基于聚类分析和主成分分析的吴茱萸及制吴茱萸指纹图谱研究

目的 建立吴茱萸生品与炮制品HPLC指纹图谱，通过计算相似度，结合化学模式识别方法，评价吴茱萸和制吴茱萸的质量，探讨炮制前后化学成分的差异，阐释吴茱萸炮制后毒性降低的机制。方法 通过高效液相色谱法建立吴茱萸及制吴茱萸的指纹图谱，运用《中药色谱指纹图谱相似度评价系统》(2012版)和SPSS17.0软件对相应图谱进行相似度评价、聚类分析(cluster analysis,CA)和主成分分析(principal analysis,PCA)。结果 吴茱萸和制吴茱萸的指纹图谱相似度均大于0.985，均含有12个相同的共有峰，指认出了其中的5个。以吴茱萸碱为参照峰，比较各共有峰的相对峰面积，发现吴茱萸炮制后除3号峰外，其余共有峰均有不同程度的下降，以2号峰和10号峰有显著性下降。聚CA结果显示，吴茱萸生品可分为2大类，其中来源于江西的为一类，来源于浙江和湖南的为一类；经PCA分析，前3个主成分因子的累积方差贡献率为87.405%，以2、5(绿原酸)、6(金丝桃苷)、7、8(柠檬苦素)、9(吴茱萸碱)、10(吴茱萸次碱)和12号峰对吴茱萸炮制前后化学成分差异变化的贡献较大。结论 吴茱萸HPLC指纹...

HPLC fingerprint of Euodia rutaecarpa and processed E. rutaecarpa based on cluster analysis and principal component analysis

Objective To establish HPLC fingerprints of Euodia rutaecarpa and processed E. rutaecarpa, evaluate the quality of E. rutaecarpa and processed E. rutaecarpa by similarity calculation and chemical pattern recognition method, explore the changes of chemical components before and after processing, and explain the mechanism of toxicity reduction after processing. Methods The fingerprints of E. rutaecarpa and processed E. rutaecarpa were established by HPLC, and the similarity evaluation system of Chinese Medicine Chromatographic Fingerprint (2012 edition) and SPSS17.0 software were used for similarity evaluation, cluster analysis (CA) and principal analysis (PCA). Results The fingerprint similarity of E. rutaecarpa and prepared E. rutaecarpa were greater than 0.985, and they all contained 12 identical common peaks, five of which were recognized. Using evodiamine as the reference peak, comparing the relative peak area of common peaks, it was found that the common peaks of processed E. rutaecarpa decreased in varying degrees except peak 3, and there were significant decreases in peak 2 and peak 10. The results of cluster analysis showed that the raw products of E. rutaecarpa could be divided into two categories, one from Jiangxi and the other from Zhejiang and Hunan. Principal component analysis results showed that the cumulative variance contribution rate of the first three principal component factors was 87.405%, with peak 2, peak 5 (chlorogenic acid) and peak 6 (hypericin), peak 7, peak 8 (limonin), peak 9 (evodiamine), peak 10 (rutacarpine) and peak 12 contribute greatly to the difference of chemical components before and after processing. Conclusion The construction of HPLC fingerprint and chemical pattern recognition of E. rutaecarpa can provide a scientific basis for explaining the mechanism of toxicity reduction after processing.