Synthesis and Evaluation of Novel Ligustrazine Derivatives as Multi-Targeted Inhibitors for the Treatment of Alzheimer’s Disease

A series of novel ligustrazine derivatives 8a–r were designed, synthesized, and evaluated as multi-targeted inhibitors for anti-Alzheimer’s disease (AD) drug discovery. The results showed that most of them exhibited a potent ability to inhibit both ChEs, with a high selectivity towards AChE. In particular, compounds 8q and 8r had the greatest inhibitory abilities for AChE, with IC50 values of 1.39 and 0.25 nM, respectively, and the highest selectivity towards AChE (for 8q, IC50 BuChE/IC50 AChE = 2.91 × 106; for 8r, IC50 BuChE/IC50 AChE = 1.32 × 107). Of note, 8q and 8r also presented potent inhibitory activities against Aβ aggregation, with IC50 values of 17.36 µM and 49.14 µM, respectively. Further cellular experiments demonstrated that the potent compounds 8q and 8r had no obvious cytotoxicity in either HepG2 cells or SH-SY5Y cells, even at a high concentration of 500 μM. Besides, a combined Lineweaver-Burk plot and molecular docking study revealed that these compounds might act as mixed-type inhibitors to exhibit such effects via selectively targeting both the catalytic active site (CAS) and the peripheral anionic site (PAS) of AChEs. Taken together, these results suggested that further development of these compounds should be of great interest.


Introduction
Alzheimer's disease (AD) is a progressive neurodegenerative brain disorder that is manifested as dementia, cognitive impairment, memory loss, severe behavioral abnormalities, and ultimately death [1][2][3][4]. To date, AD is thought to be a complex, multifactorial syndrome, with many related molecular lesions contributing to its pathogenesis. Based on the existing hypothesis, AD is characterized as amyloid plaques, neurofibrillary tangles, inflammatory intermediates, and reactive oxygen species (ROS), and imposes neuronal death via a complex array of networked pathways [5].
According to the cholinergic hypothesis, the cognitive and memory deterioration of AD is due to a loss of cholinergic function in the central nervous system. Acetylcholinesterase (AChE) and butyrylcholinesterase (BuChE) are two major cholinesterases (ChE) involved in the hydrolysis and regulation of choline in vertebrates. Indeed, current treatment of AD mainly focuses on the inhibition of AChE activity in order to rectify the deficiency of cerebral acetylcholine [6,7]. However, the role

Chemistry
The ligustrazine hybrids were conveniently synthesized according to the synthetic routes shown in Scheme 1. First, compound 5 was prepared based on a previous reported method [24]. Then, it was treated with 3-bromopropan-1-ol in the presence of EDC·HCl and DMAP to yield 3-bromopropyl 3,5,6-trimethylpyrazine-2-carboxylate 6 in a yield of 73%. Finally, intermediate 6 was reacted with the corresponding carboxylic acid derivatives 7a-r in the presence of DMF and anhydrous K 2 CO 3 to afford the desired products 8a-r. The structures of synthesized compounds were confirmed by 1 H and 13 C-NMR and HRMS. It should be noted that 3-(3,5,6-trimethylpyrazin-2-yl)acrylic acid (7q) was smoothly synthesized according to the literature, with a total yield of 60% [28]. and 13 C-NMR and HRMS. It should be noted that 3-(3,5,6-trimethylpyrazin-2-yl)acrylic acid (7q) was smoothly synthesized according to the literature, with a total yield of 60% [28].

