An online stepwise background subtraction-based ultra-high pressure liquid chromatography quadrupole time of flight tandem mass spectrometry dynamic detection integrated with metabolic molecular network strategy for intelligent characterization of the absorbed chemical-fingerprint of QiangHuoShengShi decoction in vivo

Abstract

QiangHuoShengShi decoction (QHSS) was an ancient and classical traditional Chinese medicine (TCM) prescription. In the previous study, its phytochemical fingerprint had been comprehensively characterized. However, no reports were available on its absorbed prototypes and the related metabolites in rat plasma samples. In this study, an intelligent and innovate analysis strategy was built for characterizing metabolic chemical-fingerprint in rat plasma after oral administration of QHSS extract. Firstly, a very simple and highly efficient online stepwise background subtraction (BS)-based ultra-high pressure liquid chromatography quadrupole time of flight tandem mass spectrometry (UHPLC-Q-TOF-MS/MS) dynamic detection method was established to analyze the plasma samples. Secondly, the intelligent metabolic molecular network (MMN) technology was developed and used for rapidly screening out the metabolites of interest, which was followed by prediction of chemical types using the modified deep-learning assisted mass defect filter (MDF) analysis. Thirdly, the screened metabolites with identification features (metabolic pathways and chemical classification) were deeply characterized based on the MS/MS datasets. Finally, 58 prototypes of QHSS were successfully acquired and subsequently identified, including coumarins, chromones, phthalides, phenolic acids, flavonoids, and saponins. A total of 111 metabolites of the coumarins, chromones, phthalides were filtered to be tentatively characterized. This developed qualitative strategy was very helpful to quickly target medicine-related metabolites in the complex bio-matrix and, importantly, it could further visualize medicine-metabolic pathways hidden in the messy mass spectrum datasets. In all, the innovate strategy would provide a powerful tool for effectively acquiring and decode complex metabolic fingerprint of natural products in vivo.