关键词: 指纹图谱, 质控指标, 主成分分析, 聚类分析, 发酵虫草菌粉

Abstract:

Currently, guanosine, adenosine, and uridine contents are specified as the quality criteria for related products in the quality standards for fermented Cordyceps powder preparations included in the 2020 edition of Chinese Pharmacopoeia. However, there are many other nucleosides in fermented Cordyceps powder, whose effect on the quality control has not yet been discussed. In this study, an ultra-performance liquid chromatography-ultraviolet detection (UPLC-UV) method was used for the quantitative analysis of 9 nucleosides (uracil, cytidine, guanidine, uridine, adenine, inosine, guanosine, thymidine, and adenosine) in 19 batches of fermented Cordyceps powder samples and products, and the corresponding fingerprints were established. In addition, a method for analyzing the index components was proposed based on statistics. By optimizing the sample extraction method, ultrasound-assisted extraction was selected to process 19 batches of samples. Chromatographic analysis was performed on an Agilent Eclipse Plus C18 column (150 mm×4.6 mm, 3.5 μm) using methanol and water as the mobile phases under gradient elution. The method was validated based on the calibration curves, accuracy, precision, repeatability, and recovery. The fingerprints of the 19 batches of samples were established, and 16 common peaks were obtained. Among them, nine nucleoside peaks were identified by standards, and their concentrations were determined by the external standard one-point method. Similarity evaluation of the fingerprints was conducted; the similarities of the 19 batches of samples were greater than 0.9. Then, chemical pattern recognition was performed. The same classification results were obtained by hierarchical clustering analysis (HCA) and principal component analysis (PCA). Thus, the samples could be segregated into five classes, and the fermented Cordyceps powders were classified as two types with different fermentation processes. Xinganbao capsules, Bailing capsules and Ningxinbao capsules were each separately classified into one class. This indicated that the chemical recognition pattern could effectively distinguish between the fermented Cordyceps powder and different products. PCA was used to calculate the weight value of each common peak for the first time, and the index components among the samples were selected according to the weight value. Finally, the selected index components were used to re-cluster the samples. The results were consistent with those obtained on the basis of the 16 common peaks, thus verifying the rationality of the index components. Therefore, uridine, guanosine, adenosine, adenine, and uracil are recommended for use as evaluation indicators for fermented Cordyceps powder and products, allowing for better distinction between the products on the market. In summary, the combination of liquid chromatographic fingerprints and chemical pattern recognition can provide a simple and reliable method for the analysis and quality control of fermented Cordyceps powder and products.

Key words: fingerprints, quality control index, principal component analysis (PCA), hierarchical clustering analysis, fermented Cordyceps powder