In Vitro Inhibition Studies on AChE, BuChE and Aβ (1-42) Self-Induced Aggregation.
To determine the potential of the target compounds 8a-r for the treatment of AD, their AChE (from electric eel) and BuChE (from equine serum) inhibitory activities were evaluated by using the method of Ellman [29], in which tacrine and galanthamine were employed as the reference compounds. The IC50 values of all compounds for ChEs (AChE and BuChE) and the affinity ratios were summarized as shown in Table 1. The results demonstrated that most of the tested compounds had a potent capability to inhibit AChE, even with an IC50 value at the nanomolar level. In sharp contrast, these compounds showed relatively weak inhibitory activities towards BuChE (with an IC50 value in the millimol grade). Obviously, these compounds displayed a good selectivity for AChE over BuChE, and the ratios of IC50 BuChE/IC50 AChE affinity values ranged from 1.14 × 10 2 to 1. 32  To determine the potential of the target compounds 8a-r for the treatment of AD, their AChE (from electric eel) and BuChE (from equine serum) inhibitory activities were evaluated by using the method of Ellman [29], in which tacrine and galanthamine were employed as the reference compounds. The IC 50 values of all compounds for ChEs (AChE and BuChE) and the affinity ratios were summarized as shown in Table 1. The results demonstrated that most of the tested compounds had a potent capability to inhibit AChE, even with an IC 50 value at the nanomolar level. In sharp contrast, these compounds showed relatively weak inhibitory activities towards BuChE (with an IC 50 value in the millimol grade). Obviously, these compounds displayed a good selectivity for AChE over BuChE, and the ratios of IC 50 BuChE/IC 50 AChE affinity values ranged from 1.14 × 10 2 to 1.32 × 10 7 .
As shown in Table 1, the reference compound-tacrine had a potent inhibitory activity, with an IC 50 value of 73.36 nM, which was in good agreement with previous reports [27,30]. Interestingly, among these, compound 8r, bearing a picolinic acid group, showed the most potent inhibition for AChE, with an IC 50 value of 0.25 nM, and the potency was 293-times stronger than tacrine. However, its IC 50 value against BuChE was 3.30 mM, which was much weaker than AChE. Moreover, the greatest selectivity index was also obtained for this compound, with an IC 50 BuChE/IC 50 AChE value of 1.32 × 10 7 . It indicated that compound 8r had an exclusive selectivity for AChE over BuChE. Furthermore, an SAR analysis suggested that the electron density of the benzene ring moiety in the product played a significant role in determining the inhibitory activity of AChE, and in general, the compounds bearing the electron-donating substituent exhibited more potent AChE inhibitory activity than those bearing the electron-withdrawing group. For instance, compounds 8a-8e, in which the two methoxyl substituents were installed in different positions of the benzene ring, showed strong AChE inhibitory activities, with ranges of IC 50 values from 4.12 to 387.9 nM. However, the homogeneous class of compounds 8g-h and 8m, bearing a bromo substituent in the benzene ring, exhibited relatively low AChE inhibitory activities compared with 8a-e. Moreover, compounds 8l-n exhibited AChE inhibitory effects, with IC 50 values that ranged from 303.4 µM to 7.16 mM, which was less active than 8a-k (from 4.12 nM to 993.39 µM), suggesting that the introduction of the methylene part between the ester group and the benzene ring was harmful to AChE inhibitory activity. Interestingly, the introduction of the conjugated vinyl group into 8d that gave the compound 8o, demonstrated potent AChE inhibitory effects, with an IC 50 value of 3.24 nM. A similar conclusion was observed in comparison with 8l and 8p. The results revealed that the introduction of a proper and relatively rigid group between the ester group and the benzene ring was beneficial to AChE inhibitory activity. Of note, replacement of the benzene ring moiety with the N-heterocyclic ring, such as pyrazine or pyridine, led to a dramatic increase in inhibitory activity, since the obtained compounds 8q and 8r were found to be the most potent inhibitors of AChE, with excellent selectivity towards AChE (for 8q, IC 50 = 1.39 nM, IC 50 BuChE/IC 50 AChE = 1.32 × 10 7 ; for 8r, IC 50 = 0.25 nM, IC 50 BuChE/IC 50 AChE = 2.91 × 10 6 ), indicating that the introduction of the nitrogen atom contributed to AChE inhibitory activities, probably due to the additional affinity with the active pocket of AChE.
Inspired by the above results, these synthesized compounds were further tested for their abilities to inhibit self-mediated aggregation of Aβ (1-42) by using a thioflavin-T fluorescence method [29] and employing the well-known tacrine as a standard reference. As summarized in Table 1, the results showed that most of the ligustrazine derivatives apparently prevented self-mediated Aβ aggregation. In particular, compounds 8b (IC 50 = 3.66 µM), 8j (IC 50 = 7.12 µM), and 8k (IC 50 = 5.10 µM) showed higher potency than that of the reference compound tacrine (12.21 µM). Gratefully, the two best potent AChE inhibitors 8q and 8r also possessed acceptable inhibitory activities against self-induced Aβ (1-42) aggregation, with IC 50 values of 17.36 µM and 49.14 µM, respectively. Taken together, the data from the ChEs and Aβ aggregation test revealed that these developed compounds might act as novel multi-targeted "hits" for potent anti-AD drug discovery. Thus, further development of such compounds should be of great interest.

