Novel Anti-Acetylcholinesterase Effect of Euonymus laxiflorus Champ. Extracts via Experimental and In Silico Studies

Alzheimer’s disease (AD) is the most common form of dementia, which is recorded as a global health issue. Natural acetylcholinesterase inhibitors (AChEIs) are considered a helpful therapy for the management of symptoms of patients with mild-to-moderate AD. This work aimed to investigate and characterize Euonymus laxiflorus Champ. (ELC) as a natural source of AChEIs compounds via in vitro and virtual studies. The screening parts used, including the leaves, heartwood, and trunk bark of ELC, revealed that the trunk bark extract possessed the highest activity, phenolics and flavonoid content. The in vitro anti-Alzheimer activity of ELC trunk bark was notably reclaimed for the first time with comparable effect (IC50 = 0.332 mg/mL) as that of a commercial AChEI, berberine chloride (IC50 = 0.314 mg/mL). Among various solvents, methanol was the most suitable to extract ELC trunk bark with the highest activity. Twenty-one secondary metabolites (1–21) were identified from ELC trunk bark extract, based on GCMS and UHPLC analyses. Of these, 10 volatile compounds were identified from this herbal extract for the first time. One phenolic (11) and seven flavonoid compounds (15–21) were also newly found in this herbal extract. Of the identified compounds, chlorogenic acid (11), epigallocatechin gallate (12), epicatechin (13), apigetrin (18), and quercetin (20) were major compounds with a significant content of 395.8–2481.5 μg/g of dried extract. According to docking-based simulation, compounds (11–19, and 21) demonstrated more effective inhibitory activity than berberine chloride, with good binding energy (DS values: −12.3 to −14.4 kcal/mol) and acceptable RMSD values (0.77–1.75 Å). In general, these identified compounds processed drug properties and were non-toxic for human use, based on Lipinski’s rule of five and ADMET analyses.


Introduction
Alzheimer's disease (AD) is the most common form of dementia. This disease has been recorded as a global health issue, with the number of patients worldwide reaching approximately 50 million; the number of cases of AD may double every 5 years, and it is estimated to reach 152 million by 2050 [1,2]. AD significantly affects individuals, the families of the patients, and the global economy, with the annual incurred costs estimated to reach about USD 1 trillion worldwide [1]. Up to date, no therapies may completely cure AD. However, some current therapies may significantly help to manage the symptoms, provide temporary relief, and increase the quality of life of patients with AD [3]. Among provide temporary relief, and increase the quality of life of patients with AD [3]. Among the therapies, the utilization of herbs for the management of AD has been considered since they are natural sources of bioactive compounds and are less toxic [4][5][6][7][8][9][10].
Vietnam, a tropical country, has been ranked as the sixteenth most biodiverse worldwide, with more than 10,000 herbal species recorded and 4000 species in use as traditional medicine [11]. Thus, studies based on the investigation of the biological effects and chemical profiles of medicinal plants from Vietnam have received increasing interest in recent years [12][13][14][15][16][17]. However, only a few studies and new records of herbs with potential anti-Alzheimer effects of this biodiverse area have been reported [15][16][17]; as such, the discovery of medicinal plants with biological activities related to Alzheimer's drugs from Vietnam has been received with great interest.
Euonymus laxiflorus Champ. (ELC), a medicinal plant, is wildly grown in Vietnam, India, Cambodia, Myanmar, and China [12,14,18]. ELC has been found to be a rich source of bioactivities, including antioxidant and anti-nitric oxide (anti-NO) properties [19], a potent anti-enzyme targeting anti-diabetes, such as anti-α-glucosidase and anti-α-amylase [12,14], and a significant reduction in the blood glucose of normal and diabetic rats with fewer or no side effects [20,21]. Some target compounds of α-glucosidase, α-amylase, antioxidant, and the anti-NO effect were purified and identified in our earlier works [13,19,22]. However, no data are available on anti-acetylcholinesterase (AChE) targeting anti-Alzheimer, as well as the concerning constituents of this herb.
