Font Size: a A A

Discovery Of BRD4 Inhibitors And Molecular Modeling Study On Their Mechanism Of Action

Posted on:2021-01-23Degree:DoctorType:Dissertation
Country:ChinaCandidate:H Y ZhongFull Text:PDF
GTID:1361330620977891Subject:Chemistry
Abstract/Summary:PDF Full Text Request
Epigenetic inheritance refers to the biological phenomenon that changes genetic traits directly without changing genetic material and has genetic characteristics without changing the way of DNA replication,RNA transcription and translation,and protein expression in the traditional ‘central dogma'.At present,epigenetics is mainly controlled by DNA modification,histone post-translational modification and regulation of non-coding RNAs,which have been found to be associated with a variety of diseases.In recent years,epigenetic proteins have gradually become a new class of potential drug targets and have received increasing attention in drug research and development.Bromodomain-containing protein 4(BRD4)is a well-studied member of the bromodomain and extra-terminal structural protein(BET)family and is a hot target for drug design and discovery in diseases such as solid tumors,leukemia and fibrosis.So far,a number of BRD4 inhibitors have been pushed into the clinical trials,but only one inhibitor,Fedratinib,has been successfully marketed,so it is still very urgent to discover novel BRD4 inhibitors.As a member of epigenetic family proteins,the functions of BRD4 are mainly controlled by protein-protein interaction(PPI),and therefore the binding site of BRD4 inhibitors is located on the binding interface of PPI,which poses a great challenge for the design of potent BRD4 inhibitors.In this thesis,computeraided drug design(CADD)technology has been used to advance the studies of BRD4 inhibitors,including assessing the applicability of molecular docking methods to epigenetic proteins,screening small molecule inhibitors with novel structures,and exploring the mechanism of action of the hit compounds.This paper mainly includes the following works:1.We systematically evaluated the predictive capability of nine molecular docking methods,including AutoDock,AutoDock Vina,FITTED,LeDock,UCSF DOCK,Gold,Glide,LigandFit,and Surflex Dock,for epigenetic family proteins.The results illustrated that the evaluated docking methods did not have good performance to the prediction of the binding conformation and binding affinity and identification of active molecules of epigenetic family protein inhibitors.Among the evaluated methods,Gold showed relatively better performance.The docking success rates reached 38.83% and 25.24% for the systems with and without crystal water molecules,respectively.AutoDock Vina and LeDock achieved the docking success rates of 41.75% and 42.72%,respectively,when the crystal water molecules were included,but the docking success rate was only 14.56% when the crystal water molecules were not included,suggesting that the crystal water molecules are very important for the binding of inhibitors.In terms of docking scoring capability,Gold,AutoDock,and AutoDock Vina performed relatively well,but their Pearson correlation coefficients were not ideal,suggesting that the evaluated scoring functions did not have good applicability to epigenetic protein families.2.The importance of the crystal water network in the active site of BRD4 for inhibitor binding was systematically studied,and molecular docking based on the crystal water networks was used to the virtual screening of the Specs compound database.Multiple active compounds with novel structures good activity were successfully discovered and validated by experiment.Molecular dynamics simulations and MM-GBSA binding free energy prediction results show that compounds 3 and 31 have multiple low-energy conformational spaces and still have large space for modification.Improving the substituent group of compound 3 in the WPF pocket and appropriately increasing the interaction with the ZA channel,as well as shortening the group structure of compound 31 in the ZA channel and increasing its interaction with the WPF pocket are all feasible directions for structure optimization.3.The ChemDiv database was virtually screened using a standardized processing process for ligand molecules and molecular docking based on the crystal water networks.The activity evaluation results showed that the strategy effectively improved the prediction accuracy of virtual screening,and five hit compounds with novel scaffolds could be identified.The activities of hit compound 17's analogs were evaluated and the corresponding structure-activity relationships were analyzed.The results show that the BRD4-BD1 domain is quite sensitive to the length of linker,and the structural features of the binding pocket determine the ‘L'-shaped conformation of the compound,which allows it to interact the WPF pocket while forming a key hydrogen bond with Asn140.Therefore,for the modification of hit compound 17,it is very critical to find the appropriate linker to improve the activity;for the R-group region extending into the WPF pocket,the stronger the electron withdrawing effect of the substituent on the phenyl ring,the more conducive to improve the activity of the compound;while excessive substituents will lead to decreased activity,which may be determined by its steric effect,and it is necessary to avoid the simultaneous introduction of substituents at the neighbor and spacing on one side of the phenyl ring.4.The hit compound 17 was compared with the inhibitor I-BET-762 in the clinical stage by classical molecular dynamics(CMD)and umbrella sampling(US)simulations.Their binding free energies,binding modes,and binding/dissociation processes were analyzed and compared.The results showed that the orientations of Tyr139 and the ZA channel were quite critical to the further optimization of hit compound 17,and the differences in MM-GBSA binding free energy between WPF pocket and residues Tyr139 and Asn140 in indicated that they were important amino acid residues for binding,whereas Leu92 and Leu94 were likely to play opposite roles for binding.In addition,the hydrogen bond between the ethyl amide group of inhibitor I-BET-762 and the amide group O atom of Asn140 is an important factor that causes the ZA loop moving outward to open the binding pocket.Thus,introduction of a similar binding mode is likely to bring about a qualitative improvement in the activity of hit compounds.By comparing the performance of the two inhibitors during the dissociation process,it was found that the strength and number of hydrogen bonding interactions with Asn140 directly affect the binding ability of the ligand,which means that appropriately increasing the number of hydrogen bond donors and acceptors of the ligand molecule is beneficial to the improvement of the activity ability in the further optimization.
Keywords/Search Tags:Epigenetic Inheritance, computer-aided drug design, BRD4, molecular docking, crystal water network strategy, classical molecular dynamics, umbrella sampling
PDF Full Text Request
Related items