Inhibition of Aβ (1-42) Fibril Formation Monitored by Transmission Electron Microscopy (TEM)
To further confirm the ability of these compounds in inhibiting Aβ (1-42) aggregation, the inhibitory activity of the selected compound 8q was monitored by using TEM [31]. As shown in Figure 1, after 24 h of incubation at 37 • C, Aβ (1-42) alone aggregated into well-defined Aβ fibrils and amorphous deposits were observed (Figure 1a,b). In sharp contrast, no obvious Aβ fibril was detected in the presence of either tacrine or 8q (Figure 1c,d) under identical conditions. Moreover, compound 8q could obviously alleviate the formation of Aβ amorphous deposits compared with tacrine. These results were in agreement with the data from the thioflavin-T fluorescence experiment, also suggesting that 8q can serve as a promising candidate for the treatment of AD.

Kinetic Characterization of AChE Inhibition
Encouraged by the above results, we further selected the most potent AChE inhibitor 8q for kinetic analysis to uncover the inhibition mechanism by using the graphical analysis of the reciprocal Lineweaver-Burk plot. As demonstrated in Figure 2, the results showed that the plots of 1/V versus 1/[S] gave a family of straight lines, with different slopes that intersected one another in the second quadrant. With the increase of the concentration of 8q, the values of Vmax descended, but the values of Km kept increasing, indicating that it was a mixed-type inhibitor of AChE.

Kinetic Characterization of AChE Inhibition
Encouraged by the above results, we further selected the most potent AChE inhibitor 8q for kinetic analysis to uncover the inhibition mechanism by using the graphical analysis of the reciprocal Lineweaver-Burk plot. As demonstrated in Figure 2, the results showed that the plots of 1/V versus 1/[S] gave a family of straight lines, with different slopes that intersected one another in the second quadrant. With the increase of the concentration of 8q, the values of Vmax descended, but the values of Km kept increasing, indicating that it was a mixed-type inhibitor of AChE.

Molecular Modeling Study of AChE
To gain further insight into the molecular basis, the most active compounds 8q and 8r were selected to dock into the active site of TcAChE. It is well-known that, in previous crystallographic studies, significant conformation changes were observed in the active site of TcAChE [31][32][33][34][35]. For example, bis-tacrine derivatives with a 7-carbon spacer (PDB code 2ckm) induced the rotation of side chains of Tyr70 and Trp279 to promote the sandwich interaction with the active PAS site in the pocket of AChE, while their analogues with a 5-carbon spacer (PDB code 2cmf) caused the structural alteration of Trp279-Ser291's loop in the CAS site [32]. Based on this information, we herein used 2ckm and 2cmf as molecular modeling for 8q containing a 7-carbon spacer and 8r with a 5-carbon spacer, respectively. The results showed that no large-scale structural displacement occurred in 8q or 8r binding to AChE, mainly due to their small molecular size compared with tacrine derivatives. Alternatively, the crystal structure of the HupA-TcAChE complex (PDB code 1vot), where most residues were retained in the native conformations, was chosen as the molecular docking model [33]. Gratefully, re-docking of the original ligand into the active site gave an acceptable RMSD difference of 0.5384, which verified the reliability of our strategy. Figure 3 shows that both 8q and 8r spanned the active gorge with an extended conformation. The proximal pyrazine ring was bound in the CAS at the bottom, stacking against the side chain of Trp84, while the distal pyrazine/pyridine ring was present in the PAS near the mouth, forming a π-π interaction with Trp279. Moreover, some hydrophobic or van der Waals interactions might occur between 8q or 8r and the surrounding residues, such as Tyr70, Asp72, Asn85, Gly119, Tyr121, Ser122, Gly123, Tyr130, Leu282, Ser286, Ile287, Phe288, Phe290, Phe330, Phe331, Tyr334, Gly335, Trp432, His440, and Tyr442. Such interactions all contributed to their potent affinities with AChE. It should be noted that, owing to the fact that BuChE did not have the PAS [36], these compounds could not form effective binding interactions with the active pocket of BuChE, thereby giving a specific selectivity towards AChE. Finally, the result also revealed that compound 8q could bind to both the CAS and the PAS of AChE.

