Integration of untargeted and pseudotargeted metabolomics reveals specific markers for authentication and adulteration detection of Fritillariae Bulbus using tandem mass spectrometry and chemometrics

• 全文链接
• 请求原文

作   者：

Jiang, Yan;  Li, Xin;  Zhao, Wen-Jing;  Liu, Feng-Jie;  Yang, Lu-Lu;  Li, Ping;  Li, Hui -Jun; 

作者机构：

Nanjing Forestry University;  China Pharmaceutical University;  China Pharmaceut Univ;  Nanjing Forestry Univ; 

关键词：

STEROIDAL ALKALOIDS;  Fritillariae Bulbus;  ANALOGS;  Authentication;  Partial least squares regression;  Pseudotargeted metabolomics;  IDENTIFICATION;  Untargeted metabolomics; 

期刊名称：

Journal of Pharmaceutical and Biomedical Analysis: An International Journal on All Drug-Related Topics in Pharmaceutical, Biomedical and Clinical Analysis

i s s n：

0731-7085

年卷期：

2024 年 242 卷

页   码：

116013-116013

页   码：

摘   要：

Authentication and adulteration detection of closely related herbal medicines is a thorny issue in the quality control and market standardization of traditional Chinese medicine. Taking Fritillariae Bulbus (FB) as a case study, we herein proposed a three-step strategy that integrates mass spectrometry-based metabolomics and multivariate statistical analysis to identify specific markers, thereby accurately identifying FBs and determining the adulteration level. First, an ultra-high performance liquid chromatography-quadrupole time-of-flight mass spectrometry-based untargeted metabolomics method was employed to profile steroid alkaloids in five sorts of FB and screen potential differential markers. Then, the reliability of the screened markers was further verified by the distribution in different FB groups acquired from ultra-high performance liquid chromatography triple quadrupole mass spectrometry-based pseudotargeted metabolomics analysis. As a result, a total of 16 compounds were screened out to be the specific markers, which were successfully applied to distinguish five FBs by using discriminant analysis model. Besides, partial least squares regression models based on specific markers allowed accurate prediction of three sets of adulterated FBs. All the models afforded good linearity and good predictive ability with regression coefficient of prediction (R2p) > 0.9 and root mean square error of prediction (RMSEP) < 0.1. The reliable results of discriminant and quantitative analysis revealed that this proposed strategy could be potentially used to identify specific markers, which contributes to rapid chemical discrimination and adulteration detection of herbal medicines with close genetic relationship.