Investigação de sítios inéditos e proposição de novos ligantes para a proteína de morte celular programada I através de dinâmica molecular e triagem virtual

Abstract

Cancer is a set of diseases originating from genetic modifications associated with a dysregulation in the cell's proliferation and survival functions. It is one of the leading public health problems globally, with estimates pointing to an incidence of 19 million cases in the year 2020 alone. The literature indicates that the interaction between programmed cell death protein 1 (PD-1) and Programmed death-ligand 1 (PD-L1) can provide immunological escape and cancer cell survival. Antibodies such as pembrolizumab and nivolumab act by inhibiting interaction and recovering immune system activity. Small molecules to block this interaction are an interesting alternative to antibodies since it has a lower production cost, low occurrence of adverse effects, and greater patient compliance. This work aimed to identify new PD-1 ligands in order to modulate the interaction between PD-1 and PD-L1. Molecular Dynamics simulations with co-solvents enabled the prediction of different interaction sites in PD-1. Identifying a site linked to the C’D loop of distinctive character, close to the main interface between PD-1 and PD-L1 and characterized by the DogSiteScorer server and later by the fpocket program proved to be attractive to the development of ligands. The conformational state analyses confirmed the region's greater affinity for structurally cyclic probes, such as benzene, imidazole, and phenol. The virtual screenings based on docking and ensemble docking confirmed the importance of the cyclic structure for the stability of the ligands in the site. The visual analyses based on medicinal chemistry allowed the selection of 10 ligands to perform the dynamics of the complexes, allowing to verify the greater stability of 3 ligands. The interaction profile demonstrated the importance of the cyclic and amide group present in the ligands involved in cation−π and hydrogen interactions, respectively, with PD-1 residues. PCA analysis of the PD-1 / ligand complexes demonstrated the occurrence of a particular conformational change in the major site residues with PD-L1. Thus, integrating different computational approaches, this work made it possible to identify molecules and propose structural modifications in order to optimize the structures and increase their affinity for the region of the C'D loop of PD-1 (a site predicted in this work and unexplored previously), establishing the bases for the proposition of ligands that can interfere in the interaction between PD-1 and PD-L1.