As a part of our ongoing objective to investigate novel bioactivity and the active constituents of the herbal ELC for the development of this herb as a natural and functional food and/or as drugs, in this investigation, we collected this herb from Dak Lak province of Vietnam and determined the bioactivities. This study is the first to report the anti-AChE activity of ELC. Further screening of the most active part used, including the heartwood, trunk bark, leaves of ELC, and the chemical profiles, the interaction of active molecules toward the targeting enzyme acetylcholinesterase, as well as Lipinski's rule of five and ADMET-based pharmacokinetics and pharmacology analyses via computational study are also presented in this study. All steps of our work are summarized in Scheme 1. The most bioactive extract was investigated for its chemical profiles, (D) then the virtual study was conducted for the prediction of active compounds and drug discovery.

Gas Chromatography-Mass Spectrometry (GCMS) Analysis
The herbal extract was dissolved in methanol (MeOH), then purified by solid-phase extraction using the QuEChERS method. For further analysis, GC (Thermo Trace GC Ultra, USA) and ITQ900 (Thermo, Waltham, MA, USA) were conducted. A TG-SQC capillary column (30 m × 0.25 mm × 0.25 µm) was utilized for the GCMS analysis. Helium (99.999%), a carrier gas, was set at a constant flow rate of 1 mL/min. The sample solution (1 µL) was injected in a split ratio of 10:1. The temperature of the injector and the ion source were set at 250 and 230 • C, respectively. The temperature program of the oven was set at 70 • C (isothermal for 2 min) and increased up to 280 • C with an increasing speed of 15 • C/min, ending with a 10 min isothermal at 280 • C. MS data were at 70 eV, a scanning interval time of 0.5 s, and for fragments from 50 to 650 Da. The compounds were identified via comparison with reported compounds using compounds data of the Mass Spectra Library (NIST 17.L and Wiley).

Enzyme Inhibition Assay
AChE was used to carry out the inhibitory assay using Ellman's method [23] with modifications. A 0.05 M phosphate buffer, pH 8.0, was used in this protocol. Additionally, 60 µL of the herbal extract was mixed with the same volume of 0.5 mM enzyme solution and 120 µL of phosphate buffer, and kept at 25 • C for 15 min in a flat-bottom 96-well plate. Agent 5, 5 -dithiobis-2-nitrobenzoic acid (DTNB, 30 µL of 0.003 M) and 40 µL of 0.002 M acetylthiocholine iodide (ATCI) were added into the mixture to start the reaction. The reaction was maintained at 25 • C for 10 min before the absorbance was measured at the wavelength of 415 nm. For the control group, 60 µL of phosphate buffer was used instead of 60 µL of the herbal extract solution and the same condition was used for the calculation of enzyme inhibition activity using the following equation: where A E is the absorbance value measured at 415 nm of the reaction containing herbal extract and enzyme, while A C is the absorbance value at 415 nm of the reaction containing enzyme and phosphate buffer instead of the herbal extract. The samples and berberine chloride (commercial AChE inhibitor) were dissolved in dimethylsulfoxide and eluted at various concentrations using a 0.05 M phosphate buffer and then used for the tests.

Docking Simulation
Virtual analysis was conducted using the MOE-2015.10 software to predict the active metabolite-concerning inhibition against the targeting enzyme. This virtual protocol was performed following the three typical steps mentioned in the previous reports [24,25].
Preparation of protein structure: The structure data of enzyme AChE were obtained from the Research Collaboratory for Structural Bioinformatics Protein Data Bank for the preparation of their three-dimensional structures through the use of MOE-2015. 10. The most active sites on the enzyme were found using the site finder function of the MOE software. A virtual pH 7 was used for the preparation of enzyme structure. Preparation of ligand structures: The structures of identified compounds (ligands) from the herbal extracts and commercial inhibitors were prepared using the Chem-BioOffice 2018 software and further optimized using the MOE software. The following parameters were required to prepare the structure of ligands: virtual pH 7, force field MMFF94x; R-Field 1:80; cutoff, rigid water molecules, space group p1, cell size 10, 10, 10; cell shape 90, 90, 90; gradient 0.01 RMS kcal·mol −1 A −2 . Docking ligands into enzymes and the obtained out-put data: The prepared ligands were docked into the active site of AChE using the MOE software. As major out-put for analysis, DS value, RMSD value, linkage types, compositions of amino acids, and the linkages' distances were harvested.

