High-Resolution Accurate Mass-Measurements and Metabolite Identification: an Automated Approach Using Fragment Prediction in Combination with Fragment Ion Search (FISh)

Structural elucidation of putative metabolites via LC/MS is a time-consuming process of manual interpretation of MSⁿ data and requires a strong background in metabolism pathways as well as gas-phase fragmentation pathways. In order to evaluate an automated software workflow for metabolite identification and structural elucidation, we processed high resolution accurate mass data using the Fragment Ion Search (FISh) processing tool in Thermo Scientific Mass Frontier 7.0 software. The investigated metabolites were derived from Ticlopidine, a potent thienopyridine antiplatelet drug which is known to be extensively metabolized. Ticlopidine was incubated with human liver S9 enzymes. Final experimental concentrations in the 1 mL incubation mixture were 100mM KPO₄ (pH 7.4), 3mM NADPH, and 3.8µM Ticlopidine. The incubation mixtures were homogenized gently and placed into a water bath (37°C). Aliquots of the reaction solution were withdrawn after 0, 30, and 60min and the reaction was quenched by the addition of acetonitrile. The individual samples were centrifuged and analyzed by LC/MS on a Thermo Scientific LTQ Orbitrap XL mass spectrometer equipped with an electrospray ionization source (ESI) interfaced to a Thermo Scientific Accela 600 pump and Open Accela Autosampler. The high resolution accurate mass data was directly exported to the Mass Frontier 7.0 software program. After the determination of the theoretical fragments of Ticlopidine and specification of common Phase I and Phase II modifications, component detection and FISh processing were used to screen for putative metabolites. By defining the possible modifications of Ticlopidine, the list of predicted fragments was automatically extended by the mass shifts of the modifications. All known and published metabolites were detected and identified by the automated FISh processing approach. FISh further provided the explanation of the MS/MS data including the structures of the fragments and the localization of the biotransformations for each detected component. For the metabolites M1 and M5, the putative structures of the metabolites were confirmed by a comparison between the measured and theoretical fragments. All MS/MS fragment ions were explained. In addition to these five common metabolites, an unusual S-oxide dimer metabolite M6 was detected by FISh based on an in-source fragment. The measured and simulated Full MS data of the co-eluting metabolites M3 and M4 demonstrated the need of high resolution accurate mass data. The exact mass and isotopic pattern clearly helped to verify the putative identification of both metabolites and supported the effectiveness of both in the metabolite identification workflow.