基于HPLC特征图谱、化学模式识别分析和多指标成分测定对衢枳壳及其相似药用部位的质量差异研究

Research on Quality Difference between Quzhou Aurantii Fructus and its Similar Medicinal Part based on HPLC Characteristic Fingerprint, Chemical Pattern Recognition Analysis and Multiple Compounds Determination

关键词:  衢枳壳  胡柚皮  HPLC  特征图谱  化学模式识别  质量差异评价

Key words:  Quzhou Aurantii Fructus  Huyou Peel  HPLC  characteristic fingerprint  chemical pattern recognition method  quality difference evaluation

摘要:

目的 研究并揭示新“浙八味”衢枳壳与相似药用部位胡柚皮的差异性成分及含量特征，为其质量评价提供依据。方法 建立10批衢枳壳和12批胡柚皮的HPLC特征图谱并进行相似度评价，采用主成分分析（principal component analysis, PCA）和正交偏最小二乘法判别分析（orthogonal partial least square discriminate analysis, OPLS-DA）筛选差异性成分，结合特征峰相对峰面积比值探究不同部位质量差异，通过特征性成分的含量测定及t检验和聚类分析验证方法可靠性。结果 建立了衢枳壳与胡柚皮的特征图谱，相似度均大于0.9，且组内相似度高于组间相似度；分别确认衢枳壳中12个特征峰，胡柚皮中15个特征峰，其中10个为共有峰，经对照品指认圣草次苷等8个色谱峰。化学模式识别分析中，PCA提取出2个主成分，累积方差贡献率可达83.841%；OPLS-DA分析筛选出新北美圣草苷、芸香柚皮苷、柚皮苷、新橙皮苷4个差异性成分。统计分析结果显示，峰8、峰13的峰面积比值可区分两种不同药用部位。含量测定结果表明特征成分新北美圣草苷、芸香柚皮苷、柚皮苷、新橙皮苷在10批衢枳壳中含量范围分别为0.25%~0.55%、0.84%~2.29%、4.24%~8.57%、3.40%~10.13%；在12批胡柚皮中含量范围分别为0.66%~1.40%、0.30%~0.45%、3.60%~5.52%、2.19%~3.69%，且其在衢枳壳与胡柚皮中的平均含量均有极显著差异（P<0.01）。基于4个特征成分对22批样品的聚类分析可将不同批次的样本准确地分为衢枳壳与胡柚皮两类。结论 本研究建立的方法能特异性识别衢枳壳及其相似药用部位的差异成分并对两者进行有效区分，为衢枳壳质量控制提供依据。

Abstract:

OBJECTIVE To study and reveal the differential components and content characteristics of new 'Zhebawei' Quzhou Aurantii Fructus (QZQ) and its similar medicinal parts Huyou Peel (HYP), and provide a basis for the quality evaluation of QZQ. METHODS The HPLC characteristic fingerprints of 10 batches of QZQ and 12 batches of HYP were established, and similarity were evaluated. Principal component analysis (PCA) and orthogonal partial least square discriminate analysis (OPLS-DA) were applied for differential components screening. The relative peak area ratios of characteristic peaks were analyzed to explore quality differences between QZQ and HYP. The contents of the selected characteristic components were determined, and t test and hierarchical cluster analysis (HCA) were performed to verify method reliability. RESULTS The HPLC characteristic fingerprints of QZQ and HYP were established, and the similarity (all greater than 0.9) within the group was higher than that between the groups. Totally 12 characteristic peaks in QZQ and 15 characteristic peaks in HYP were confirmed, including 10 common peaks, and 8 chromatographic peaks were identified by reference substances. Chemical pattern recognition analysis was carried out based on the area of 17 characteristic peaks in each batch. Two principal components were extracted by PCA, and the cumulative variance contribution rate was 83.841%. Four characteristic components, neoeriocitrin, narirutin, naringin and neohesperidin were screened by OPLS-DA. The results of statistical analysis showed that the peak area ratio of peak 8 and peak 13 could be used for QZQ and HYP classification. In addition, the contents of neoeriocitrin, narirutin, naringin and neohesperidin were 0.25%?0.55%、0.84%?2.29%、4.24%?8.57% and 3.40%?10.13% in 10 batches of QZQ, and 0.66?1.40%, 0.30?0.45%, 3.60?5.52% and 2.19?3.69% in 12 batches of HYP, which showed significant difference (P<0.01) in QZQ and HYP. HCA based on 4 characteristic components could clearly divided 22 batches of samples into QZQ and HYP. CONCLUSION The method established in this study can specifically identify the characteristic components of QZQ from its similar medicinal parts and effectively distinguish them, and this work would provide a basis for the quality control of QZQ.