Novel and Potent Acetylcholinesterase Inhibitors for the Treatment of Alzheimer’s Disease from Natural (±)-7,8-Dihydroxy-3-methyl-isochroman-4-one

Alzheimer’s disease (AD) is a neurodegenerative disease that causes memory and cognitive decline as well as behavioral problems. It is a progressive and well recognized complex disease; therefore, it is very urgent to develop novel and effective anti-AD drugs. In this study, a series of novel isochroman-4-one derivatives from natural (±)-7,8-dihydroxy-3-methyl-isochroman-4-one [(±)-XJP] were designed and synthesized, and their anti-AD potential was evaluated. Among them, compound 10a [(Z)-3-acetyl-1-benzyl-4-((6,7-dimethoxy-4-oxoisochroman-3-ylidene)methyl)pyridin-1-ium bromide] possessed potent anti-acetylcholinesterase (AChE) activity as well as modest antioxidant activity. Further molecular modeling and kinetic investigations revealed that compound 10a was a dual-binding inhibitor that binds to both catalytic anionic site (CAS) and peripheral anionic site (PAS) of the enzyme AChE. In addition, compound 10a exhibited low cytotoxicity and moderate anti-Aβ aggregation efficacy. Moreover, the in silico screening suggested that these compounds could pass across the blood–brain barrier with high penetration. These findings show that compound 10a was a promising lead from a natural product with potent AChE inhibitory activity and deserves to be further developed for the prevention and treatment of AD.


Introduction
Alzheimer's disease (AD) is a neurodegenerative disease and the most common type of dementia with the symptoms of progressive memory loss, cognitive impairment, changes of personality, and linguistic disorders [1][2][3]. There are currently more than 50 million AD patients worldwide, with an estimated increase to 139 million patients by 2050, leading to a significant financial burden on families and society [4][5][6][7][8][9].
The senile plaque (SP) generated by β-amyloid (Aβ) and neurofibrillary tangles (NFTs) made of phosphorylated tau proteins in the hippocampus are considered as the main pathogenic etiologies of AD [10]. The pathological mechanism of AD is complex, and several hypotheses have been proposed, including cholinergic injury hypothesis [11,12], β-amyloid protein cascade hypothesis [13], tau protein hyperphosphorylation hypothesis [14], metal ion poisoning hypothesis [15], calcium imbalance hypothesis [16], chronic inflammation hypothesis [17], and oxygen free radical damage hypothesis [18]. Among these hypotheses, the cholinergic injury is the most widely accepted and its clinical effectiveness has been approved. In the human brain, acetylcholine functions as a cholinergic The indenone fragment of donepezil occupies PAS of AChE. Due to the structural similarity between indenone and isochromanone, the indenone fragment of donepezil was cleverly designed by replacing it with the isochromanone structure of (±)-XJP. We hoped that the isochromanone structure could occupy PAS as well. The benzyl piperidine fragment of donepezil is the key pharmacophore that occupies CAS, and, combined with our previous studies [30][31][32], we retained the cyclic structure of this fragment and replaced the piperidine ring with a pyridine ring. The central region of the AChE active pocket becomes narrower and consequently restricts the passage of ring structures containing larger substituents; the bottom region of the pocket becomes relatively larger and therefore contains several lateral cavities that could be further occupied to improve the affinity of a compound on AChE ( Figure 2B). In a previous work [31], we fused (±)-XJP with a donepezil analog possessing a pyridine ring to obtain a series of new compounds with no substituents on the pyridine ring, and the most potent compound had good AChE inhibitory activity with IC 50 value of 21.1 ± 0.8 nM. On this basis, when designing new molecules, we introduced acetyl or aminocarbonyl groups that would occupy less space on the pyridine ring for the following reasons: the smaller acyl structure does not affect benzyl pyridine fragment to access CAS; and the spatial configuration of acyl group and hydrogen binding donors or acceptors that they contain has the potential to form more interactions with the active pocket. In our previous study, the structural modification of (±)-XJP provided an analog of it, named (±)-XJP-B, which displayed more potent activity than natural (±)-XJP. Therefore, by fusing the pharmacophores of (±)-XJP-B and donepezil, we designed and synthesized thirty-five new 4-isochromanone derivatives and evaluated their AChE inhibition to investigate the potential for AD treatment (Figure 2A).

