Figure 1.
Workflow of combining molecular docking and molecular dynamics simulation approaches for indentifying the reasonable binding site and generating the proper pharmacophore model.
Figure 2.
Sequence alignment and homology modeling for S. cerevisiae α-glucosidase using a template B. cereus oligo-1,6-glucosidase.
(A) Sequence alignment of S. cerevisiae α-glucosidase (represented as YEAST) with B. cereus oligo-1,6-glucosidase (1UOK). Sequence identities are denoted by asterisks (*), conservative substitutions by colons (:), and semi-conservative substitutions by dots (.). The catalytic residues are indicated in a red box. Comparative view of the homology modeled structure S. cerevisiae α-glucosidase (B), the template structure of B. cereus oligo-1,6-glucosidase (PDB ID: 1UOK) (C) with the conserved catalytic residues represented as sticks. The N-terminal, subdomain, and C-terminal domains are shown in blue, orange, and yellow, respectively. (D) Ramachandran plot of the φ/ψ distribution of the homology model as obtained by PROCHECK. (E) Z-score plot for our modeled structure shows that the score is within the range of scores typically found for native proteins of similar size.
Figure 3.
2D chemical structures of stilbene derivatives with experimental binding affinity value such as Ki and IC50.
Figure 4.
Results of initial molecular docking simulation.
(A) Validation of molecular docking simulation by comparison between crystal structure (gray) of S. cerevisiae isomaltase (PDB ID: 3A4A) co-crystallized with the α-D-glucose and homology modeled structure (blue) of the isomaltase with its docking pose. Hydrogen bonds are represented as dotted lines in crystal (cyan) and homology modeled (yellow) structures. 2D structure of α-D-glucose is shown in the right box. (B) Initial molecular docking results of compound 6 (orange) and compound 12 (green) with representing interacting residues which shown as sticks. 2D interaction diagram of compound 6 (C) and compound 12 (D) with representing charged (pink plate), π (orange line), and hydrophobic (light green plate) interacting residues.
Figure 5.
MD simulation results of four systems (Apo, first, second, and third trials) for structure adjustment of compound 12-bound state.
Cα RMSD plot of three trials and Apo systems (A), RMSD of ligand in three trial systems (B). First trial (C), second (D), third (E), and all three representative structures (F) with initial docking pose colored by cyan.
Table 1.
Averaged interaction energy of compound 12 obtained from MD simulation.
Figure 6.
Molecular docking results of compound 12.
The best docking poses (light green) of compound 12 in adjusted protein structure of Apo (A), first (B), second (C), third (D) trial systems with comparing the MD results which are represented by dark green.
Table 2.
Interaction energy and negative CDOCKER energy of compound 12 obtained from molecular docking simulation.
Figure 7.
Results of MD simulations of stilbene derivatives starting with docked structure in the second trial system.
(A) Cα RMSD plot of different stilbene derivative-bound systems. (B) Overlapped structure of compound 6 (orange), compound 7 (red), compound 10 (bluish green), compound 11 (violet), compound 12 (green), compound 13 (sky blue), compound 14 (light blue), and compound 16 (light violet). (C) Interaction energy plot of all systems during the 1.5 ns simulation time. (D) Correlation graph between experimental Ki value and interaction energy.
Table 3.
Correlation between exp. Ki and calculated energy.
Figure 8.
Binding modes of compound 6 (orange) and compound 12 (green) obtained from 10 ns MD simulation.
(A) Hydrogen bonding interactions (light blue line) of compound 6 with Glu276, Ser308, and Arg312 are displayed with π-interacting residues (orange line). (B) Interactions of compound 12 with four hydrogen bonding residues including Glu276, Val303, Thr307, and His348 are represented with π-interacting residues: Glu304 for π-sigma and Phe311 for π-π interactions. 2D interaction diagram of compound 6 (C) and compound 12 (D) with representing charged (pink plate), π (orange line), and hydrophobic (light blue plate) interacting residues.
Table 4.
Hydrogen bonding and hydrophobic contacting residues between protein and compound.
Figure 9.
Distance of π-sigma interaction between Glu304 and each compound.
Distance of compound 12 with β-carbon and γ-carbon of Glu304 is represented by blue and red lines, respectively (upper left). Distance between γ-carbon of Glu304 and compound 12 is shown as green line (upper right). These three distance values are compared during the simulation time (bottom). The threshold of π-sigma interaction is highlighted by green dotted box.
Figure 10.
Receptor-ligand pharmacophore model generation and validation.
(A) Four featured pharmacophore model consists of two hydrogen bond donors (HBD), two hydrophobic (HPhob), and excluded volumes. Mapping of generated pharmacophore model on compound 12 (B) and compound 14 (C). (D) Correlation graph between experimental Ki and scale fit values.
Table 5.
Correlation between exp. Ki and scale fit value.