An integrated two-step filtering strategy of collision cross-section interval predicting and mass defect filtering for targeted identification of analogues in herbal medicines using liquid chromatography-ion mobility-mass spectrometry

Rapid identification of chemical analogues in herbal medicines using liquid chromatography-mass spectrometry was an efficient tool for discoveries of potentially active ingredients. Multi-dimensional combination of various separation technologies could significantly enhance the capacities for detection of trace components and discrimination of multiple isomers. In this study, an integrated two-step filtering strategy on liquid chromatography-ion mobility tandem with quadrupole-time-of-flight mass spectrometry (LC-IM-QTOF MS) was developed for identification of analogues in complex matrixes. The extracted raw data were preliminarily filtered by a collision-cross section (CCS) interval generated from power regression with confidence level at 99% for prediction of analogues. Then, the remained ions were further screened using a mass defect filtering (MDF) window based on m/z and decimal m/z of potential skeletons and substituents. By applying this strategy, 86, 102, 73, and 57 isoquinoline alkaloids were identified in herbal materials of Coptis chinensis Franch (CC), C. deltoidea C.Y.Cheng et Hsiao (CD), C. teeta Wall (CT), and Corydalis yanhusuo W.T.Wang (CY). The integrated two-step filtering presented higher efficiencies on exclusion of the background interference and reducing the false-positive rates than previously reported approaches. This study facilitated the application of LC-IM-MS on small molecular analysis and promoted the discoveries of bioactive components of herbal medicines for further pharmacological researches and quality control. (c) 2021 Elsevier B.V. All rights reserved.

Automated summary

The study developed an integrated two-step filtering strategy using LC-IM-QTOF MS for efficient identification of analogues in herbal medicines. By utilizing power regression and mass defect filtering, the method successfully identified multiple isoquinoline alkaloids in different herbal materials, showing higher efficiencies in excluding background interference and reducing false-positive rates compared to previous approaches.