Adventures in natural products drug discovery : metabolomic, proteomic, synergy and deep learning studies

Abstract

In the present project, integrated untargeted metabolomic profiling and antiproliferative assays were implemented for smart identification of the antiproliferative metabolites of Australian propolis against MCF7 breast cancer cells. The synergistic chemotherapeutic combinations of the Australian propolis or cannabinoids such as cannabidiol were investigated and validated against different synergy quantitation models and designs. Moreover, label-free quantification proteomics studies were employed to elucidate the potential mechanism of action and biological targets of the mono treatment or the synergistic chemotherapeutic combinations. The holistic metabolomic profiling was performed for black cohosh roots and rhizomes, purslane seeds, and chasteberry fruits with the exploration of its potential effects and molecular targets in vivo for polycystic ovarian syndrome (PCOS), acrylamideinduced neurotoxicity and indomethacin-Induced gastric injury, respectively. Synpredict, a deep machine learning model, was developed to outperform the current state-of-the-art deep learning predictive model of the pairwise anticancer synergistic combinations. Intermediate and early fusion architectures were explored by comparing two publicly available anticancer drug interaction data sources over different synergy metrics. Several natural products were reviewed for their potential pharmacological effects and possible interactions, including ginger, cannabis, and avocado. Firstly, the potential ameliorative and protective effects of ginger and its metabolites were recapitulated against natural, chemical, and radiation-induced toxicities. Secondly, the potential interactions of medicinal cannabis were reviewed together with the recent advances in the mass spectrophotometric-based quantitation of phytocannabinoids in different biological matrices. Finally, the synergistic effects of Chinese herbal medicine (CHM) and biological networks were summarised to highlight the quantitative system pharmacology methods either at molecular or network levels and their applications in CHM. The interdisciplinary approaches and advanced technologies implemented in the current project will help to develop the evidence base for Traditional and Complementary Medicines and their integration with the conventional healthcare systems.

Date of Award	2021
Original language	English