LC−MS-Based Chlorine-Signature Query Engine Strategy for Automated Discovery of Phosphate-Containing Metabolites

Abstract Comprehensive profiling of phosphate-containing compounds (PCCs) in untargeted metabolomics remains challenging due to their high hydrophilicity, poor separation, and low stability, which hampers high-throughput screening from large-scale liquid chromatography−mass spectrometry (LC−MS) datasets. Herein, an LC-MS-based chlorine-tagging strategy integrated with a Chlorine Signature Query Engine (Cl-SQE) was established for the rapid identification of potential PCCs in complex matrices. The novel derivatization reagent, [5-(2-chlorophenyl)isoxazol-3-yl]methyl amine (ClXIMA), can react with phosphate groups on PCCs, such as nucleotides, sugar phosphates, glycerophosphates, and nucleotide sugars under mild conditions, thereby markedly enhancing reversed-phase retention and separation resolution. More importantly, the incorporation of chlorine, identifiable by its characteristic isotopic peak pattern, provides a unique digital signature in the mass spectra. Leveraging these features, Cl-SQE, a data mining program, was developed to systematically and automatically extract candidate PCCs from large-scale datasets. By applying this strategy to liver tissues from a mouse model of cholestasis, 318 potential ClXIMA derivatives were screened out, and 189 metabolites were identified or tentatively assigned. Statistical analysis revealed significant disturbances in nucleotide metabolism, addressing a critical gap in phospho-metabolomic profiling for cholestatic disease and revealing dysregulated nucleotide homeostasis as a potential feature of cholestasis-associated metabolic remodeling. Overall, this platform facilitates the discovery of PCCs in complex matrices and provides new insights into the pathogenesis of cholestasis and other clinical diseases.