High‐resolution chromatography/time‐of‐flight MS ^E with <i>in silico</i> data mining is an information‐rich approach to reactive metabolite screening

Electrophilic reactive metabolite screening by liquid chromatography/mass spectrometry (LC/MS) is commonly performed during drug discovery and early-stage drug development. Accurate mass spectrometry has excellent utility in this application, but sophisticated data processing strategies are essential to extract useful information. Herein, a unified approach to glutathione (GSH) trapped reactive metabolite screening with high-resolution LC/TOF MS(E) analysis and drug-conjugate-specific in silico data processing was applied to rapid analysis of test compounds without the need for stable- or radio-isotope-labeled trapping agents. Accurate mass defect filtering (MDF) with a C-heteroatom dealkylation algorithm dynamic with mass range was compared to linear MDF and shown to minimize false positive results. MS(E) data-filtering, time-alignment and data mining post-acquisition enabled detection of 53 GSH conjugates overall formed from 5 drugs. Automated comparison of sample and control data in conjunction with the mass defect filter enabled detection of several conjugates that were not evident with mass defect filtering alone. High- and low-energy MS(E) data were time-aligned to generate in silico product ion spectra which were successfully applied to structural elucidation of detected GSH conjugates. Pseudo neutral loss and precursor ion chromatograms derived post-acquisition demonstrated 50.9% potential coverage, at best, of the detected conjugates by any individual precursor or neutral loss scan type. In contrast with commonly applied neutral loss and precursor-based techniques, the unified method has the advantage of applicability across different classes of GSH conjugates. The unified method was also successfully applied to cyanide trapping analysis and has potential for application to alternate trapping agents.