Metabolomic analysis among ten traditional “Arnica” (Asteraceae) from Brazil

• Published in: Journal of ethnopharmacology 2021-01, Vol.265, p.113149-113149, Article 113149
• Main Authors: de Athayde, Amanda Ellen, de Araujo, Carlos Eduardo Salles, Sandjo, Louis Pergaud, Biavatti, Maique Weber
• Format: Article
• Language: English
• Subjects: Arnica - chemistry; Asteraceae; Asteraceae - chemistry; Brazil; Brazilian arnica; Chromatography, High Pressure Liquid; Cluster analysis; Mass Spectrometry; Metabolomics; Multivariate Analysis; Plant Extracts - analysis; Plant Extracts - chemistry; Plant Extracts - metabolism; Plant Leaves
• ISSN: 0378-8741; 1872-7573
• DOI: 10.1016/j.jep.2020.113149

Extracts of several Asteraceae species in Brazil are popularly used as anti-inflammatory. Some of these species are popularly recognizes as “arnica” because of the morphological and sensorial analogy with the traditional European Arnica montana. These used species in Brazil were identified as Calea uniflora Less, Chaptalia nutans (L.) Polák, Lychnophora ericoides Mart. Lychnophora pinaster Mart. Lychnophora salicifolia Mart. Porophyllum ruderale (Jacq.) Cass, Pseudobrickellia brasiliensis (Spreng.) R. M. King & H. Rob. Sphagneticola trilobata (L.) Pruski and Solidago chilensis Meyen. However, the comparative chemical profile of these so-called “arnicas” has never been reported in the literature. This work aimed to compare the main plants recognized as “arnica” in Brazil by using metabolomic analysis, based on UPLC-ESI-QTof-MS2 data and multivariate statistical analysis. The metabolites profiling of 10 “arnica” species were established by UPLC-ESI-QTof-MS2. Three tinctures of each species (dry leaves) were produced and one aliquot of each tincture was injected and analyzed three times by UPLC-ESI-QTof-MS2. Data were acquired both in negative and positive modes and processed by MassLynx®, MarkerLynx® and Matlab® softwares. Principal component analysis (PCA) was used to reduce dimensionality and data redundancy; hierarchical trees helped to identify and eliminate contaminated or misplaced injections/samples. To achieve the objectives both hierarchical and k-means clustering techniques were employed to group similar samples or species. Diagnostic analysis of MS data allowed the identification of 54 metabolites. The identification was supported with the use of an external standard, fragmentation pattern and data from the literature. The main classes of identified compounds included phenolic acids, coumarin, flavonoids, heterosides, terpenoids and nitrogen compounds. Cluster analysis revealed that Sphagneticola trilobata, Solidago chilensis and Lychnophora pinaster have some chemical features similar to those of Arnica montana. In contrast, the same statistical analysis also showed that Pseudobrickellia brasiliensis, Porophyllum ruderale and Chaptalia nutans are chemically diverse from Arnica montana. The variability of the samples relied principally on nitrogenated compounds (confidence level 4) found in P. brasiliensis and P. ruderale, three phenolic compounds (level 2) detected in P. brasiliensis and in C. nutans and triterpenes (level 3) found in L. salicifo