基于氢核磁共振与偏最小二乘法对酸枣仁及其掺伪品的鉴别

Differentiation between Authentic and Adulterated Ziziphi spinosae Semen by 1H NMR Spectroscopy Combined with Partial Least Squares

采用氢核磁共振(1H nuclear magnetic resonance,1H NMR)代谢组学技术对酸枣仁、理枣仁以及不同比例的掺伪粉末进行对比,结合多元统计分析建立一种快速鉴别酸枣仁粉末及其掺伪品的方法。主成分分析显示掺伪比例较低的样品与酸枣仁有部分的重叠。偏最小二乘法-判别分析(partial least squares-discriminant analysis,PLS-DA)显示,当掺伪比例不低于40%时,酸枣仁粉和掺伪品沿t[1]轴完全分开。在此基础上,采用PLS对酸枣仁及其不同掺伪比例粉末样品训练集数据进行回归分析,并建立掺伪判别模型。1H NMR与PLS相结合的模型预测结果准确性较高,可实现酸枣仁中理枣仁掺假的快速鉴别。该方法为酸枣仁及含酸枣仁的保健食品、复方制剂质量评价提供了一条新途径。

In the present study,the chemical composition of pure powders of Ziziphi spinosae Semen(ZSS)and Ziziphi mauritianae Semen(ZMS)and their mixtures in different proportions was compared by 1H nuclear magnetic resonance(1H NMR)-based metabolomic fingerprinting.Then the obtained spectral data were analyzed by multivariate analysis to develop a new method for rapid detection of ZSS adulteration.Principal component analysis(PCA)performed on the 1H NMR data revealed that only a few samples at low adulteration ratios were overlapped with ZSS.In addition,the data were also analyzed by partial least squares-discriminant analysis(PLS-DA).A clear separation was seen between ZSS and 40%adulterated samples.Then a PLS model was established based on the 1H NMR data and was applied to detect adulteration in ZSS falsified by intentional addition of different concentrations of ZMS,and a quantitative calibration model was established.The PLS regression approach provided a highly reliable and accurate prediction model.The results demonstrated that 1H NMR spectroscopy coupled with PLS is a fast and effective technique for checking adulteration in ZSS.Meanwhile,the method provides a potential tool for quality control of ZSS.