Identification and Investigation of the Intrinsic Receptor Activation Potential and Metabolization of the New Oxo‐Pyridyl Synthetic Cannabinoid Receptor Agonist CH‐FUBBMPDORA
Laboratory of Toxicology, Department of Bioanalysis, Faculty of Pharmaceutical Sciences Ghent University Ghent Belgium
Ufficio Laboratori Agenzia delle Dogane e dei Monopoli Rome Italy
Food Chemistry Lab, Department of Chemistry and Technology of Drugs Sapienza University of Rome Rome Italy
Center for Forensic Science Research and Education Fredric Rieders Family Foundation Horsham Pennsylvania USA
Laboratoire de Paris Service Commun de Laboratoire DGCCRF‐DGCCI (SCL) Massy France
Joint Research Centre (JRC) European Commission Ispra Italy
* Correspondence:Christophe P. Stove (christophe.stove@ugent.be)
Claude Guillou (claude.guillou@ec.europa.eu)
ABSTRACT
CH‐FUBBMPDORA (CHO‐4′Me‐5′Br‐FUBOXPYRA), a recent addition to the recreational drug market, bypasses the Chinese generic ban (2021) on synthetic cannabinoid receptor agonists (SCRAs) due to its new 5‐bromo‐4‐methylpyridin‐2(1H)‐one core. Its pharmacological properties are currently undefined, and it is yet to be found in biological samples. However, it is unclear whether this is due to low prevalence or hampered detection. The aim of this study was twofold. First, we used a powder seized by customs as a case study to evaluate the utility of low‐field nuclear magnetic resonance (LF‐NMR) to unequivocally identify CH‐FUBBMPDORA. This demonstrated the potential of this technique, which is increasingly used by customs and forensic laboratories for substance identification. High‐field nuclear magnetic resonance (HF‐NMR), Fourier transform infrared spectrometry (FTIR), gas chromatography–mass spectrometry (GC–MS), liquid chromatography coupled to time‐of‐flight mass spectrometry (LC‐QTOF‐MS), and Raman spectroscopy were used as complementary techniques for identification and characterization. Second, we investigated the potential to activate CB1 and CB2 and the metabolism of CH‐FUBBMPDORA. Potencies and efficacies were assessed using βarr2 recruitment assays. Metabolite studies were conducted via human liver microsome (HLM) incubation followed by LC‐QTOF‐MS. CH‐FUBBMPDORA showed a limited activation potential at both cannabinoid receptors. The seized powder exhibited a pronouncedly higher activity, suggesting the potential presence of contaminants with higher cannabinoid activity. Although analytical characterization revealed minor impurities, it is uncertain whether these explain the bioassay findings. Four metabolites were identified, which were all the result of hydroxylation of either the cyclohexyl head group or of the methyl group on the core.
Graphical
Another Chinese generic ban‐evading SCRA, CH‐FUBBMPDORA, has emerged. This study reports its first identification in Europe, its analytical characterization (NMR, Raman, FTIR, GC–MS, and HRMS), its cannabinoid activation potential in a βarr2 recruitment assay, and its metabolization profile. Furthermore, the potential of low‐field NMR for customs and forensic laboratories for substance identification was explored using a seized powder containing CH‐FUBBMPDORA as a case example.
Boxed Text
Article notes
M. Deventer , M. Emanuele , B. Stang , et al., “Identification and Investigation of the Intrinsic Receptor Activation Potential and Metabolization of the New Oxo‐Pyridyl Synthetic Cannabinoid Receptor Agonist CH‐FUBBMPDORA,” Drug Testing and Analysis 17, no. 9 (2025): 1623–1639, 10.1002/dta.3854.PMC1240163839943794
Footnote Group
1Introduction
The United Nations Office on Drugs and Crime (UNODC) defines “new psychoactive substances” (NPSs) as “substances of abuse, either in a pure form or a preparation, that are not controlled by the 1961 Single Convention on Narcotic Drugs or the 1971 Convention on Psychotropic Substances, but which may pose a public health threat” [1]. NPS is generally available to drug users through offers on internet vendor sites, with shipments of small parcels complicating the monitoring of market evolutions of new recreational drugs. Fast identification of new molecules is necessary for efficient monitoring, information sharing, early warning, and risk awareness to respond efficiently to the potential health and social threats posed by the consumption of new substances [2]. The health effects of newly emerging NPS are generally unknown. In most cases, assumptions regarding their psychoactive effects are drawn from the similarity of their chemical structure with those of known substances. This paradigm has also been driving the production and design of new molecules for the recreational drug market: small modifications of core‐based chemical structures led to numerous synthetic cannabinoid receptor agonists (SCRAs) and synthetic cathinones present on the market of recreational synthetic drugs in Europe. The objective could be the replacement of a substance subject to control measures in certain countries or even enlarging the offer to satisfy a demand of customers with new “attractive” products [3]. While potential psychoactive effects can be attributed to new molecules on the basis of structural similarities with known substances for which such effects are known, ideally, experimental data becomes available to support these hypotheses.
This study presents the case of a new substance, not showing structural similarity with typical NPS, present in 5 g of white powder in a suspicious package labelled “5CP.” The parcel was sent from the Netherlands and seized by French customs in June 2022. Structure elucidation using various analytical techniques (discussed below) confirmed the identity of the substance as CH‐FUBBMPDORA (CHO‐4′Me‐5′Br‐FUBOXPYRA).
CH‐FUBBMPDORA (also referred to as 6TP/SGT and N‐[5‐bromo‐1‐(4‐fluorobenzyl)‐4‐methyl‐2‐oxo‐1,2‐dihydropyridin‐3‐yl)cyclohexanecarboxamide) (Figure 1) was introduced on the recreational drug market as a new addition to the SCRA family. Its name, CH‐FUBBMPDORA, was established in relation to the structure showing a cyclohexyl (CH) linked group, a fluorobenzyl (FUB) tail, a 5‐bromo‐4‐methylpyridin‐2(1H)‐one (BMPDO) core, and a retroamide (RA) linker [4]. SCRAs typically act on the central nervous system and have the cannabinoid 1 receptor (CB1) as their prime target, and SCRA users aim to experience psychotropic effects such as relaxation and euphoria, similar to those exerted by Δ9‐tetrahydrocannabinol (Δ9‐THC), the predominant psychoactive substance found in cannabis [5, 6]. However, due to their often higher potency, SCRA use has been associated with severe side effects such as cardiovascular, renal, and gastrointestinal toxicity, as well as respiratory depression, psychosis, seizures, stroke, coma, and even death [6, 7, 8, 9]. In addition, SCRAs often also act on the cannabinoid 2 receptor (CB2), expressed at sites of active inflammation and present on cells of the immune system [10, 11].
Since the enactment of a seemingly blanket ban on SCRAs in China in mid‐2021, which controlled a wide array of current and potential future substances, the worldwide SCRA market has expanded and diversified even more [12, 13, 14]. Gradually moving away from the conventional head‐core‐tail building block structure, a plethora of new compounds have emerged on the SCRA scene, aiming to circumvent legislation by including several structural modifications such as the absence of features (e.g., the tail‐less precursor ADB‐INACA [15]), the presence of altered linkers (e.g., ADB‐FUBIATA [16]), the introduction of halogen atoms to the core moiety (e.g., ADB‐5′Br‐INACA [17]), and the overall inclusion of new core structures (e.g., BZO‐HEXOXIZID [18]) [19, 20, 21]. CH‐FUBBMPDORA can be classified as part of the latter group, carrying an oxo‐pyridyl structure, not mentioned as one of the seven core structures covered by the Chinese generic legislation. Interestingly, this substance also carries an additional bromine atom on the core, a structural modification also seen in other recently emerging SCRAs (e.g., ADB‐5′Br‐INACA, MDMB‐5′Br‐INACA, ADB‐5′Br‐PINACA, and ADB‐5′Br‐BINACA [17, 21, 22, 23, 24, 25]). Furthermore, a methyl substitution of the core moiety as a way to evade the generic ban on indazole‐carrying SCRAs has been noticed recently (MDMB‐5′Me‐INACA) and is also present in the structure of CH‐FUBBMPDORA [26]. It is the first SCRA reported to the EU Early Warning System of the European Drugs Agency (EUDA, formerly EMCDDA) with a BMPDO core.
Following its first identification in Europe (FIE) (presented in this study), CH‐FUBBMPDORA has been detected in several countries in Europe and in the United States (Table 1). It was identified for the first time in France, Germany, and Italy in 2022 and in Romania in 2023, according to information from EUDA [27]. As of November 2022, the substance has been added to the 1971 Schedule I of the Convention on Psychotropic Substances in Lithuania [28]; however, in 2023, it was seized in a Lithuanian prison. Following this, the Belarusian monitoring platform AIPSIN reported CH‐FUBBMPDORA as a substance with high potential for use in June of 2023 [29]. In March 2023, the Slovenian National Forensic Laboratory (NFL) included the substance in their NPS database as a detection based on a test purchase [30]. Also in March 2023, the US‐based Center for Forensic Science Research & Education (CFSRE) received a white powder containing the substance and reported the finding in June 2023 through their NPS Discovery program [31]. Furthermore, the compound was included as a Tier 2 (“Recommend”) substance in their Q4 2023 testing scope recommendations [32] (but was classified as a Tier 3 substance (“Consider”) at the beginning of 2024 and later removed from the scope recommendations [33, 34]). Still in 2023, by means of a decree on August 2, 2023, issued by the Ministry of Health in Italy, CH‐FUBBMPDORA was included in Table 1 of prohibited NPSs [35]. Also in August, the Swedish Public Health Agency received a request from the unit for drug prevention to declare the substance as a narcotic or health hazard [36] after it had been detected in an airport seizure in June 2023. After this, CH‐FUBBMPDORA has been seized again in Italy and Spain.
