T-ReXing Synthetic Cannabinoids─Unveiling Metabolite Features Suitable for Urine Screening by Trapped Ion Mobility HRMS

Authors:
Open Access
Publisher website
Google Scholar
Abstract
Targeted and untargeted mass spectrometric analyses of urine, a preferred matrix for toxicological screening, require comprehensive spectral libraries of drug metabolites. To detect new synthetic cannabinoids (SCs), biomarkers for newly emerging compounds must be incorporated rapidly. Unfortunately, reference standards for metabolites of new psychoactive substances (NPS) such as SCs are scarce. Since parent SCs are rarely detectable in urine, elucidating their metabolism poses a significant challenge for forensic toxicology laboratories. Untargeted high-resolution mass spectrometry (HRMS) approaches, often complemented by in silico methods, are increasingly vital to forensic toxicology and metabolomics. A pooled human liver microsome (pHLM) assay was used for in vitro generation of phase I metabolites of three prevalent SCs (ADB-BUTINACA, MDMB-BUTINACA, and MDMB-4en-PINACA). Analysis was performed by trapped ion mobility spectrometry-time-of-flight mass spectrometry (timsTOF-MS) using parallel accumulation serial fragmentation (PASEF) acquisition and MetaboScape software with the T-ReX (time-aligned region complete extraction) 4D workflow including in silico metabolite prediction, fragmentation, and collision cross section (CCS) forecasting. In addition, hydrolyzed and nonhydrolyzed SC-positive urine samples were reanalyzed with this workflow. Results showed successful annotation of predicted metabolites for all three SCs, including monohydroxylations and ester hydrolysis. Qualitative pHLM and urine findings were in good agreement. Compound-specific and most abundant in vivo metabolites were incorporated into a targeted UHPLC-QTOF-MS method to analyze 42 urine samples from the casework. Although complex multistep biotransformation reactions (e.g., ADB-BUTINACA dihydrodiol formation) continue to pose a challenge for in silico prediction, this approach is significantly less time-consuming and less labor-intensive than manual evaluation of known metabolic patterns.
Funding Information
  • Albert-Ludwigs-Universität Freiburg (NA)

This publication has 72 references indexed in Scilit: