目的 建立川芎生品和不同炮制程度的酒炙品HPLC指纹图谱，通过指纹图谱动态变化，从客观上确认最佳炮制时间，同时结合化学模式识别方法比较酒炙前后化学成分指纹峰的差异，寻找差异标志物，为川芎的炮制工艺优化及炮制机制研究提供科学依据。方法 采用HPLC法，Hypersil Gold C₁₈色谱柱（250 mm×4.6 mm，5 μm），乙腈-0.3%甲酸水溶液为流动相进行梯度洗脱，体积流量0.8 mL/min，全波长扫描，检测波长254 nm。采用中药色谱指纹图谱相似度评价系统（2012版）计算相似度，运用聚类分析（hierarchical cluster analysis，HCA）、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 squares-discriminant analysis，OPLS-DA）对川芎生品与酒炙品进行化学模式识别研究。结果 建立了川芎生品和不同炮制程度酒炙品的指纹图谱，共标定22个共有峰，通过与对照品比对，指认了6个成分，分别为阿魏酸（7号峰）、洋川芎内酯I（8号峰）、阿魏酸松柏酯（15号峰）、洋川芎内酯A（17号峰）、藁本内酯（19号峰）和丁烯基苯酞（20号峰）；通过指纹图谱动态变化监测，确定了最佳炮制时间为15～20 min；通过HCA、PCA和OPLS-DA可明显将川芎生品与酒炙品分为2类，藁本内酯、丁烯基苯酞等6个成分可能是影响二者质量的差异标志物。结论 建立的指纹图谱方法精密度、重复性、稳定性良好，结合化学模式识别可用于川芎炮制前后整体质量控制，进行炮制过程监测和炮制程度判断。

Objective HPLC fingerprints of raw and processed Chuanxiong (Chuanxiong Rhizoma) under different processing degrees were established. The dynamically changing of fingerprints could help to confirm the best processing time from an objective point of view, meanwhile chemical pattern recognition methods were combined to compare the differences of chemical components before and after processing, which provided scientific basis for the process optimization and processing mechanism of Chuanxiong Rhizoma. Methods A HPLC method was established on a Hypersil Gold C₁₈ column (250 mm×4.6 mm, 5 μm) with acetonitrile- 0.3% formic acid as the mobile phase for gradient elution. The flow rate was 0.8 mL/min, meanwhile, full wavelength scanning was used and chromatogram fingerprints was established at 254 nm. The fingerprints of raw and processed Chuanxiong Rhizoma were determined respectively and analyzed by the software of "Similarity Evaluation System for Chromatographic Fingerprint of Traditional Chinese Medicine". The discriminant analysis of hierarchical cluster analysis (HCA), principal component analysis (PCA) and orthogonal partial least squares-discriminant analysis (OPLS-DA) were employed to distinguish raw and wine processed Chuanxiong Rhizoma. Results The fingerprints of raw and processed Chuanxiong Rhizoma under different processing degrees were established, a total of 22 common peaks were marked. By comparison with the standard, six ingredients were identified, they were ferulic acid (F7), senkyunolide I (F8), coniferyl ferulate (F15), senkyunolide A (F17), ligustilide (F19) and butylidenephthalide (F20), respectively. The results of dynamically changing fingerprints showed that the optimal processing time was 15-20 min. The results of HCA, PCA and OPLS-DA demonstrated that there were obvious distinctions between raw and wine processed Chuanxiong Rhizoma. Six components containing ligustilide and butylidenephthalide may be the markers of quality difference between them. Conclusion The established fingerprint method is accurate, repeatable and reliable. Furthermore, combined with chemical pattern recognition, it could be used for the overall quality control of Chuanxiong Rhizoma before and after processing, and also could be used for processing monitoring and judgment.

R283.1

国家自然科学基金资助项目（81703704）；国家自然科学基金资助项目（81673600）；浙江省重点建设高校优势特色学科（中药学）开放基金项目（ZYAOX2018023）；山西省应用基础研究项目（201801D221436）；山西省脑病重点实验室开放课题（CME-OP-201003）；山西省教育厅科技创新计划项目（2019L0720）；山西省“1331”工程项目（中药微乳与生物新制剂研发平台）；山西中医药大学研究生创新项目（2020CX042）