State Key Laboratory of Chemo/Biosensing and Chemometrics, College of Chemistry and Chemical Engineering, Hunan University, Changsha 410082, People's Republic of China.
State Key Laboratory of Chemo/Biosensing and Chemometrics, College of Chemistry and Chemical Engineering, Hunan University, Changsha 410082, People's Republic of China.
J Chromatogr A. 2022 Oct 11;1681:463426. doi: 10.1016/j.chroma.2022.463426. Epub 2022 Aug 18.
Quantitation of protoberberine alkaloids is an essential guarantee for efficacy control and medication safety of Coptidis Rhizoma (CR) related medicines. Traditional univariate chromatography faced challenges with co-elution, unknown interferences, and retention time shift when analyzing isomeric analytes in varying sample matrices. We presented a chemometrics-enhanced high-performance liquid chromatography-diode array detection (HPLC-DAD) strategy for simultaneous quantification of six protoberberine alkaloids and processed multi-channels chromatographic-spectral data with four second-order calibration algorithms. Chromatographic conditions were firstly optimized. Four groups of predicted samples were modeled individually with the designed calibration set. Mathematical resolutions were then obtained, and pseudo-univariate regression gave the quantitative concentration of each analyte. Four models were scored on fit, linearity, recovery, and robustness, where alternating trilinear decomposition assisted multivariate curve resolution (ATLD-MCR) exhibited an optimal and stable performance. Besides, the resolved spectra presented high consistency with the actual spectra (r≥0.9993). Limits of quantification (LOQ) fully met the pharmacopoeia stipulation and were 0.17, 0.60, 0.19, 0.74, 0.15, and 0.38 µg mL for columbamine, epiberberine, jatrorrhizine, coptisine, palmatine, and berberine, respectively. The importance of this strategy is to exploit collinearity resolution and additional selectivity that permit accurate quantitation at poor chromatographic resolutions, avoiding individual pretreatment and HPLC optimizations for different samples. This study provides a universal alternative for routine quality assessment of protoberberine alkaloids in CR-related medicines.
定量分析原小檗碱类生物碱是控制黄连（CR）相关药物疗效和用药安全的重要保证。传统的单变量色谱法在分析不同样品基质中的同系物分析物时，面临着共洗脱、未知干扰和保留时间漂移的挑战。我们提出了一种化学计量学增强的高效液相色谱-二极管阵列检测（HPLC-DAD）策略，用于同时定量分析六种原小檗碱类生物碱，并采用四种二阶校正算法处理多通道色谱-光谱数据。首先优化了色谱条件。用设计的校准集分别对四组预测样品进行建模。然后得到数学分辨率，并通过伪单变量回归给出每个分析物的定量浓度。对四个模型的拟合度、线性度、回收率和稳健性进行了评分，其中交替三线性分解辅助多元曲线分辨（ATLD-MCR）表现出最佳和稳定的性能。此外，分辨光谱与实际光谱高度一致（r≥0.9993）。定量限（LOQ）完全符合药典规定，分别为 0.17、0.60、0.19、0.74、0.15 和 0.38 µg·mL 用于小檗碱、表小檗碱、黄连碱、小檗碱、巴马汀和盐酸巴马汀。该策略的重要意义在于利用共线性分辨率和额外的选择性，在色谱分辨率较差的情况下实现准确的定量分析，避免对不同样品进行单独的预处理和 HPLC 优化。该研究为黄连相关药物中原小檗碱类生物碱的常规质量评估提供了一种通用的替代方法。
