目的 采用超高效液相色谱法（UPLC）分别对18个产地川赤芍Paeonia veitchii的根、茎和叶中的芍药苷、氧化芍药苷、1，2，3，4，6-O-五没食子酰葡萄糖3种成分进行含量测定，并建立其指纹图谱，基于多元统计法综合分析川赤芍根、茎和叶中主要化学成分的整体分布以及在不同部位中的差异性，并对川赤芍品质进行综合评价。方法 采用UPLC法，色谱条件：C₁₈色谱柱（100 mm×2.1 mm，1.8 μm），流动相为乙腈-0.1%磷酸水溶液，梯度洗脱，柱温30℃，体积流量0.2 mL/min，进样量1 μL，检测波长230 nm；采用2012版《中药色谱指纹图谱相似度评价系统》软件建立图谱；并采用差异分析、相似度分析、主成分分析（PCA）、TOPSIS分析和偏最小二乘法判别分析（PLS-DA）对川赤芍品质进行综合评价。结果 根据含量测定结果，川赤芍中氧化芍药苷含量大小为根>茎>叶；芍药苷、1，2，3，4，6-O-五没食子酰葡萄糖含量大小为根>叶>茎。其中，根样品中的芍药苷含量均大于1.8%，符合《中国药典》2020年版规定；茎、叶中芍药苷含量与根较为接近。并建立了根、茎、叶的UPLC指纹图谱，共标定了17个共有峰，并指认出3个共有峰（芍药苷、氧化芍药苷、1，2，3，4，6-O-五没食子酰葡萄糖）。相似度评价结果显示，不同产地川赤芍根、茎、叶的相似度均较好，质量稳定。结论 指纹图谱可有效评价不同产地川赤芍品质。川赤芍茎、叶中芍药苷含量与根中相近，基于《中国药典》2020年版以芍药苷为质量控制指标，为综合评价川赤芍的品质和开发新的入药部位等研究提供相关理论基础。

Objective To determine the content of paeoniflorin, oxypaeoniflorin and 1,2,3,4,6-O-pentagalloyl glucose in different parts of 18 batches of Paeonia veitchii and establish an UPLC method for the fingerprint analysis to make a comprehensive analysis on the overall distribution and the differences of the main chemical components in different parts based on the multivariate statistical analysis, which was used to evaluate the medicinal materials quality of P. veitchii. Methods The separation was performed on a chromatographic Diamonsil C₁₈column (100 mm×2.1 mm, 1.8 μm), with acetonitrile-0.1% phosphoric acid as the mobile phase for gradient elution. Column temperature was 30℃. Volume flow rate was 0.2 mL/min. Injection was 1 μL and the detection wavelength was 230 nm. The fingerprints were established and evaluated by the Similarity Evaluation System of TCM (version 2012). Difference analysis, similarity analysis (SA), principal component analysis (PCA), TOPSIS analysis, and orthogonal partial least squares-discriminant analysis (OPLS-DA) were used to comprehensively evaluate the quality of P. veitchii. Results According to the content determination results, the content of oxypaeoniflorin in P. veitchii was as follows:root>stem>leaf, and the content of paeoniflorin, 1,2,3,4,6-O-pentagalloyl glucose was as follows:root>leaf>stem. Among them, the content of paeoniflorin in root samples was more than 1.8%, which was in line with the provisions of the 2020 edition of China Pharmacopoeia. The content of paeoniflorin in stems and leaves was close to that of in roots. In this study, the UPLC fingerprint of roots, stems and leaves was established, a total of 17 common peaks were identified, and three common peaks (paeoniflorin, paeoniflorin, 1,2,3,4,6-O-pentagalloyl glucose) were identified. The results of similarity evaluation showed that the similarity of roots, stems and leaves of P. veitchii from different habitats was good and the quality was stable. Conclusion Fingerprint could effectively evaluate the quality of P. veitchii from different habitats. Based on the quality control index of paeoniflorin (C₂₃H₂₈O₁₁) in the 2020 edition of the China Pharmacopoeia, the content of paeoniflorin in the stems and leaves of P. veitchii was similar to that of in the roots, this study provided relevant theoretical basis for the comprehensive evaluation of the quality of P. veitchii and the development of new medicinal parts.

R286.2

国家重点研发计划（2018YFC1708005）；四川省科技农村领域重点研发（2019YFN0089）；中央高校基本科研业务费专项资金研究类项目（2020NGD01）