Molecular Modeling Study of AChE
To gain further insight into the molecular basis, the most active compounds 8q and 8r were selected to dock into the active site of TcAChE. It is well-known that, in previous crystallographic studies, significant conformation changes were observed in the active site of TcAChE [31][32][33][34][35]. For example, bis-tacrine derivatives with a 7-carbon spacer (PDB code 2ckm) induced the rotation of side chains of Tyr70 and Trp279 to promote the sandwich interaction with the active PAS site in the pocket of AChE, while their analogues with a 5-carbon spacer (PDB code 2cmf) caused the structural alteration of Trp279-Ser291's loop in the CAS site [32]. Based on this information, we herein used 2ckm and 2cmf as molecular modeling for 8q containing a 7-carbon spacer and 8r with a 5-carbon spacer, respectively. The results showed that no large-scale structural displacement occurred in 8q or 8r binding to AChE, mainly due to their small molecular size compared with tacrine derivatives. Alternatively, the crystal structure of the HupA-TcAChE complex (PDB code 1vot), where most residues were retained in the native conformations, was chosen as the molecular docking model [33]. Gratefully, re-docking of the original ligand into the active site gave an acceptable RMSD difference of 0.5384, which verified the reliability of our strategy. Figure 3 shows that both 8q and 8r spanned the active gorge with an extended conformation. The proximal pyrazine ring was bound in the CAS at the bottom, stacking against the side chain of Trp84, while the distal pyrazine/pyridine ring was present in the PAS near the mouth, forming a π-π interaction with Trp279. Moreover, some hydrophobic or van der Waals interactions might occur between 8q or 8r and the surrounding residues, such as Tyr70, Asp72, Asn85, Gly119, Tyr121, Ser122, Gly123, Tyr130, Leu282, Ser286, Ile287, Phe288, Phe290, Phe330, Phe331, Tyr334, Gly335, Trp432, His440, and Tyr442. Such interactions all contributed to their potent affinities with AChE. It should be noted that, owing to the fact that BuChE did not have the PAS [36], these compounds could not form effective binding interactions with the active pocket of BuChE, thereby giving a specific selectivity towards AChE. Finally, the result also revealed that compound 8q could bind to both the CAS and the PAS of AChE. Surrounding residues are depicted in grey (carbon), red (oxygen), and blue (nitrogen).

Cytotoxicity Assay for 8q and 8r
One of the best documented side-effects of the drug discovery and development of AD is cytotoxicity [37][38][39]. To probe whether such a type of target compounds involves cytotoxic effects, we respectively treated HepG2 cells and SH-SY5Y cells with several concentrations of the corresponding compound. Cell viability was determined by a CCK-8 assay. As reported, the referenced compound tacrine showed dose-dependent cytotoxicity ( Table 2 and 3). Excitedly, our potent "hits" 8q and 8r displayed no obvious cytotoxicity in the two cell lines, even at a high concentration of 500 μM, suggesting that potent compounds 8q and 8r have a potent safe window for the next stage of the development of anti-AD drug discovery.

