Metabolic profiling of the Brazilian medicinal plant Erythrina velutina (Willd) Fabaceae

Erythrina velutina (EV) is popularly used in Brazil against central nervous system disorders. Recently, anticholinesterase activity (ACA) of the plant has been detected in mice brain, suggesting a potential therapeutic usage for EV to combat Alzheimer disease (AD) symptoms. To get insight into the variation in metabolite composition between different EV trees and to study the effect of growing area and harvest season, we profiled leaf extracts using an untargeted LC-QTOF-MS metabolomics approach. Multivariate data analyses tools were subsequently applied to identify differences and similarities between samples and to link the variation in their metabolic profiles to variation in bioactivity. Principal component analysis of the LC-MS data showed a clear separation of the extracts, which, however, was independent of growth location or harvest time. As a measure of bioactivity, we assayed in vitro both ACA and antioxidant activity, as antioxidant compounds have been shown to exert neuroprotective effects in AD models. Individual antioxidants were subsequently profiled by HPLC with an on-line antioxidant detection system and identified by LC-MS/MS. The antioxidant profiles showed the presence of three main antioxidants present in all leaf material analyzed though at varying levels. Using partial least square regression-discriminant analysis (PLS-DA), the metabolic profiles could be clustered into two groups related to differential bioactivity, for both ACA and antioxidants, and the metabolites most significantly contributing to the clustering were selected. Currently, the structural elucidation of the most active compounds is underway.

Acknowledgments: CAPES, CNPq, PRI.