| Case identification | Disclosed circumstances | Country | Reporting organization | Sample type |
|---|---|---|---|---|
| Date of report | ||||
| EDND‐CR‐2022‐718 08/19/2022 | FIE in a seizure by French customs 06/17/2022 | France | Service Commun des Laboratoires (SCL) Paris and Joint Research Centre (JRC) | White powder (5 g) |
| EDND‐CR‐2022‐1163 12/28/2022 | FIC in a seizure 03/31/2022 | Germany | Landeskriminalamt Schleswig‐Holstein (LKA‐SH) | Paper sheet |
| EDND‐CR‐2023‐5 01/03/2023 | Police seizure at the airport 10/15/2022 | Italy | Instituto Superiore di Sanità (ISS) | Herbal material |
| EDND‐CR‐2023‐347 03/17/2023 | Test purchase sample marked as 5CP 07/29/2022 | Slovenia | National Forensic Laboratory (NFL) | White powder (3 g) |
| EDND‐CR‐2024‐643 06/26/2024 | Seizure at the airport, letter shipment from Spain 06/17/2023 | Sweden | Customs laboratory | Herbal material |
| New‐drug‐monograph‐NPS‐discovery‐230,628 06/28/2023 | Sample originated from the state of Florida March 2023 | United States | CFSRE | White powder |
| EDND‐CR‐2023‐1058 09/14/2023 | Seizure in prison | Lithuania | Narkotikų, tabako ir alkoholio kontrolės departamentas (NTAKD) | Herbal material |
| EDND‐CR‐2024‐64 02/05/2024 | FIC in police seizure 10/2023 | Romania | Romanian Observatory on Drugs and Drug Addiction (ORDT) | Powder (6.05 g) |
| EDND‐CR‐2024‐193 03/14/2024 | Large‐scale seizure in a distribution facility in Spain 10/16/2023 | Spain | Institut Nacional de Toxicologia i Ciències Forenses (INTCF), Barcelona | CH‐FUBBMPDORA was identified in a seizure among 42 other different NPS, containing powders, tablets, and herbal materials |
| JRC‐24040004 04/04/2024 | Seizure analyzed by Milan police laboratory | Italy | Police lab and JRC | White powder (0.3 g) |
| EDND‐CR‐2024‐356 04/24/2024 | NPS exports from Spain via postal mail to several countries | Spain | INTCF, Barcelona | Powder in a shipping parcel (0.4 g) |
In this study, seized powder (indicated in gray in Table 1) was analyzed for structure identification by high‐field nuclear magnetic resonance (HF‐NMR), gas chromatography–mass spectrometry (GC–MS), Fourier transform infrared spectrometry (FTIR), liquid chromatography coupled to time‐of‐flight mass spectrometry (LC‐QTOF‐MS), Raman spectroscopy, and benchtop low‐field nuclear magnetic resonance (LF‐NMR), as discussed below. In particular, the potential of LF‐NMR for chemical elucidation was investigated. Modern LF‐NMR instruments use compact permanent magnets with magnetic fields typically below 2.5 Tesla. The high magnetic fields (typically above 7 Tesla) of HF‐NMR are produced in cryomagnets, making use of the phenomenon of supraconductivity. Higher magnetic fields allow for achieving higher resolution and better sensitivity, facilitating the structural analysis of complex molecules. On the other hand, HF‐NMR is an expensive technology. It requires the use of cryofluids and poses several technical constraints to the laboratory premises and infrastructure for the location of large‐dimension cryomagnets and electronic consoles. In contrast, LF‐NMR instruments are much more affordable, do not require cryofluids, and their small dimensions are comparable to those of most routine instruments used in analytical laboratories. Consequently, the LF‐NMR technique is becoming increasingly attractive to enforcement laboratories. Despite lower sensitivity and resolution, it is expected that LF‐NMR can be applied for the quantitative determination of certain controlled substances and, in combination with other analytical techniques, may contribute to the chemical identification of illicit unknown seized materials such as NPS [37, 38]. This topic is of high interest to enforcement customs and forensic laboratories. In this work, the spectra measured by LF‐NMR in the customs laboratory of Paris were used as a “challenge” to assess whether the information provided by this technique could allow partial or complete elucidation of the chemical structure. We present and discuss the interpretation of LF‐NMR data and the reasoning supporting the elucidation of the chemical structure of this new SCRA, CH‐FUBBMPDORA, which was established together with the customs laboratories of Paris and Rome, which are both equipped with such benchtop LF‐NMR instruments.
Despite the structural similarity of CH‐FUBBMPDORA with compounds described in the literature, the cannabinoid receptor activation potential of CH‐FUBBMPDORA itself was previously not investigated, leaving users (and legislators) unaware of the effects and potential harms associated with this substance, a topic that has been discussed on social media websites such as Reddit [39]. Although relatively new in the recreational drug landscape, oxopyridine structures were already described in a patent published in 2003 as compounds with an affinity for CB2 [40]. Of relevance is that, in 2018, Chicca et al. studied the cannabinoid receptor affinity and activity of a series of oxopyridine carboxamides for future potential therapeutic purposes. One of the investigated compounds (“B2,” Figure 1) resembles CH‐FUBBMPDORA, as its retroamide, with the CH carboxamide moiety being switched to a cycloheptyl carboxamide. B2 showed nanomolar activity and affinity at both cannabinoid receptors [41]. Similarly, in 2019 and 2021, Gado et al. published the synthesis of “C2” (Figure 1), reportedly the first synthetic allosteric CB2 modulator, which only differs from CH‐FUBBMPDORA in that it carries a cycloheptyl instead of a CH moiety. Furthermore, this compound displayed antinociceptive activity in an in vivo mouse model for neuropathic pain [42, 43]. Of relevance here is that it is not uncommon for clandestine labs or manufacturers to draw inspiration from (older) published literature describing research compounds for therapeutic purposes, as was seen before for nitazenes, a potent class of opioids [44]. At this point, it is unclear whether this was also the case for CH‐FUBBMPDORA.
To gain more insight into the pharmacological properties of CH‐FUBBMPDORA, in vitro CB1 and CB2 β‐arrestin 2 (βarr2) recruitment assays were performed using a reference standard as well as a sample of the seized powder. Moreover, metabolite studies via human liver microsome (HLM) incubation were carried out to facilitate the detection of use of this new SCRA, should it (further) emerge on recreational drug markets. Although CH‐FUBBMPDORA has not been detected in biological samples to this day, it is unclear whether this is due to a low prevalence of the substance, the absence of serious intoxication cases, or the inability of laboratories to detect the use of this compound because it—and its metabolites—are not included in the utilized targeted mass spectral libraries. Identifying the most plausible potential metabolites may thus aid in a more accurate mapping and monitoring of the use of this new SCRA.
