河南牡丹叶的HPLC指纹图谱及化学模式识别研究  

HPLC fingerprint combined with chemical pattern recognition for quality evaluation of poeny leaves from Henan province

作　　者：宁二娟 陈玲[1] 王学方 张丽先 李晓[1] 张立攀[2] 王法云[2] NING Er-juan;CHEN Ling;WANG Xue-fang;ZHANG Li-xian;LI Xiao;ZHANG Li-pan;WANG Fa-yun(Henan Natural Product Biotechnology Co.,Ltd,Henan Academy of Sciences,Zhengzhou 450002,China;Henan Bussiness Research Institute Co.,Ltd,Henan Academy of Sciences,Zhengzhou 450011,China)

机构地区：[1]河南省科学院河南省纳普生物技术有限公司,郑州450002 [2]河南省科学院河南省商业科学研究所有限责任公司,郑州450011

出　　处：《药物分析杂志》2023年第3期486-493,共8页Chinese Journal of Pharmaceutical Analysis

基　　金：河南省科学院重大科研项目聚焦专项(No.200111002)。

摘　　要：目的:建立河南牡丹叶的HPLC指纹图谱及其化学模式识别方法,为牡丹叶的质量评价研究提供参考。方法:采用Agilent C18色谱柱(250 mm×4.6 mm, 5μm),流动相为乙腈-0.1%甲酸水溶液,梯度洗脱,流速为1.0 mL·min^(-1),检测波长270 nm,柱温35℃,采用“中药色谱指纹图谱相似度评价系统”(2012版)建立对照指纹图谱,考察21批牡丹叶相似度,结合主成分分析和聚类分析对各批牡丹叶进行整体质量评价。结果:21批牡丹叶HPLC指纹图谱,有18个共有峰,分析确定其中的5个化学成分,分别为没食子酸、没食子酸乙酯、芍药苷、鞣花酸和木犀草苷;样品相似度在0.907~0.979,主成分分析将21批样品聚为3类,聚类分析结果与主成分分析结果一致。结论:利用指纹图谱与化学模式识别相结合的方法简便、可靠,可快速评定河南牡丹叶的品质和一致性,可用于牡丹叶的整体质量评价。Objective:To establish an HPLC method and chemical pattern recognition model of poeny leaves from Henan province,providing quality evaluation of poeny leaves.Methods:Analysis was performed on an Agilent C18 column(250 mm×4.6 mm,5μm)with acetonitrile-0.1%formic acid for gradient elution.The flow rate was 1.0 mL·min^(-1),the detection wavelength was set at 270 nm,and the column temperature was 35℃.“Similarity Evaluation System of Fingerprint of Tradition Chinese Medicine(2012 edition)”was adopted to establish refer-ence fingerprint and to calculate the similarities of 21 batches of poeny leaves.Combining with principal compo-nent analysis and cluster analysis,the overall quality of poeny leaves was evaluated.Results:There were 18 com-mon characteristic peaks in the fingerprints of the 21 batches of samples,5 peaks were identified as gallic acid,ethylgallate,paeoniflorin,ellagic acid and luteoloside.Similarities of 21 batches of samples were between 0.907-0.979.Samples were clustered into 3 types by principal component analysis which consistented with clus-ter analysis result.Conclusion:The method combining HPLC fingerprint with chemical pattern recognition is sim-ple and reliable,which can quickly evaluate the quality and consistency of different resources of poeny leaves from Henan province,and is suitable for the evaluation of the overall quality evaluation of poeny leaves.

关 键 词：牡丹叶 高效液相色谱 指纹图谱 化学模式识别 主成分分析 聚类分析 

分 类 号：R917[医药卫生—药物分析学]