Unveiling the absorbed bioactive constituents of Cuscuta seeds: A systematic strategy integrating automated MS annotation, in vivo chemometric screening and bioactivity evaluation

The strategy integrated MATLAB automated MS annotation, <i>in vivo</i> chemometric screening, and progesterone assessment under oxidative stress was established for the identification of absorbed bioactive constituents of <i>Cuscuta</i> seed (CS). MATLAB platform characterized 203 components (flavonols, alkaloids, phenolic acids, etc) in CS extract, outperforming commercial software through its combinatorial "parent molecules + group fragments" database and automated neutral loss/diagnostic ion matching. This approach was designed to effectively minimize in-source fragmentation false positives, while its dual-dimension similarity algorithm refined molecular networking. Furthermore, 20 prototypes and 46 metabolites were discovered and identified in plasma and urine after oral administration of CS by OPLS-DA and MATLAB analysis platform. Cuscutamine and <i>p</i>-coumaric acid exhibited high systemic exposure. Based on <i>in vivo</i> metabolic analysis, 9 major absorbed constituents were revealed. Hyperoside, ferulic acid, cuscutamine, kaempferol, and quercetin demonstrated significant bioactivity by attenuating H₂O₂-induced oxidative damage in R2C Leydig cells and restoring progesterone levels.