From comprehensive profiling to targeted authentication: an FBMN-guided metabolomics strategy for authentication of Pseudostellaria heterophylla Dao-di and non-Dao-di herbs.

Abstract

Pseudostellaria heterophylla (PH), a high-value medicinal herb with significant economic and clinical importance, faces quality standardization challenges due to wild resource depletion and inconsistent cultivated materials. To lay the foundation for establishing reliable quality standards, we developed an integrated metabolomics platform combining ultra-high performance liquid chromatography-high resolution mass spectrometry (UHPLC-HRMS) with feature-based molecular networking (FBMN) for phytochemical profiling. FBMN enabled relatively comprehensive metabolite annotation, identifying cohesive clusters of alkaloids/organic acids/purines via conserved fragmentation pathways, while cyclopeptides formed isolated nodes due to structural heterogeneity. Multivariate statistical analysis (PCA, OPLS-DA) and machine learning (K-means clustering) revealed significant chemometric differences between geo-authentic (Fujian) and other major commercial (Anhui, Guizhou, Jiangsu) samples, which were further validated through reference crude drug-controlled intergroup comparative analysis. Notably, six signature metabolites (pseudostellarins A/F, heterophyllin J, ferulic acid, dihydro ferulic acid, and azelaic acid) exhibited markedly higher concentrations in Fujian PH, serving as robust geographical marker candidates. These results lay the foundation supporting Fujian's status as a Dao-di region and establish a systematic framework for geographical traceability. This FBMN-integrated metabolomics approach not only offers a viable strategy for quality control of Dao-di herbs but also proposes a transferable analytical paradigm for geo-authenticity assessment of other herbs.