The Lipinski Rule of Five and ADMET Analyses
The online software accessed at (http://www.scfbioiitd.res.in/software/drugdesign/ lipinski.jsp (accessed on 15 May 2023)) was utilized for the Lipinski Rule of Five performance. A web tool called SwissADME (http://www.swissadme.ch/ (accessed on 15 May 2023)) was used for the prediction of some pharmacokinetic parameters of AD-MET. The out-put data of theoretical interpretations of pharmacokinetic parameters have been previously described [26] and used as a public reference online, accessed at (http://biosig.unimelb.edu.au/pkcsm/theory (accessed on 15 May 2023)).

New Records of the Potential Anti-Acetylcholinesterase Effect of Euonymus laxiflorus Champ
To evaluate ELC as a source of anti-Alzheimer drugs, the ELC trunk bark (ELCTB) was extracted with methanol, butanol, ethyl acetate, n-hexane, and water, and then tested for their inhibition against AChE, a key enzyme in the discovery of anti-Alzheimer drugs [27]. The biological effect was recorded and shown under the IC 50 value, which is the sample concentration that reduces 50% of enzymatic effect; the lower the value a sample yields, the greater the effect it displays. The results are shown in Table 1, and Figure A1 shows that the MeOH extract of ELCTB could most efficiently inhibit AChE with a low IC 50 value (0.323 mg/mL), comparable to that of a commercial inhibitor compound, berberine chloride (IC 50 = 0.314 mg/mL). This meOH extract was also found to inhibit AChE with a high max inhibition value of 97.0%, while other extracts showed a max effect of less than 80.1%. To screen for the most functional part to be used, the heartwood, trunk bark, and leaves of ELC were also extracted by meOH and used for the activity tests. The trunk bark MeOH extract showed the most AChE inhibitory activity with the smallest IC 50 values of 0.336 mg/mL (Table 2 and Figure A2). This part used was also rich in total content of polyphenol (567.3 GAE/g dry extracts) and flavonoid (335.2 QE/g dry extracts). Thus, the MeOH extract of the trunk bark was further examined. Recently, ELCTB has been found showing several medical effects, including antioxidant, anti-NO, anti-α-glucosidase, anti-α-amylase, and antidiabetic effects, and ELCTB has also been used by folks for the treatment of several diseases [12,14,[19][20][21]. However, the potential AChE inhibitory effect of ELC was newly recorded in this study; as such, the findings herein contributed to enriching the catalog of medical effects of this plant.
Clinical trials using herb extracts in AD treatment were conducted in some works. Some herbal extracts in China achieved good effects in clinical trials for treating dementia [28]. Ginkgo biloba extracts were trialed widely in AD treatment and trials were conducted for almost 3-6 months [29]. In the Ginkgo One Tablet A Day (GOTADAY) trial, the dose of the one-daily drug containing 240 mg of G. biloba extract showed a significant improvement in neuropsychiatric symptoms and cognition for dementia patients [30]. Overall, other reports of this herb showed non-potential or unreliable effects in trials [29]. Furthermore, a patient group using a high dose of Panax ginseng recorded significant results based on clinical dementia rating and Alzheimer disease assessment scale when compared with the control [31]. In Iran, Salvia officinalis extract was tested with a fixed dose of 60 drops/day for mild-to-moderate Alzheimer's patients in a 4-month period. It supported improved cognition after 16 weeks of treatment; however, some of its side effects need further confirmation [32]. Based on some of the above literature, using herb extracts as AD treatment agents is also considered a promising orientation. ELC extracts can also be suggested as a potential source for in-depth trials. This herb was demonstrated non-toxic for normal cells and mouse models and it has also been used as a traditional medicine by ethnic minorities in Vietnam for a long time [22].

