Previous studies provide substantial evidence that catechins, polyphenol bioactive compounds, exhibit medicinal benefits. These polyphenols are found in abundance in green teas, including a combination of the four major types of catechins: (-)-Epicatechin (EC), (-)-Epicatechin-3-gallate (ECG), (-)- Epigallocatechin (EGC), and (-)-Epigallocatechin-3-gallate (EGCG). Although all four exhibit medicinal benefits, the catechin cited in the literature the most is EGCG, so derivatives of this catechin were selected for these studies. Literature searches identified catechins as biologically active compounds for a diverse set of diseases ranging from cancer, metabolism, neurological, and neuromuscular ailments. A diverse set of potential protein targets for docking with catechin derivatives was first identified as a list (n=48). The targets were then selected based on the presence of 3D protein coordinates for these targets provided by the Rutgers Consortium for Structural Biology (RCSB) Protein Data Bank (PDB) (n=10). The surfaces of the 3D protein targets were evaluated with computational methods to identify potential binding sites for the EGCG catechin derivatives. Static and flexible docking was done using target protein binding sites performed with the catechin derivatives followed by molecular dynamics (MD). MD protocols were run to confirm binding in the physiological range and environment. In summary, the results of computational protocols confirmed predicted binding by docking with MD of several catechin derivatives to be used as scaffolds once validated in lab-based assays. Possible changes to these scaffolding molecules that could result in tighter, more specific binding is discussed.

A three‐phase computational protocol for drug discovery has been developed that includes 1) 3D shape characterization of binding sites of a protein to predict optimum performance with selected molecular docking programs; 2) Implementation of predictive algorithms for pharmacokinetic properties and ADME profiling for targeted searches in the ZINC15 database; 3) Flexible molecular docking/dynamics protocols of small molecule/protein complexes for drug discovery. Focusing on cancer immunotherapy to implement this computational approach, the allosteric site of ecto‐5’‐nucleotidase (CD73), a regulator of the extracellular concentration of adenosine, is targeted for large scale small molecule searches. CD73 which hydrolyzes AMP to adenosine, is a GPI‐anchored cell surface protein cleaved and activated in the extracellular environment as a homodimer. CD73 exhibits higher expression on lymphocytes, endothelial, and epithelial cells during a stress response and his been reported activated for various types of cancer. CD73 participates in blocking antitumor responses and enhances tumor growth and metastasis by producing high levels of adenosine in the extracellular environment. Protein coordinates of CD73 were downloaded for these from Rutgers Consortium of Structural Biology (RCSB) Protein Data Bank (PDB) and the potential inhibitory molecules from the ZINC15 Database. AutoDock Vina was selected as the ideal docking program for the target based on comparing predicting binding constants with actual Ki's and RMSD analysis of the co‐crystallized ligand with the re‐docked ligand. Once molecules with high affinity predicted scores were identified from docking, molecular dynamics studies were performed on the highest scoring protein/small molecule complexes. Studies were carried out using GROMACS, implemented with AMBER force field. Dissociation constants (Ki's) for the protein‐ligand complex were calculated throughout MD runs. Conformational changes in the binding sites were observed visually and quantitified with PCA analysis. Novel hydrogen bonding networks and hydrophobic networks of residues in the 3D structures were identified to connect the active sites to the allosteric binding site using a MD protocol.