Hits Identification For A Few Anti-virus And Anti-cancer Drug Targets And Their Mechanism Of Action Studies

In the pharmaceutical study,molecular docking is very important for the discovery and optimization of active compounds.Molecular docking predicts the interaction between protein and ligand by computational simulation.With the advance of computer methods,the precision accuracy of molecular docking is improved greatly.Because of its rapidity and efficiency,molecular docking has been widely applied into the scientific research process of large-scale virtual screening.For most systems,molecular docking can get desired results.Targeting EV71 3C protease and PRMT5,we successfully found active compounds with novel scaffold.For dengue NS2B/NS3,it is not suitable to use the molecular docking for virtual screening because of the flat and charge pocket.Therefore,we used the ligand-based virtual screening.It is very important to choose a suitable virtual screening method to discover inhibitors for different targets.The first chapter describes the basic process of molecular docking,and reviews the latest advances in flexible docking and covalent docking,then gives advices for improvement;finally introduces scoring function,especially machine learning scoring function,which is compared with the traditional scoring functions in scoring ability,docking ability,sorting ability and screening ability.Through the review of the latest advances in molecular docking,it is helpful to understand the advantages and disadvantages of different docking softwares,and choose a suitable one for docking.In addition,it may be useful to further improve and develop the molecular docking software.In chapter 2,we introduce our work of the discovery and biochemical validation of an inhibitor targeting EV71 3C protease.Hand,foot and mouth disease(HFMD)caused by EV71 virus can lead to serious neurological complications for patients.EV71 3C protease is an important target for the treatment of HFMD.First,a small molecule inhibitor DC07090 was discovered by docking-based virtual screening,with an IC50 value of 21.72?M.Molecular docking and MD simulation have been used to predict the binding mode of DC07090.The result of mutant assay is in accordance with our prediction from the binding mode.Through sequence alignment,we found that the residues which involved in the interaction with DC07090,is conserved in the Picornaviridae family.Antiviral experiment results show that DC07090 has antiviral activity for both EV71 and CVA16,with EC50 values of 22.09?M and 27.76 ?M,respectively.DC07090 has no toxicity(CC50 > 200?M).Different from the existing inhibitors,such as peptidomimetics and flavonoids,the scaffold of DC07090 is nonpeptidyl,novel and non-toxic.Therefore,DC07090 has a promising prospect for research.In chapter 3,we conducted the research work targeting dengue NS2B/NS3 protease.Dengue NS2B/NS3 protease is critical to viral replication,thus is the main target against dengue virus.Because NS2B/NS3 protease catalytic pocket is flat and polar,we adopted the method of ligand-based virtual screening.First of all,we collected active and inactive compounds,divided them into training set and testing set.The training set was used to build the pharmacophore model ADRR261.The test set validated the precision of the model.Enrichment rate test shows that the model ADRR261 can successfully distinguish active molecules from decoys.Then,we used the model ADRR261 to screen the SPECS database,obtaining DC16,with an IC50 value of 14.8?M.Based on the chemical similarity,we found three more active compounds,which are the analogues of DC16.Structure-activity relationship of the four compounds,shows that all of them can inhibit dengue virus replication.The antiviral activity of DC16 is 6.82?M.But these compounds have certain toxicity,which need further optimization.In this study,we used the pharmacophore model to find novel inhibitors against dengue NS2B/NS3 protease.These research findings would be helpful to the development of antiviral drugs.In chapter 4,we introduced the discovery of a series of selective inhibitors against PRMT5.Protein arginine methylational transferase 5(PRMT5)is overexpressed in some tumor cells.Inhibition of PRMT5 becomes an important method for the treatment of cancer.First of all,we built a virtual screening model and tested the ability to distinguish between actives and decoys.Using the docking model,we successfully found a PRMT5 inhibitor DC_P33 with novel scaffold,with an IC50 value of 35.6?M.We found 14 analogues of DC_P33 based on the chemical similarity.According to the binding mode of DC_P33,the structure-activity relationship of analogues was analyzed,which gave us some clues for further modification.Then,we synthesized 19 analogues of DC_P33.DC_C01 is the most potent with IC50 of 2.8?M.We further explored the structure-activity relationship.Through sequence and structure alignment,we predicted that our compounds have high selectivity against PRMT5.The inhibitory experiment verified our estimation.In the end,the cell-based experiment showed that the compound DC_C01 has a concentration dependent antiproliferative activity against three kinds of tumor cell line Z-138,Jeko-1 and Maver-1,with an EC50 of 12.7?M,12.7?M,and 10.5?M.In this study,we found a novel scaffold and selective inhibitors against PRMT5.These research results may provide some ideas for further optimization and development of the inhibitors.