2Materials and Methods
2.1Chemicals and Reagents
All reagents used for the analytical characterization of the seized sample were of high‐performance liquid chromatography (HPLC) grade. For nuclear magnetic resonance (NMR) analysis, all reagents were obtained from Sigma‐Aldrich (Darmstadt, Germany). For the analytical characterization performed at the Laboratory of Toxicology in Ghent, liquid chromatography coupled to mass spectrometry (LC–MS) grade methanol and formic acid were procured from Chem‐Lab NV (Zedelgem, Belgium), and ammonium formate, ortho‐phosphoric acid (85%), and potassium dihydrogen phosphate were purchased from Sigma‐Aldrich.
A CH‐FUBBMPDORA reference standard (purity ≥ 98%) was procured from Cayman Chemical (Ann Arbor, MI, United States). N‐(5‐bromo‐1‐(4‐fluorobenzyl)‐4‐methyl‐2‐oxo‐1,2‐dihydropyridin‐3‐yl)cycloheptanecarboxamide, further referred to as C2, was previously synthesized at the Department of Pharmacy, University of Pisa, Italy, and was a kind gift from Professor Clementina Manera [42]. The reference standards for (−)‐CP55,940 and JWH‐018 were purchased from Cayman Chemical and LGC (Wesel, Germany), respectively. Dimethylsulfoxide (DMSO), poly‐D‐lysine, and fetal bovine serum (FBS) were procured from Sigma‐Aldrich. Dulbecco's modified Eagle's medium (DMEM) (GlutaMAX), Opti‐MEM I Reduced Serum, penicillin, streptomycin, amphotericin B, and trypsin–EDTA (0.05%) were from Thermo Fisher Scientific (Waltham, MA, United States). The Nano‐Glo Live Cell reagent and the Nano‐Glo LCS Dilution buffer were obtained from Promega (Madison, WI, United States).
Pooled HLMs for metabolite identification were procured from Thermo Fisher Scientific, stemming from 50 donors, and pooled at 20 mg/mL. Sodium phosphate dibasic (anhydrous), sodium phosphate monobasic (monohydrate), and magnesium chloride hexahydrate were procured from VWR (Radnor, PA, United States). Nicotinamide adenine dinucleotide phosphate (NADPH) was from Cayman Chemical. All reagents used for metabolite identification using LC–MS were of LC–MS grade purity. Solvents were bought from Honeywell Chemicals (Charlotte, NC, United States). Ammonium formate (99%) and formic acid ampoules were from Alfa Aesar (Ward Hill, MA, United States) and Thermo Fisher Scientific, respectively.
2.2NMR Spectroscopy (Customs Laboratory, France, and European Commission, Italy)
For benchtop LF‐NMR analysis, 30 mg of powder was completely dissolved in 0.6 mL of deuterated DMSO‐d6. An LF‐NMR spectrometer, Spinsolve 60 MHz, from Magritek, equipped with a 1H/13C/19F probe, was used to record the NMR spectra in DMSO‐d6, 1H, 19F, 13C DEPT‐135, and bidimensional 1H/1H COSY and TOCSY, 1H/13C HMBC, 1H/13C HSQC‐DEPT experiments. The assignment of the peaks, the structure of the multiplets, and the associated coupling constants were analyzed using ACD/Labs Spectrus. LF‐NMR 1H and 19F spectra are acquired within a few minutes for the concentrations indicated in sample preparation. Bidimensional homonuclear spectra (COSY, TOCSY) are obtained in about 1 h or less, while for hetero‐correlated spectra (HSQC, HMBC), several hours are generally required (in routine 2–4 h). In the present case, a longer acquisition time (overnight) was applied for HSQC and HMBC in order to enable the detection of weaker signals (for instance, for quaternary carbon atoms in HMBC). Monodimensional 13C spectra also require long‐time acquisition (in the present case the 13C DEPT‐135 was also recorded overnight) to obtain a sufficient signal‐to‐noise ratio. However, this is generally not always sufficient to allow the detection of quaternary carbons. HMBC spectra is an alternative strategy to obtain information regarding the quaternary carbons with about the same experimental acquisition time.
For HF‐NMR analyses, 8.9 mg of powder was completely dissolved in 0.6 mL of deuterated DMSO‐d6 and approximately 10 mg in 0.6 mL of CDCl3. The instruments and methods used for HF‐NMR measurements were the same as published before [45]. Supporting Information S8 and S9 present the interpretation of the HF‐NMR analyses performed at the Joint Research Centre (JRC), which allowed the elucidation of the chemical structure presented in Figure 1.
2.3Additional Analytical Characterization (Customs Laboratory, France, and European Commission, Italy)
Besides LF‐ and HF‐NMR analysis, additional analytical characterization was performed using GC–MS, LC‐QTOF‐MS, FTIR, and Raman spectroscopy, as described in Supporting Information S1.
2.4Search of Information About a New Substance (European Commission, JRC, Italy)
The workflow and cheminformatics tools used in the project CLEN2SAND for the management of chemical data and for search information have been described previously [2, 46]. Once a new substance is identified, forensic chemists need to obtain relevant information, more specifically about its biological properties. CAS SciFinder or its alternative open community resource, CAS Common Chemistry (limited to nearly 500,000 chemical substances from CAS REGISTRY), are well‐known sources of chemical information. Upon input of a chemical identifier such as InChiKey, the JRC ChemAgora Portal allows the search of information in several chemical databases [47]. The database ChEMBL is especially relevant for information about bioactive molecules with drug‐like properties [48, 49]. Moreover, through its advanced search option, it also allows searches based on similarity of chemical structure.
