Pymol 1

PRDX1 models with point mutations were prepared based on homology modeling procedures, using Modeller [37 (link)] and evaluated with MetaMQAP [38 (link)] and PROQ [39 (link)]. Protein structures superpositions were prepared with Swiss-PDBViewer [40 (link)]. Figures were prepared with PyMOL 1.5 (Schrödinger). For docking procedures we selected crystal structures-derived models (PDB codes 1QQ2 and 2Z9S) and corresponding models with point mutations, with all rat-specific residues substituted with human equivalents. Docking simulations were carried out with Surflex-Dock 2.6 software [41 (link)] based on the “anchor-and-grow” algorithm for optimized structures of small molecules. We used elNémo web server [42 (link)] to calculate proteins normal modes. Files with coordinates are available in Supplementary Data.

A model of eGFP-OSK1 was created using the UCSF Chimera 1.10.1 interface40 (link) to Modeller 9.1441 (link), using the structure of eGFP (PDB ID: 2Y0G) and OSK1 (1SCO) as templates. The N-terminal region containing the His-tag and the (SG₄)₃ linker between the eGFP and OSK1 modules were presented as disordered.
3D alignment of OSK1 from the generated eGFP-OSK1 model and ChTx from the spatial structure of its complex with the K_V1.2-K_V2.1 paddle chimera (4JTA) was performed using PyMOL 1.7.4 (Schrödinger).

The H-subunit (residues 2–392) of deoxy-form mushroom tyrosinase protein was obtained from the RCSB Protein Data Bank (ID: 2Y9X) [42 (link)]. Small molecules, including water, holmium (Ho) atoms, and tropolone, excluding copper (II) ions, were removed from the target enzyme. Binuclear copper-binding catalytic site of H-subunit of 2Y9X was slightly modified to fulfill the oxy-form enzyme [43 (link)]. All hydrogens were added. AutoDock 4.2 program was used to predict the ligand-protein interactions [44 (link),45 (link)]. The copper ion parameters were prepared and added to run AutoGrid 4. The 3D structures of KG, MG, AB, l-tyrosine, kojic acid, and cinnamic acid were downloaded from PubChem Compound (NCBI), with compound CIDs of 5281667, 196583, 480819, 6057, 3840, and 444539, respectively. Kojic acid and cinnamic acid [34 (link)] were used as the reported catalytic and mixed type inhibitors against mushroom tyrosinase, respectively, and their binding sites were used to validate the results of AutoDock 4.2 docking analysis. The protein-ligand interactions were visualized and analyzed using PyMOL 1.7.4 (Schrodinger, LLC, Cambridge, MA, USA) for 3D models, and Discovery Studio Visualizer 16.1 (Accelrys, San Diego, CA, USA) was used for 2D diagrams.

The 3D structure of crystallized antigen-binding fragment (Fab) of the broadly neutralizing antibody, VRC01, in complex with core gp120 of HIV-1 clade A/E recombinant 93TH057 (PDB ID: 3NGB), with a resolution of 2.68 Å was downloaded from the Protein Data Bank (PDB). The core gp120 trimer consisting of outer and inner domains with truncated N- and C-termini, as well as V1/V2 and V3 variable loops, were extracted from the complex as protein chains A, G, and D using PyMOL 1.74 (Schrödinger, Inc., NY, USA) [82 (link)].

In order to find the behavior of the small molecule in the binding pockets of target proteins, molecular docking simulation was performed. For docking studies, the crystal structure of α-glucosidase and BACE1 protein targets were obtained from the RCSB Protein Data Bank with the respective accession codes 3A4A and 2WJO, respectively. The co-crystallized ligands, acarbose and (Z)-3-butylidenephthalide (BIP) for α-glucosidase and 2-amino-3-{(1R)-1-cyclohexyl-2-[(cyclohexyl-carbonyl)amino]ethyl}-6-phenoxyquinazolin-3-ium (QUD) and 3,5,7,3′,4′-pentamethoxyflavone (PMF) for BACE1, were used to generate the grid box for catalytic and allosteric inhibition mode respectively with compounds CIDs of 41774, 5352899, 44631815, and 97332, respectively. The 3D structures of the isolated compounds luteolin, luteolin 5-O-β-d-glucopyranoside, and luteolin 7-O-β-d-glucopyranoside were downloaded from PubChem Compound (NCBI), with compound CIDs of 5280445, 15559460, and 5280637, respectively. The results were visualized and analyzed using PyMOL 1.7.4 (Schrödinger, LLC, New York, NY, USA) and LigPlot+ 1.4.5 (European Bioinformatics Institute, London, UK).

Molecular modeling and graphics manipulations were performed using Maestro 10.5 (Schrödinger, LLC, New York, NY, 2016) and UCSF-CHIMERA 1.8.1 software packages, (http://www.cgl.ucsf.edu/chimera), running on an E4 Computer Engineering E1080 workstation E4 Computer Engineering E1080 workstation provided with an Intel Xeon processor. GOLD Suite 5.4.1 docking package (CCDC Software Limited: Cambridge, U.K., 2008)^{49 (link)} was used for all docking calculations. Figures were generated using Pymol 1.8.2 (Schrödinger, LLC, New York, NY, 2016).

Diffraction data were collected at the European Synchrotron Radiation Facility (ESRF, Grenoble, France) at different beam lines (ID14-4, ID23-1, ID23-2 and ID29-1). All data sets were processed with XDS [62 (link)], intensities of integrated reflections were scaled using XSCALE, and structure factors were calculated using XDSCONV. The structures were solved with the CCP4 suite [63 (link)] using the recombinant BChE structure (PDB entry 1P0I and 4AQD) as the starting model. The initial models were refined by iterative cycles of model building with Coot [64 (link)], then restrained and subjected to TLS refinement with Phenix [65 (link)]. The ligands and their descriptions were built using phenix.elbow included in Phenix. Data collection and refinement statistics are reported in Table S1. Protein structures were illustrated using the program PyMOL 1.8.2 (Schrödinger, Mannheim, Germany). 2Fo-Fc electron density maps of the ligands are represented in Figure S1. The molecular surfaces of the gorges were represented in PyMol with the help of the program Hollow 1.2 [66 (link)].