目的 采用多指标成分定量结合化学计量学对不同产地野生白鲜Dictamnus dasycarpus皮样品进行综合质量评价并建立产地识别模型。方法 采用高效液相色谱法同时测定野生白鲜皮中汉黄芩素、柠檬苦素、白鲜碱、黄柏酮、梣酮5种活性成分的含量。结合相关性分析、差异性分析及主成分分析对不同产地野生白鲜皮进行质量评价；利用系统聚类、正交偏最小二乘法-判别分析及Fisher线性判别建立野生白鲜皮的产地识别模型。结果 不同产地野生白鲜皮化学成分含量测定结果存在差异，可通过5种成分含量进行区分。其中内蒙古的样品质量较好，辽宁的样品其次，吉林的样品质量稍差。利用系统聚类可初步对产地进行识别，但辽宁仍有部分样品无法准确识别；正交偏最小二乘法-判别分析可准确将样品产地进行分类，Fisher线性判别模型结果可准确判别未知产地的野生白鲜皮样品，准确率为100%。结论 以多指标成分定量结合化学计量学建立的评价方法明确了野生白鲜皮质量较好的产地，评价了不同产地野生白鲜皮的质量，建立的野生白鲜皮产地识别模型准确可靠，为野生白鲜皮质量控制及产地快速鉴别提供依据。

Objective To undertake a comprehensive quality assessment of wild Baixianpi (Dictamni Cortex) samples from diverse origins through multi-index component quantification in conjunction with chemometrics and establish an origin identification model. Method The contents of five active components, namely limonin, wogonin, dictamnine, obacunone and fraxinellon, in wild Dictamni Cortex were concurrently determined by high performance liquid chromatography. The quality of wild Dictamni Cortex from different origins was evaluated by integrating correlation analysis, difference analysis, and principal component analysis. The origin identification model of wild Dictamni Cortex was constructed by applying systematic clustering, orthogonal partial least squares-discriminant analysis, and Fisher's linear discriminant. Results There were disparities in the determination outcomes of chemical components in wild Dictamni Cortex from different origins, which could be discriminated based on the contents of the five components. Among them, the samples from Inner Mongolia exhibited superior quality, followed by those from Liaoning, while the samples from Jilin presented slightly inferior quality. The origin could be initially identified through systematic clustering, yet some samples from Liaoning could still not be precisely identified. Orthogonal partial least squares-discriminant analysis could accurately categorize the origin of the samples, and the results of the Fisher's linear discriminant model could precisely discriminate the samples of wild Dictamni Cortex from unknown origins, with an accuracy rate of 100%. Conclusion The evaluation approach established by multi-index component quantification combined with chemometrics clarified the origins with better quality of wild Dictamni Cortex, evaluated the quality of wild Dictamni Cortex from different origins, and the established origin identification model of wild Dictamni Cortex was precise and reliable, providing a foundation for the quality control and rapid origin identification of wild Dictamni Cortex.

R286.2

辽宁省“兴辽英才计划”教学名师项目(XLYC2211011)；辽宁省科技厅科技特派行动计划项目—辽宁省喀左县中药材科技特派团（2024JH5/10400152）；通辽市科技计划项目“通辽市6种道地药材生态适宜性区划和品质区划研究”（TL2023YF035）