In silico evaluation of the interaction of peptide ligands with bevacizumab by structural analysis and molecular docking

JFO ORLOWSKI; GR BARREDO VACCHELLI; SL GIUDICESSI; SA CAMPERI

Congreso; 12vo Congreso Argentino de Bioinformática y Biología Computacional; 2022

Bevacizumab, monoclonal antibody used for cancer treatment, binds vascular endothelial growth factor (VEGF) preventing angiogenesis. Bevacizumab must be highly purified to be intravenously administered by an expensive process based in affinity chromatography (AC) with protein A immobilized.. Alternatively, AC with short peptides as ligands reduce the cost of the process and are of higher physically and chemically stability.Recently, synthetic peptides have been evaluated for bevacizumab AC purification. The best results were obtained with a peptide corresponding to the 85-PHQGQHIG-92 VEGF. However, the screening process was very expensive and laborious.The aim of this study was to probe if in silico studies could predict peptide interaction and reduce the time and cost of the screening assay.RESULTSThe 3D peptide structure was modeled with PEP-FOLD3. The molecular docking was executed with AutoDock Vina software. Five models were found through directed molecular docking simulation with good performance in molecular coupling when evaluated for their binding energy.The conformations obtained against the same fragment within the native crystallized structure were compared. These models’ root mean square deviation (RMSD) values were of about 2 Å which implies that they were positioned in a similar way to the crystallized structure.The comparison of the crystallographic structure of the complex VEGF/bevacizumab against the models, conducted in LigPlot+, allowed to identify the key amino acids and the type of interactions involved. The H-bond interactions between the peptide and Gly33 and Thr53 of bevacizumab were preserved in both the simulated models and the native structure together with the hydrophobic interactions.CONCLUSIONSHere it was demonstrated that molecular docking analysis is a useful tool to evaluate in silico ligand-target interactions. The correlation of the bioinformatic results with the experiments may save time and resources in ligand design to optimize protein purification as well as drug developing.