目的: 建立南非叶的多成分测定方法及高效液相色谱（HPLC）指纹图谱,并结合化学模式识别的方法,综合评价不同产地的南非叶质量,为今后制定南非叶质量控制标准及其他相关南非叶成分研究奠定基础。方法: 使用HPLC法,采用Welch Ultimate^® AQ-C₁₈色谱柱（250 mm×4.6 mm,5 µm）,以0.2%磷酸水溶液-乙腈为流动相进行梯度洗脱,检测波长258 nm,体积流量1 mL · min^-1,柱温35 ℃。同时建立24批不同产地的南非叶指纹图谱及9个成分HPLC测定方法,并结合化学模式识别对各批不同产地的南非叶进行质量评价。结果: 24批南非叶的指纹图谱的相似度均在0.9以上,建立的对照指纹图谱可稳定有效地对南非叶进行定性判别,指纹图谱中10个共有峰,指认出9个峰。24批南非叶中尿嘧啶、绿原酸、木犀草苷、木犀草素-7-O-β-D-葡萄糖醛酸苷、异绿原酸B、3,5-O-二咖啡酰奎宁酸、1,5-二咖啡酰奎宁酸、芹菜素-7-O-β-D-吡喃葡萄糖苷、4,5-O-二咖啡酰奎宁酸的质量分数分别为0.007 314~0.084 30 mg · g^-1、0.619 6~9.763 mg · g^-1、0.303 8~4.031 mg · g^-1、0.984 6~8.146 mg · g^-1、0.043 29~0.438 7 mg · g^-1、0.537 5~11.57 mg · g^-1、0.437 6~13.78 mg · g^-1、0.032 19~0.720 1 mg · g^-1、0.190 0~1.931 mg · g^-1。通过聚类分析,将24批南非叶分为2类,广东和海南产的南非叶为一类,广西产的南非叶独自为一类;主成分分析的分类结果与聚类分析一致,通过主成分得分进一步研究各产地南非叶质量差异,发现P5、P24、P22地区得分位居前三质量优于其他地区。通过正交偏最小二乘-判别分析,发现造成各地南非叶质量差异的影响能力由强到弱排序为1,5-二咖啡酰奎宁酸、绿原酸、木犀草素-7-O-β-D-葡萄糖醛酸苷、3,5-O-二咖啡酰奎宁酸。结论: 南非叶指纹图谱及多成分测定方法稳定可靠,可弥补南非叶质量控制标准方面的空白,结合化学模式识别方法,指出了南非叶质量优质的产地及影响不同产地南非叶质量应着重控制的成分,为南非叶质量控制及评价提供了坚实的基础。

Objective: To establish a multi-component determination method and HPLC fingerprint for Vernonia amygdalina Del., combined with chemical pattern recognition to comprehensively evaluate the quality of Vernonia amygdalina Del. from different producing, laying a foundation for the future development of quality control standards for Vernonia amygdalina Del. and other related research on Vernonia amygdalina Del.’s components. Methods: The HPLC using Welch Ultimate^® AQ-C₁₈ column (250 mm×4.6 mm,5 µm) and 0.2% phosphoric acid aqueous solution - acetonitrile as moblie phase by gradient elution at the flow rate of 1 mL · min^-1, wavelength was set at 258 nm and the column temperature was 35 ℃. Simultaneously establish 24 batches of fingerprint spectra of Vernonia amygdalina Del. from different origins and a multi-component HPLC determination method for 9 components of Vernonia amygdalina Del.. Combine chemical pattern recognition to evaluate the quality of 24 batches of Vernonia amygdalina Del. from different origins. Results: The similarities of the fingerprint spectra of 24 batches of Vernonia amygdalina Del. were above 0.9. The established control fingerprint spectra could stably and effectively distinguish Vernonia amygdalina Del. qualitatively. There were 10 common peaks in the fingerprint spectra, and 9 peaks were identified. The mass fractions of 24 batches of Vernonia amygdalina Del. components that were uracil, chlorogenic acid, luteolin, luteolin-7-O-β-D-glucuronide, isochlorogenic acid B, 3,5-O-dicaffeoyl quinic acid, 1,5-dicaffeoyl quinic acid, apigenin-7-O-β-D-glucopyranoside, 4,5-O-dicaffeoyl quinic acid were 0.007 314-0.084 30 mg · g^-1,0.619 6-9.763 mg · g^-1,0.303 8-4.031 mg · g^-1,0.984 6-8.146 mg · g^-1,0.043 29-0.438 7 mg · g^-1,0.537 5-11.57 mg · g^-1,0.437 6-13.78 mg · g^-1,0.032 19-0.720 1 mg · g^-1,0.190 0-1.931 mg · g^-1. Through cluster analysis, 24 batches of Vernonia amygdalina Del. were divided into 2 categories. Vernonia amygdalina Del. from Guangdong and Hainan were classified as one category, while Vernonia amygdalina Del. from Guangxi were classified as one category alone. The classification results of principal component analysis were consistent with cluster analysis. Further research on the differences in leaf quality among different regions of Vernonia amygdalina Del. was conducted through principal component scores, and it was found that the P5, P24 and P22 regions ranked in the top three with better quality than other regions. Through orthogonal partial least squares discriminant analysis, it was found that the reasons for the differences in leaf quality among different regions of Vernonia amygdalina Del. were ranked in descending order: 1,5-dicaffeoyl quinic acid, chlorogenic acid, and luteolin-7-O-β-D-glucuronide, 3,5-O-dicaffeoyl quinic acid. Conclusion: The fingerprint and multi-component determination methods of Vernonia amygdalina Del. are stable and effective through methodological detection, which can fill the gap in quality control standards for Vernonia amygdalina Del.. Combined with chemical pattern recognition methods, the high-quality production areas of Vernonia amygdalina Del. and the components that should be emphasized in controlling the quality of Vernonia amygdalina Del. from different production areas are pointed out. A comprehensive evaluation of the quality of Vernonia amygdalina Del. has been conducted, providing a solid foundation for future quality control and evaluation of Vernonia amygdalina Del..