural differences between hAChE (PDB entry 4EY7) and hBuChE (PDB entry 5K5E). Carbon atoms are represented in pink for hAChE and in yellow for hBuChE. The oxygen atoms are in red, and the nitrogen atoms are in blue.

Minocycline, a tetracyclic class of antibiotic, has been proposed as a candidate for AD due to its blood–brain barrier (BBB) permeability potential and promising results in clinical studies.[37,38] This report provides a basis for us to utilize various antibiotic structures to identify innovative treatments for AD. Recently, we identified a novel library of DNA gyrase-targeted antimicrobial compounds from the CoCoCo chemical database[39] and assessed their antimicrobial potential. Seven compounds demonstrated potent activity against multiple pathogenic bacteria.[40] In the present work, we aimed to utilize a structure-based scaffold repurposing approach using our designed DNA gyrase inhibitors library and performed molecular docking studies with AChE and BuChE. Some of the compounds demonstrated important interactions within the active sites of AChE and BuChE (Table ). As such, we then screened the compounds for their AChE and BuChE inhibitory potential using Ellman’s assay.[41] The in silico absorption, distribution, metabolism, and excretion (ADME) properties of the identified inhibitors were calculated, and the potential cytotoxic effects were tested in a PC12 cell line (Pheochromocytoma) at a concentration of 10 μM.

Table 1

AChE and BuChE Inhibitory Potential of Compounds 1–25 and the Standard Compound Tacrinea,b

compound no.	AChE % inhibition at 10 μM	AChE docking score (kcal/mol)	BuChE % inhibition at 10 μM	BuChE docking score (kcal/mol)
1	–	–8.49	14.6 ± 2.40	–5.88
2	–	–6.62	–	–4.07
3	58.80 ± 0.60c	–11.54	34.3 ± 1.80c	–6.03
4	8.16 ± 1.25	–9.98	–	–6.50
5	34.30 ± 1.60	–9.67	–	–5.14
6	3.62 ± 2.26	–10.42	–	–6.81
7	25.90 ± 1.70	–9.05	–	–5.65
8	–	–8.16	–	–5.44
9	22.70 ± 1.11	–10.84	–	–5.70
10	8.07 ± 2.17	–8.08	–	0.98
11	–	–9.91	–	–2.54
12	4.69 ± 1.62	–9.73	–	–5.47
13	31.60 ± 2.20	–9.21	–	–5.61
14	2.95 ± 1.71	–8.09	–	–5.93
15	1.79 ± 0.59	–6.71	–	–6.65
16	–	–6.42	–	–6.33
17	–	–6.32	–	–0.93
18	8.28 ± 2.16c	–14.2	55.3 ± 1.58c	–9.14
19	–	–6.94	–	–4.52
20	–	–7.27	–	–7.13
21	–	–9.74	–	–6.31
22	9.80 ± 3.40	–9.63	5.71 ± 2.14	–8.22
23	–	–7.86	–	–7.42
24	–	–8.88	–	–1.80
25	–	–6.46	–	–2.81
tacrine	66.40 ± 0.80c,d	–10.9	91.9 ± 0.40c,d	–7.08

Compounds were checked for pan assay interference compounds (PAINS). None were found to be PAINS.[45]

Values are the mean of three independent experiments.

IC50 of the compound (compound, EeAChE IC50, eqBuChE IC50): compound 3, 6.10 ± 1.01, 21.7 ± 4.4 μM; compound 18, 32.3 ± 4.2, 5.50 ± 0.007 μM; tacrine, 0.152 ± 0.006, 0.0150 ± 0.0001 μM.

Inhibition was tested at 1.67 μM. “–” refers to no inhibition.