Synthesis
The synthesis of new isochroman-4-one derivatives 10a-10s and 13a-13p is depicted in Scheme 1. Briefly, the synthetic route is divided into two parts: the first part is synthesis of key intermediate 7, and the second part is synthesis of the target compounds. Firstly

In vitro Inhibition of AChE Enzyme and Structure-Activity Relationship
The inhibitory potencies of compounds 10a-10s and 13a-13p toward AChE were evaluated using the Ellman's assay [23]. Donepezil was used as a positive control. The anti-AChE activities for all target compounds expressed as IC50 values were summarized in Table 1. The results in Table 1 indicated that most of the target compounds exhibited potent inhibitory activities against AChE. Overall, fourteen compounds had better inhib-

In Vitro Inhibition of AChE Enzyme and Structure-Activity Relationship
The inhibitory potencies of compounds 10a-10s and 13a-13p toward AChE were evaluated using the Ellman's assay [23]. Donepezil was used as a positive control. The anti-AChE activities for all target compounds expressed as IC 50 values were summarized in Table 1. The results in Table 1 indicated that most of the target compounds exhibited potent inhibitory activities against AChE. Overall, fourteen compounds had better inhibitory activities than positive control donepezil (IC 50 = 12.06 nM). Among them, compound 10a (IC 50 = 1.61 nM) was the most potent compound in the first series of compounds 10a-10s, showing the best inhibitory activity of all the target compounds. Compound 13b (IC 50 = 3.54 nM) was the best compound in the second series of compounds 13a-13p. Analyzing the results of activity from the perspective of substituents on the benzyl group, it could be found that the substitution of different groups and positions on the benzene ring has a significant differentiation on the inhibitory activity. On the one hand, based on the substituted position in benzene ring, the order of inhibitory potency against AChE was: ortho > meta > para. Furthermore, for the different substituents at the same position, the activity order of the first series of compounds was H > F > Me > Cl > Br > NO 2 > CF 3 , and the activity order of the second series was F > H > Me > Cl > Br > CF 3 ; it could be seen that the more appropriate substituents were H (10a) or F (13b), and the less active substituents were CF 3 (10s and 13p) in these two series, respectively, indicating the order of activity was not much related to the electrical effect of the substituent groups and it may be also related to the steric effect of substituent groups. By analysis of the substituents on the pyridine ring, we found that the first series of compounds 10a-10s performed better in activity overall. In summary, the above results suggested that the acetyl substitution on the pyridine ring (10a-10s) was better than the carbamoyl substitution (13a-13p), and the carbonyl moiety may be a key group in the series of skeleton structures. We chose the most potent compound 10a for further investigations.

Kinetic Study of AChE Inhibition
In order to analyze the AChE inhibitory mechanism of these compounds, the most potent compound 10a was selected for kinetic testing. The mechanistic inhibition study of compound 10a on the AChE was carried out through kinetic parameters such as maximal velocity (Vmax), inhibitory concentration (Ki) and Michalis-Menten dissociation constant (Km). The reciprocal Lineweaver-Burk plot of compound 10a on AChE demonstrated that as the concentration of inhibitor increased, both slopes and intercepts increased, and the lines generated by compound 10a with different concentrations eventually converged in the second quadrant ( Figure 3). The results in Figure 3 clearly demonstrated that compound 10a is a mixed-type inhibitor for AChE, suggesting that compound 10a bound both CAS and PAS of AChE.  Figure 3 clearly demonstrated that co pound 10a is a mixed-type inhibitor for AChE, suggesting that compound 10a bound b CAS and PAS of AChE.

In Vitro Cytotoxicity of Compounds 10a and 13b
To evaluate the cytotoxicity of these new compounds, cell viability was tested on SY5Y cells. The representative compounds 10a and 13b in these two series were selec for cytotoxicity. As shown in Figure

In Vitro Cytotoxicity of Compounds 10a and 13b
To evaluate the cytotoxicity of these new compounds, cell viability was tested on SH-SY5Y cells. The representative compounds 10a and 13b in these two series were selected for cytotoxicity. As shown in Figure 4, both compounds 10a and 13b did not significantly alter cell viability at concentrations of 10, 25 and 50 µM, indicating that compounds 10a and 13b had no neurotoxicity up to the concentration at 50 µM.
To evaluate the cytotoxicity of these new compounds, cell viability was tested on SH-SY5Y cells. The representative compounds 10a and 13b in these two series were selected for cytotoxicity. As shown in Figure 4, both compounds 10a and 13b did not significantly alter cell viability at concentrations of 10, 25 and 50 μM, indicating that compounds 10a and 13b had no neurotoxicity up to the concentration at 50 μM.

