Exploration of mephedrone metabolism by molecular networking in human urine samples

Abstract

There is a lack of tools for visualizing and organizing MS data of unknown compounds detected by untargeted toxicological screening. Molecular networking allows exploring and organising MS/MS data without prior knowledge of the sample's chemical composition. The organization of spectral data is based on spectral similarity. Hence, important information can be obtained even before the annotation step. The link established between molecules enables the propagation of structural information. We applied molecular networking to untargeted toxicological screening for the metabolites of mephedrone, a popular new psychoactive substance. We analysed 5 urine samples that tested positive for mephedrone. Sample preparation was performed by mixing 50 μL urine, 50 μL internal standard (trimipramine-D 3 ) and 200 μL methanol:acetonitrile (1:1 v/v). After shaking and centrifugation, the supernatant was transferred into a vial and 10 μL was injected into the LC-HR-MS/MS, a Thermo Ultimate 3000 UHPLC coupled to with a Thermo Q-exactive. The column was a Thermo Accucore phenyl hexyl 100 × 2.1 mm, 2,6 μm. The mobile phase consisted of a gradient of 2 mM ammonium formate, 0.1% formic acid and 2 mM ammonium formate, 0.1% formic acid and 1% water in MeOH:ACN 50:50, from 1–99% in 15.5 minutes, with a flow of 0.5 mL/min. The instrument was used in positive/negative switching ionisation mode with full scan (FS) and a subsequent data-dependent acquisition (DDA) mode. After conversion to a .mzXml open mass format and pretreatment with the MZmine 2 software, data were uploaded to the GNPS web platform ( http://gnps.ucsd.edu/ ) in order to generate a molecular network based on spectral similarity between acquired fragmentation spectra. The network displays each node as an individual ion with its associated MS/MS spectrum and edges connecting nodes indicate the degree of similarity between spectra, thus clustering structurally similar molecules. Nodes were labeled with corresponding exact mass and edges with the exact mass shift between each node. The mephedrone cluster was spotted within the entire network and metabolites were annotated based on spectral similarity, exact mass shifts and fragmentation spectra analysis. In addition to mephedrone and seven already described metabolites (1-dihydro-normephedrone, 4-carboxymephedrone, dihydromephedrone, hydroxylmephedrone-3-O-glucuronide, hydroxytolylmephedrone, normephedrone, normephedrone-3-carboxylic acid) [1] , 11 new putative metabolites with m/z 160.112, 164.107, 176.107, 178.086, 178.123, 182.118, 190.086, 192.102, 196.097, 196.133 and 210.113 (4-carboxydihydromephedrone) were detected. We observed different metabolic patterns, possibly the result of different times after intake or genetic polymorphism (mephedrone is metabolised by CYP 2D6). The use of molecular networking is promising for identifying new metabolites of NPS. This could be a useful complement to liver microsome studies.