Results

Molecular Docking and Enzyme Inhibition Studies

To determine lead compounds using the scaffold repurposing approach, initial docking studies were performed with 25 compounds and the following parameters were considered significant: (a) AChE—H-bonding interaction of the molecule with His447,[42,43] π–π interaction with Trp86 and Tyr337,[42,43] and entry of the molecule within the acyl pocket;[42,43] (b) BuChE—H-bonding interaction of the molecule with His438,[14,44] π–π interaction between the molecule Trp82,[14,34] Trp231,[14,26] and Phe329,[14,26] and entry of the molecule within the acyl pocket.[14,26] Seventeen compounds in AChE and eight compounds in BuChE showed promising interactions and were found to have comparable docking scores with tacrine. As such, in the next stage, we screened all 25 compounds for their percentage inhibition against Electrophorus electricus-derived AChE (EeAChE) and equine-derived BuChE (eqBuChE). Of those tested, compounds 3, 5, 7, 9, 13, and 18 showed >20% inhibition at a concentration of 10 μM. Of these, compounds 3 and 18 (Figure ) were potent inhibitors of AChE and BuChE, respectively. In contrast, compounds 5, 7, 9, and 13 were identified as selective AChE inhibitors (Table ). In the next stage, IC50 values were calculated for molecules 3 and 18. Compound 3 was the most potent AChE inhibitor within the series with an IC50 of 6.10 μM. In contrast, compound 18 had the highest potency against BuChE with an IC50 of 5.50 μM.

Figure 3

AChE and BuChE inhibitors identified through screening of the library of compounds. Refer to Table for percentage (%) inhibition and IC50 values of compounds 3 and 18.

Molecular Interactions of Compounds 3 and 18

A proposed docking model of compound 3 with the active site of AChE (hAChE) shows the carbonyl oxygen of one benzyl ring, and the nitrogen of another benzyl ring forming hydrogen-bond interactions with the hydroxy group of Tyr72 (Figure A). Further, the carbonyl oxygen of the quinazolinone ring is shown to interact via hydrogen bonding with Phe295. In addition, π–π stacking between the aromatic rings of compound 3 and Trp286 and Tyr337 was also predicted. A surface view of the active site of AChE and compound 3 (Figure B) reveals the binding orientation of the compound within the active site of AChE. In this representation, it can be observed that two benzylamine chains and an aliphatic chain on the quinazolinone ring are pointing out of the active site pocket of AChE. Therefore, removal of one benzylamine chain and the aliphatic chain might confer AChE inhibition and aid the future development of a compound 3-based AChE inhibitor.

Figure 4

(A) Docking model showing proposed interactions of 3 within the active site of hAChE (PDB entry 4EY7). Active site residues of hAChE are presented as sticks with carbon atoms represented in turquoise (orange for 3). The yellow dashed line represents a hydrogen bond, and the light blue dashed lines represent π–π stackings. (B) Surface view of the binding orientation of compound 3 within the active site of hAChE. Colors represent the electrostatic potential.

A proposed docking model of compound 18 with the active site of BuChE (hBuChE) depicts that compound 18 is well fitted with the active site of hBuChE (Figure A), with Figure B (surface view) demonstrating the binding orientation. Moreover, it is observed that the compound is positioned in such a way that the trifluorophenyl ring is within a choline-binding pocket, whereas the pyridine rings enter directly into the catalytic site of BuChE. In this confirmation, the methyl ether chain on the pyridine ring of compound 18 is pointed toward the catalytic active site residue Ser198. Rivastigmine, a carbamate-based clinically approved anti-AD drug, acts via carbamylation of the active site Ser198 residue of BuChE and ultimately halts hydrolysis of ACh. Therefore, modification at the methyl ether chain of compound 18 into the carbamate group could prove to be an efficacious carbamate-based BuChE inhibitor. Taken together, we hypothesized that designed carbamate-based analogues of 18 would demonstrate a similar mechanism of action to rivastigmine.

