基于GC指纹图谱及化学计量学预测鱼腥草挥发油的质量标志物及其定量分析

Prediction of Quality Markers and Quantitative Analysis of Volatile Oil from Houttuynia cordata Thunb.Based on GC Fingerprint andChemometrics

目的建立鱼腥草挥发油GC指纹图谱,结合化学计量学预测其质量标志物。方法采用GC对20批干燥鱼腥草(地上及地下部分)挥发油指纹图谱进行相似度评价,结合层次聚类分析(HCA)、主成分分析(PCA)及正交偏最小二乘法-判别分析(OPLS-DA)进行化学模式识别,预测鱼腥草挥发油的潜在质量标志物,并对14种挥发性成分进行定量分析。结果指纹图谱中地上部分有26个共有峰,相似度为0.727~0.993;地下部分有29个共有峰,相似度为0.801~0.997;地上和地下部分有19个共有峰。HCA和PCA可将地上与地下部分聚为两类,OPLS-DA筛选出7个潜在的质量标志物。14种挥发性成分在各自范围内线性关系良好,平均加样回收率为99.2%~102.1%,RSD为1.38%~2.61%。结论该方法稳定可靠,可用于鱼腥草挥发油的质量控制和评价。

OBJECTIVE To establish the GC fingerprint of volatile oil from Houttuynia cordata Thunb.,and to predict its quality markers b y chemometrics.METHODS GC was used to evaluate the similarity of volatile oil fingerprints of 20 batches of dried Houttuynia cordata Thunb.(a boveground and u nderground parts).Combined with hierarchical cluster analysis(HCA),principal component analysis(PCA)and orthogonal partial least squares-discriminant anal ysis(OPLS-DA),chemical pattern recognition was used to predict the potential quality markers of volatile oil of Houttuynia cordata Thunb.,a nd 14 volatile c omponents were quantitatively analyzed.RESULTS There were 26 c ommon peaks in the aerial part of the fingerprints,and the similarity was 0.727-0.993.There were 29 common peaks in the underground part,and the similarit y was 0.801-0.997.There are 19 common peaks in the above-ground and undergr ound parts.HCA and PCA could cluster the aboveground and underground parts into two categories,and OPLS-DA screened out seven potential quality markers.The 14 volatile components showed a good linear relationship in their respective ran ges,whose average recoveries were 99.2%-102.1%with the RSD of 1.38%-2.61%.CONCLUSION The method is stable and reliable,and can be u se d for quality control and evaluation of volatile oil of Houttuynia corda ta Thunb.