Design Of Virtual Library Based On Linear Encoding And Discovery Of Plasmepsin Ⅱ Inhibitors

Computer-aided drug design(CADD) has been widely used in the whole pipeline of drug discovery, which greatly shorten the research cycle and reduce the expenses. This thesis mainly introduces various approaches of CADD and puts emphasis on the discovery of lead compunds using receptor-based approaches, design of vitual combinatorial database, as well as research on virtual screening and quantitative structure-activity relationship(QSAR) of inhibitors. The first charpter introduced the common approaches of CADD, especially the application of molecular docking in structure-based drug design. Also, we introduced the de novo drug design, virtual combinatorial library design and new chemical entities(NCEs) identification, as well as QSAR and the application of molecular similarity calculation in ligand-based drug design.The second charpter introduced the application of molecular docking in the virtual screening process which discovered a novel Plasmepsin II inhibitor. We successfully discovered five non-peptide inhibitors with novel structures and medium inhibitory activity from commercial databases by using structure-based virtual screening strategy. By analyzing their binding modes, results show that these compounds can form strong hydrogen bonds with key amino acids such as Asp34, Asp214, Val78, Ser218and Gly36, and these key interactions play a vital role in boosting the inhibitory activities.The third charpter introduced two approaches of virtual library design using SMILES encoding. The first one is reaction-based virtual library design, which apply uncommon but valid rules in SMILES. We built the focused library of p38MAP inhibitors and successfully simulated the reaction route of these inhibitors. The second one is product-based virtual library design. We built a large-size focused library of COX-2inhibitors and successfully realized the enrichment of bioactive molecules. At last, we built a vitual library based on some hits using the above-mentioned Plasmepsin II inhibitors, against which further synthesis, structural modification as well as bioactivity tests will be performed in our future work.The forth charpter introduces the research on3D-QSAR of brequniar inhibitors of human DHODH. As the key enzyme involved in the de novo biosynthesis of pyrimidine, the human dihydroorotate dehydrogenase (DHODH) has emerged as a promising target for the development of novel immunosuppressive and antiproliferative agents. In this study, the combination of molecular docking, CoMFA and CoMSIA were employed to investigate the quantitative structure-activity relationship (QSAR) for a series of brequinar derivatives. Based on the proposed binding conformations with molecular docking method, CoMFA and the best CoMSIA models were successfully built with the cross-validated coefficients (q2) of0.667and0.708, cross-validated coefficients (r2) of0.980and0.994and the predictive correlation coefficient (r2pre) of CoMFA and the best CoMSIA models were obtained as0.724and0.732, respectively. The count maps of both models agreed well with the structural characteristics of the tunnel-like ubiquinone-binding site, and outlined the steric, electrostatic, hydrophobicity effects on the potencies of the inhibitors, which provided detailed insights for the development of novel potent DHODH inhibitors.