Figure 5

(A) Docking model showing proposed interactions of 18 within the active site of hBuChE (PDB entry 6QAA). Active site residues of hBuChE are represented as sticks with carbon atoms shown in blue (primarily) or yellow (for compound 18). (B) Surface view of the binding orientation of compound 18 within the active site of hBuChE. Colors represent the electrostatic potential.

Prediction of ADME Properties and Inherent Toxicity Studies

Poor pharmacological profiles are the principal reason for late-stage failure in drug discovery. Hence, it is essential to discern the inherent medicinal properties of the target molecules earlier in the pipeline.[46] Moreover, various physicochemical parameters must also be taken into account during drug development. These include molecular weight (MW), log P, number of hydrogen-bond (HB) donors and acceptors, and polar surface area (PSA). Furthermore, an ideal drug candidate should have good oral bioavailability and Caco-2 and MDCK cell permeability. In addition, since in this instance the target disease is AD, the drug must have the ability to cross the BBB and reach the site of action.[47] Current ADME evaluation methods are ultimately time-consuming and costly due to the high level of animal testing required. Further, evaluation of a large number of molecules makes in vivo methodologies highly ineffectual.[48] In contrast, in silico determination of ADME properties is a promising screening tool as it reduces the time, cost, and labor required to perform such testing.[49] These in silico techniques were also found to be useful in the development of novel cholinesterase inhibitors.[27,28,50−52] Currently, various online tools and computer-aided methods such as molinspiration are available to assess ADME properties. Among them, QikProp software[53] calculates various physiochemical descriptors, provides an estimate regarding BBB permeability potential, oral bioavailability, Caco-2 and MDCK cell permeability, and the number of metabolic reactions involved during metabolism. The QikProp software provides ranges for these properties by comparing a particular molecule’s properties with 95% of known drugs. As such, numerous significant descriptors and pharmacologically relevant properties were calculated and analyzed for all six identified inhibitors (Table ). Our results show that the physicochemical values of our identified inhibitors are within the range suggested by QikProp software. Further, all inhibitors possess druglike characteristics based on their MW, HD, hydrogen acceptors (HA), and PSA. The significant feature of central nervous system (CNS) active drugs is dependent upon their ability to cross the BBB and display CNS activity. QP log BB is the predicted brain/blood partition coefficient calculated by QikProp and indicates compound BBB permeability, which must be within the range of −3.0 to 1.2.[53] Specifically, the higher the QP log BB value, the greater the ability of the compound to pass through the BBB. QP log BB values of all inhibitors were found to be within the designated range, thus implying the target compounds have the potential to penetrate the CNS. PSA is another crucial predictor for BBB permeability.[54] All inhibitors were within the set limit, further confirming their BBB permeability potential. Along with QP log BB, apparent Caco-2 (QPPCaco) and MDCK cell (QPPMDCK) permeability are considered vital parameters to predict a compound’s distribution within the human body. Importantly, all identified inhibitors exhibited high permeability in both Caco-2 and MDCK cells. Overall, our two lead inhibitors (compounds 3 and 18) possess the appropriate pharmacokinetic profiles required for distribution in the human body and a high probability of being able to successfully penetrate the BBB. As such, both have the potential as lead candidates in AD drug development.

Table 2

Physicochemical Properties of Compounds 3, 5, 7, 9, 13, and 18 Used for the Prediction of ADME Profile

descriptors (recommended values(53))	3	5	7	9	13	18
MW (130.0–725.0)	715.8	584.6	584.6	407.4	573.7	467.5
QP log Po/w (−2.0 to 6.5)	7.75	6.65	6.73	4.07	–7.49	3.83
SASA (300.0–1000.0)	1112	874	890	702	920	771
PSA (7.0–200.0)	139	116	116	122	107	75
donorHB (0.0–6.0)	1	1	1	2	1	1
accptHB (2.0–20.0)	9.5	7	7	4	6.2	7
% HOA (<25, poor; >80, high)	89.9	68.9	70	89.8	78.2	95.8
QP log BB (−3.0 to 1.2)	–1.66	–1.5	–1.52	–1.77	–1.71	–0.22
QPPCaco (<25, poor; >500, great)	331	41	45	151	72	393
QPPMDCK (<25, poor; >500, great)	1353	87.4	96.6	64.3	139	1275
Met (1–8)	5	5	3	2	3	7

