Page 54 - 《中国药房》2022年14期

藏药鸭嘴花的指纹图谱建立、化学模式识别及含量测定 Δ

干志强 1，2，3＊ ，熊双凤 ，钟 镥 2，3，4 ，罗晴方 2，3，4 ，张 艺 2，3，4 # （1.成都中医药大学药学院，成都 611137；2.成都中医
4
药大学民族医药学术传承创新研究中心，成都 611137；3.成都中医药大学中成药质量评价重点实验室，成都
611137；4.成都中医药大学民族医药学院，成都 611137）

中图分类号 R917；R284 文献标志码 A 文章编号 1001-0408（2022）14-1712-06
DOI 10.6039/j.issn.1001-0408.2022.14.09

摘 要 目的 建立藏药鸭嘴花药材的指纹图谱，并测定鸭嘴花碱、鸭嘴花酮碱的含量，结合化学模式识别分析综合评价其质量。
方法 以鸭嘴花碱为参照，采用《中药色谱指纹图谱相似度评价系统（2012版）》建立11批鸭嘴花药材的高效液相色谱（HPLC）指纹
图谱，指认共有峰并进行相似度评价；采用SPSS 25、SIMCA 14.1软件进行聚类分析、主成分分析、正交偏最小二乘法-判别分析，
以变量重要性投影（VIP）值＞1.0为标准筛选影响鸭嘴花药材质量的差异性成分；采用HPLC法同时测定其中鸭嘴花碱、鸭嘴花酮
碱的含量。结果 11批鸭嘴花样品共有23个共有峰，指认2号峰为鸭嘴花碱、4号峰为鸭嘴花酮碱；相似度为0.920～0.994。聚类
分析结果显示，11批样品可分为3类（距离为14），S1～S8为一类（产地云南、西藏）、S9为一类（产地云南），S10～S11为一类（产地
四川）；主成分分析、正交偏最小二乘法-判别分析结果显示，S9、S10～S11分别为一类，S1～S8被进一步分为2类，S1、S4为一类，
S2～S3、S5～S8 为一类；VIP 值＞1.0 的共有峰包括 2、16、21、17、1、13 号峰。11 批样品中，鸭嘴花碱、鸭嘴花酮碱的含量分别为
4.12～10.22、0.60～3.26 mg/g。结论 所建HPLC指纹图谱和含量测定方法操作简单、准确，结合化学模式识别分析可用于评价藏
药鸭嘴花的整体质量。鸭嘴花碱等成分可能是影响药材质量的差异性成分。
关键词 藏药；鸭嘴花；鸭嘴花碱；鸭嘴花酮碱；指纹图谱；化学模式识别；含量测定；高效液相色谱法

Fingerprint establishment，chemical pattern recognition and content determination of Tibetan medicine
Adhatoda vasica
GAN Zhiqiang 1，2，3 ，XIONG Shuangfeng ，ZHONG Lu 2，3，4 ，LUO Qingfang 2，3，4 ，ZHANG Yi 2，3，4 （1. College of
4
Pharmacy，Chengdu University of Traditional Chinese Medicine， Chengdu 611137， China； 2. Academic
Inheritance and Innovation Research Center of Ethnic Medicine，Chengdu University of Traditional Chinese
Medicine，Chengdu 611137，China；3. Key Laboratory of Quality Evaluation of Chinese Patent Medicine，
Chengdu University of Traditional Chinese Medicine，Chengdu 611137，China；4. Ethnic Medicine School，
Chengdu University of Traditional Chinese Medicine，Chengdu 611137，China）

ABSTRACT OBJECTIVE To establish the fingerprint of Tibetan medicine Adhatoda vasica，and determine the contents of
vasicine and vasicinone，so as to comprehensively evaluate its quality combined with chemical pattern recognition. METHODS
Using vasicine as control，HPLC fingerprints of 11 batches of A. vasica were established with Similarity Evaluation System for
Chromatographic Fingerprints of TCM（2012 edition）. The common peaks were identified and their similarities were evaluated.
Cluster analysis（CA），principal component analysis（PCA）and orthogonal partial least squares-discriminant analysis（OPLS-DA）
were performed by using SPSS 25 software and SIMCA 14.1 software. The variable importance in the projection（VIP）value＞1.0
was used as the standard to screen the differential components affecting the quality of A. vasica；the contents of vasicine and
vasicinone were determined by HPLC simultaneously. RESULTS A total of 23 common peaks were found，and peak 2 was
identified as vasicine，and peak 4 was identified as vasicinone. Their similarities ranged 0.920-0.994. The results of CA showed that
11 batches of samples were clustered into 3 categories（distance was 14）：S1-S8 as one category（origin：Yunnan，Tibet），S9 as
one category（origin：Yunnan），S10-S11 as one category（origin：Sichuan）；the results of PCA and OPLS-DA showed that S9 and
S10-S11 were divided into one category respectively， and
Δ 基金项目 国家重点研发计划-中医药现代化研究重点专项
S1-S8 were further divided into 2 categories：S1，S4 as one
（No.2017YFC1703900）；四川省药品监督管理局中药（民族药）标准提
category，S2-S3，S5-S8 as one category；the common peaks
升项目（No.510201201903121）
with VIP value＞1.0 included peak 2，peak 16，peak 21，peak
＊第一作者 硕士研究生。研究方向：中药、民族药药效物质基础
及质量标准化。E-mail：931125098@qq.com 17，peak 1 and peak 13. Among 11 batches of samples，
# 通信作者 研究员，博士生导师。研究方向：民族药药效物质基 contents of vasicine and vasicinone were 4.12-10.22 and
础 及 质 量 标 准 化 。 电 话 ：028-61932600。 E-mail：zhangyi@cdutcm. 0.60-3.26 mg/g， respectively. CONCLUSIONS Established
edu.cn HPLC fingerprint and content determination method are simple

·1712 · China Pharmacy 2022 Vol. 33 No. 14 中国药房 2022年第33卷第14期