2.5In Vitro CB1 and CB2 βarr2 Recruitment Assays (Ghent University, Belgium)
Cannabinoid activity profiling of the CH‐FUBBMPDORA reference standard, compound C2 (reported in the literature by Gado et al. [42, 43]), and a sample of the seized powder was achieved using CB1 and CB2 activity‐based bioassays. Human embryonic kidney (HEK) 293 T cells stably expressing either the CB1‐βarr2 or CB2‐βarr2 constructs were maintained at 37°C, at 5% CO2, and under a humidified atmosphere in DMEM (GlutaMAX) supplemented with heat‐inactivated FBS (10%), penicillin (100 IU/mL), streptomycin (100 μg/mL), and amphotericin B (0.25 μg/mL). The development of the stable cell lines and the setup of the assay have been published previously [50, 51, 52]. In summary, on the day prior to the assay, cells were trypsinized and seeded at 5 × 104 cells/well in white opaque‐walled poly‐D‐lysine–coated 96‐well plates, followed by overnight incubation. For each assay, test solutions were freshly prepared via serial dilution in Opti‐MEM I Reduced Serum and used within the next 24 h. The next day, cells were rinsed twice with Opti‐MEM, after which 100 μL of this medium was added to each well. The substrate mix was prepared by diluting the Nano‐Glo Live Cell Reagent (containing furimazine) 20‐fold in Nano‐Glo LCS Dilution Buffer. Then, 25 μL of this mix was added to the wells, and the plate was immediately placed into a Tristar2 LB 942 Multimode Microplate Reader (Berthold Technologies GmbH & Co., Germany). Luminescence was monitored for an equilibration period of 10–15 min until the signal stabilized, and this data was then used to correct for inter‐well variability. Then, 10 μL of a 13.5× concentrated test solution (compensating for the final volume in well) was added to the plate, and luminescence was recorded for 2 h. (−)‐CP55,940 was used as a reference standard for normalization, whereas JWH‐018 was taken along for easy comparison with earlier published activity data, and a concentration gradient of both substances was included on each plate. Appropriate solvent controls for all tested compounds were also included. All tested concentrations were run in duplicate in minimally three independent experiments. Using data recorded during the equilibration period (before adding the test compounds), absolute luminescence values were corrected for inter‐well variability in Microsoft Excel 2019. For each tested concentration, area under the curve (AUC) values were calculated, which were then blank‐corrected by subtracting the AUC values of the appropriate solvent control. Using GraphPad Prism ([Version 10.1.2], San Diego, CA, United States), data were normalized to the maximal receptor activation (E max) of (−)‐CP55,940, which was set at 100%. Data points represent the AUC ± standard error of the mean (SEM). AUC values for the highest concentrations were consistently excluded in case of a reduction of at least 20% compared to the closest lower dilution, as this potentially indicates cell toxicity or solubility issues. Concentration‐response curves were generated using the GraphPad Prism software, followed by the calculation of E max (efficacy) and EC 50 (potency) values, relative and nonrelative parameters for intrinsic receptor activation, respectively, by curve fitting via nonlinear regression (three‐parameter logistic fit). Grubbs' test was used to identify potential outliers, leading to the omission of 6 data points from the dataset (912 points) (p value < 0.05).
2.6Characterization of Potential Impurities in the Seized Powder (Ghent University, Belgium)
To further investigate potential impurities in the seized powder, high‐performance liquid chromatography coupled to diode‐array detection (HPLC‐DAD), GC–MS, and LC‐QTOF‐MS were performed at the Laboratory of Toxicology at Ghent University. An overview of the used techniques and the obtained results can be found in Supporting Information S11–S13.
3Results and Discussion
3.1GC–MS Analysis, QTOF Mass Spectrometry, FTIR Analysis, and Raman Spectroscopy (European Commission, Italy)
The complete analytical reports of these analyses can be found in Supporting Information S2–S5. In essence, the GC–MS spectrum of the sample revealed the presence of bromine. The accurate mass m/z = 421.0930 for [M + H]+ was determined by high‐resolution QTOF mass spectrometry, which also confirmed the presence of bromine.
3.2LF‐NMR Experiments (Customs Laboratory, France)
The 19F NMR spectra obtained by LF‐NMR showed the presence of organic fluorine in the sample. This information and the accurate mass allowed us to establish the raw formula C20H22BrFN2O2 as the only one with a mass difference below 5 ppm.
The 1H LF‐NMR spectrum presented in Figure 2 indicates the presence of aromatic protons in the spectral range between 7 and 8.5 ppm and a number of aliphatic protons in the 1–2 ppm spectral range. Noncoupling protons, singlets, were also observed at 2, 5.07, and 9.18 ppm, suggesting the presence of protons very likely close to or borne by heteroatoms. In 1H‐13C HSQC experiments (Figure 3), the signal at 2 ppm for 1H and 19.9 ppm for 13C was attributed to a noncoupled methyl group (18). The signal at 5.07 ppm is clearly recognized as a methylene (7) with a 13C chemical shift at 51.5 ppm.
Another methyl group was observed as a 1H triplet at 1.17 ppm, with 13C at 8.6 ppm. The measured coupling constant 1JHH = 7 Hz and also the COSY spectrum shows a correlation between this signal and the quadruplet at 3.1 ppm between the peaks of residual water and that of the DMSO‐d6 solvent (see LF‐NMR COSY spectrum in Supporting Information S6). It is clear that this ethyl group cannot belong to the molecule producing the other signals observed in the proton spectrum. The NMR signals in 1H‐13C HMBC (Figure 4) and COSY are characteristic of triethylamine, which is a common impurity often encountered in seized samples (see below). This pattern of signals is easily recognizable, allowing experienced analysts to detect its presence.