Due to the promising computational ADME data, we assessed the inherent cytotoxic effects of all six inhibitors using the (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium) (MTS) assay at 10 μM concentration. Interestingly, none of the identified inhibitors were found to be toxic at 10 μM (Figure ). Thus, from the enzymatic inhibition studies, in silico ADME properties, and in vitro experimental toxicity studies, we propose that all identified inhibitors could be potential candidates in the development of future ChE inhibitors for the treatment of AD.

Figure 6

Viability of PC12 cells post-exposure to compounds 3, 5, 7, 9, 13, and 18 and the standard. PC12 cells were treated with compounds 3, 5, 7, 9, 13, and 18 and standard (tacrine) for 48 h. Then, the cell viability was determined using MTS analyses and the percent viability of PC12 cells was calculated against growth media control (untreated). Tacrine (10 μM) was used as the standard AD drug. Data are represented as % mean ± SEM of three independent experiments, each tested in triplicate (n = 9).

Discussion

Traditional AD drug development has faced numerous obstacles as the associated methods are laborious, costly, and compounded by a low success rate.[55] Scaffold repurposing is an efficient way to detect novel leads from existing scaffolds known for other biological activities. For example, topiramate, originally approved as an antiepileptic, is now prescribed for the treatment of obesity, while ketoconazole, an antifungal drug, is currently used for Cushing’s syndrome.[56] Similarly, different classes of drugs such as angiotensin receptor blockers and calcium channel blockers have shown promising results for the treatment of AD.[57]In silico scaffold repurposing is an inexpensive and emerging tool to ascertain new applications of existing drugs.[58,59] Moreover, structure-based virtual screening identified raltegravir, an HIV-1 integrase inhibitor, as an adjuvant in the treatment of cancer.[60] Hence, it is crucial to use an in silico scaffold repurposing approach to discover novel lead pharmaceuticals for any disease, including AD.

AChE and BuChE both play an important role in AD pathogenesis and, to date, have been thoroughly investigated and considered as potential targets for AD drug discovery.[11] In the present study, we utilized a structure-based scaffold repurposing approach to discover novel ChE inhibitors from our DNA gyrase inhibitor library. Initially, we performed docking studies with both AChE and BuChE to determine binding efficacy, and then assessed inhibitory potential in vitro. With this approach, we discovered that compounds 3 and 18 are not only a novel chemical class of ChE inhibitors but also nontoxic at 10 μM against the PC12 cell lines. Further, both molecules showed excellent in silico ADME properties, thereby indicating their potential as possible treatments of AD.

This study identified six novel ChE inhibitors via structure-based scaffold repurposing with compounds 3 and 18 being most potent AChE and BuChE inhibitors, respectively. However, some of the screened compounds that showed higher docking scores in their respective targets were inactive when tested in the enzyme inhibition assays. Therefore, to avoid false-positive docking results, it is important to perform docking validation studies and check binding orientation differences with regard to the docked and co-crystallized ligands. Our results showed that both ligands demonstrated similar docking orientations, hence further validating our results. Such false-positives might be a result of conformational changes in the ChE enzymes upon binding of ligands and the case-dependent performances of docking algorithms in prioritizing true-positives.[61] Further, receptor plasticity also plays a significant role in false-positive docking results.[62] Interestingly, a study conducted by Pahl et al. reported that compounds that showed higher docking scores were found to have zero inhibitory action.[63] Therefore, other approaches such as consensus scoring and/or a combination of multiple scoring functions are now utilized by researchers to avoid false-positive docking results.[64]

Recently, Kumar et al. used in silico repurposing on antipsychotic drugs for the identification of novel leads for the treatment of AD.[59] However, the authors did not perform any enzymatic inhibition assays, but instead suggested in vitro enzymatic assays to validate their in silico results. In contrast, we used our in-house designed DNA gyrase inhibitor library to identify novel leads for AD and prove the inhibitory potential of the lead compounds using in vitro enzyme inhibition assays.

