Page 155 - EJMO-9-1

Eurasian Journal of Medicine and
Oncology
Potential of flavonoids against glioblastoma

in cancer progression were selected for analysis. These default parameters were maintained for site constraints,
proteins included UPAR (PDB ID: 2FD6), P38 (PDB ID: rotatable groups, and excluded volumes, ensuring an
3ZYA), NRF (PDB ID: 4XMB), ERK (PDB ID: 5NHJ), optimized environment for ligand docking. 24
mTOR (PDB ID: 5OQ4), STAT (PDB ID: 6NUQ), BCL-XL
(PDB ID: 6RNU), MMP-9 (PDB ID: 4WZV), NRAS (PDB 2.4.6. Molecular docking
ID: 6ZIR), and AKT1 (PDB ID: 3O96). These targets were Molecular docking was conducted to explore the interaction
chosen based on specific criteria: they originate from dynamics between the ligands and their respective protein
Homo sapiens, possess a resolution of less than 3 Å, and targets. This was accomplished using the Glide module
are devoid of mutations, except for NRAS (PDB ID: 6ZIR). within the Maestro interface of Schrodinger 2020-3. Both
The proteins were downloaded in PDB format, while the Compounds 1 and 2, along with the CCLs, were subjected
co-crystal ligands (CCLs) of each protein were saved in to flexible docking simulations using the energy-minimized
SDF format for subsequent docking analyses. conformations derived from the earlier steps. The docking
mode was set to extra precision (XP), which is designed
2.4.3. Preparation of ligands
to yield highly reliable binding affinity predictions. In
The ligands, including Compounds 1 and 2, along with the addition, the root mean square deviation (RMSD) was
CCLs, were subjected to optimization using the Ligprep tool calculated for the input ligand geometries to validate the
within Maestro, Schrodinger 2020-3 (version 12.5.139). accuracy of the docking procedure. Finally, the top-
23
The aim was to achieve energy-minimized 3D structures scoring ligands were visualized in the BIOVA Discovery
with accurate chiralities at a pH level of 7.0 ± 2 while Studio Visualizer (version 21.1.0.20298), enabling a
maintaining default settings for consistency. To enhance detailed examination of the molecular interactions within
the accuracy of the molecular modeling, the OPLS3e the active site of each target protein and providing valuable
(Optimized Potentials for Liquid Simulations) force field insights into their binding mechanisms.
was employed during the ligand minimization process,
which significantly improved the conformational stability 2.4.7. Prediction of absorption, distribution,
of the compounds. 22 metabolism, and excretion properties
2.4.4. Preparation of proteins Understanding the absorption, distribution, metabolism,
and excretion (ADME) characteristics of phytochemicals
Protein preparation was conducted using the Protein is a crucial step in the drug discovery process. These
Preparation Wizard tool available in the Maestro interface pharmacokinetic properties play a vital role in
of Schrodinger 2020-3 (version 12.5.139). Initially, each determining the effectiveness of a potential therapeutic
protein was individually imported into the workspace agent’s effectiveness, as many drug candidates fail due
and underwent preprocessing, which involved filling in to unfavorable ADME profiles, leading to issues in drug
missing loops and chains using the Prime job module. development and clinical outcomes. To mitigate these
25
Subsequent steps included the deletion of extra side risks and ensure the viability of the compounds, we
chains and the assignment of zero-order bonds to metal conducted a comprehensive ADME analysis following the
atoms. Optimization was carried out at a specific pH of molecular docking studies.
7.0 ± 2, following the PROPKA predictions, to ensure
optimal ionization states. In addition, water molecules For this purpose, we utilized SwissADME, an advanced
located beyond 3 Å from the binding site were removed tool specifically designed to assess the pharmacokinetic
to refine the protein’s active site environment. The final attributes of chemical compounds. The 2D molecular
step involved energy minimization using the OPLS3e structures of the phytochemicals were transformed
force field to ensure the protein’s stability for docking into Simplified Molecular Input Line Entry System
procedures. 23 (SMILES) strings, providing a simplified yet highly
precise representation of their chemical makeup. This
2.4.5. Receptor grid generation conversion enabled a more rigorous and detailed analysis
Receptor grid generation was a critical step in the of the compounds’ potential drug-like properties. During
docking process, executed using Maestro’s Receptor Grid this evaluation, each phytochemical was systematically
Generation tool. The active binding site of each protein- screened against Lipinski’s Rule of Five (LRF), a widely
ligand complex was pinpointed, and the grid was centered recognized standard that predicts the oral bioavailability of
on the centroid of the CCL for each target protein. To refine chemical entities based on key molecular parameters, such
this grid, a scaling factor of 1.0 was applied to the van der as molecular weight, lipophilicity, hydrogen bond donors,
Waals radii alongside a partial charge cutoff of 0.25. The and acceptors. 26

Volume 9 Issue 1 (2025) 147 doi: 10.36922/ejmo.5768