Development of non-targeted metabolomic approaches for the screening of bioactive molecules in Clematis species.
Résumé
Recent advancements in analytical techniques and data treatment have led to the development of new workflows for discovering natural bioactive compounds. UHPLC-HRMS/MS crude extract profiling allows rapid and sensitive analysis of small amounts of plants, but annotation of complex mixture, even using molecular networking (MN) (Wang et al., 2016) can be time-consuming. Furthermore, increasing annotation confidence is a challenge, requiring the integration of diverse approaches, such as implementing taxonomic information through filters, scoring (Tima-R) and comparison to experimental or predicted
databases (Rutz et al., 2019).
Clematis genus, third largest of Ranunculaceae, is widespread throughout the northern hemisphere, used in traditional medicines or ornamental purposes. These different uses can be explained by a wide diversity of compounds leading to different bioactivities (Chawla et al., 2012). To highlight phytochemical differences and identify potential way of valorizing the huge biomass from these crops, 10 species of Clematis were analysed by UHPLC-HRMS/MS. Data were represented using MN, and annotation tools as SIRIUS4 (Dührkop et al., 2019) and Tima-R were employed to refine the compound identification.
Shared clusters of O-glycosylated flavonoids, triterpenoids and lignans had been determined between the 10 species. However, specific clusters had been also identifying for different species for example acetylated O-glycoside flavonoids for C. chinensis and C-glycoside flavonoids for C. viticella which may lead to a modulation of the bioactivity of plant extract (Xiao et al., 2016), while C. koreana, is the only species with a chemical profile rich in alkaloids directing to a different way of valorization