New approaches to plant-based drug discovery

Natural products have been used to treat or prevent diseases since time immemorial. Although their value is widely recognised, plant-based drug discovery presents several challenges. New methods are needed for the rapid analysis of plant extracts, with minimal consumption of starting plant material, resources and time. Moreover, techniques enabling the extrapolation of many types of information with a single analysis are required. Considering these challenges, new approaches for the study of plant extracts and the identification of the active metabolites are presented in this thesis. Among the recent techniques, metabolomics coupled with chemometrics is particularly outstanding. The metabolomic-chemometric approach was applied in a screening of thirty-six plant extracts for antibacterial activity. The PCA model enabled the identification of the most promising extracts and provided an insight allowing several considerations and inputs for future studies. The same technique was used to reap phytochemical variations of Arbutus unedo leaves samples, collected under different conditions. The most phytochemically diverse samples exhibited different antibacterial activities, demonstrating that the composition and biological activity of plant extracts are highly dependent on the conditions under which samples are collected. One of the compounds responsible for antibacterial activity, kaempferol-3-O-rhamnoside, was identified. Another important issues of plant-based drug discovery is the rapid identification of known molecules, that is referred to dereplication approach. Amaryllidaceae alkaloids are secondary metabolites with a wide range of biological activities and structural variety. The most occurring alkaloids are widely known, hence, they were selected for the development of a dereplication protocol. The method allowed the characterization of the bulb extract of Urceolina peruviana working directly on the extract fractions without resorting to purification processes, saving time and resources. Only the fraction containing an alkaloid whose NMR profile was absent in the databases was purified. This alkaloid was named 6-dehydroxy-6-acetamido-nerinine.