RESUMO
Baccharis trimera is a species recognized by health agencies and recommended by the Brazilian Pharmacopoeia by having medicinal properties. In this work, HR-MAS NMR spectroscopy combined with chemometric tools, such as Principal Components Analysis (PCA), Partial Least Squares Discriminant Analysis (PLS-DA) and Soft Independent Modeling by Class Analogy (SIMCA), were used to evaluate the quality control and authenticity of commercial samples of Baccharis, as well as to discriminate B. trimera samples from other species of the Caulopterae section (B. articulata, B. trimera, B. junciformis, B. milleflora, and B. myriocephala). The high morphological similarity of these species makes it difficult for their identifications and discriminations, even by taxonomists. Different PCA pre-processing (autoscaling, Pareto scaling, and mean centering) allowed to discriminate B. trimera and B. myriocephala from the other species, mainly due to the presence of carquejyl acetate, indicated their chemical similarity. The 1H HR-MAS NMR spectral profile offers the possibility of tracking not only the chemical markers, such as the presence of carquejyl acetate, which can also be helpful in the B. trimera authentication/identification. The application of classification methods in standard samples revealed that PLS-DA models showed better performance on the calibration and validation sets than SIMCA model. However, PLS-DA and SIMCA applied to commercial samples showed that none of the commercial samples were classified as B. trimera, which suggested the lack of strict quality control regarding these products. The methodology developed in the present work might contribute to chemotaxonomy of the genus Baccharis.
