Enzyme‐Based Chemical Identification for Metabolomics

Abstract

Background The diversity of chemicals detectable in human samples by high-resolution mass spectrometry (HRMS) exceeds the number of readily available chemical standards, limiting our ability to identify minor metabolites of drugs and other xenobiotics in vivo. Standard workflows for identification of unknown metabolites typically begin by interpretation of ion dissociation spectra (MS/MS) and comparison with library spectra and/or authentic standards – many of which are not available and do not incorporate enzymatic precursor-product relationships as identification criteria. Objective To develop a high-throughput strategy to generate human biotransformation products from diverse xenobiotic and drugs for analysis with LC-HRMS. Hypothesis We hypothesize that unidentified metabolites are generated from enzymatic reactions of known metabolites. Therefore, strategies to generate and characterize metabolic products for diverse arrays of xenobiotics will facilitate identification of metabolites in vivo. Methods Here, we adopt a strategy to generate human biotransformation products of diverse xenobiotics in a high-throughput 96-well plate format using incubations with human liver S9 fractions. Extracts from these reactions were collected at 0 and 24 hour time points and analyzed using LC-HRMS (Thermo Scientific Fusion/High-Field Q-Exactive) to characterize metabolic products generated in a time-dependent manner. Data-dependent MS/MS was performed to collect MS/MS spectra. Expected biotransformation products were characterized by retention time, accurate mass m/z (MS1), MS/MS, and an increase in signal intensity for predicted or previously unreported biotransformation products. Incubations with stable-isotope precursors aided the identification of previously unreported biotransformation products. Results Our data show that known and previously unidentified Phase 1 and Phase 2 metabolites were produced from a range of xenobiotics in a 96-well format using human liver S9 fractions. S9 reaction extracts of 138 drug or xenobiotic precursors were analyzed generating a total of 502 validated metabolites. Selected metabolites were used to support their identification in mouse and human samples based on matched accurate mass MS1, retention time, and co-occurrence with the parent xenobiotic and/or related metabolites in the samples. Conclusion We have developed a scalable tool to generate metabolites from thousands of xenobiotics to facilitate identification of drug and other xenobiotic metabolites in human samples.