Quality marker prediction in Trillium tschonoskii based on UHPLC-MS chemical characterisation and network pharmacology.

Abstract

As a folk herbal medicine, Trillium tschonoskii has been used for thousands of years. However, due to the complexity of the chemical constituents of this herb, few investigations have acquired a comprehensive understanding of its quality markers. This study was conducted to characterise the chemical composition of T. tschonoskii and identify its potential quality markers. A systematic analytical method based on ultra-high-performance liquid chromatography coupled with mass spectrometry (UHPLC-MS) was used to characterise the constituents of T. tschonoskii. Multivariate statistical analysis was performed to investigate the chemical differences between different tissues, as well as the relationship between chemical compositions and habitats. The potential quality markers were predicted via network pharmacology and molecular docking, then confirmed by cellular assays. A total of 77 compounds were co-isolated and identified, and among them, 26 were discovered from the genus Trillium for the first time. Ten batches of roots/rhizomes were explicitly clustered into five groups according to the climate types of the habitats, and the clusters of the fruits and roots/rhizomes from the same plants were independent due to the significant difference in chemical composition. Diosgenin had a good docking affinity with the relevant targets within the IL-17 pathway and cytokine pathway and could significantly inhibit TNF-α expression in hypoxic brain microvascular endothelial cells (BMECs). This is the first study to establish the chemical composition profile of T. tschonoskii by UHPLC-MS systematically, and diosgenin was confirmed to be a potential quality marker of T. tschonoskii for the treatment of headaches.