Dual 13 C NMR and LC-MS 2 analysis for chemical profiling of Polynesian plant extracts
Résumé
Five Polynesian plants among the most traditionally used in cosmetopeia and pharmacopeia were studied: Calophyllum inophyllum, Gardenia taitensis, Curcuma longa, Cordia subcordata, and Ficus prolixa. Their extracts were obtained by eco-extraction performed by ultrasoundassisted extraction (UAE) using ethanol:water (7:3, v/v) solvent. Liquid-liquid partition of crude extracts using stepwise gradients (heptane, EtOAc, n-butanol, and water) led to four extracts, and EtOAc extracts were fractionated by centrifugal partition chromatography (CPC). The dereplicative chemical profiling method, combining 13 C NMR and LC-MS 2 analysis, was performed on CPC fractions. NMR analysis were carried out and hierarchical clustering analysis (HCA) was performed on 13 C signals with the Permut Matrix application. The resulting map included several clusters corresponding to molecular structures (Hubert et al., 2014). Custom databases were created with VersaDB (Cordonnier et al., 2024) and used to assign compound structures for each cluster. Right structures were then confirmed manually by 2D NMR analysis. LC-MS 2 data were processed via MZmine 3. Molecular networks were built using GNPS Feature-based molecular networking (Nothias et al., 2020) and annotation was performed with SIRIUS. The intelligent use of these two complementary methods allowed accelerated and efficient characterization of main compounds in complex mixtures. The known chemical composition of plant extracts such as C. longa and C. inophyllum allowed the validation of this process. Compounds identified and annotated for F. prolixa and C. subcordata were described for the first time in these two indigenous Polynesian tree extracts. This powerful strategy led to the quick characterization of cosmetics with antioxidant and inflammatory properties.