天然产物研究与开发 ›› 2025, Vol. 37 ›› Issue (5): 930-943.doi: 10.16333/j.1001-6880.2025.5.014 cstr: 32307.14.1001-6880.2025.5.014

• 开发研究 • 上一篇    下一篇

基于指纹图谱及多成分含量测定结合加权TOPSIS的红芪及其不同炮制品质量评价

龚文慧1,高 磊1,辛 力1,刘明贵2*,张 文1,金姝含1,陈小燕1,张金莲1*
  

  1. 1江西中医药大学,南昌 330004;2江西景德中药股份有限公司,南昌 330038
  • 出版日期:2025-05-28 发布日期:2025-05-26
  • 基金资助:
    国家重点研发计划中医药现代化研究专项(2018YFC1707206);江西省中医药标委会标准化项目(2021B01);江西省高水平本科教学团队项目(2252201013);横向课题(53522058/2004)

Quality evaluation of Hedysari Radix and its different processed products based on fingerprint and multi-component content determination combined with weighted TOPSIS

GONG Wen-hui1,GAO Lei1,XIN Li1,LIU Ming-gui2*,ZHANG Wen1,JIN Shu-han1,CHEN Xiao-yan1,ZHANG Jin-lian1*#br#

#br#
  

  1. 1Jiangxi University of Chinese Medicine,Nanchang 330004,China;2Jiangxi Jingde Chinese Traditional Medicine Co.,Ltd.,Nanchang 330038,China
  • Online:2025-05-28 Published:2025-05-26

摘要:

建立红芪生品及不同炮制品(药典法炙红芪、蜜糠炙红芪)的HPLC指纹图谱及多成分含量测定方法,综合评价红芪及其不同炮制品质量。采用HPLC法建立红芪不同炮制品指纹图谱并对5种主要成分进行含量测定,通过主成分分析(principal component analysis,PCA)、正交偏最小二乘-判别分析(orthogonal partial least squares-discrimination analysis,OPLS-DA)等方法对结果进行分析,并结合灰色关联度分析理论计算指标权重值,建立加权-逼近理想解排序法(technique for order preference by similarity to ideal solution,TOPSIS)算法模型,对红芪及其不同炮制品进行质量评价。结果表明,红芪不同炮制品指纹图谱中有22个共有峰,指认出5个共有峰,分别为香草酸(峰5)、毛蕊异黄酮苷(峰8)、芒柄花苷(峰12)、毛蕊异黄酮(峰17)、芒柄花素(峰18);PCA共提取出5个主成分,综合评分中15批次蜜糠炙红芪排名靠前;加权TOPSIS模型结果表明红芪不同炮制品各15批次的相对接近度值分别为生红芪0.116 1~0.447 3,炙红芪0.238 3~0.501 1,蜜糠炙红芪0.318 9~0.654 2。本研究建立的HPLC指纹图谱及多成分含量测定方法操作简便,稳定可靠,加权TOPSIS算法模型合理,可用于红芪及其不同炮制品的综合质量评价。

关键词: 红芪, 指纹图谱, 蜜糠炙红芪, 灰色关联, 熵权, 逼近理想解排序

Abstract:

HPLC fingerprint of Hedysari Radix and its different processed products (Hedysari Radix processed by pharmacopoeia method and Hedysari Radix processed with honey chaff) and multi-component determination method were established to evaluate the quality of Hedysari Radix and different prepared products comprehensively. HPLC method was used to establish fingerprints of different Hedysari Radix processed products, and the contents of five main components were determined. principal component analysis was performed. PCA), orthogonal partial least squares discrimination analysis (OPLS-DA) and other methods were used to analyze the results, and the weights of indicators were calculated according to the grey relational degree analysis theory. The technique for order preference by similarity to ideal solution (TOPSIS) was established to evaluate the quality of Hedysari Radix and its different processed products. The results showed that there were 22 common peaks in the fingerprints of different Hedysari Radix processed products, and five common peaks were identified, which were vanilic acid (peak 5), calycosin-7-O-β-D-glucoside (peak 8), ononin (peak 12), calycosin (peak 17) and ononin (peak 18). A total of five principal components were extracted from PCA, and 15 batches of honey bran and Hedysari Radix were ranked first in the comprehensive score. The results of weighted TOPSIS model showed that the relative proximity values of 15 batches of different prepared Hedysari Radix were 0.116 1-0.447 3, 0.238 3-0.501 1, and 0.318 9-0.654 2, respectively. The HPLC fingerprint and multi-component content determination method established in this study is simple, stable and reliable, and the weighted TOPSIS algorithm model is reasonable, which can be used for comprehensive quality evaluation of Hedysari Radix and its different processed products.

Key words: Hedysari Radix, fingerprint, Hedysari Radix processed with honey chaff, grey correlation, entropy weight, TOPSIS

中图分类号:  R283.1