Resolving lipid isomers up to the double bond position level using reversed-phase chromatography, EAD fragmentation, and high-resolution mass spectrometry: Application to pancreatic cancer

Background: Electron-activated dissociation (EAD) is a radical-based fragmentation technique that provides detailed structural information on lipids in a single spectrum, including double-bond positions, sn-1/sn-2 assignment, and molecular species. Its broader application, however, remains limited by the low intensity of diagnostic fragments. An alternative computational approach, called LC=CL, which is an extension of the software Lipid Data Analyzer (LDA), leverages retention time (RT) information obtained by reversed-phase liquid chromatography (RPLC) to identify ω-positions in intact lipid species. Results: Here, we present an integrated RPLC-EAD-TOF-MS/MS workflow that combines LC=CL's RT-based with EAD's fragmentation-based annotation for deep structural characterization using a ZenoTOF 7600. This strategy was validated using both unlabeled and uniformly ¹³C-labeled yeast extracts. By combining the benefits of retention-time identification and radical-induced fragmentation, our method enables the reliable identification of double-bond locations and sn-positional isomers across 15 lipid classes. Significance: In the analysis of plasma samples from pancreatic ductal adenocarcinoma (PDAC) patients and healthy controls, we demonstrate the power of our strategy for unambiguously resolving lipid isomers, revealing structure-specific patterns of dysregulated lipids. A total of 353 lipids were identified across 15 classes composed of glycerolipids, glycerophospholipids, and sphingolipids. Our findings not only confirmed known biomarkers, but also revealed additional chain isomers (e.g., SM 18:1; O2/20:0 and SM 16:1; O2/22:0), where only one of them was differentially regulated. Moreover, we discovered novel double-bond location-specific regulations, such as differential regulation of PC 16:0/18:3(n-6) in PDAC patients, whereas the isomer PC 16:0/18:3(n-3) did not exhibit any significant changes. Such an observation would have remained concealed using conventional methods. Accordingly, the presented RPLC-EAD-TOF-MS/MS platform facilitates detailed structural lipidomics in biologically and clinically relevant samples.