The purpose of the study was to confirm the utility of multivariate technique to predict antioxidant activity of selected Eryngium extracts based on chromatographic data. Five Eryngium species were analyzed using Thin Layer Chromatography (TLC) with silica gel to obtain their fingerprints. The highest relative content of quercitrin (marker) was observed for E. giganteum. The chemical similarity between extracts was evaluated by Principal Component Analysis (PCA) and Hierachical Cluster Analysis (HCA), and the results confirm the similarity between the same three E. planum from various natural sites. The total phenolic content with Folin–Ciocalteu reagent and antioxidant activity with DPPH reagent by spectrophotometry were determined, and the highest content of phenolics (concentration of gallic acid, mg⋅mL−1) was obtained for E. campestre, E. amethystinum and E. planum T. The highest antioxidant activity (concentration of gallic acid and inhibition %) were obtained for E. planum T. and E. amethystinum, while the lowest was obtained for E. planum Dębówka. High value of correlation coefficient between TLC and spectrophotometric data using Partial Least Squares Regression (PLS) confirms the usefulness of statistical method for the predicting the biological activity. In conclusion, the chromatographic and chemometric methods were suitable for prediction of antioxidant activity.
Read full abstract