Inhibitory Mechanisms Study And Virtual Screening Towards BRD4 As A Novel Therapeutic Target For Liver Fibrosis

Epigenetic regulation of gene expression is currently the focus of intensive research in the postgenomic era. Histones are subjected to a range of post-translational modifications.The histone modifications, usually namely ?histone code’, are found mainly on the N-terminal histone tails that protrude from the globular core of the nucleosome. Histone modifications include acetylation, methylation, phosphorylation, ribosylation, biotinylation, citruillination and SUMOylation. Each modification can affect chromatin structure, and the overall state of chromatin is ultimately determined by combinations of these modifications. “Readers” of epigenetic marks, which recognize covalent modifications of histone proteins or DNA, are structurally diverse proteins and containing one or more evolutionarily conserved effector modules. Bromodomain(BRD) is a protein module that selectively recognize histones as a “reader” by binding to acetylated lysine substrate. The human BRD4 has emerged as a promising drug target for a number of disease pathways. Liver fibrosis is one of the most important disease which regulated by BRD4 for enhancer-mediated gene expression in hepatic stellate cells(HSCs) and several potent BRD inhibitors are discovered recently. However, the detailed inhibition mechanism especially for the inhibitor binding kinetics are not clear.This dissertation aims to study the inhibitory mechanism and discover several optimum leads by computational simulations and virtual screening strategies.Chapter one reviews the epigenetic, histone modifications and BRDs. Especially the classification, structures and functions of BRD. In addition, several computational methods and its concise theory applied in medicinal chemistry are reviewed, including the molecular dynamics, quantum chemistry, substructure or similarity search, virtual screening, and machine learning methods.In chapter two, the dynamic characteristics of ZA-loop in BRD4 are revealed by employing classical molecular dynamics(MD) and state-of-the-art density functional QM/MM MD simulations. The correlation between binding pocket size and ZA-loop motion are also elucidated. Moreover, our simulations found that the compound(-)-JQ1 could be accommodated reasonable in thermodynamics whereas it is infeasible in binding kinetics against BRD4. Its racemate(+)-JQ1 is proved to be both thermodynamically reasonable and kinetics achievable against BRD4, which could explain the previous experimental results that(+)-JQ1 shows high inhibitory effect towards BRD4(IC50 is 77 n M) while(-)-JQ1 is inactive(> 10,000 n M). Furthermore, the L92/L94/Y97 in ZA-loop and Asn140 in BC-loop are identified to be critical residues in(+)-JQ1 binding/releasing kinetics.In chapter three, by structure based virtual screening towards BRD4 based on ZINC chemical library and above mentioned inhibitory mechanism. Fifty compounds are found as active agents against BRD4.Specifically,compound ZINC03870415、ZINC39929324、ZINC85340727、ZINC77256922、ZINC03870413、ZINC03870412、ZINC96316287 shown high potent to be lead compounds by further scaffold-similarity search, while further bioassay validation is required.Chapter four summarizes the pros and cons of computational study on the inhibitory mechanisms and virtual screening towards bromodomain, as well as some useful perspective views are proposed to facilitate BRD selective inhibitor design in future.