Cytotoxicity Assay for 8q and 8r
One of the best documented side-effects of the drug discovery and development of AD is cytotoxicity [37][38][39]. To probe whether such a type of target compounds involves cytotoxic effects, we respectively treated HepG2 cells and SH-SY5Y cells with several concentrations of the corresponding compound. Cell viability was determined by a CCK-8 assay. As reported, the referenced compound tacrine showed dose-dependent cytotoxicity (Tables 2 and 3). Excitedly, our potent "hits" 8q and 8r displayed no obvious cytotoxicity in the two cell lines, even at a high concentration of 500 µM, suggesting that potent compounds 8q and 8r have a potent safe window for the next stage of the development of anti-AD drug discovery.

Chemistry
All commercially available compounds and dried solvents were used without further purification. The NMR ( 1 H and 13 C) spectra were recorded on a Bruker Avnace III 400 spectrometer (Bruker Avnace, Billerica, MA, USA) or Varian Mercury-Plus 300 spectrometer (Palo Alto, CA, USA). Chemical shifts (δ) were referenced to TMS as an internal standard. Coupling constants are given in Hz. MS spectra were obtained on an Agilent 6330 Ion Trap Mass Spectrometer (Palo Alto, CA, USA). HRMS spectra were recorded on a Shimazu LCMS-IT-TOF mass spectrometer (Kyoto, Japan). Microscopy images were captured on a HITACHI H-7650 transmission electron microscope (Tokyo, Japan). TLC was monitored by precoated silica gel GF254 plates purchased from Merck& Co., (Darmstadt, Germany) and the spots were detected through a UV lamp (Yuhua Co. Ltd., Guangzhou, China) at 254 nm. Flash column chromatography was performed with silica gel (200-300 mesh) purchased from Qingdao Haiyang Chemical Co. Ltd., Qingdao, China.

Synthesis of Intermediate 6
3-Bromopropan-1-ol was added dropwise to an ice-cold solution of 5 (20.0 mmol), DMAP (10 mg), and EDC·HCl (52 mmol) in CH 2 Cl 2 (20 mL). After the addition had been completed, the mixture was allowed to adjust to room temperature and was stirred for 24 h. When the reaction had been monitored by TLC, the solvent was removed in vacuum. The obtained brown residue was dissolved by adding distilled water, extracted with EtOAc three times (50 mL × 3), and washed with water. The organic layer was washed with brine dried over anhydrous Na 2 SO 4 . It was concentrated under reduced pressure to give compound 6 (2.08 g) as tan jelly without further purification. The yield was 73%.

Kinetic Study of AChE Inhibition
The assay solution (100 µL) was afforded by the addition of 30 µL of DTNB (0.01 M), 10 µL of 2 units/mL AChE, and 30 µL of substrate (ATC) to the phosphate buffer (0.1 M, pH 7.0). Three different concentrations of 8q were added to the assay solution with the AChE and pre-incubated for 20 min at 37 • C, followed by the addition of substrate in different concentrations. Kinetic characterization of the hydrolysis of ATC catalysed by AChE was carried out at 410 nm. A parallel control was made with the assay solution of no inhibitor for each time. The plots were assessed by a weighted least square analysis that assumed the variance of V to be a constant percentage of V for the entire data set. Slopes of these reciprocal plots were then plotted against the concentration of 8q in a weighted analysis, and Km was determined as the intercept on the negative x-axis.

