基于化学模式识别的夜仁合有效部位HPLC指纹图谱研究  

Study on HPLC Fingerprints of the active part of YERENHE based on Chemical pattern recognition

作　　者：吕涵 葛金环 卞宏生[1] 黄莉莉[1] 王艳艳[1] 于爽[1] 李廷利[1] 金阳[1] LÜHan;GE Jin-huan;BIAN Hong-sheng;HUANG Li-li;WANG Yan-yan;YU Suang;LI Ting-li;JIN Yang(Heilongjiang University of Chinese Medicine,Heilongjiang Haerbin150040,China)

机构地区：[1]黑龙江中医药大学,黑龙江哈尔滨150040

出　　处：《时珍国医国药》2023年第1期103-106,共4页Lishizhen Medicine and Materia Medica Research

基　　金：黑龙江省博士后科研启动金资助项目(LBH-Q19057)。

摘　　要：目的建立夜仁合有效部位高效液相色谱(HPLC)指纹图谱,并结合化学模式识别技术对其进行质量评价。方法采用HPLC法,以峰5(槲皮苷)为参照峰,采用《中药色谱指纹图谱相似度评价系统(2012版)》建立10批夜仁合有效部位的HPLC指纹图谱并进行相似度评价,经混合对照品标定并指认共有峰。分别采用SPSS 26.0软件和SIMCA 14.0软件进行聚类分析、主成分分析和正交偏最小二乘法-判别分析对10批夜仁合有效部位的指纹图谱进行质量评价,筛选影响其质量的差异性标志物。结果10批夜仁合有效部位指纹图谱共确认17个共有峰,相似度均大于0.9,并指认了峰3(斯皮诺素)、峰4(阿魏酸)、峰5(槲皮苷)、峰13(五味子醇甲)、峰16(五味子甲素)、峰17(五味子乙素)共6个色谱峰。聚类分析与主成分分析分析结果一致,均将10批夜仁合有效部位分为2类,表明10批夜仁合有效部位的质量存在一定差异;结合正交偏最小二乘法-判别分析筛选出导致批次间质量差异的13个差异性标志物(阿魏酸、五味子醇甲、五味子甲素等)。结论基于化学模式识别所建立的HPLC指纹图谱分析方法准确、简便、稳定,可用于夜仁合有效部位的质量控制。Objective To establish high performance liquid chromatography(HPLC)fingerprint of the effective parts of Yurenhe and to evaluate its quality with chemical pattern recognition technology.Methods HPLC method was adopted,with peak 5(quercetin)as the reference peak,HPLC fingerprint of 10 batches of Effective parts of Yurenhe was established and evaluated by Similarity Evaluation System of TCM Chromatographic Fingerprint(2012 edition),and the common peak was identified after calibration of mixed reference substance.SPSS 26.0 software and SIMCA 14.0 software were used for cluster analysis,principal component analysis and orthogonal partial least squares discriminant analysis,respectively,to evaluate the quality of fingerprints of 10 batches of Effective parts of Yurenhe,and screen the difference markers affecting its quality.Results A total of 17 common peaks were identified in the fingerprints of the effective parts of 10 batches of Yurenhe,with the similarity greater than 0.9,and 6 chromatographic peaks were identified,including peak 3(spinosin),peak 4(ferulic acid),peak 5(quercetin),peak 13(schisandrae a),peak 16(schisandrae a)and peak 17(schisandrae B).The results of cluster analysis and principal component analysis were consistent,and the 10 batches of Yerenhe effective parts were divided into 2 categories,indicating that there were some differences in the quality of the 10 batches of Yerenhe effective parts.Combined with orthogonal partial least squares and discriminant analysis,13 differential markers(ferulic acid,schisandrae methyl,schisandrae methyl,etc.)leading to quality difference between batches were screened out.Conclusion The HPLC fingerprint analysis method based on chemical pattern recognition is accurate,simple and stable,and can be used for the quality control of the effective parts of Yerenhe.

关 键 词：夜仁合 高效液相指纹图谱 化学模式识别 质量控制 

分 类 号：R284.1[医药卫生—中药学]