野生与栽培夏枯草HPLC指纹图谱研究及模式识别分析  

HPLC Fingerprint Analysis and Pattern Recognition of Wild and Cultivated Prunella vulgaris from Different Habitats

作　　者：皮胜玲 胡玉珍[1,2,3] 彭曦[1,2,3] 李亚梅[1,2,3] 林丽美[1,2,3] 夏伯候[1,2,3] 吴萍[1,2,3] 

机构地区：[1]湖南中医药大学药学院,长沙410208 [2]湘产大宗药材品质评价湖南省重点实验室,长沙410208 [3]湖湘中药资源保护与利用协同创新中心,长沙410208

出　　处：《中国药学杂志》2017年第5期367-371,共5页Chinese Pharmaceutical Journal

基　　金：国家自然科学基金资助项目(81503041);中医药行业科研专项(201507002);教育部高等学校博士学科点专项科研基金(20124323120004);湖湘青年科技创新创业平台资助项(2013);一方创新课题(2015YF04);研究生创新课题[2016(X)4]

摘　　要：目的建立野生和栽培夏枯草的HPLC指纹图谱,为夏枯草药材的鉴别和质量控制提供依据。方法采用高效液相色谱法对28批不同产地的野生和栽培药材进行测定,运用主成分分析和最小偏二乘法-判别分析对其进行模式识别研究,分类并筛选其主要差异性组分。结果建立了野生与栽培夏枯草的HPLC指纹图谱,得到12个共有峰,各产地指纹图谱相似度较高。主成分分析不能完全区分野生与栽培夏枯草,最小偏二乘法-判别分析能进行明确的区分,主要导致差异的成分有6个,其中4个为芦丁、迷迭香酸、咖啡酸和木犀草素。结论指纹图谱与模式识别方法相结合可以区分野生和栽培夏枯草,可以作为夏枯草的质量控制和评价的有效手段之一。OBJECTIVE To develop a method of fingerprint analysis of Prunella vulgaris by HPLC for the quality control of P. vulgaris. METHODS Twenty-eight samples of wild and cultivated P. vulgaris, obtained from different habitats were analyzed. The difference in the chromatographic fingerprints of P. vulgaris, samples between the two varieties was identified by chemometric methods including principal component analysis （PCA） and partial least squares discriminate analysis （PLS-DA）. The selected biomarkers were identified by comparing with reference standards. RESULTS The fingerprint of P. vulgaris was established, and 12 common peaks and good similarities were found in the HPLC fingerprints of P. vulgaris from different habitats. PLS-DA result showed obvious distinction between the two varieties of P. vulgaris, while PCA showed poor distinguishing ability. Six compounds were screened as biomarkers, representing major differences between the two varieties. Four of them were identified as rutinum, rosmarinic acid, caffeic acid, and luteolin. CONCLUSION The fingerprint analysis combined with chemical pattern recognition can be applied as a measure for the quality control and comprehensive evaluation of P. vulgaris.

关 键 词：夏枯草 HPLC指纹图谱 模式识别 主成分分析 偏最小二乘法-判别分析 

分 类 号：R284[医药卫生—中药学]