Comparative Investigation of the Stems, Leaves, Flowers, and Roots of Centipeda Minima Based on Fingerprinting–Multivariate Classification Techniques

Abstract

Abstract Background Centipeda minima (L.) A. Br. et Aschers, known as Ebushicao (EBSC) in Chinese, has long been used in traditional Chinese medicine for dispelling wind, clearing orifices, detoxification, and swelling. Although the traditional use of EBSC involves the whole plant, during harvesting and processing, separation of the stems, leaves, flowers, and roots often occurs. However, there are few studies on its medicinal parts. Objective A strategy combining high-performance liquid chromatography (HPLC) fingerprinting and multivariate classification techniques are here proposed for the comparison of roots, stems, leaves, and flowers of EBSC. Method The roots, stems, leaves, and flowers of EBSC samples were analyzed and compared based on HPLC fingerprints combined with chemometrics, including hierarchical cluster analysis (HCA), principal component analysis (PCA), partial least-squares discriminant analysis (PLS-DA), and back propagation artificial neural network (BP-ANN). Chemical markers were screened using PLS-DA, and the contents of representative ingredients were determined by an HPLC method. Results The HCA and PCA provided clear discrimination of roots, stems, leaves, and flowers. Moreover, the PLS-DA model and BP-ANN were established to verify the classification results and showed a greater ability to predict new samples. Four representative chemical markers were screened out, and the content of these markers in flowers and leaves was higher than that in stems and roots, and the difference was significant. Conclusions Combining HPLC fingerprinting and multicomponent chemical pattern recognition technology can be used to distinguish different parts of EBSC. The results indicated that brevilin A, quercetin, rutin, and chlorogenic acid, the important active components of EBSC, were mainly present in the leaves and flowers. This is of great significance for the differentiation and identification of the different medicinal parts of EBSC, as well as for the effectiveness of drug usage in clinical practice. Highlights HP LC was used to quickly obtain chemical for fingerprint analysis. HCA, P CA, P LS-DA were used to visualize the discrimination of roots, stems, leaves and flowers of EBSC. P LS-DA model was established to verify the classification results and obtained the chemical marker. BP-ANN model was used to further improve the discrimination accuracy.