Phytochemicals as Potential Anti-Alzheimer’s Agents- An In-Silico Evidence

  


    
Sl.    No.

    
Molecule

    
Binding    Affinity (Kcal/mol)

    
Chain

    
Amino    acid Residues with interactions

  

  


    
1. 

    
Anaferine 

    
-3.5

    
B

    
Met-35(alkyl    interaction)

  

  


    
2. 

    
Cuscohygrine

    
-4.0

    
B

    
Met-35    (alkyl interaction); Gly- 33 (H-Bonding)

  

  


    
3. 

    
Docosahexaenoic acid

    
-3.7

    
A

    
Leu-34,    Val-36 (alkyl interactions)

  

  


    
4. 

    
Genistein

    
-5.6

    
A

    
Asp-    23( H-Bonding); Gly-25 (pi-sigma interaction); Leu-34 (pi-alkyl interaction)

  

  


    
5. 

    
Ginkgolide B

    
-6.1

    
D

    
Val-39    (alkyl interaction)

  

  


    
6. 

    
Huperzine A

    
-5.5

    
E

    
Phe-19,    Leu-34, Val-36 (alkyl and pi-alkyl interactions); Ala-21 (pi-sigma    interaction); Glu-22 (H-Bonding)

  

  


    
7. 

    
Isopelletierine

    
-4.2

    
A

    
Phe-19    (pi-sigma interaction); Ala-21 (pi-alkyl interaction)

  

  


    
8.

    
Kaempferol

    
-5.6

    
A,B

    
Gly-25    (H-Bonding,pi-sigma interaction); Ile-32 (pi-alkyl interaction); Lys-28 :B    (pi-cation interaction); Leu-34 (pi-sigma interaction)

  

  


    
9.

    
Methyl Quercetin

    
-5.7

    
A,    B

    
Asp-23,    Gly-29 (H-Bonding); Ile-32 (pi-alkyl interaction); Leu-34(pi-sigma    interaction); Gly-25 (pi-sigma interaction); Lys-28:B( pi-cation interaction)

  

  


    
10.

    
Paclitaxel

    
-6.5

    


    
No    Interaction

  

  


    
11.

    
Piceatannol 

    
-5.4

    
A

    
Asp-23(H-Bonding);    Ala-21(amide-pi stacking); Leu-34 (pi-sigma interaction); Val-36,Ile-32    (pi-alkyl interaction)

  

  


    
12.

    
Protocatechuic Acid

    
-4.4

    
A

    
Ala-21,Gly-37    (H-Bonding); Phe-19 (pi-pi stacking); Ala-21 (pi-alkyl interaction)

  

  


    
13.

    
Resveratrol

    
-5.9

    
A,B

    
Ala-21,Val-36,    Val-40, Val-40:B( pi-alkyl interactions); Phe-20 (H-Bonding, pi-pi stacking)

  

  


    
14.

    
Vinpocetine

    
-5.3

    
B,C

    
Met-35:B,    Met:35:C (alkyl interactions); Ile-31 (pi-sigma interaction)

  

  


    
15.

    
Withanolide A

    
-6.8

    
C,D,E

    
Val-36:C    (H-Bonding); Val-39:D,Val-39:E, Ile-41:D, Ile-41:E (alkyl interactions)

  

Table 6: Analysis of Docking results of Aβ protofibril (PDB ID- 2BEG) [21] against the selected molecules.

Materials and Methods

Molecular Docking studies

Preparation of protein: The 3D structure of all the proteins human AChE enzyme (PDB ID- 5HF6), BChE enzyme (PDB ID- 6EP4), BACE1 enzyme (PDB ID- 1W51), Aβ monomer (PDB ID- 1LYT) and Aβ protofibril (PDB ID- 2BEG) were downloaded in .pdb format from the Protein Data Bank (https://www.rcsb.org/).

Preparation of ligands structure: The ligands ie. All the phytochemicals were selected from various dietary supplements and the structures were drawn by ChemDraw14.0 tool. The Chem Draw 3D tool was then used for conversion of the structures to 3D. Vega ZZ program was used to fix the charges of the ligands as gasteiger. The energy of the ligands was minimized using Vega ZZ program and the ligands were saved in .pdb format.

Docking studies: AutoDock Tools (ADT v1.5.6) and AutoDock Vina, developed by the Scripps Research Institute (http://www. scripps.edu/downloads) has been used to perform the docking studies of selected molecules against all the targets of interest. The AutoDock Tools program consists of Lamarckian Genetic Algorithm search engine [22]. The docking study was performed by following the standard docking protocol, grids of 126, 126, and 126 points in x, y, and z directions were built for the protein preparation and the grid spacing was set to 0.375Å. The exhaustiveness was set to default. AutoDock Tools (ADT v1.5.6) was used to save the proteins and ligands as .pdbqt format. The docking of selected phytochemicals was performed by using Auto Dock Tools program. Docking output and log file for the docking study was generated once the docking was completed. The docking output was optimized with the minimum energy. The visualization of docking output and generation of 2Ddiagram for the docking studies were done using Discovery Studio 2017 R2 tool. The binding interactions were observed for almost all the selected phytoconstituests towards the targets of interest.

ADME Prediction by Using Swiss-ADME Tool

The bioavailability potential of the selected molecules can be detected through absorption, distribution, metabolism and excretion (ADME). Computational prediction using ADME software is an in silico method, suitable due to quick screening, financially affordable, no animal testing, etc. Swiss ADME online tool (freely accessible at http://www.swissadme.ch) was used for the prediction of in silico studies, especially pharmacokinetics, bioavailability, drug likeness and medicinal chemistry friendliness. The tool presents an early estimation of these ADME parameters (for Absorption, Distribution, Metabolism and Excretion) and provides sufficient information about the physicochemical parameters which is needed to be improved during formulation [23]. Human gastrointestinal absorption (HIA) and Blood-Brain Barrier (BBB) permeation is represented as Boiled Egg model provided by this tool [24]. The Boiled egg representation of the selected phytochemicals is depicted in (Figure 2). In this model some the molecules were found to be permeable through the Blood-Brain Barrier (BBB) that is very essential in the treatment of AD and most of the molecules were found to be absorbed through gastrointestinal tract and Ginkgolide B was found out of range. The tool also gives the idea about the details of chemistry information of the molecules such as molecular weight, Molecular formula, hydrogen-bond donor, hydrogen-bond acceptor etc. The tool also provides detailed information on the inhibitory action of the molecules towards the enzyme of cytochrome enzyme series. The chemistry information of the molecules is described in (Table S1; See supplementary information) while Pharmacokinetics prediction, Bioavailability prediction, Druglikeness prediction and Medicinal chemistry prediction are depicted in (Table S2, Table S3, S4 & S5; See supplementary information) respectively. All the tabular information relating to Swiss ADME prediction are provided in supporting information. The Swiss ADME data reviles that most of the phytochemicals are water soluble, highly GI absorbable and some are even permeable through BBB. Most of the phytochemicals follow the rules of Drug likeness and have a good synthetic accessibility score.

    


    


    Figure 2: BOILED EGG representation of the selected phytochemicals where the yolk denotes the molecules that are blood-brain-barrier permeable, while the

white portion denotes the molecules that can be absorbed through gastrointestinal tract. The graph is plotted against WLogP [a log P (n-octanol/water partition

coefficient) method developed by Wildman and Crippen] versus TPSA or Topological Polar Surface Area.