The other aliphatic signals observed in the HSQC are typical of cyclic methylene groups (22–26). The methylene group (7), the four aromatic protons (9, 10, 12, and 13), and the two signals at 8 and 9 ppm (6, 15) make a total number of 11 hydrogen atoms. This confirms that the remaining aliphatic signals correspond to the CH group (22–26), as also evidenced by the pattern in the 13C spectrum (Figure 5). The couplings 19F‐13C on the aromatic ring can be easily observed in the 13C spectrum, as shown in Figure 5B,C. Similarly, the carbon atoms of the CH ring could also be distinguished, as shown in Figure 5D. Because of its multiplicity and its overlap with the peak of DMSO‐d6, the CH signal (21) of the CH group is not easily identified in the proton spectrum. However, it can be observed through the HSQC spectrum, showing the correlation with the 13C peak at 44 ppm, and in the TOCSY spectrum through its correlations with protons of the CH group (see LF‐NMR TOCSY spectrum in Supporting Information S7).
The quaternary carbons were not visible in the acquired monodimensional carbon spectra but could be detected through the signals obtained in the long‐range hetero‐correlated spectra. The HMBC spectrum shows that the noncoupling methylene group (7) is clearly linked in para to the FUB and shows a correlation with the CH (6) and with a quaternary carbon at 158.6 ppm, likely a carboxyl (2). Its chemical shifts are also in agreement with binding to a nitrogen atom. The CH (6) shows HMBC correlations with four quaternary carbons (2, 3, 4, and 5). Considering the raw formulae, the remaining part of the structure is necessarily a ring bearing the methyl group (18), which also shows a correlation with 3 of the quaternary carbons (3, 4, and 5) already seen by the CH (6). Finally, the proton signal at 9.17 ppm showing no correlation in the HSQC spectrum is therefore an NH (15) (Figure 3). In the HMBC spectrum (Figure 4), this NH (15) shows a correlation with the quaternary carbon (3) and with a quaternary carbon (16) with a chemical shift of 174.2 ppm, typical of a carbonyl group. An overview of the LF‐NMR chemical shifts of CH‐FUBBMPDORA is presented in Table 2.
| Atom label | δ13C (ppm) | δ 1H (ppm) | H multiplicity | ||
|---|---|---|---|---|---|
| CH‐FUBBMPDORA | CH2 | 23, 25 | 25.4 | [1.00, 1.85] | Broad multiplet |
| CH2 | 24 | 25.7 | [1.00, 1.85] | Broad multiplet | |
| CH2 | 22, 26 | 29.5 | [1.10, 1.95] | Broad multiplet | |
| CH3 | 18 | 19.3 | 2.00 | s | |
| CH | 21 | 43.9 | [2.24, 2.44] | Broad multiplet | |
| CH2 | 7 | 51.2 | 5.07 | s | |
| C | 5 | 100.8 | |||
| CH | 10, 12 | 115.5 d (21.4 Hz) | [6.98, 7.30] | Broad multiplet | |
| C | 4 | 126.5 | |||
| CH | 9, 13 | 130.4 d (8.4 Hz) | [7.27, 7.50] | Broad multiplet | |
| C | 8 | 133.4 | |||
| CH | 6 | 135.2 | 8.19 | s | |
| C | 3 | 143.5 | |||
| C | 2 | 158.4 | |||
| C | 11 | 162.6 | |||
| C | 16 | 174.2 | |||
| NH | 15 | 9.17 | s | ||
| Triethylamine | CH3 | t2, t5, t7 | 8.5 | 1.17 | t (7.3 Hz) |
| CH2 | t1, t4, t6 | 45.5 | 3.04 | q (7.3 Hz) | |
All these correlations allowed us to establish the chemical structure in Figure 1 as the first suggestion for this unknown substance. While alternative structures for the arrangements of the groups of the central ring could be considered, their predicted NMR spectra, in particular their carbon‐predicted chemical shifts, allowed us to rule out these alternative hypothetical structures.
3.3HF‐NMR Analysis (European Commission, JRC, Italy)
The HF‐NMR analysis (600 MHz) of the sample allowed the full structural elucidation (see Supporting Information S8 and S9). The presence of triethylamine was observable in the 1H spectra recorded in both deuterated solvents, DMSO‐d6 and CDCl3. Better resolution and less overlap were obtained with CDCl3. In both solvents, the relative intensities measured in the proton spectra agree with a molar proportion of ca. 10% of triethylamine in the sample (corresponding to 2.4% w/w). The NMR spectra of the test‐purchased material in Slovenia [30] and of the seized powder in the United States [31] show the presence of triethylamine in roughly the same amount (around 2.5% w/w). Another powder sample (case JRC‐2404004 in Table 1), recently seized in April 2024 by the police in Milan and analyzed by the JRC, also showed the presence of triethylamine in the same proportion. It is noteworthy to remind that triethylamine is not detected by usual analytical techniques applied for forensic investigations (i.e., GC–MS, LC–MS, and FTIR) while it is easily identified in NMR spectra.
