*2.4. Molecular Modeling Studies*

The X-ray crystal structures of MMP-2 (PDB ID: 1CK7), MMP-8 (PDB ID: 5H8X) and MMP-9 (PDB ID: 2OW1) were retrieved from RCSB protein data bank and the electrostatic surface potential (ESP) maps calculated using PyMOL (Molecular Graphics System, Schrödinger, NY). MMP-8 was prepared for docking by removing water molecules, adding polar hydrogens, Kollman charges and solvation parameters using MGL (Molecular Graphics Laboratory) Tools of AutoDock 4.2 (Scripps Research, CA) to generate pdbqt files. Docking was performed in two stages. First, to identify a potential binding site of NSGMs, a blind docking was performed using a grid box of 100 × 100 × 100 Å with 0.375 Å grid spacing to include all the amino acid residues. A short search of 10 runs with 250,000 evaluations per run was performed with Lamarckian Genetic Algorithm (LGA) docking. Later, based on the results of this initial blind docking, a more stringent grid with a radius of 16 Å was defined and the evaluations increased to 25000000. The top 10 binding poses were then individually analyzed, and the pose favoring the best score with least root-mean-square difference was chosen. The interactions of the docked poses were analyzed using PyMOL or LigPlot+ (European Molecular Biology Laboratory, Heidelberg, Germany).
