Evaluation of Novel Dual Acetyl- and Butyrylcholinesterase Inhibitors as Potential Anti-Alzheimer's Disease Agents Using Pharmacophore, 3D-QSAR, and Molecular Docking Approaches

Abstract

DL0410, containing biphenyl and piperidine skeletons, was identified as an acetylcholinesterase (AChE) and butyrylcholinesterase (BuChE) inhibitor through high-throughput screening assays, and further studies affirmed its efficacy and safety for Alzheimer's disease treatment. In our study, a series of novel DL0410 derivatives were evaluated for inhibitory activities towards AChE and BuChE. Among these derivatives, compounds 6-1 and 7-6 showed stronger AChE and BuChE inhibitory activities than DL0410. Then, pharmacophore modeling and three-dimensional quantitative structure activity relationship (3D-QSAR) models were performed. The R² of AChE and BuChE 3D-QSAR models for training set were found to be 0.925 and 0.883, while that of the test set were 0.850 and 0.881, respectively. Next, molecular docking methods were utilized to explore the putative binding modes. Compounds 6-1 and 7-6 could interact with the amino acid residues in the catalytic anionic site (CAS) and peripheral anionic site (PAS) of AChE/BuChE, which was similar with DL0410. Kinetics studies also suggested that the three compounds were all mixed-types of inhibitors. In addition, compound 6-1 showed better absorption and blood brain barrier permeability. These studies provide better insight into the inhibitory behaviors of DL0410 derivatives, which is beneficial for rational design of AChE and BuChE inhibitors in the future.

Keywords: 3D-QSAR; Alzheimer’s disease; DL0410; cholinesterase inhibitor; kinetics; molecular docking.

Figure 1

DL0410 showed best alignment with Hypo4 pharmacophore model. Hypo4 consisted of two hydrogen bond acceptors, two ring aromatic features, and one hydrophobic feature.

Figure 2

Plot of the experimental versus predicted AChE and BuChE inhibitory activities of the training set and test set. The correlation coefficient R² of AChE and BuChE 3D-QSAR models between determined and predicted activities of training set were 0.925 and 0.883 (A,C). The R² of AChE and BuChE test set were found to be 0.850 and 0.881 (B,D), respectively.

Figure 3

3D-QSAR model coefficients on electrostatic potential and Van der Waals (VDW) grids. Figure A and B described the coefficients on electrostatic potential grids. Blue represents positive coefficients; red represents negative coefficients. (A: AChE, B: BuChE). Figure C and D presented the coefficients on VDW grids. Green represents positive coefficients; yellow represents negative coefficients. (C: AChE, D: BuChE).

Figure 4

Visualization of the non-bond interactions between the protein and ligands. Figure A and D showed the interaction between AChE and DL0410. Biphenyl and piperidine rings of DL0410 interacted with TRP286, TYR341 and TRP86 via π-π stacking or π-alkyl interaction. Compound 6-1 showed a strong interaction with AChE (Figure B and E). Except for TRP286, TYR341 and TRP86, compound 6-1 could form hydrogen bonds with TRP72 via carbonyl group. Compound 7-6 also had powerful interactions with AChE, mainly through π-π stacking, π-alkyl and hydrogen bond interaction (Figure C and F). TRP286, TYR341 and TRP86 were the key amino acid residues for AChE interacting with the three compounds.

Figure 5

Visualization of the non-bond interactions between the protein and ligands. The figures of A and D showed the interaction between BuChE and DL0410. Biphenyl and piperidine rings of DL0410 interacted TRP332, PHE329 and TRP82 via π-π stacking or π-alkyl interaction. The carbonyl group could form hydrogen bonds via interacting ASN289. Figure B and E suggested compound 6-1 had the strongest interaction with BuChE. The piperidine of compound 6-1 had a stronger electrostatic attraction with APS70 and TRP82, and π-Alkyl interaction with ALA328, PHE329, and TYR332. From Figure C and F, we observed that compound 7-6 also had stronger interaction with BuChE than that of DL0410. There existed more hydrogen bonds for compound 7-6, and a highly electronegative fluorine atom caused its strong interaction with GLY116.

Figure 6

The Lineweaver-Burk plots of DL0410, compound 6-1 and compound 7-6 on AChE inhibitory activity (A–C). The Lineweaver-Burk plots of DL0410, compound 6-1 and compound 7-6 on BuChE inhibitory activity (D–F). Kinetic analysis of AChE and BuChE was performed by using three effective concentrations of DL0410, compound 6-1 and compound 7-6 (0.1, 0.3, 1 μM for AChE, and 0.3, 1, 3 μM for BuChE) with the incubation of five different substrate concentrations (1.25, 2.5, 5, 10, 20 mM for AChE, and 2, 4, 8, 16, 32 mM for BuChE). The intersections of fitting lines with different concentrations of the three compounds were neither on the X axis or Y axis of Lineweaver-Burk plots of both AChE and BuChE, suggesting that DL0410, compounds 6-1 and 7-6 were mixed-type of inhibitors. (n = 3; a = mM⁻¹; b = min/ΔA).

Figure 7

Ki values were determined by Dixon plots. Dixon plots of DL0410, compound 6-1 and compound 7-6 on AChE inhibitory activity were showed as Figure A–C, respectively. The AChE Ki values of them were, separately, 1.65 ± 0.14 μM, 0.42 ± 0.04 μM and 0.55 ± 0.2 μM. Figure D–F presented the Dixon plots of DL0410, compound 6-1 and compound 7-6 on BuChE inhibitory activity, respectively. The BuChE Ki values of them were, separately, 1.81 ± 0.08 μM, 0.74 ± 0.16 μM and 1.63 ± 0.18 μM. (n = 3; b = min/ΔA).