Identification of glycerophospholipids using self-built recognition software based on positive and negative ion high-resolution mass spectrometric fragmentation experiments

Glycerophospholipids (GPs) have a wide variety and complex structure, which makes their identification challenging. Our software affords a novel tool for the automated identification of non-target GPs in biological mixtures. Here, we explored the multi-stage fragmentation processes of GPs in positive and negative ion modes, and then constructed multi-stage fragment ion databases. This database includes 8214 simulated GP molecules from a random combination of fatty acids corresponding to 42,439 self-built predicted multi-stage fragment ions in positive ion mode and 31,487 self-built predicted multi-stage fragment ions in negative ion mode (MS ≤ 3). The automatic GP identification (AGPI) software can screen out GP candidates utilizing the MS 1 accurate mass. The isomers of fatty acid chains and the phosphoryl head group can be distinguished using the MS 2 and MS 3 fragment spectra in positive-ion and negative-ion modes. All of the selected 45 GP standards were putatively identified using AGPI software; however, there were false positives because the software cannot distinguish positional isomers of fatty acids. Therefore, the AGPI software could be applied to identify GPs in samples, such as cancer cells; we successfully identified 41 GPs in cancer cells.