In Vitro Antioxidant Activity of Compound 10a
DPPH assay was used to evaluate the free radical scavenging ability of the representative compound 10a, while Trolox was used as a control [33]. As shown in Table 2, compound 10a displayed moderate radical scavenging activity (31.89% ± 3.23% at 50 µM). The result showed that compound 10a possessed moderate antioxidant activity at a concentration of 50 µM.

Inhibition of Aβ Self-Aggregation of Compound 10a
Aggregation of β-amyloid peptide (Aβ) is one of the main histopathological hallmarks of AD [34]. Therefore, we selected the optimal compound 10a using a thioflavin T assay to determine the prevention of Aβ aggregation. As shown in Table 3, compound 10a had a moderate inhibitory effect on the Aβ aggregation, with an inhibitory rate of 56.95 ± 2.68% at a concentration of 25 µM. The result showed that compound 10a exhibited moderate anti-Aβ aggregation efficacy.

Molecular Docking
In order to assess how ligand interacts with enzyme, docking simulations were performed using the Glide program of Schrodinger. The X-ray crystal structure of human AChE (PDB entry 4EY7) was downloaded from PDB. Based on the in vitro inhibitory results, compound 10a was selected as a typical ligand for the evaluation. The docking results  Figure 5. The isochromanone fragment was located at the PAS site outside pocket, and the benzyl-substituted pyridine fragment was located at the CAS site inside pocket. Some previous studies showed a proposed H-bonding of donepezil's C=O group with human AChE; however, we did not find the proposed H-bonding of C=O group of compound 10a with human AChE, but we found that the pharmacophore of compound 10a based on isochromanone structure of ±(XJP-B) could occupy PAS well. Among them, the benzene ring of the isochromanone fragment forms a π-π interaction with Trp-286 residue, the pyridine ring forms a π-π interaction with Tyr-337 residue, and the benzyl ring formed a π-π interaction with Trp-86 residue (brown-dashed line); the acetyl group on the pyridine ring formed a 2.2 Å hydrogen bond with Tyr-124 residue (black-dashed line); the N atom on pyridine ring was connected to Try-337 residue and Trp-86 residue, the two aromatic rings of the radical formed three π-cations interaction (blue-dashed line). All these results indicated that compound 10a can simultaneously bind to the PAS and CAS of AChE, which is in accordance with the results of the kinetic study.

Prediction of Toxicity and Drug-Likeness Properties
To evaluate the drug-likeness characteristics and predict toxicity of the target compounds, blood-brain barrier (BBB) penetration, human intestinal absorption (HIA), caco-2 cells permeability, AMES mutagenesis, and rat acute toxicity were calculated using ad-metSAR web-based application tool [35]. Some compounds with better activity were selected for this experiment. As shown in Table 4, according to the predicted values for BBB penetration, all compounds would be able to penetrate into the CNS. However, in the case of HIA and Caco-2 permeability, all compounds performed less well. Compound 10n with an interposition nitro substitution on the benzyl group showed the only exception to the AMES mutagenicity, which showed potential genotoxicity ("+" sign), indicating that nitro was not a suitable substituent. Overall, the results indicated that these new compounds exhibit good drug-like properties, but there was scope for further improvement.

Prediction of Toxicity and Drug-Likeness Properties
To evaluate the drug-likeness characteristics and predict toxicity of the target compounds, blood-brain barrier (BBB) penetration, human intestinal absorption (HIA), caco-2 cells permeability, AMES mutagenesis, and rat acute toxicity were calculated using admet-SAR web-based application tool [35]. Some compounds with better activity were selected for this experiment. As shown in Table 4, according to the predicted values for BBB penetration, all compounds would be able to penetrate into the CNS. However, in the case of HIA and Caco-2 permeability, all compounds performed less well. Compound 10n with an interposition nitro substitution on the benzyl group showed the only exception to the AMES mutagenicity, which showed potential genotoxicity ("+" sign), indicating that nitro was not a suitable substituent. Overall, the results indicated that these new compounds exhibit good drug-like properties, but there was scope for further improvement.

