Enhancing coverage in LC–MS-based untargeted metabolomics by a new sample preparation procedure using mixed-mode solid-phase extraction and two derivatizations

Qian Wu,Yamei Xu,Hongchao Ji,Yang Wang,Zhimin Zhang,Hongmei Lu

DOI: https://doi.org/10.1007/s00216-019-02010-x

2019-08-14

Analytical and Bioanalytical Chemistry

Abstract:It is a challenge to expand the metabolome coverage of liquid chromatography (LC)–electrospray ionization (ESI) mass spectrometry (MS) based untargeted metabolomics analysis. The limited coverage is attributed to the weak signal of hydroxyl and carboxyl groups in ESI-MS and the limited capacity of LC separation for metabolites with a wide range of polarities. Here a new sample preparation procedure is proposed to solve these problems. Mixed-mode (reversed-phase and anion-exchange) solid-phase extraction sorbents were used to separate metabolites into hydrophilic amine, hydrophobic amine/alcohol, and organic acid groups. Then, alcohols and carboxylic acids in separated groups were tagged with pyridine with use of two derivatization systems for signal enhancement. Finally, hydrophilic amines were analyzed by LC–MS with a hydrophilic interaction LC column, and the two hydrophobic compound groups were analyzed by LC–MS with a C18 column. From the results for standard samples, the detection limits of the new method are lower than those of the classic solvent extraction–protein precipitation method by 3.3–70 times for five amino acids and by 65–1141 times for five fatty acids. Moreover, the detection limit of this new method is 125 ng mL-1 for cholesterol, which has no signal with the classic method even at 10 μg mL-1. In seminal plasma samples, 110 more metabolites were identified by this new method than by the traditional solvent extraction–protein precipitation method in positive-mode ESI (new method vs traditional method, 65 vs 22 identified by comparing MS/MS spectra with those of standards, 203 vs 136 identified by searching MS spectra in a published database). Among them, 53 carboxylic acids and 21 alcohols were identified only by the new method, and more hydrophilic amine metabolites, such as amino acids and nucleosides, were identified by the new method than by the classic method. Finally, in application to the study of male infertility, more potential biomarkers of oligoasthenoteratospermic infertility were found with the new method (46 potential biomarkers) than with the classic method (19 potential biomarkers) and previously reported methods (10–30 potential biomarkers). Thus, it is demonstrated that this new sample preparation method expands the detection coverage of LC–MS-based untargeted metabolomics methods and has application potential in biological research.