Insight into the Interactions and Energy Binding Bioactive Compounds toward Targeting Enzyme-Acetylcholinesterase via Docking Study
The active compounds, the interaction and binding energy of the bioactive inhibitors toward AChE were predicted through the docking study using the MOE-2015.10 software. The data of AChE protein structure was obtained from the Worldwide Protein Data Bank. The most active site (on AChE for docking ligands) was found using the site finder function of the MOE software. Based on the out-put data of MOE, 25 binding sites on AChE were determined. The sizes, residues, and 3D structures of these binding sites were mapped and are presented in the Appendix A (Table A1). The ligands may bind to various sites on the enzyme; however, only the most active site was chosen to be presented and discussed in detail. Based on the site finder function of MOE, and the pre-screening results, binding site 1 is suggested as the most active binding site for further investigation. This active site was found (Figure 3a) to contain 39 residues. To determine whether binding site 1 was covered by the catalytic site of enzyme or not, the CASTp3.0 server was used in the prediction of the AChE catalytic site, which has the volume and the surface area of 904.278 Å 3 and 529.676 Å 2 , respectively (Figure 3b, Table A2). Figure 3 indicated that binding site 1 was not located in the catalytic site. Thus, all these inhibitors show a high possibility of no binding to the catalytic site of AChE.  In the virtual study, RMSD and DS are important parameters to determine whether a compound (ligand/inhibitor) may bind to the target protein and inhibit its enzymatic activity [51,52]. For a successful binding, when a ligand interacts with a target protein with an RMSD value less than 2.0 Å, the binding is considered to be significant and widely accepted for further virtual analysis [51]. As shown in Table 5, all the identified compounds (1-21) interacted with AChE with low RMSD values in the range of 0.63-1.75 Å, while the interaction of the commercial inhibitor berberine chloride had an RMSD value of 1.65 Å. This result suggested the successful binding of all the ligands to the target protein with acceptable RMSD values. For virtual evaluation of effective inhibition, DS was commonly used. When a compound binds to the target enzyme with a DS value less than −3.20 kcal/mol, it is proposed as an enzyme inhibitor [52]. In comparison, the lower the DS value an inhibitor has, the greater is its inhibitory effect. As summarized in Table 5, all the compounds could bind to AChE with DS values greater than −6.3 kcal/mol, indicating that they may be possible AChE candidates. The commercial inhibitor berberine chloride showed high inhibition against AChE, with a DS value of −12.1 kcal/mol. Compared with the commercial inhibitor, eleven compounds (1-10, 20) exhibited weaker AChE inhibition (DS value in the range of −6.3 to −11.0 kcal/mol) than berberine chloride (22). Other compounds (11-19 and 21) demonstrated higher activity than berberine chloride (22) with low DS values of −12.3 to −14.4 kcal/mol. Overall, the order of inhibition of the AChE protein by active compounds was as follows: 17 > 13 > 18 > 16 > 14 > 11 = 19 > 15 > 12 > 21 > 22 (commercial inhibitor) > 20 > 9 > 7 > 1 > 3 > 2 > 8 > 6 > 10 > 4 > 5. The result of the in vitro test and virtual analysis could nearly fit and corroborate each other, indicating that the meOH extract of ELC trunk bark showed a high effectiveness against AChE, due to harboring various major active compounds (11)(12)(13)(14)(15)(16)(17)(18)(19)21). In the virtual study, RMSD and DS are important parameters to determine whether a compound (ligand/inhibitor) may bind to the target protein and inhibit its enzymatic activity [51,52]. For a successful binding, when a ligand interacts with a target protein with an RMSD value less than 2.0 Å, the binding is considered to be significant and widely accepted for further virtual analysis [51]. As shown in Table 5, all the identified compounds (1-21) interacted with AChE with low RMSD values in the range of 0.63-1.75 Å, while the interaction of the commercial inhibitor berberine chloride had an RMSD value of 1.65 Å. This result suggested the successful