Comprehensive metabolic profiling of Dictamni Cortex in normal and rheumatoid arthritis rats using UHPLC-Q/TOF-MS/MS coupled with polygonal mass defect filtering.

Due to the low efficiency and time-consuming nature of detecting multiple metabolites from complex backgrounds-especially those with unpredictable metabolic types-as well as significant methodological limitations and data processing complexities, the in vivo metabolite profile analysis of herbs remains a major challenge. Therefore, this study developed an integrated strategy combining VBA background subtraction, polygonal mass defect filter (p-MDF), and untargeted metabolomics for identifying metabolites of quinoline alkaloids and limonoids. Through applying VBA background subtraction and p-MDF, the filtration ranges achieved for the biological samples were 19.30-24.34% and 78.32-85.46%, respectively. Finally, 108 compounds were efficiently and accurately identified from the water extract of Dictamni Cortex (DDW), including 84 quinoline alkaloids and 24 limonoids, among which 24 previously undescribed and 11 first-identified compounds were revealed. And a total of 158 compounds were detected in the plasma, urine, and feces of rheumatoid arthritis (RA) and normal rat following oral administration with DDW, including 69 prototype compounds and 89 metabolites. Untargeted metabolomics was performed to analyse plasma, urine, and feces samples from RA and normal rat. A total of 51, 28, and 9 differential parent compounds and metabolites were identified in the three sample types, respectively. It was speculated that the impact on the hepatointestinal environment and bioavailability under the RA state caused metabolic abnormalities in rats. In conclusion, this study provides a conceptual framework and methodological reference for the comprehensive analysis of herbal metabolites in complex biological matrices.