High resolution LC/MS has revolutionised the ability to characterize drug metabolism and stability, and has greatly accelerated pharmaceutical R+D. Developments to instrumentation and software have enabled the convenient and rapid elucidation of molecular formulae of chromatographically separated metabolites and degradants. During such studies isobaric species are commonly formed and this necessitates further work to establish the complete structure of each isomer using tandem MS and NMR. Metabolite ID can be challenging and time-consuming - particularly when products are only generated at low concentrations relative to the substrate. We propose that in silico collisional cross sectional area (CCS) prediction coupled with ion mobility measurements can help build confidence in structural assignments, facilitate the characterization of isobaric products and help confirm identity during method development.
In this study we generated mixtures of oxidation products of naltrexone and naloxone using an electrochemical flow cell and then studied them by UPLC-IMS-MS in order to assess the feasibility of relating measured CCS to in silico predicted CCS and degradant profiles to help identify and track isomers. A systematic rotor conformation search was conducted using the Open Babel 2.3.2 as implemented in Avogadro v1.0.1 at the MMFF94 level of theory. Lowest energy conformers were selected as the starting geometry for further geometry optimisation. Final DFT calculations of all the structures, and associated energies were then carried out with the Gaussian 09 program using the hybrid SCF-DFT B3LYP method and 6-31+G(d,p) basis set and additional keywords pop=(mk,dipole) to generate Merz-Kollman electrostatic potential partial atomic charges
LC-IMS/MS and LC-IMS/MS/MS chromatograms of electrochemically oxidized samples of naloxone and naltrexone were screened against a set of predicted degradation products. IMS-resolved product ion spectra were matched to theoretical fragmentation patterns for the precursor or predicted degradation products. In addition to the high energy fragmentation data, collisional cross section areas (CCS) were measured for the precursors and associated fragment ions. In cases where isomeric products were observed , particularly for the mono-hydroxylated and dimeric (inter-molecular crosslinked) species, the measured CCS values were compared to a set of calculated values generated for the postulated structures.
On-line electrochemical generation of drug degradants and elucidation using molecular modelling and tandem MS data.