Assessing the Pharmacological Properties of Novel Psychoactive Substances (NPS) Identified Online: In Silico Studies on Designer Benzodiazepines and Novel Synthetic Opioids

Abstract

Background By 2022, a total of 1,127 of Novel Psychoactive Substances (NPS) have been identified worldwide and officially reported by the United Nations Office on Drugs and Crime (UNODC) and the European Monitoring Centre for Drugs and Drug Addiction (EMCDDA). An analysis of the surface web via the use of a web crawler, NPSfinder®, indicated that the number of NPS could be almost four times higher than that known to both the UNODC and EMCDDA. This is of particular concern, especially if one considers the public health risks and harms associated with NPS use/abuse and the paucity of data related to their pharmacological/toxicity profiles. In particular, in the last few years two NPS classes, i.e. novel synthetic opioids (NSOs) and designer benzodiazepines (DBZDs) were associated with serious side-effects and life-threatening scenarios (i.e., fatalities and overdoses). Gaps in knowledge Hence, with online NPS numbers exceeding those reported by official sources, there is a strong need to address the gap in knowledge concerning the discrepancies between the online and the evidence based NPS market(s); as well as the gap in knowledge concerning lack of pharmacological profiles for most of the newly-identified NPS. Objectives This programme of research aimed to: use data available from NPSfinder®, the UNODC and EMCDDA to assess the current general NPS scenarios, and in particular for DBZDs and NSOs; use in silico computational techniques to predict the biological activity of the emerging NPS; use the predicted values to infer possible health threats associated with the consumption of these substances, underscoring which of the NPS identified online could indeed represent a serious threat to public health; assess the potential of in silico methodologies as preliminary risk assessment tools; and subsequently inform relevant stakeholders about the risks associated with these new NPS. Methods The NPSfinder® web crawler was used to identify NPS which are available/discussed online. A comparison with UNODC and EMCDDA databases was then carried out to assess the extent of the total NPS scenario, and the numbers of the NSOs and DBZDs classes. To appreciate and predict the biological activities of NSOs and DBZDs, in silico models (e.g., quantitative structure-activity relationship (QSAR), Molecular Docking (MD) and pharmacophore mapping) were used as reliable, time- and cost- effective alternatives to the classical approaches such as in vivo, in vitro or preclinical studies. Results and Discussion A total of 4,231 NPS were identified on the surface web, almost four times the numbers reported by both UNDOC and EMCDDA databases (circa 1,127). These results suggest how the online content analysis should be considered as an important source for the assessment of the NPS scenario. The same discrepancy in the total NPS numbers was observed for each NPS class and a total of 115 DBZDs and 371 NSOs were identified compared to 33 and 123 reported by the UNODC respectively. To assess pharmacological profiles of these NSOs and DBZDs identified online, specific QSAR models were developed in MOE® and Forge™. For the prediction of biological activities of DBZDs, the γ-aminobutyric acid A receptor (GABA-AR) was used; the mu opioid receptor (MOR) was used for the NSOs. In addition, for the DBZDs, a set of new potential ligands resulting from “scaffold hopping” exercises conducted with MOE® was also evaluated. The generated QSAR models returned good performance statistics confirming their strong reliability in predicting the biological activity of an unknown or a newly-identified molecule. The DBZDs predicted to be the most active were flubrotizolam, clonazolam, pynazolam and, fluclotizolam, consistently with reported literature and/or drug discussion forums. In particular with flubrotizolam and fluclotizolam, it was found they were discussed on drug fora but not previously identified either by the UNODC or EMCDDA (flubrotizolam only). This suggests the possible presence on the market of very potent NPS which are still unknown to international agencies, potentially representing a serious threat to public health. Worrisome results were also obtained for the class of NSOs, with the identification of new and potent analogues of carfentanyl (10,000 more potent than morphine), i.e., 2-methyl carfentanyl, n-methyl-carfentanyl and butyryl-carfentanyl. Moreover, the scaffold hopping exercise conducted for the DBZDs class, strongly suggested that structural replacement of the pendant phenyl moiety could increase biological activity and highlighted the existence of a still unexplored chemical space for this NPS class. The results obtained with QSAR analysis were supported by molecular docking exercises, which gave an indication of the binding affinity of these NPS towards their respective receptors. Moreover, the binding affinity of a set of DBZDs was assessed for the MOR, in an attempt to assess a possible multi-receptor activity of these molecules. Conclusions The online identification of a great number of NPS, including very potent central nervous system depressants, represents a serious challenge, in particular if one considers that DBZDs and NSOs are usually consumed either together or in combination with stimulants for recreational purposes and self-medication. The high numbers of available molecules, their patterns of use and the paucity of pharmacological data could lead to worrisome outcomes, including the synergy of each NPS class side-effects, which could (and are) increasing the likelihood of respiratory depression, coma, and deaths. To retrieve an extensive picture of the current NPS drug scenario, the online analysis has proven very useful, if not fundamental. Its ability to identify novel mentioned NPS, in a timely manner, makes it a very important tool for a range of activities, including informing law-enforcement and public health stakeholders, supporting the European and United Nations Early Warning Systems impacting and influencing law-making and guiding monitoring/surveillance. Moreover, in silico methodologies, proven as reliable tools for a fast prediction of biological activity, could be used in describing the activity/toxicity profile of novel NPS, aiming at supporting both law enforcement in scheduling process and public health stakeholders in drafting treatment/management educational packages. Finally, the combination of online and in silico analysis could support and improve the risk assessment procedures currently in place for NPS.