目的 采用多指标成分定量、化学计量学和Logistic回归分析建立不同产地紫珠叶质量的评价方法。方法 采用HPLC法测定不同产地紫珠叶中木犀草苷、大波斯菊苷、野漆树苷、木犀草素、芹菜素、鼠尾草素、毛蕊花糖苷、芝麻素、白桦脂酸、熊果酸、大叶紫珠萜酮、豆甾醇、β-谷甾醇，同时测定醇溶性浸出物、总灰分。结合化学计量学筛选不同产地紫珠叶的质量差异关键指标，利用Logistic回归分析法综合评价紫珠叶的质量。结果 提取了2个主成分，代表了15个指标87.654%的信息量，能明确区分49批紫珠叶。因子分析法结果显示S34～S49的排名靠前，说明这16批样品品质较优。以VIP值大于1为标准筛选质量差异标志物，结果筛选出毛蕊花糖苷、熊果酸、木犀草苷和大波斯菊苷4个成分。Logistic回归分析结果表明，49批紫珠叶分为3个等级，其中优级和良级各16批，差级17批。结论 多指标定量、化学计量学和Logistic回归分析用于评价紫珠叶质量的方法快速、准确，为不同产地紫珠叶的质量差异评价、质量等级评价提供技术支撑。

Objective To establish a method for evaluating the quality difference of Callicarpae Formosanae Folium from from different producing areas by multi-index quantification, chemometrics, and Logistic regression analysis. Methods The content determination of luteolin-7-O-glucoside, apigenin, rhoifolin, luteolin, apigenin, salvigenin, verbascoside, sesamin, betulinic acid, ursolic acid, calliterpenone, stigmasterol, and β-sitosterol in Callicarpae Formosanae Folium from different producing areas were carried out by HPLC method, and the amount of alcohol-soluble extract and total ash was measured at the same time. Based on the chemical identification model, the key indicators of the quality difference of Callicarpae Formosanae Folium from different producing areas were screened, and the quality difference was comprehensively evaluated by Logistic regression analysis. Results Two principal components were extracted, representing 87.654% of the information content of 15 indicators, which can clearly distinguish 49 batches of Callicarpae Formosanae Folium. The results of factor analysis show that the rankings of S34—S49 were high, indicating that the quality of these 16 batches of samples was relatively good. Using VIP values greater than 1 as the standard for screening quality difference markers, four components were identified: verbascoside, ursolic acid, verbascoside, and cosmos glycosides. The results of logistic regression analysis showed that 49 batches of Callicarpae Formosanae Folium leaves were divided into three grades, with 16 batches of excellent and good grades, and 17 batches of poor grades. Conclusion The method of multi on multi-index quantification, chemometrics, and logistic regression analysis for evaluating the quality differences of Callicarpae Formosanae Folium based is fast and accurate, providing technical support for evaluating the quality differences and grades of Callicarpae Formosanae Folium from different production areas.

R284.1

国家中医药管理局全国中药特色技术传承人才培训项目[国中医药人教函(2023)96号