Page 54 - 202018

南、北五味子药材的HPLC指纹图谱建立及化学模式识别分析 Δ

郭 丽 ，杨忠杰 ，于晓涛 ，贾 陆 ，金少举 ，王 瑞 （1.漯河市中心医院药学部/河南省工程研究中心，河南
1 #
1＊
3
1
2
1
漯河 462005；2.郑州大学药学院，郑州 450002；3.漯河医学高等专科学校药理学教研室/漯河市肿瘤发生与
防治研究创新型科技团队，河南 漯河 462002）
中图分类号 R284.1 文献标志码 A 文章编号 1001-0408（2020）18-2224-06
DOI 10.6039/j.issn.1001-0408.2020.18.09

摘 要 目的：建立南、北五味子药材的高效液相色谱（HPLC）指纹图谱，并进行化学模式识别分析。方法：采用HPLC法，以五味
子甲素为参照，绘制南、北五味子各10批样品（编号分别为N1～ N10、S1～ S10）的HPLC指纹图谱；采用《中药色谱指纹图谱相似
度评价系统（2012版）》进行相似度评价，确定共有峰；采用SPSS 20.0和SIMCA 14.1软件进行聚类分析（HCA）、无监督模式主成分
分析（PCA）、有监督模式正交偏最小二乘法-判别分析（OPLS-DA）；以变量重要性投影值（VIP）大于1为标准，筛选影响南、北五味
子质量的差异标志物。结果：南、北五味子分别指认出32、33个共有峰；10批南五味子和10批北五味子的相似度均大于0.9，南、
北五味子的相似度为0.50；南、北五味子共有19个共有特征峰，共指认出其中的5个，即五味子醇甲、五味子醇乙、五味子酯甲、五
味子甲素、五味子乙素。HCA 结果显示，N1～N10 聚为一类，S1～S10 聚为一类，其中 N1、N3、N8、N9 聚为一类，其余聚为一类；
S1、S3、S6、S9聚为一类，其余聚为一类。无监督模式PCA结果显示，前2个主成分因子的累积方差贡献率为87.19％。有监督模
式OPLS-DA结果显示，五味子甲素、五味子醇甲、五味子酯甲、五味子乙素为影响南、北五味子药材质量的差异标志物（VIP值分
别为2.29、2.24、1.73、1.48）。结论：本研究所建立南、北五味子HPLC指纹图谱准确、科学、简便易行，结合多元统计分析可用于评
价南、北五味子药材的质量。南、北五味子成分有所不同，五味子甲素等为差异标志物。
关键词 南五味子；北五味子；高效液相色谱法；指纹图谱；聚类分析；主成分分析；正交偏最小二乘法-判别分析；差异标志物

Establishment of HPLC Fingerprint of Schisandra sphenanthera and S. chinensis and Analysis of Chemical
Pattern Recognition
1
1
1
1
3
2
GUO Li ，YANG Zhongjie ，YU Xiaotao ，JIA Lu ，JIN Shaoju ，WANG Rui（1. Dept. of Pharmacy，Luohe
Municipal Central Hospital/Henan Engineering Research Center，Henan Luohe 462005，China；2. School of
Pharmaceutical Sciences，Zhengzhou University，Zhengzhou 450002，China；3. Dept. of Pharmacology，Luohe
Medical College/Tumor Occurrence and Prevention Research Innovation Team of Luohe，Henan Luohe 462002，
China）
ABSTRACT OBJECTIVE：To establish HPLC fingerprint of Schisandra sphenanthera and S. chinensis，and to analyze chemical
pattern recognition. METHODS：HPLC method was adopted. Using schizandrin A as reference，HPLC fingerprints of 10 batches of
S. sphenanthera and S. chinensis （N1-N10， S1-S10） were drawn. Similarity Evaluation System of TCM Chromatographic
Fingerprint （2012 edition） was adopted for similarity evaluation to determine the common peaks. SPSS 20.0 and SIMCA 14.1
software were used for HCA，unsupervised madel of PCA，supervised model of OPLS-DA. Using variable importance projection
（VIP）value greater than 1 as the standard，the differential markers that affected the quality of S. sphenanthera and S. chinensis
were screened. RESULTS：S. sphenanthera and S. chinensis were identified 32 and 33 common peaks，respectively. The similarity
of 10 batches of S. sphenanthera and 10 batches of S. chinensis were all higher than 0.9，and the similarity of S. sphenanthera and
S. chinensis was 0.05. A total of 19 characteristics peaks were identified，among which five common peaks were identified as
schisandraol A，schisandraol B，schisantherin A，schizandrin A and schisandrin B by reference. HCA results showed that N1-N10
were clustered into one category，and S1-S10 were clustered into one category，of which N1，N3，N8，and N9 were clustered into
one category，and the rest were clustered into one category；S1，S3，S6，and S9 were grouped together，and the rest were
grouped together. The results unsupervised model of PCA showed that the cumulative variance contribution rate of the first two
principal component factors was 87.20％. Supervised model of OPLS-DA showed that schizandrin A，schisandraol A，schisantherin
A and schisandrin B were the differential markers that affected
Δ 基金项目：河南省科技发展计划（No.192102310090）；漯河市科
the quality of S. sphenanthera and S. chinensis（VIPs were
技创新卷项目（No.20181000009）
2.29，2.24，1.73，1.48，respectively）. CONCLUSIONS：The
＊药师，硕士。研究方向：中药质量评价与药效物质基础。电话：
established fingerprint is accurate，scientific，simple and easy
0395-3356189。E-mail：guoliyy@163.com
# 通信作者：主任药师，硕士。研究方向：医院药学与临床。电 to use，combined with multivariate statistical analysis can be
话：0395-3356116。E-mail：wangrui56116@163.com used to evaluate the quality of S. sphenantherae and S.

·2224 · China Pharmacy 2020 Vol. 31 No. 18 中国药房 2020年第31卷第18期