Molecular Networking: A Useful Tool for the Identification of New Psychoactive Substances in Seizures by LC–HRMS

Abstract

New Psychoactive Substances (NPS) are a global concern since they are spreading at an unprecedented rate. Despite their commerce still being limited compared to traditional illicit drugs, the identification of NPS in seizures may represent a challenge because of the variety of possible structures. In this study we report the successful application of molecular networking (MN) to identify unexpected fentanyl analogs in two seizures. The samples were extracted with 1 mL of methanol and analyzed with an untargeted data-dependent acquisition approach by LC–HRMS. The obtained data were examined using the MN workflow within the Global Natural Product Search (GNPS). A job was submitted to GNPS by including both seizures and standard mixtures containing synthetic cannabinoids and fentanyls raw files; spectra obtained from standards were used to establish representative networks for both molecular classes. All synthetic cannabinoids in the mixture were linked together resulting in a molecular network despite their different fragmentation spectra. Looking at fentanyls, all the molecules with the typical 188.143 and 105.070 fragments were combined in a representative network. By exploiting the standard networks two unexpected fentanyls were found in the analyzed seizures and were putatively annotated as para-fluorofuranylfentanyl and (iso)butyrylfentanyl. The identity of these two fentanyl analogs was confirmed by NMR analysis. Other m/z ratios in the seizures were compatible with fentanyl derivatives; however, they appeared to be minor constituents, probably impurities or synthetic byproducts. The latter might be of interest for investigations of common fingerprints among different seizures.