In view of the scaffold novelty of the identified inhibitors, we found 18 references within SciFinder that contain the concept “Quinazoline cholinesterase.” The majority of the references belong to the quinazoline scaffold fused with other rings or substitution at the second carbon position of the quinazoline ring. However, our identified compound 3 contains a quinazolinone ring system with alkyl chain attached to its amide nitrogen and substitution at the seventh carbon position of the quinazolinone ring, which marks this as a unique scaffold finding for an AChE inhibitor. We performed a similar SciFinder search for compound 18, and we discovered 241 references that contain the concept of “pyridine cholinesterase.” However, most references belonging to the molecular hybridization approach showed that the pyridine ring was linked to the known ChE pharmacophores, such as in tacrine or carbazole. In addition, we performed structural similarity (>75%) search of both identified inhibitors on SciFinder and found that none of the references reported ChE activity, further confirming the scaffold novelty of the inhibitors. Overall, this work has successfully demonstrated that a structure-based scaffold repurposing approach can be used for the discovery of novel ChE inhibitors.

Conclusions

In conclusion, we successfully utilized a scaffold repurposing approach and identified six compounds that demonstrated ChE enzymatic inhibition at 10 μM concentration. Of these, compounds 3 and 18 were identified as the most potent and showed important molecular interactions with AChE and BuChE, respectively. In computational pharmacokinetic prediction studies, both compounds were identified as BBB-permeable, orally bioavailable with no cytotoxic effects. However, other approaches, such as cell-based BBB studies and in vivo models are needed to validate the pharmacological and toxicological properties of our target compounds (3 and 18).

Materials and Methods

Materials and General Methods

All general reagents were supplied from various commercial suppliers unless stated otherwise. All experiments were conducted at room temperature (23 °C) unless specified. Analytical grade commercial chemicals and reagents were used to conduct all experiments. Milli-Q water was also utilized during the study. Tacrine (A3773), acetylthiocholine iodide (ATCI) (01480), 5,5′-dithiobis (2-nitrobenzoic acid) (D8130; Ellman’s reagent; DTNB), dimethyl sulfoxide (DMSO)-d6 (151874), acetylcholinesterase from E. electricus (EeAChE) (C3389), butyrylcholinesterase from equine serum (eqBuChE) (C7512), and Trizma HCl buffer (pH 7.4, 93313) were purchased from Sigma-Aldrich. DMSO (AJA2225) was purchased from Ajax Finechem. Compounds (25) used in this study were purchased from Enamine, ChemDiv, and Princeton Biomolecular Research. Absorbances were measured at 400 nm using a POLARstar OPTIMA microplate reader (BMG Labtech, Offenburg, Germany). Low-protein-binding Eppendorf tubes and Nuclon 96-well plates were used for Ellman’s assay. Statistical analysis was performed using GraphPad Prism 8 (version 8.2.1). 1H nuclear magnetic resonance (NMR) experiments were performed on a Bruker AV600 spectrometer. Chemical shifts (δ) are reported in parts per million (ppm) and are referenced to the DMSO-d6 and CDCl3 peak, which were calibrated at 2.50 and 7.26 ppm, respectively. Mass was recorded with a Thermo Scientific LTQ XL linear ion trap mass spectrometer. ChemBioDraw Ultra (version 18.1.0.535) was used to calculate accurate theoretical mass values. The acquired 1H NMR spectra and mass values are consistent with the structures reported for compounds 3, 5, 7, 9, 13, and 18. All computational calculations were performed on a Dell workstation with Intel i7-8700 3.2 GHz processor; 16 GB RAM; Intel UHD Graphics 630 card; 500 GB primary solid-state storage drive, while running the 64-bit Windows 10 operating system.