In Vitro Cytotoxicity of Tacrine and Ligustrazine Hybrids 8q, 8r in HepG2 Cells
HepG2 cells (human hepatocellular liver carcinoma cell line from American Type Culture Collection, ATCC) were cultured in Eagle's minimum essential medium (EMEM) supplemented with 10% heat-inactivated fetal bovine serum (FBS), and 100 U/mL of penicillin/streptomycin (Gino Biomedical Technology Co. LTD., Hangzhou, China). Cultures were seeded into flasks containing supplemented medium and maintained at 37 • C in a humidified atmosphere of 5% CO 2 . For assays, cells (1.0 × 10 5 cells/well) were seeded in a 96-well plate in complete medium, the medium was removed after 24 h, and cells were exposed to the increasing concentrations of compounds 8q or 8r (0, 25, 50, 125, 200, 300, 400, and 500 µM) in DMEM with no serum for a further 24 h. Cell survival was measured through a CCK-8 assay. All compounds were dissolved in pure DMSO.
3.4.6. In Vitro Cytotoxicity of Tacrine and Compounds 8q, 8r in SH-SY5Y Cells SH-SY5Y cells (human neuroblastoma cell line from American Type Culture Collection, ATCC, Manassas, VA, USA) were seeded into 96-well plates at a density of 4 × 10 4 cells/mL in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% FBS and 100 U/mL of penicillin/streptomycin (Gino Biomedical Technology Co. LTD., Hangzhou, China), and incubated in a humidified atmosphere containing 5% CO 2 at 37 • C. For the experiments, cells (1.0 × 105 cells/well) were seeded in a 96-well plate in complete medium; after 24 h, cells were treated with the test compound in different concentrations (0, 25, 50, 125, 200, 300, 400, and 500 µM) at 37 • C for 24 h. After this incubation, 10 µL/well of CCK-8 (5 mg/mL) was added and incubated at 37 • C for 4 h. The optical density (OD) of each well was measured using an Epoch Microplate Spectrophotometer (BioTek Instruments, Inc., Winooski, VT, USA) with a test wavelength of 450, 630 nm.

Molecular Docking
For the docking studies, the Surflex dock in Sybyl X-2.1.1 (Certara, L.P., Princeton, NJ, USA) was used. The structures of 8q and 8r were built in Sybyl and the energy was minimized with the conjugate gradient method. For the protein, all the water molecules were extracted and the AMBER FF99 force field was applied. The standard docking procedure was used with the parameters set in default. Re-docking the original ligand into the crystal structure was performed to test the reliability of the docking strategy. Binding modes and interactions were analyzed in Pymol.

Conclusions
In summary, a series of novel ligustrazine derivatives 8a-8r have been designed, synthesized, and evaluated for the first time as multi-targeted inhibitors for anti-AD drug discovery. The results demonstrated that: (1) most of the target compounds exhibited potent AChE inhibitory activities, even at a nanomolar level, with excellent selectivity towards AChE. In particular, compounds 8q and 8r were found to be the most active inhibitors against AChE, with IC 50 values of 1.39 and 0.25 nM, respectively, and the highest selectivity for AChE (for 8q, IC 50 BuChE/IC 50 AChE = 2.91 × 10 6 ; for 8r, IC 50 BuChE/IC 50 AChE = 1.32 × 10 7 ). Additionally, (2) the two most potent AChE inhibitors 8q and 8r possessed acceptable inhibitory activities against self-induced Aβ (1-42) aggregation, with IC 50 values of 17.36 µM and 49.14 µM, respectively. The SAR analysis indicated that the introduction of a proper and relatively rigid group between the ester group and the benzene ring was beneficial to AChE inhibitory activity, and the replacement of the benzene ring moiety with a pyrazine unit or pyridine core also led to a dramatic increase in AChE inhibitory activity. A combined Lineweaver-Burk plot and molecular docking study revealed that these compounds might act as mixed-type inhibitors to exhibit such effects via selectively targeting both the CAS and the PAS of AChE. Moreover, the investigations of both the HepG2 cells and the SH-SY5Y cells showed that the potent "hits" 8q and 8r have a potent safe window for the next stage of the development of anti-AD drug discovery. Taken together, these results suggested that 8q and 8r could serve as promising candidates for the treatment of AD disorders and also provide a rational basis for the development of future multi-targeted inhibitors as anti-AD drugs with validated advantages over the classical anti-AD drug tacrine.