| Receptor-based drug design can design highly targeted and specific drugs based on the characteristics of the receptor,reduce the blindness of drug synthesis,and greatly shorten the drug design cycle.Novel drug creation from natural active compounds directly or indirectly have become the most convenient and popular pathway recognized today.Currently,tumor immunotherapy represented by PD-1/PD-L1 antibodies is the third revolution in cancer therapy,and the six drugs that have been marketed are all monoclonal antibodies,which have many disadvantages such as long half-life,immunogenicity and poor penetration into solid tumors,and high production cost,while small molecule inhibitor drugs have the advantages of higher bioavailability,high penetration,and low production cost.Due to the fact that PD-L1 and PD-1 binding surface sites are flat and do not have obvious active cavities,the development of small molecule drugs lagged behind monoclonal antibodies,and no drugs have been marketed yet.Therefore,the development of novel and efficient PD-1/PD-L1 small molecule inhibitors based on the structural features of the receptor molecules by means of computer-aided drug design will provide the necessary theoretical basis and design guidance for the discovery of innovative drugs to broaden the PD-1/PD-L1 pathway-based tumor immunotherapy.The following four main areas of research were carried out in this project.Part Ⅰ:PD-L1 protein has a very flat surface and no obvious active cavity,which makes small molecules difficult to bind to it.However,it has been shown that individual types of small molecules can inhibit PD-1/PD-L1 binding well,and further crystallographic experiments have demonstrated that one small molecule binds two PD-L1 molecules instead of small molecules binding to PD-L1 protein in a 1:1 ratio.The mechanism of this novel binding mode is still unclear,so in this section,systematically simulates of the whole process of small molecule inhibitor induced PD-L1 dimer formation by molecular dynamics method were conducted,and the kinetic mechanism of its induced PD-L1 dimer formation was proposed:the inhibitor molecule first binds to the chain A of PD-L1,then induces the chain B binding,and finally forms a stable dimer complex.Molecular dynamics simulations also confirmed that the dimeric proteins formed after small molecule induction are structurally stable,and the pockets formed by dimeric proteins can be used for designing small molecule drug based on PD-L1 receptor.Part Ⅱ:Since the most important feature of natural glycoconjugates is that most of them act on the cell surface.Such drugs are much less disruptive to the whole cell and thus to the whole than drugs that enter the nucleus and cytoplasm of the cell.In this section,four types of glycans(glycosides)compounds with antitumor activity synthesized in our laboratory were subjected to molecular docking screening with PD-L1 protein and PD-L1 inhibition activity experiments.The first group is the compounds after sulfation modification of monosaccharides,the second group is the compounds after modification with the backbone of iridoid glycoside(catalpol),the third group is the compounds after modification with the backbone of iridoid glycoside(picrosidei II),and the fourth group is the compounds after modification with the backbone of iridoid glycoside(Geniposide).The results showed that the sulfated monosaccharides had the weakest binding to the target proteins,geniposide were slightly stronger,catalpol also had inhibitory effects on PD-L1 dimer proteins and had some room for design and modification.The inhibitory effect of picroside Ⅱ on PD-L1 dimer protein was stronger than that of other compounds.The four compounds P Ⅱ-2,P Ⅱ-3,P Ⅱ-5,and P Ⅱ-7 with higher docking scores and activity were selected for kinetic validation.The complexes formed by P Ⅱ-2 and PD-L1 protein were stable in RMSD values during kinetic simulations,which were consistent with the results of activity experiments.However,the overall molecular docking scores of all four classes of compounds were lower than those of the reported inhibitor molecules,and they all had greater scope for modification design.Part Ⅲ:In view of the overall low anti-PD-L1 activity of the glycans(glycosides)tested in Part Ⅱ,in this part,the cavity of the PD-L1 dimer protein was analyzed,which was hydrophilic at the outer edge and hydrophobic at the inner edge,and a targeted series of compounds were designed based on iridoid glycoside(catalpol)by using a "dual pharmacodynamic nucleation hybridization combination".A series of novel compounds were designed by adding different types of hydrophobic groups to the backbone of iridoid glycoside(catalpol).The designed compounds were screened by molecular docking method to identify compounds with low binding energy,and the docking score was combined with the results of further PD-L1 small molecule inhibitor design modification strategy.Compounds with docking scores≤-14 kcal/mol were selected for kinetic simulation studies.The results showed that the small molecule inhibitors 2-1,2-2 and 2-5 formed by hydroxyl substitution at the C10 position of iridoid glycoside(catalpol)backbone were strongly and stably bound to the PD-L1 dimer under the same docking parameter settings,and the three small molecules were expected to be the lead molecules for further modification and synthesis.Among them,the lowest binding energy of compound 2-2 was-14.66 kcal/mol(the corresponding Ki is17.85 pM),which was 1.52 × 106 fold higher compared than the value of catalpol(-6.72 kcal/mol,corresponding to Ki is 11.77 μM).Further docking experiments were performed on the new glycosides formed by replacing the sugar ligands of iridoid glycoside and the results showed that the binding ability of the compounds with PD-L1 protein was ranked as follows:ribose>fucose>galactose,mannose,rhamnose>maltose>fructose>lactose.The results provide a theoretical basis for the next step of synthesizing small molecule inhibitors of glycans with high inhibitory activity and diverse structural types.Part Ⅳ:Screening of receptor-based inhibitors from small molecule databases is an effective tool for innovative drug discovery.In this part,the screening of small molecule inhibitors of PD-L1 protein was expanded by downloading a database containing 1 million small molecule compounds from the Lead-Like module of the ZINC database provided by Irwin and Shoichet’s laboratory in the Department of Medicinal Chemistry,University of California,San Francisco,according to the criteria of molecular weight 300-400 and lipid-water partition coefficient-1－4.The small-molecule inhibitors of PD-L1 protein were screened in high throughput using the Autodock Vina program,and 3 5 compounds whose docking scores lower than the original ligand in crystal(binding energy:-12.4 kcal/mol)were successfully screened.Among them,four compounds,ZINC000008021169,ZINC000013 63 9809,ZINC0000094213 87 and ZINC000018223 166,had binding energy to PD-L1 protein less than or equal to-13 kcal/mol.The screened compounds provide a new backbone and design ideas for the design of PD-L1 small molecule inhibitors. |
PDF Full Text Request