| Cancer has a huge adverse effect on human life and health.Anti-tumor therapy has undergone tremendous changes since its invention,especially in recent years immunotherapy has been a research hotspot in the field of tumor therapy.When programmed cell death protein 1(PD-1,also known as CD279)was discovered,scientists opened a new chapter in the field of tumor therapy.It is negatively regulated by programmed cell death ligand 1(PD-L1,CD274)and participates in the key processes of T cell proliferation and metabolism,which ultimately changes the T cell signaling pathway to prevent excessive activation of T cells.However,tumor cells use the immunosuppressive function of PD-L1 to avoid being killed by T cells that recognize the new antigen.Therefore,it is generally believed that immune checkpoint blocking antibodies targeting PD-1/PD-L1 are a promising strategy for the treatment of cancer.Currently,monoclonal antibody drugs targeting PD-1/PD-L1 have achieved great success.However,research on its small molecule inhibitors is very slow,and no small molecule inhibitors have been approved yet.In the discovery of drug-lead compounds and new skeletons,the use of computational methods to accurately predict the binding free energy between proteins and small molecules and to specifically understand the interactions between different groups of small molecules and important residues of proteins is of great significance for guiding drug discovery.This article is divided into two parts: First,the alanine scanning binding interaction entropy(AS-IE)method is applied to the binding energy and binding energy of the PD-L1 dimer protein and the 35 inhibitors with experimental activity data in the BMS company patent.A specific analysis and research were carried out on the combination method.Alanine scanning calculation has the advantage of quantitatively decomposing the total binding energy to each residue.The binding energy of 35 systems was calculated by the AS-IE method,and it was analyzed that these 35 systems made outstanding contributions to the total binding energy.The hot spot residues of the protein and the important groups of small molecules that bind to the protein.By comparing the binding energies calculated by the traditional MM/GBSA method,it is found that the AS-IE method has a better correlation with the experimental values and smaller errors in the ranking prediction results of the ligand binding free energy of the PD-L1 dimer protein system.In the second part of the thesis,the PD-L1 dimer was screened for inhibitors and experimentally verified by combining two methods of virtual screening with Schrodinger commercial software and AS-IE.Using Gilde to perform virtual inhibitor screening on the commercial compound library Specs containing 210,000 small molecules,combined with the AS-IE method to predict the binding energy,40 small molecules were selected for experimental verification,and finally one brand new with enzymatic activity was obtained.Skeletal inhibitor.And through molecular dynamics simulation,the possible conformation of binding to PD-L1 dimer protein and the difference in binding free energy with BMS-200 were obtained.This creates favorable conditions for the further modification of the compound. |