10 Applications of Molecular Dynamics Simulations …
203
acting residues from MD simulation, a predictive model was developed to predict
tobacco constituents binding activity. Training and test set compounds from PDB
informed and evaluated the model (Fig. 10.8). The predictive model successfully
differentiated α7 nAChR binders from non-binders. Of 12 experimental test compounds, 11 were predicted correctly as nAChR α7 binders by the predictive model.
Hence, this predictive model might be helpful to screen tobacco constituents which
are responsible for addiction.
10.6.4 Antagonist Binding-Induced AR Structural Changes
Androgen receptor (AR) is as prostate cancer target is gaining traction in the drug
discovery field. AR activity is blocked both by non-steroid and steroid antagonists.
Prolonged use of AR antagonists causes mutations in AR which paradoxically cause
AR antagonists act as AR agonists. Driven by the binding of agonist/antagonist,
AR undergoes a considerable conformational change which impacts DNA and coregulatory protein binding. The X-ray crystal structure of antagonist binding in the
ligand binding pocket of wild-type (WT)-AR is required to understand the mechanism. Unfortunately, such a crystal structure does not exist. To address this challenge,
induced fit molecular docking and MD simulations were used to characterize AF2
site structural changes and the mechanism of antagonist binding in the ligand binding
pocket of WT AR. Induced fit molecular docking and a long-term MD simulation
were leveraged to construct the WT-AR-antagonist complex and to analyze WT-AR
structural changes, respectively. Three molecular systems (WT-AR bound by agonist
R1881, WT-AR bound by antagonist bicalutamide, and Mutant-AR bound by bicalutamide) were utilized to study the agonist binding-induced structural changes in the
WT-AR AF2 site. WT-AR bound to agonist R1881 and mutant-AR bound to bicalutamide were selected from PDB. Induced fit docking (IFD) and MD simulations
generated the structure of WT-AR bound to antagonist bicalutamide. Twenty-five
binding poses of bicalutamide in the ligand binding pocket of WT-AR were obtained
from IFD. Among the 25 poses, the best pose of bicalutamide in the ligand binding
pocket of WT-AR was selected based on both the IFD scores and also critical residue
interactions around the active site. The three systems were subject to 1 microsecond
of an MD simulation using AMBER 14 to optimize the three complexes. The RMSD
and RMSF plots revealed that the systems were stable throughout the simulations
(Fig. 10.9a, b). Five residues had shown a deviation of more than 2 Å. As expected,
those residues are present in the loop regions (Fig. 10.9c). The representative structure
from the trajectory files was superimposed on the X-ray crystal structures of WT-AR
with R1881 and mutant-AR with bicalutamide, which demonstrates the reliability of
the MD simulations (Fig. 10.10). The binding of bicalutamide in the ligand binding
pocket moved residues V716, K720, Q733, M734, Q738, and E897 and changed
the structure of the AF2 site of WT-AR, rendering AF2 unsuitable for co-activator
binding. The electrostatic potential map revealed that residues V716/K720/Q733 or
M734/Q738/E897 played a vital role in the formation of the positive and negative
203
acting residues from MD simulation, a predictive model was developed to predict
tobacco constituents binding activity. Training and test set compounds from PDB
informed and evaluated the model (Fig. 10.8). The predictive model successfully
differentiated α7 nAChR binders from non-binders. Of 12 experimental test compounds, 11 were predicted correctly as nAChR α7 binders by the predictive model.
Hence, this predictive model might be helpful to screen tobacco constituents which
are responsible for addiction.
10.6.4 Antagonist Binding-Induced AR Structural Changes
Androgen receptor (AR) is as prostate cancer target is gaining traction in the drug
discovery field. AR activity is blocked both by non-steroid and steroid antagonists.
Prolonged use of AR antagonists causes mutations in AR which paradoxically cause
AR antagonists act as AR agonists. Driven by the binding of agonist/antagonist,
AR undergoes a considerable conformational change which impacts DNA and coregulatory protein binding. The X-ray crystal structure of antagonist binding in the
ligand binding pocket of wild-type (WT)-AR is required to understand the mechanism. Unfortunately, such a crystal structure does not exist. To address this challenge,
induced fit molecular docking and MD simulations were used to characterize AF2
site structural changes and the mechanism of antagonist binding in the ligand binding
pocket of WT AR. Induced fit molecular docking and a long-term MD simulation
were leveraged to construct the WT-AR-antagonist complex and to analyze WT-AR
structural changes, respectively. Three molecular systems (WT-AR bound by agonist
R1881, WT-AR bound by antagonist bicalutamide, and Mutant-AR bound by bicalutamide) were utilized to study the agonist binding-induced structural changes in the
WT-AR AF2 site. WT-AR bound to agonist R1881 and mutant-AR bound to bicalutamide were selected from PDB. Induced fit docking (IFD) and MD simulations
generated the structure of WT-AR bound to antagonist bicalutamide. Twenty-five
binding poses of bicalutamide in the ligand binding pocket of WT-AR were obtained
from IFD. Among the 25 poses, the best pose of bicalutamide in the ligand binding
pocket of WT-AR was selected based on both the IFD scores and also critical residue
interactions around the active site. The three systems were subject to 1 microsecond
of an MD simulation using AMBER 14 to optimize the three complexes. The RMSD
and RMSF plots revealed that the systems were stable throughout the simulations
(Fig. 10.9a, b). Five residues had shown a deviation of more than 2 Å. As expected,
those residues are present in the loop regions (Fig. 10.9c). The representative structure
from the trajectory files was superimposed on the X-ray crystal structures of WT-AR
with R1881 and mutant-AR with bicalutamide, which demonstrates the reliability of
the MD simulations (Fig. 10.10). The binding of bicalutamide in the ligand binding
pocket moved residues V716, K720, Q733, M734, Q738, and E897 and changed
the structure of the AF2 site of WT-AR, rendering AF2 unsuitable for co-activator
binding. The electrostatic potential map revealed that residues V716/K720/Q733 or
M734/Q738/E897 played a vital role in the formation of the positive and negative