binding of all the ligands to the target protein with acceptable RMSD values. For virtual evaluation of effective inhibition, DS was commonly used. When a compound binds to the target enzyme with a DS value less than −3.20 kcal/mol, it is proposed as an enzyme inhibitor [52]. In comparison, the lower the DS value an inhibitor has, the greater is its inhibitory effect. As summarized in Table 5, all the compounds could bind to AChE with DS values greater than −6.3 kcal/mol, indicating that they may be possible AChE candidates. The commercial inhibitor berberine chloride showed high inhibition against AChE, with a DS value of −12.1 kcal/mol. Compared with the commercial inhibitor, eleven compounds (1-10, 20) exhibited weaker AChE inhibition (DS value in the range of −6.3 to −11.0 kcal/mol) than berberine chloride (22). Other compounds (11-19 and 21) demonstrated higher activity than berberine chloride (22) with low DS values of −12.3 to −14.4 kcal/mol. Overall, the order of inhibition of the AChE protein by active compounds was as follows: 17 > 13 > 18 > 16 > 14 > 11 = 19 > 15 > 12 > 21 > 22 (commercial inhibitor) > 20 > 9 > 7 > 1 > 3 > 2 > 8 > 6 > 10 > 4 > 5. The result of the in vitro test and virtual analysis could nearly fit and corroborate each other, indicating that the meOH extract of ELC trunk bark showed a high effectiveness against AChE, due to harboring various major active compounds (11)(12)(13)(14)(15)(16)(17)(18)(19)21). To investigate the interaction of ligands and AChE, the detailed binding at the active site were recorded, as presented in Table 6 and Figure 4. The active inhibitor compounds (11)(12)(13)(14)(15)(16)(17)(18)(19)21) and commercial inhibitor (22) were examined. Rutin (17) demonstrated the highest inhibition against AChE by interacting with four amino acids, including His400, Glu199, Trp84, and Asp72 of this enzyme, resulting in the creation of seven linkages (4 Hdonor, 2 H-pi, 1 pi-H). Among these, this ligand (17) was found to interact with His400 and Asp72 and formed one H-donor linkage and one pi-H linkage, respectively, while it interacted with Glu199 and Trp84 to create two linkages (H-donor), and three linkages (2 H-pi, 1 pi-H), respectively. This was followed by epicatechin (13), which showed efficient inhibition against AChE at a low DS value (−14.3 kcal/mol) by binding and interacting with this enzyme through four amino acids, Ser81, Asn85, Ser200, and Glu199, to generate four H-donor linkages. The next five compounds (11, 14, 16, 18, and 19) could also bind tightly to AChE with DS values lower than −13.1 kcal/mol by interacting with five, seven, two, two, and two amino acids to form five, seven, four, three, and four linkages, respectively. The binding of apigetrin (18) to AChE was through the least interaction (three linkages) but possessed the best binding energy (DS value of −13.9 kcal/mol), while other ligands had greater interaction with AChE (four to seven linkages) but showed weaker bind energies (DS values in the range of −13.1 to −13.8 kcal/mol). These results indicated that the DS values may be independent of the number of interactions between the ligand and the enzyme. The next three ligands (12,15, and 21) also showed a slightly higher binding effect to AChE than the commercial inhibitor (22). These compounds, 12, 15, and 21, could bind to AChE by interacting with two, four, and three amino acids and generated four, four, and three linkages, respectively, while the commercial inhibitor (22) binds to AChE through only one amino acid, forming only one linkage. Among the prominent amino acids contained in the binding site, Glu199 had an important role in the interaction with all the above-mentioned ligands (11-19, and 21). This amino acid (Glu199) was bound to the  ligands 11, 13, 14, 15, 20, and 21 to generate one H-donor linkage (binding linkage energy in the range of −0.6 to −5.8 kcal/mol) with each ligand, while it interacted with ligands 12, 17, and 18 to generate two H-donor linkages (binding linkage energy in the range of −0.5 to −3.8 kcal/mol) with each ligand. Especially, Glu199 was found to bind with ligands 16 and 19, and for each ligand, up to three H-donor linkages (binding linkage energy in the range of −1.1 to −4.0 kcal/mol) were formed.