Overall, the information obtained from HF‐NMR is more detailed than that from LF‐NMR. This is due to the nonoverlapping of proton signals and the easy detection of all quaternary carbons in HF‐NMR, both of which facilitate the structural assignment. For instance, the identification of the para‐fluoro benzyl group was evident from the aromatic signals in the proton spectrum, and the NOESY experiment allowed for easy identification of the correct isomer. In HF‐NMR, the structural elucidation is typically performed using data from the proton spectrum, carbon spectrum, and homonuclear 2D 1H‐1H correlation experiments. In contrast, the overlapping of proton signals and the lower sensitivity of the carbon spectrum in LF‐NMR limits researchers to primarily use heteronuclear 2D experiments to obtain the structural information.
3.4Further Remarks
An internet search for the string “5CP powder” led to an NPS vendor web page where “5CP” was referring to a product described as “cycloheptanecarboxylic acid [5‐bromo‐1‐(4‐fluoro‐benzyl)‐4‐hydroxy‐6‐methyl‐2‐oxo‐1,2‐dihydro‐pyridin‐3‐yl]‐amide.” The web page also presented an image of the compound in agreement with that IUPAC chemical name (see Supporting Information S10).
However, QTOF analysis was not in agreement with that structure, which led SCL Paris to carry out LF‐NMR analyses, sending in parallel the sample to the JRC for HF‐NMR (600 MHz) analyses. This allowed the elucidation of the chemical structure presented in Figure 1. At the moment of its identification, the search using the InChIKey chemical identifier did not return any information. A search in the ChEMBL database based on similarity of structure (with a threshold > 95%) led to the Compound Report Card CHEMBL4467500 describing (N‐[5‐bromo‐1,2‐dihydro‐1‐(4′‐fluorobenzyl)‐4‐methyl‐2‐oxopyridin‐3yl]cycloheptanecarboxamide), the close heptyl analog of the identified substance, referred to as “C2” in the cited scientific research [42, 43] and this article. The analytical reports (presented in Supporting Information S2–S5, S8, and S9) were sent with this information to the EUDA at the end of August 2022 by the French National Focal Point [54]. The fact that the structure advertised on the vendor website and the actual identified structure did not correspond, a phenomenon that has been reported before [55, 56], adds another layer of complexity to the recreational drug problem. As products purchased online might not contain the displayed product, users are potentially exposed to new, unknown substances, entailing additional health hazards.
3.5Pharmacological Characterization
At the time of its identification, no information regarding the pharmacology or toxicology of CH‐FUBBMPDORA was available. Therefore, intrinsic activity profiling of this newly emerging SCRA, a structural analog found in the literature (“C2”) and the seized powder discussed earlier, was done using live‐cell cannabinoid receptor‐based bioassays. These allow quantification of the agonist‐induced recruitment of the intracellular signaling protein βarr2 to activated CB1 and CB2. The assay concept is based on the functional complementation of a split nanoluciferase enzyme, as adopted by the NanoLuc Binary Technology (NanoBiT, Promega). In essence, one (inactive) subunit is fused to the receptor (either CB1 or CB2), whereas the complementary subunit is linked to βarr2. Once the receptor is activated, the recruitment of βarr2 allows the two subunits to come into close proximity, resulting in the functional complementation of the enzyme and restoration of the luciferase activity. The addition of the furimazine substrate results in the emission of bioluminescence, which is used to quantify the intrinsic receptor activation potential of the test compounds.
Figure 6 represents the concentration‐response curves of CH‐FUBBMPDORA and the seized powder. In Table 3, the corresponding values for potency and efficacy are indicated. In line with earlier published work [19, 20, 57] regarding the cannabinoid NanoBiT assay, (−)‐CP55,940 had a potency of 0.89 nM at both CB1 and CB2. For clarity and to facilitate comparison with earlier work and literature, JWH‐018 was also included in the test panel. At CB1, it had an EC50 value of 14.3 nM, with an E max of 378%, whereas at CB2, a potency and (relative) efficacy of 4.36 nM and 63.2%, respectively, were found [20]. Overall, CH‐FUBBMPDORA only exhibited a limited activation potential at both cannabinoid receptors, with a potency in the micromolar range (3.782 μM at CB1 and 4.267 μM at CB2) and efficacies not exceeding 22% (relative to (−)‐CP55,940). Furthermore, a sample of the seized powder was analyzed using the same bioassays. Extensive analytical characterization (described above) revealed no significant impurities. However, interestingly, when applied at high concentrations, this seized powder showed a very steep increase in activity at CB1, with the efficacy clearly exceeding that of CP55,940 at the highest evaluated concentration of 50 μM. A plateau of maximal activation could not be reached as higher test concentrations surpassed the solubility of the powder. For this reason, accurate EC 50 values could not be calculated. Similarly, at CB2, the seized powder exhibited a higher activation potential than the reference standard, exemplified by a higher potency, although here the difference in efficacy was less pronounced (EC 50 of 6.079 μM and E max of 45.3%). We hypothesize that trace amounts of a more potent and efficacious SCRA may be present in the powder, explaining the profiles that were observed upon testing high concentrations. This clearly underscores the danger of some powders circulating on the market: contamination of a powder with trace amounts of other substances might not only go unnoticed, even when using in‐depth analytical characterization, but it may also lead to unpredictable and potentially severe health hazards if the contaminants have an intrinsically very high receptor‐activating potential.
