The Application Of Biological Network Regulations In Drug Development

In the studies of traditional drug development and biology, researchers always focus on a sole gene or protein target to design corresponding experiments, so they can not know overall interactions between drugs and organisms. With the sharp development of modern biology theory and experimental technology, it is known that organisms often exhibit physical functions by functional pathways consisting of compositions like kinases and transcription factors executing certain interactions in order, which might be more complicated when multiple pathways make up a biological regulation network by cross-talks to each other. Thus the focus of biological researches turns to the identification of potential targets from a biological regulation network by network topology analysis or dynamics in regulations of key cell progresses, which initiates a new era of drug development.In general, biological network regulations use mathematic models to describe biological observations and thus integrate multidisciplinary researches, which involve biological chemistry, molecular biology, chemical reaction kinetics, graphic theory, computational science and medicinal chemistry. According to the concrete objects to be investigated, it could be divided into two branches:one is the systems biology to quantitatively describe the behavior of every component in the network by reaction kinetic models, and the other is the network biology to differentiate nodes by their geometric characteristics derived from the graphic theory. Currently, they are both used to identify a sole key target or optimal combinations of targets. Based on the predictions, researchers can perform "multi-target drug design" studies or discover novel drug candidates with certain biological activities by multiple technologies in modern drug development. Further, biological network regulations could be applied to explore how compositions in a biological network respond to the stimulus of a drug candidate with known activities, which should be a new way to evaluate the efficiency and safety of a drug candidate from a system viewpoint.Consequently, based on previous achievements in our group, we planned to do researches on "The application of biological network regulations in drug development" and divided it into three issues:1. Systems biology studies on the RSK2signaling transduction pathway,2. Discovery and structural modifications on RSK2inhibitors,3. Predicting the drug safety for Traditional Chinese Medicine (TCM) through a comparative analysis of withdrawn drugs using pharmacological network. In details, we constructed a computational model for the RSK2signaling transduction pathway, and verified that RSK2could induce the phosphorylation of IκBα to result in the activation of the NF-κB pathway by the stimulus of EGF. Moreover, we identified RSK2, EGFR, MEK and PDK1as the key regulators in this pathway through parameter sensitivity analysis, parameter scannings, initial concentration scannings of key factors or kinases and kinase knockdown in silico studies. Then we focused on the RSK2kinase to do corresponding drug development researches including the discovery of RSK2inhibitors by a ligand-based virtual screening, structural modifications, structure-activity relationship explorations and potential binding modes investigations on RSK2inhibitors, which contributed to abundant RSK2inhibitors of natural products, benzohydrazide series, barbituric acid series and indolin-2-one series with both potent activities and novel structures, especially some hits of indolin-2-one series showed superior selectivity against RSK2and moderate anti-proliferation effects on human breast cancer cells and prostate cancer cells. Furthermore, we downloaded drug and target records from the DrugBank database and integrated them into a withdrawn drug-target network and an approved drug-target network respectively. The consequent network topology analysis results implicated that approved drugs had higher degrees, connected component numbers and average neighbor node numbers. Further, we used structures of withdrawn drugs as probes to do pharmacovigilance studies on approved drugs and natural products from a local traditional Chinese medicine database by2D fingerprint similarity calculations, which uncovered47approved drugs already in the discontinued phase or withdrawn in certain regions or countries and made drug safety indications on75natural products.In summary, the dissertation clearly described biological network regulations in the drug development from different aspects, where we firstly focused on the dynamics of a typical cell signal transduction pathway, function studies of a key node in a network and drug safety problems by topological characteristic analysis on drug-target networks as follows. We hope that it would help medicinal chemists understand the mechanism of drug-target interactions in a thorough way, providing useful information and novel aspects to study diseases and drug development.