Abstract

An ideal sample-preparation method for LC-MS metabolomic analysis should be as non-selective as possible for metabolites but still capable to remove the interferences such as salts and proteins. Microextraction methods have not been widely used in this application despite their potential to reduce ionization suppression and/or increase metabolite coverage. The main goal of this M.Sc. project was to develop a new dispersive solid phase microextraction (D-SPME) sample preparation method and investigate whether this approach can improve the coverage of the metabolome from human plasma. Different types of poly-N-isopropylacrylamide hydrogel extraction phases functionalized with vinyl acetate (VAC), acrylic acid (AAC) or N-3-aminopropyl methacrylamide hydrochloride (APMAH) were tested. Sample analysis was performed using three complementary liquid chromatography–high resolution mass spectrometry methods for high, intermediate and low polarity sub-metabolomes respectively. The extraction conditions were optimized in terms of desorption solvent, influence of pH, centrifugation time, extraction time, sorbent to sample ratio, increasing portion of functional monomer and evaluation of reproducibility. Finally, the performance of the optimized D-SPME method was compared against protein precipitation using methanol, which is currently the gold standard method for global metabolomics of human plasma, and commercial core-shell nanoparticles (CERES Nanotrap) functionalized with acrylic acid or Cibacron blue cores. The main criteria used for the comparison were ionization suppression, metabolite coverage and precision. Hydrogel microparticles performed as well as nanotraps in terms of extraction and performed better than nanotraps and methanol precipitation in terms of ion suppression. Hydrogel D-SPME had lower total coverage than methanol precipitation, as expected for a microextraction method, but successfully revealed more than 568 low abundance metabolites in positive ESI mode and 48 metabolites in negative mode that could not be observed using the methanol method. Therefore, hydrogel D-SPME appears to be a promising new method for global metabolomics.