Science Inventory

Untargeted LC-MS/MS Analysis of Drug Metabolites in Human Hepatocyte Suspensions for Comparison to Rat Liver S9 Fractions

Citation:

Larson, T., K. Ravindra, B. Wetmore, AND D. Macmillan. Untargeted LC-MS/MS Analysis of Drug Metabolites in Human Hepatocyte Suspensions for Comparison to Rat Liver S9 Fractions. ASMS, Anaheim, CA, June 02 - 06, 2024. https://doi.org/10.23645/epacomptox.25988194

Impact/Purpose:

N/A

Description:

Introduction (120/120): The US EPA is developing new approach methodologies (NAMS) to rapidly evaluate thousands of data-poor chemicals for human health hazards. These NAMs include in vitro exposures to investigate chemical metabolism in novel cell systems to inform hazard and risk. Incorporation of metabolism in these NAMs is important to ensure both parent compound and metabolite exposures are considered during risk evaluations. While hepatocytes are widely used to study chemical biotransformation, attempts to retrofit assays with metabolism had better success when using rat liver S9 fractions as the metabolic source. Here, untargeted analysis is used to characterize metabolite formation (i.e., biotransformation) in human hepatocyte suspensions treated with chemicals with known toxicokinetics for subsequent comparison to human and rat liver S9 fraction exposures. Methods (118/120) Primary human hepatocyte suspensions and metabolically inactive controls were treated with 10 µM rosiglitazone, verapamil, or coumarin for 0-4 hours. Biotransformation products were predicted using BioTransformer 3.0. Untargeted LC-MS/MS analysis was performed using a Shimadzu Prominence LC coupled to a Sciex X500R Q-TOF, operating in a data-dependent acquisition mode. Mobile phases A and B were water and acetonitrile, respectively, both containing 0.1% formic acid. A gradient separation was used from 0-100% B on a Phenomenex Kinetex C18 column (2.1 × 100 mm). MS1 level peak picking, grouping, and alignment were done in Sciex MarkerView 1.3.1. MS/MS spectra for unknown features were annotated using MetFrag 2.5.0. Metabolite annotation confidence was assigned using the scale described by Schymanski et al. Preliminary Data (293/300) For all three drugs (rosiglitazone, verapamil, and coumarin), a decrease in abundance was observed over the four-hour dosing period, suggesting each precursor compound was metabolized. No decrease in abundance was observed for the inactive controls. Potential metabolites were identified as features that were not present in the 0 h timepoint and the inactive controls and showed a higher abundance in the active cells. All metabolites were within ± 6.0 ppm error of their predicted m/z values, regardless of the precursor. For rosiglitazone, three metabolites were detected. The first, at m/z 344.1063 (RT 5.68 min) could be annotated as N-desmethylrosiglitazone. The second two were chromatographically resolved isomers (formula C18H19N3O4S) at m/z 374.1175 (RT 5.82 min) and 374.1152 (RT 6.60 min). Using MetFrag, all three metabolites could be annotated as Level 2A: Probable Structure. For coumarin, two predicted metabolites were annotated as Level 2A. The first, m/z 163.0390 (RT 6.25 min) was annotated as 7-hydroxycoumarin, a well-known Phase I metabolite. The second, at m/z 339.0690 (RT 5.60), was annotated as 7-hydroxycoumarin glucuronide, a Phase II metabolite. For verapamil, nine features were annotated as metabolites with two of those features not previously predicted. The predicted features included m/z 196.1325 (RT 5.22 min), 291.2052 (RT 6.07 min), two features at m/z 441.2735 (formula C26H36N2O4; RT 6.50 and 6.70), and three features at m/z 471.2857 (formula C27H38N2O5; RT 6.15, 6.60, and 6.41 min). All features could be annotated as Level 2A, except for m/z 196.1325 for which no MS/MS spectrum was acquired (Level 5: Exact Mass of Interest). Two features were not predicted for verapamil, m/z 427.2593 (RT 6.51) and 457.2692 (RT 6.58). These metabolites, annotated as Level 2A, were likely further transformations of the features at m/z 441.2735. Disclaimer: This abstract does not necessarily reflect Agency policy. Novel Aspect (10/20) – Use of untargeted analysis to aid development of non-animal-based methods to predict chemical toxicity in vitro.