Integrated work-flow for quantitative metabolome profiling of plants, Peucedani Radix as a case

Abstract

Universal acquisition of reliable information regarding the qualitative and quantitative properties of complicated matrices is the premise for the success of metabolomics study. Liquid chromatography-mass spectrometry (LC-MS) is now serving as a workhorse for metabolomics; however, LC-MS-based non-targeted metabolomics is suffering from some shortcomings, even some cutting-edge techniques have been introduced. Aiming to tackle, to some extent, the drawbacks of the conventional approaches, such as redundant information, detector saturation, low sensitivity, and inconstant signal number among different runs, herein, a novel and flexible work-flow consisting of three progressive steps was proposed to profile in depth the quantitative metabolome of plants. The roots of Peucedanum praeruptorum Dunn (Peucedani Radix, PR) that are rich in various coumarin isomers, were employed as a case study to verify the applicability. First, offline two dimensional LC-MS was utilized for in-depth detection of metabolites in a pooled PR extract namely universal metabolome standard (UMS). Second, mass fragmentation rules, notably concerning angular-type pyranocoumarins that are the primary chemical homologues in PR, and available databases were integrated for signal assignment and structural annotation. Third, optimum collision energy (OCE) as well as ion transition for multiple monitoring reaction measurement was online optimized with a reference compound-free strategy for each annotated component and large-scale relative quantification of all annotated components was accomplished by plotting calibration curves via serially diluting UMS. It is worthwhile to highlight that the potential of OCE for isomer discrimination was described and the linearity ranges of those primary ingredients were extended by suppressing their responses. The integrated workflow is expected to be qualified as a promising pipeline to clarify the quantitative metabolome of plants because it could not only holistically provide qualitative information, but also straightforwardly generate accurate quantitative dataset.