Materials and Methods
All commercially available chemicals and reagents were analytical grade and used without further purification only if otherwise noted. The 1 H NMR and 13 C NMR were recorded on Bruker-300 spectrometers (Bruker Company, Karlsruhe, Germany) using TMS as an internal standard. High-resolution mass spectrometry was accomplished on an Agilent 1100-LC-MSD-Trap/SL mass spectrometer (Agilent Technologies Inc., Santa Clara, CA, USA). Huanghai HSGF 254 silica gel plates (Yantai, China) were used for TLC. Silica gel 60 H (200-300 mesh), general chromatography was performed manufactured by Qingdao Haiyang Chemical Group Co., Ltd. (Qingdao, China).

Synthesis of (2-Bromo-4,5-dimethoxyphenyl)methanol (2)
To a solution of 2-bromo-4,5-dimethoxybenzaldehyde (1) (81.61 mmol, 1 eq) in 100 mL methanol, sodium borohydride (40.80 mmol, 0.5 eq) was slowly added at 0 • C and the mixture was stirred at the same temperature. After the completion of the reaction was detected by TLC, crushed ice was added into the mixture and stirred for another 1 h. After filtration, the filtrate was washed with cold water and collected as a white solid [32].

Synthesis of 2-((2-Bromo-4,5-dimethoxybenzyl)oxy)acetic Acid (4)
To a solution of tert-butyl 2-((2-bromo-4,5-dimethoxybenzyl)oxy)acetate (3) in 100 mL methanol, a solution of sodium methanol (101.04 mmol, 2.5 eq) in methanol was slowly added; the mixture was stirred at room temperature for about 15 min, and an appropriate amount of H 2 O was slowly added to the mixture with vigorous stirring for 5 min. After the reaction was completed, the solvent was concentrated and the aqueous layer was extracted with ethyl acetate (3 × 200 mL). The pH of the aqueous layer was adjusted to 2 by slowly adding concentrated hydrochloric, and then the aqueous layer was extracted three times with ethyl acetate. The combined organic layers were washed with brine (100 mL) and dried over anhydrous Na 2 SO 4 . The organic layers were removed in vacuo to give crude white solid 4 [32].
3.1.6. Synthesis of 6,7-Dimethoxyisochroman-4-one (6) Intermediate 2-((2-bromo-4,5-dimethoxybenzyl)oxy)-N-methoxy-N-methylacetamide 5 (14.36 mmol, 1 eq) was dissolved in anhydrous THF under an N 2 atmosphere and placed at a low temperature of −78 • C. A solution of 1.3 M tert-butyllithium (35.90 mmol, 2.5 eq) in n-pentane was added and the mixture was stirred for another 10 min. At the end of the reaction, a saturated aqueous ammonium chloride solution was added to quench the reaction and the mixture was brought to room temperature. The residue was extracted with ethyl acetate (EA) (3 × 200 mL), the organic layers were combined, washed with brine, dried over anhydrous Na 2 SO 4 , and the organic layers were concentrated and purified by column chromatography to give pale yellow solid 6. 3.1.7. Synthesis of (Z)-3-((3-Bromopyridin-4-yl)methylene)-6,7-dimethoxyisochroman -4-one (7) To a solution of Intermediate 6 (4.80 mmol, 1 eq) in dry toluene, 3-bromopyridine-4aldehyde (7.20 mmol, 1.5 eq) and p-toluenesulfonic acid monohydrate (7.20 mmol, 1.5 eq) were added sequentially and the mixture stirred at reflux for 4 h. At the end of the reaction, the reaction system was cooled to room temperature, and the pH was adjusted to 8 by adding saturated sodium bicarbonate. The mixture was extracted with dichloromethane (3 × 200 mL). The organic layer was washed with brine and dried over anhydrous Na 2 SO 4 , and purified by column chromatography to give yellow solid 7. 1 H NMR (300 MHz, CDCl 3 ) To a solution of intermediate 7 (3.46 mmol, 1 eq) in dry toluene at nitrogen atmosphere, p-(diaziridine acetonide)palladium (0.05 mmol, 0.04 eq), tributyl(1-ethoxyethylene)tin (1.37 mmol, 1.2 eq) and triphenylphosphine (0.09 mmol, 0.08 eq) were added sequentially and stirred at reflux for 10 h, then brought to room temperature. After filtration, the filtrate was concentrated and purified by column chromatography to give yellow solid, which was further recrystallized by dichloromethane/petroleum ether to give intermediate 8.