Cholinesterase Inhibitory Assay

The cholinesterase inhibitory potential of all compounds was determined using a modified Ellman’s method according to previously published protocols.[14,22] Tacrine was used as a standard positive control. Briefly, to the 96-well plate, 120 μL of DTNB, 40 μL of test compounds, and 40 μL of EeAChE or eqBuChE were added. The plate was incubated at room temperature for 30 min; then, the absorbance was measured at 400 nm every 2 min for 10 min. Next, 40 μL of ATCI was added, and the absorbance was measured at 400 nm every 1 min for 10 min. The reaction velocity (V) was obtained by fitting the equation A = 13V + A0 to a plot of absorbance (A) versus time (t) by linear regression.

The % inhibition at each concentration was estimated using the following equationwhere V0 is the reaction velocity before the addition of ATCI and V1 is the reaction velocity calculated after the addition of ATCI. The IC50 was calculated using the following equationwhere X is the log compound concentration; Y is % inhibition; “top” and “bottom” are plateaus in the same unit as Y; and log IC50 is the same log unit as X.

Docking Studies

Molecular modeling simulations of all screened compounds were performed using Maestro (version 12.1.013), implemented from the Schrödinger Molecular Modelling Suite-2019-3. Initially, low energy conformers and tautomers of all molecules were performed by the LigPrep module. Structural coordinates of hAChE and hBuChE were taken from the Protein Data Bank (PDB) with PDB entries 4EY7 and 6QAA used, respectively. The raw PDB protein structures were prepared by giving preliminary treatment, including the addition of hydrogen atoms and missing side chains, refinement of loops, and then finally minimizing using an OPLS-2005 force field. The grids for docking simulations were generated using the structural coordinates of bound co-crystallized ligands [donepezil in 4EY7; (S)-N-(1-((2-cycloheptylethyl)amino)-3-(1H-indol-3-yl)-1-oxopropan-2-yl)butan-1-aminium for 6QAA]. All compounds were docked using the Glide module in extra-precision (XP) mode.[65] The ligands were kept flexible, whereas the receptor remained rigid throughout the docking studies. The lowest energy conformations were selected, and ligand interactions with the target proteins were determined. In validation studies, the co-crystallized ligands, donepezil for 4EY7 (root mean square deviation (RMSD) = 0.42 Å) and (S)-N-(1-((2-cycloheptylethyl)amino)-3-(1H-indol-3-yl)-1-oxopropan-2-yl)butan-1-aminium for 6QAA (RMSD = 1.55 Å), showed an acceptable superposition with a docking pose of respective ligands.

Assessment of Inherent Cytotoxic Effects of Compounds 3, 5, 7, 9, 13, and 18 Using PC12 Cells

The ability of selective compounds to promote or maintain PC12 cell viability was assessed in cell growth assays utilizing a tetrazolium compound (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium) (MTS)-based assay. The CellTiter 96 Aqueous One Solution Cell Proliferation Assay kit (Promega Corporation) was used to determine the number of viable cells. The cells were prepared in growth media, and 10 000 cells (50 μL) in each well were added to 96-well plates and incubated for 16 h at 37 °C and 5% CO2. Subsequently, treatments were prepared in growth media at 2× concentration and 50 μL was added to each well and incubated at 37 °C, 5% CO2 for 48 h. At the end of the assay, 20 μL of MTS reagent was added to each well, and the cells were incubated for a further 2 h at 37 °C, 5% CO2 to allow the tetrazolium salt color to develop. Absorbance was read at 490 nm in a microplate reader (Molecular Devices), and the results were expressed as percentage viability compared to untreated control cells cultured in complete growth media.