Enhanced detection and identification in metabolomics by use of LC-MS/MS untargeted analysis in combination with gas-phase fractionation.

Abstract

Liquid chromatography coupled to tandem mass spectrometry is one of the most widely used analytical platforms for profiling analysis in metabolomics. One weakness of untargeted metabolomic analysis, however, is the difficulty of identifying metabolites. In fact, the process typically involves mass-based searching of LC-MS and LC-MS/MS data and requires using MS/MS data for unequivocal identification. Current strategies use LC-MS analysis in the scan mode prior to acquiring MS/MS information about targeted metabolites or the "auto MS/MS" mode to fragment automatically the most intense precursor ions. Therefore, in both cases additional injections are required to obtain MS/MS data after data treatment to identify significant compounds whose signals are not so intense. Because an additional procedure is needed to enhance the fraction of metabolites with MS/MS data, in this work, the effectiveness of utilizing different MS/MS parameters across an analytical batch or repetitions of the same sample by using exclusion or inclusion criteria to select precursor ions is assessed. The procedure, known as "gas-phase fractionation (GPF)", was used here for untargeted analysis of serum. The joint use of four methods with a different mass range for selection of precursor ions each provided useful MS/MS information for at least 80% of all molecular entities detected in the MS scan replicates. By contrast, the conventional "auto MS/MS" mode of data acquisition provided MS/MS data for only 48-57% of entities and was therefore less effective toward identifying metabolites. The additional use of GPF improved the detection and annotation of metabolite families such as phospholipids, amino acids, bile acids, carnitines, and fatty acids and their derivatives.