Isomerization of cannabidiol and Δ9-tetrahydrocannabinol during positive electrospray ionization. In-source hydrogen/deuterium exchange experiments by flow injection hybrid quadrupole-time-of-flight mass spectrometry

RATIONALE Liquid chromatography/tandem mass spectrometry (LC/MS/MS) is frequently used for analysis of cannabinoids in drug abuse control. Despite differences in structure, the isomers ?9-tetrahydrocannabinol (THC) and cannabidiol (CBD) provide identical fragment spectra after positive electrospray ionization (ESI). For elucidation of the reason, hydrogen/deuterium (H/D) exchange experiments were performed. METHODS Solutions of THC and CBD in D2O/acetonitrile (50:50, v/v) were flow-injected into acetonitrile as the mobile phase and measured by hybrid quadrupole-time-of-flight mass spectrometry (FI-QTOF-MS) in targeted MS/MS mode. The MS and collision-induced dissociation (CID) spectra at 10, 20 and 40 eV were interpreted with respect to number and position of exchanged hydrogen atoms. For comparison the same measurements were preformed in H2O, after addition of 0.5% formic acid and with negative ESI. RESULTS Depending on injected volume and position in the response curve, up to 7 or 8 hydrogen atoms were exchanged by deuterium in THC or CBD. Positive ESI CID spectra were available for precursors with up to 4 exchanged D-atoms and showed that besides the OH groups also an H/D exchange at carbon atoms of the non-aromatic part of the molecules occurred for both THC and CBD. After negative ESI, no H/D exchange in addition to the OH groups and different CID spectra of both substances was found. CONCLUSIONS Injection of the investigated substances in D2O and measurement by FI-QTOF-MS proved to be an efficient way to perform H/D exchange experiments. The results were interpreted as an acid-catalyzed in-source equilibration between THC and CBD leading to the same precursor ions and to an H/D exchange in the methyl groups under the increased acidic conditions in the positive ESI droplets. Therefore, in positive LC/ESI-MS/MS, peak identification by CID spectra or by abundance ratio of multiple reaction monitoring (MRM) transitions is not sufficient for unambiguous discrimination between THC and CBD and must be supported by retention time or other experimental evidence. Copyright (C) 2012 John Wiley & Sons, Ltd.