Yihang He

Microcos paniculata L. (ML) is an edible herbal tea plant with regional differences in flavor and bioactive quality. However, the chemical and microbial factors associated with these differences remain unclear. This study integrated HS-GC-Orbitrap-MS/MS, 16S rRNA sequencing, sensory evaluation, color analysis, and antioxidant assays to characterize ML samples from Guangdong, Guangxi, and Yunnan, China. Guangdong samples were characterized by the highest total phenolic and flavonoid contents and superior antioxidant capacity, whereas Yunnan samples showed the best sensory performance, particularly for aroma and taste attributes. A total of 133 volatile constituents were tentatively identified, mainly comprising olefins and cycloolefins, aromatic compounds, terpenes, alcohols, and ketones. Chemometric analysis revealed clear origin-dependent separation and identified 12 differential volatile compounds, including p-cymene, L-limonene, α-phellandrene, dimethyl sulfide, and ethylbenzene. Bacterial community analysis showed distinct origin-associated patterns, and several bacterial taxa were correlated with characteristic volatile compounds and quality traits. Functional prediction and multivariate correlation analysis suggested potential associations among environmental factors, bacterial communities, volatile metabolites, and antioxidant properties. These findings provide a foodomics-based framework for understanding geographical quality variation in ML and offer potential markers for origin traceability and quality evaluation of edible herbal tea resources.

In the regulatory market, it is not uncommon for ginseng radix et rhizoma (GR) to be adulterated with panacis quinquefolii radix (PR). Amid the digital transformation, this study puts forward a new method for the identification of GR adulterated with PR. Ultra-high-performance liquid chromatography–quadrupole time-of-flight mass spectrometry (UPLC-QTOF-MS) was used to detect multiple batches of GR and PR to obtain mass spectrometry data. The common ions were isolated from multiple batches of GR and PR, serving as GR and PR’s “ion matrices”. Furthermore, GR and PR’s “ion matrices” were used to eliminate intersecting ion data to extract the top-100 ions as GR and PR “matrix identity cards” (MICs). Then, GR and PR’s MICs were employed as a reference for identification, yielding contrast credibility (CC) as feedback. The results indicated that leveraging the MICs of GR and PR enables efficient and precise digital identification of the two herbs: pure GR showed CC ≥ 95% when matched with GR MIC (≤2% with PR MIC), pure PR showed CC ≥ 93% with PR MIC (≤3% with GR MIC), and non-parametric analysis confirmed significant differences between groups (p < 0.01). Even in 5% PR-adulterated samples, CC ranged from 24% to 28% (avg. 25.8%) when matched with PR MIC, leading to a 26% adulteration detection threshold. Moreover, two adulterated batches were identified among ten GR blind samples, which was consistent with verification via PR-specific pseudo-ginsenoside F₁₁. This research is practically valuable for distinguishing between GR and PR, combating adulteration, and reinforcing GR quality management. It also informs the digital identification of GR via UPLC-QTOF-MS and “MICs”, contributing to the digital quality control of traditional Chinese medicines (TCMs).