Synthesis of (Z)-3-((3-Acetylpyridin-4-yl)methylene)-6,7-dimethoxyisochroman -4-one (9)
Intermediate 8 dissolved in 6 N aqueous HCl and stirred for 1 h at 50 • C. After the reaction completed, the hydrochloric acid was neutralized by adding saturated sodium bicarbonate solution and the organic layer was extracted with dichloromethane. The organic layer was washed with saturated salt water, dried over anhydrous sodium sulphate, concentrated and recrystallized using a mixture of dichloromethane/petroleum to give yellow solid 9.
3.2.5. Cytotoxicity Assay on SH-SY5Y Cells SH-SY5Y cells were grown in a 25 cm 2 culture flask in fresh MEM/F12 medium, supplemented with 10% fetal bovine serum, 100 U/mL penicillin, and 100 g/mL streptomycin, and incubated at 37 • C with 5% CO 2 . Cells were seeded at a density of 1 × 10 4 cells per well in a 96-well plate. After incubation for 24 h, medium was removed. Compounds 10a or 13b (10, 20, 50 µM) were prepared with serum-free media and added into each well of the plate for 24 h. After 24 h treatment, 10 µL of 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2-H-tetrazolium bromide was added into each well. After 4 h of incubation at 37 • C, 100 µL of DMSO was added to dissolve the formazan crystals, and the mixture's absorbance was measured at 490 nm under a microplate reader (FC/K3, Thermo, Waltham, MA, USA). The survival rate was estimated using the formula Ae/Ab × 100%, where Ae and Ab represent the absorbance of SH-SY5Y cells in the presence and absence of the tested drugs, respectively. Each experiment was replicated three times.

Molecular Docking Study
The protein structure of AChE (PDB ID: 4EY7) was downloaded from Protein Data Bank (https://www.rcsb.org) (accessed on 15 January 2022). AChE was refined and an active pocket for AChE was defined using the protein preparation and receptor grid generating components of Schrodinger software. To improve tiny molecule structures, the ligand preparation module was employed. Finally, AChE was docked using the ligand docking module.

Conclusions
In conclusion, by fusing the pharmacophores of (±)-XJP-B and donepezil, a series of novel isochroman-4-one derivatives were designed, synthesized and evaluated as AChE inhibitors. The AChE inhibitory activity of these new compounds was overall strong, with IC 50 values in the nanomolar range, and representative compound 10a [(Z)-3-acetyl-1-benzyl-4-((6,7-dimethoxy-4-oxoisochroman-3-ylidene)methyl)pyridin-1-ium bromide] was the most potent molecule (IC 50 = 1.61 nM). Diverse substituents, including electrondonating groups and electron-drawing groups, were introduced into different positions on the benzene ring of the target compounds to investigate the substituent effects (the electronic effect, steric effect and position of substituted groups) and systematically analyze the structure activity relationships. Further molecular docking studies revealed that compound 10a occupies the active pocket of AChE well, forming secondary bonds with the pocket's amino acid residues. With the results of low cytotoxicity, anti-oxidation and inhibition of self-induced Aβ aggregation, it is strongly suggested that compound 10a is a promising lead compound to be further studied for discovery and development of novel and effective anti-AD drugs.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/molecules27103090/s1, The spectra of 1 H-NMR and 13 C-NMR for the target compounds.
Author Contributions: X.L., P.Z. and J.X. conceived and designed the experiments; X.L., P.Z., Y.J. and J.L. (Junda Li) performed the synthesis; T.L., L.L. and Z.Z. performed the activity tests, and X.L. and Y.J. performed molecular docking studies and performed analysis studies; X.L., J.X., H.Y., J.L. (Jie Liu) and Z.Z. wrote the paper. All authors have read and agreed to the published version of the manuscript.

Funding:
The authors thank the National Natural Science Foundation of China (No. 81874289) for financial support. This study was financed in part by the "Double First-Class" University project CPU2018GY04, China Pharmaceutical University.

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
The data presented in this study are available within the article and Supplementary Materials.