目的 基于指纹图谱、网络药理学及分子对接技术初步分析夏枯草Prunella vulgaris 4个部位（穗、茎、叶、根）的潜在质量标志物（quality Markers，Q-Marker）并测定其含量。方法 建立8个产地夏枯草的穗、茎、叶和根的指纹图谱，标定不同部位指纹图谱的共有峰和非共有峰；采用相似度分析、主成分分析（principal component analysis，PCA）、TOPSIS分析和正交偏最小二乘法判别分析（orthogonal partial least squares discriminant analysis，OPLS-DA），运用网络药理学预测夏枯草Q-Marker，同时进行定量分析，对夏枯草品质进行综合评价。结果 8批夏枯草穗、茎、叶和根分别标定共有峰17、22、21和16个。指认出5、11、12、13、14、16和20号峰分别为咖啡酸、芦丁、金丝桃苷、异槲皮苷、异迷迭香酸苷、迷迭香酸和木犀草素。其中咖啡酸、迷迭香酸为4个部位的共有成分，同一药用部位相似度均大于0.97，不同药用部位指纹图谱有明显差异。网络药理学筛选得到包括PTGS2、EGFR等在内的14个核心靶点，对以上靶点进行基因本体（gene ontology，GO）和京都基因与基因组百科全书（Kyoto encyclopedia of genes and genomes，KEGG）功能分析可知涉及癌症中的蛋白多糖、新冠肺炎、脂质和动脉粥样硬化等通路，最后构建“成分-靶点-通路”网络图，选取14个核心靶点与2个成分进行分子对接，结果显示，成分与蛋白之间具有较好的结合性能。结论 所建立的指纹图谱方法准确、可靠，同时结合网络药理学和分子对接技术预测了夏枯草4个部位中2个潜在Q-Marker的活性，为综合评价夏枯草的品质和开发研究提供相关理论基础。

Objective To initially analyze and determine the levels of potential quality Markers (Q-Markers) in four medicinal parts (spikes, stems, leaves, and roots) of Prunella vulgaris based on fingerprinting, network pharmacology, and molecular docking techniques.Methods Fingerprint profiles of the spikes, stems, leaves, and roots of P. vulgaris were established for eight different origins. The common and non-common peaks in the fingerprints of different parts were identified. To predict Q-Markers of P. vulgaris, similarity analysis, principal component analysis (PCA), TOPSIS analysis, and orthogonal partial least squares discriminant analysis (OPLS-DA) were employed, utilizing network pharmacology. Additionally, quantitative analysis was conducted to provide a comprehensive quality evaluation. Results A total of 17 common peaks were identified in the spikes of eight batches of P. vulgaris, 22 in the stems, 21 in the leaves, and 16 in the roots. Peaks 5, 11, 12, 13, 14, 16, and 20 were identified as caffeic acid, rutin, hyperoside, isoquercitrin, salviaflaside, rosmarinic acid, and luteolin, respectively. Caffeic acid and rosmarinic acid were present in all four parts, with similarity values exceeding 0.97 within the same medicinal part, while the fingerprints of different medicinal parts showed significant differences. Network pharmacology screening identified 14 core targets, including PTGS2 and EGFR. Gene ontology (GO) and Kyoto encyclopedia of genes and genomes (KEGG) pathway analysis revealed involvement in protein polysaccharides in cancer, COVID-19, lipid metabolism, and atherosclerosis. A “component-target-pathway” network diagram was developed. For molecular docking, 14 core targets and two components were chosen. The results indicated that the components exhibited strong binding properties with the proteins. Conclusion The established fingerprinting method is both accurate and reliable. Combined with network pharmacology and molecular docking technology, it predicted the activity of two potential Q-Markers in the four medicinal parts of P. vulgaris. This study provides a theoretical foundation for the comprehensive quality evaluation of P. vulgaris and the exploration of new medicinal parts.

R286.2

辽宁省科技厅博士科研启动项目（2023-BS-138）