These potential molecules were further investigated in their frontier molecular orbitals. The data of the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) of these compounds are presented in Figure 5. These molecular structures showed their low E HOMO values in the range of − 5.65 to − 8.65 eV. This indicates their significant electronic stability (commonly accepted under − 5 eV). In the previous investigation [53], theoretical complexes of the tetrylone family with E HOMO values of − 3 to − 7 eV were also found as highly stable. It has been evidenced that all structures possess an insulation-to-semiconduction energy gap (3.2 eV < EG < 9 eV), showing good intermolecular binding capability toward protein structures [54], and this was further explained based on the super-exchange theory [55,56] and the electron hopping model [57]. In this study, the energy gap values of the inhibitors were found in the range of 3.77 to 5.61 eV; as such, they have potential intermolecular binding capability toward the targeting enzyme.
Based on RMSD, DS values and some data of frontier molecular orbitals recorded, some secondary metabolites (compounds 11-19, 21) contained in the methanol extract of ELC trunk bark may be suggested as potential candidates of AChE inhibitors. However, some works summarized by Pagadala et al. [58] indicated unreliable binding affinity predictions using docking studies. Several reports also indicated some drawbacks of the docking study [59][60][61]. Thus, further works should be performed, including purifications of active compounds and testing activity via in vitro, in vivo, and clinical trials for development of these compounds to be drugs.

Lipinski's Rule of Five and ADMET-Based Pharmacokinetics and Pharmacology
Lipinski's rules have been applied to evaluate the drug-likeness of compounds; the five rules include "molecular mass must be less than 500 Da (rule 1), high lipophilicity with LogP value < 5 (rule 2), hydrogen bond donors < 5 (rule 3), hydrogen bond acceptors < 10 (rule 4), and the molar refractivity should be between 40-130 (rule 5)". A compound is considered to have drug-likeness properties and has a high possibility of being a drug when it satisfies at least 2/5 of Lipinski's rules. As presented in Table 7, all the identified compounds complied with five of Lipinski's rules, except compound 2, which complied with four of Lipinski's rules. Thus, these compounds have a high probability of successfully being developed as a drug.
The ADMET properties of these compounds and commercial inhibitors were also compared, and the data are presented in Appendix A (Figures A3 and A4). In general, these compounds in the MeOH extract of ELCTB also showed good ADMET properties in the required allotted limitation. In addition, all the active inhibitor compounds (11-19 and 21) and the commercial inhibitor (22) were not toxic for human use. Furthermore, the evidence of safety on normal cells of some compounds was indicated via in vitro or in vivo tests. Chlorogenic acid was recorded to affect cancer cells without normal cell influence [62]. EGCG was found as toxic on HuCC-T1 cancer cells but safe on the viability of 293T normal cells [63]. In another report, EGCG from green tea at 40-200 microM caused a significant death for some tumor cell lines, but only 1% of the WI38 normal cells growth was affected in the same condition [64]. Epicatechin and Epicatechin gallate were discovered to cause death for various cancer cells but had no effect on normal cells in some reports [65][66][67][68]. Vitexin has a non-effect on bronchial epithelial 16HBE normal cells [69]. Rutin, Apigetrin, and Myricetin all are potential and safe adjuvant chemotherapeutic agents with trivial toxicity and side effects via in vivo and clinical Life 2023, 13, 1281 12 of 25 trials [70][71][72]. The scientific proof also demonstrated that Quercetin and Apigenin have no or low effects on normal cells [71,73]. Almost phenolics in this study were showed safe for normal cells. However, the safe evidence for all compounds detected still needs to be performed continually in further research via in vitro, in vivo, and clinical trials.    Based on RMSD, DS values and some data of frontier molecular orbitals recorded, some secondary metabolites (compounds 11-19, 21) contained in the methanol extract of ELC trunk bark may be suggested as potential candidates of AChE inhibitors. However, some works summarized by Pagadala et al. [58] indicated unreliable binding affinity predictions using docking studies. Several reports also indicated some drawbacks of the docking study [59][60][61]. Thus, further works should be performed, including purifications