Joint GC–MS and LC–MS platforms for comprehensive plant metabolomics: Repeatability and sample pre-treatment

Abstract

Metabolomics nowadays mostly comprises the application of both LC–MS and GC–MS based approaches. Here we investigate different extraction set-ups for these two established analytical platforms in the field of plant metabolomics. Six extraction approaches for Arabidopsis thaliana leaves, varying in extraction solvent composition, extraction temperature and order of solvent addition within the extraction sequence, were analyzed on the two platforms. Our aim was to establish the most suitable analysis protocol, practicable for both LC–MS and GC–MS analysis, in order to obtain as extensive as possible metabolome coverage. One single sample preparation procedure would save time and valuable sample while still offering the complementary datasets generated by GC–MS and LC–MS. All extraction approaches were evaluated based on the following criteria: number of detected m/ z-retention time pairs, heat maps of the detected peaks, and residual enzymatic activity of invertase and phosphatase in the plant leaf extracts. Unsupervised principal component analysis (PCA) was used to evaluate grouping trends between the different extraction approaches. Quality controls, a blend of aliquots of the different extracts, were used to establish a paired evaluation of the repeatability performance of the GC–MS and LC–MS analysis. We conclude that the use of chloroform in the extraction solvent is counterproductive in an untargeted LC–MS metabolomics approach as is heating. Below room temperature (instead of heated) extraction does not significantly degrade GC–MS performance but one should be more cautious with respect to residual enzymatic activity in the plant extract.