Molecular networks and CPC fractionation for rapid screening of bioactive natural molecules

Natural products of plant origin have emerged as major source of bioactive compounds for the cosmetic and pharmaceutical industries. However, determination of active principle i.e. identifying, isolating and testing each molecule is complex, long and tedious taking into account molecular diversity of the extracts.

The objective of this study is to associate i) the dereplication information obtained by means of molecular networks and ii) a bioguided fractionation by Centrifugal Partition Chromatography (CPC). This combination allows developing a simple and fast approach to obtain and test simplified fractions, even pure molecules, thus facilitating screening and identification of the bioactive principles.

Firstly, the aerial parts of Artemisia annua and Artemisia vulgaris were extracted and their antioxidant activity was evaluated by DPPH, ABTS, CUPRAC and FRAP assays. Extracts with positive response were subjected to UHPLC-HRMS with autoMS/MS experiments in order to build molecular networks using GNPS platform. Molecular networks linked compounds according to spectra similarities (same ions and/or neutral losses) then grouping the numerous analytes by molecular families. Secondly, crude extracts were fractionated using CPC with an adapted Arizona solvent system according to molecular families (e.g. sesquiterpernes; phenolic derivatives). The numerous fractions were evaluated for antioxidant activity. Active fractions of the two plants were analyzed by FIA-HRMS in order to identify the isolated molecules, which can be located on the molecular network.

Therefore, this approach has rapidly revealed the few major phenolic compounds that are responsible for the antioxidant activity of the extracts: Artemisia annua (one compound) and Artemisia vulgaris (four compounds).