Integrated approach for confidence-enhanced quantitative analysis of herbal medicines, Cistanche salsa as a case

Although far away from perfect, it is practical to assess the quality of a given herbal medicine (HM) through simultaneous determination of a panel of components. However, the confidences of the quantitative outcomes from LC-MS/MS platform risk several technical barriers, such as chemical degradation, polarity range, concentration span, and identity misrecognition. Herein, we made an attempt to circumvent these obstacles by integrating several fit-for-purpose techniques, including online extraction (OLE), serially coupled reversed phase LC-hydrophilic interaction liquid chromatography (RPLC-HILIC), tailored multiple reaction monitoring (MRM), and relative response vs. collision energy curve (RRCEC) matching. Confidence-enhanced quantitative analysis of Cistanche salsa (Csa), a well-known psammophytic species and tonic herbal medicine, was conducted as a proof-of-concept. OLE module was deployed to prohibit chemical degradation, in particular E/Z-configuration transformation for phenylethanoid glycosides. Satisfactory retention took place for each analyte regardless of polarity because of successive passing through RPLC and HILIC columns. Optimum parameters for the minor components, at the meanwhile of inferior ones for the abundant ingredients, ensured the locations of all contents in the linear ranges. The unequivocal assignment of the captured signals was achieved by matching retention times, ion transitions, and more importantly, RRCECs between authentic compounds and suspect peaks. Diverse validation assays demonstrated the newly developed method to be reliable. Particularly, the distribution of mannitol rather than galactitol was disclosed although these isomers showed identical retention time and ion transitions. The contents of 21 compounds-of-interest were definitively determined in Csa as well as two analogous species, and the quantitative patterns exerted great variations among not only different species but different Csa samples. Together, the fortification of OLE-RPLC-HILIC-tailored MRM with RRCEC matching could fully address the demands from confidence-enhanced quantitative analysis of HMs.