Cathinones are new psychoactive substances (NPS) with sympathomimetic and amphetamine-like psychoactive effects. Among this family, 4-chloro-pyrrovalerone (4-Cl-PVP) has a poorly known metabolism and toxicity. The objectives of this study were to investigate the metabolism of 4-Cl-PVP to identify new consumption biomarkers, increment our identification databases and characterize its metabolism pathway.

This work was carried out using in silico, in vitro (HepaRG cells) and in vivo (authentic human intoxication case) models. In silico studies were performed using open access metabolism prediction softwares (Biotransformer 3.0® and GloryX®). In vitro experiments were performed on a metabolically competent human hepatic cell model (differentiated HepaRG cells) at three 4-Cl-PVP concentrations (1, 10 and 18.8μM) for three different durations (3, 8 or 24h) with or without cytochrome P450 (CYP) inhibitors (ketoconazole as CYP3A4/5 inhibitor and quinidine as CYP2D6 inhibitor). Cell viability test based on extracellular ATP measurement was also performed to evaluate the 4-Cl-PVP-induced hepatotoxicity. In vivo biological samples (urine and blood) were collected from a 4-Cl-PVP intoxication case. In vivo samples and in vitro supernatants were extracted and analyzed using untargeted liquid chromatography coupled to high-resolution tandem mass spectrometry (LC-HRMS/MS) (Q-Exactive™ Orbitrap mass spectrometer). LC-HRMS/MS data were reprocessed using molecular networking, a powerful bioinformatics tool for grouping structurally related molecules in a sample into a cluster, thus grouping the parent molecules and their metabolites together.

In vitro, 13 metabolites were identified, including a 4-Cl-PVP hydroxylated compound (m/z 282.125) and its glucuronoconjugate derivative (m/z 458.157). The kinetic study showed a progressive appearance of metabolites in cell supernatants where some of them were only identified at 48h (e.g. a dihydroxylated compound m/z 298.12). Some metabolites remained in equal proportions as a function of time, suggesting that they are intermediate metabolites (e.g. hydroxy-4-Cl-PVP m/z 282.125). The use of cytochrome inhibitors allowed the metabolism pathways study of 4-Cl-PVP. We found that ketoconazole (CYP3A4 inhibitor) markedly decrease metabolite formation, unlike quinidine (CYP2D6 inhibitor). Cell viability was not impaired by 4-Cl-PVP treatments. In the event of in vivo intoxication, we quantified 4-Cl-PVP in biological samples at 134.4ng/ml in blood and 1018ng/ml in urine at hospital admission. Most of metabolites identified in vitro were also found in vivo, mostly in urine. Lastly, the in silico approach was unsuccessful, finding few irrelevant metabolites.

In silico approach does not currently appear to be suitable to the NPS metabolism prediction as exemplified in this study. This reflects the importance of in vitro and in vivo metabolism studies. On the other hand, in vitro metabolic kinetic studies using molecular networking showed (i) the late appearance of some metabolites at 48h and (ii) the constant presence of some compounds suggesting their potential use in obtaining a large detection window in vivo. These molecules represent reliable biomarkers such as hydroxy-4-Cl-PVP (m/z 282.125). The use of enzymatic inhibitors showed that metabolization was mainly mediated by CYP3A4. The low involvement of CYP2D6 (highly subject to polymorphism) in 4-Cl-PVP metabolism suggests a low risk of interindividual variability these biomarker levels during intoxication. These results showed that the differentiated HepaRG model is relevant for studying NPS metabolism. Most of metabolites detected in vitro were also found in vivo highlighting the relevance of the identified metabolites. Quantitation of 4-Cl-PVP in biological samples coupled with molecular networking analysis revealed renal elimination of this molecule and its metabolites. These results suggest that urine is a matrix of choice for the documentation of recent 4-Cl-PVP consumption and the discovery of 4-Cl-PVP metabolites allowed the identification of relevant consumption biomarkers, thus expanding the detection window for this compound.