| Compound | CB1 | CB2 | ||
|---|---|---|---|---|
| Relative to CP55,940 (100%) | ||||
| EC 50 (nM) (95% CI) | E max (%) (95% CI) | EC 50 (nM) (95% CI) | E max (%) (95% CI) | |
| CH‐FUBBMPDORA | 3782 (1495–9310) | 21.8 (17.9–26.3) | 4267 (2060–8346) | 17.0 (14.4–19.8) |
| CH‐FUBBMPDORA (seized powder) | ND | 351 a | 6079 (3927–9260) | 45.3 (40.4–51.0) |
| (−)‐CP55,940 | 0.89 (0.45–1.76) | 99.8 (90.6–109) | 0.89 (0.49–1.63) | 96.6 (91.1–108) |
| JWH‐018 | 14.3 (8.09–25.9) | 378 (348–409) | 4.36 (1.72–10.6) | 63.2 (55.8–70.9) |
To further investigate this, a sample of the powder was analyzed at the Laboratory of Toxicology (Ghent University) via HPLC‐DAD and GC–MS. No impurities were detected using those two techniques. Spectra and more detailed results can be found in Supporting Information S11–S12. LC‐QTOF‐MS was then performed to screen for potential contaminants at low concentrations. Minor impurities (< 5% relative to the amount of CH‐FUBBMPDORA) were found, two of which were tentatively identified as either the nonbrominated counterpart of CH‐FUBBMPDORA or a structural analog that has a linear heptane head group replacing the cyclohexane moiety, based on the common fragment ions (Supporting Information S13). No other (known) SCRAs could be identified using the HighResNPS library [58]. In the absence of reference standards for the tentatively identified contaminants, we cannot make a formal statement about their activity, although we consider it less likely that these compounds would have caused such a pronounced increase in cannabinoid activity. For instance, a comparison between the cannabinoid activity of ADB‐INACA and its brominated counterpart ADB‐5′Br‐INACA showed that the presence of additional bromine on the core did not result in a substantial change in activity [21]. With this in mind, and given the remarkably low CB1 activity of CH‐FUBBMPDORA, we hypothesize that a major shift in activity (as seen for the seized powder) is not expected. Alternatively, we cannot rule out the possible presence of another (very) minor contaminant (< 0.1%), with a very high cannabinoid activity, which may explain the findings of the bioassay. To investigate this hypothesis, we tested a mixture of CH‐FUBBPMDORA containing 0.1% 5F‐MDMB‐PINACA (molar ratio 999:1), as an example of a highly potent and efficacious SCRA. A very pronounced increase in efficacy could be observed. From these data, we conclude that the presence of very active compounds, even at very low concentrations, could indeed overrule the activity of a weakly active compound such as CH‐FUBBMPDORA. Data can be found in Supporting Information S14. Although these data confirm the validity of our hypothesis, it remains uncertain whether this is the real reason behind the discrepancy between the CH‐FUBBMPDORA reference standard and the seized powder. Anyhow, it is clear that applying a combined set of approaches, both analytical and activity‐based testing, is essential to obtain the full picture of suspect samples.
As readily mentioned above, CH‐FUBBMPDORA shows a remarkable structural resemblance with the compounds synthesized by Chicca et al. and Gado et al., which either had a promising pharmacological profile at CB1 and CB2 or were reported to behave as a positive allosteric modulator at CB2, investigated by means of [35S]GTPγS, HitHunther cAMP and PathHunter βarr2 assays [41, 42, 43, 59]. Therefore, the functional activity of the analog structurally best resembling CH‐FUBBMPDORA (N‐(5‐bromo‐1‐(4‐fluorobenzyl)‐4‐methyl‐2‐oxo‐1,2‐dihydropyridin‐3‐yl)cycloheptanecarboxamide, C2) was investigated at both CB receptors (Figure 7). C2 carries a cycloheptyl head group, whereas CH‐FUBBMPDORA carries a CH moiety at this position (Figure 1). At both receptors, the activity profiles of both compounds were relatively similar. At CB1, C2 was more active than CH‐FUBBMPDORA (Figure 7). On the other hand, based on the overlapping concentration‐response curves, the CB2 receptor activation potential was almost identical and equally weak for both compounds. The results for C2 are in line with earlier published work by Gado et al. and Ferrisi et al., showing a (very) limited activation potential at CB2 using a [35S]GTPγS, cAMP, and βarr2 recruitment assay [42, 59]. Overall, given the quite similar pharmacological parameters obtained for both substances, slightly increasing the size and bulkiness of the head group does not seem to have a major impact on the CB activation potential of these oxopyridine carboxamides.
4Conclusion
The evolving and diversifying SCRA market remains a significant challenge for forensic toxicologists, healthcare professionals, and law enforcement agencies. As new chemicals with unexpected structural features are continuously being released on the recreational drug market, control measures are constantly lagging behind while drug users may be unaware of the potential health hazards they are exposed to. The present work reports on the first identification and full analytical characterization of CH‐FUBBMPDORA in a powder seized by French customs. Furthermore, the data presented here show that interpretation of the spectra obtained by LF‐NMR, in combination with information from other analytical techniques, can allow the identification of the chemical structure of new substances present in the drug market. After this first identification, CH‐FUBBMPDORA was found in several EU countries and in the United States. Also, as already observed in previous cases, the difference in the chemical structure shown on the vendor website and the actual molecule that was identified in the seized sample highlights the precarious reliability of online sellers, entailing additional risks for consumers of recreational drugs. This study is also the first to characterize the intrinsic cannabinoid receptor activation potential and metabolite profile of the newly emerging oxopyridine carboxamide CH‐FUBBMPDORA. Although this substance activated both CB1 and CB2, its receptor activation potential was limited. A heptyl analog found in literature exhibited a very similar activity profile. The seized powder, on the other hand, considered to be of high purity, showed a pronounced increase in activity at CB1, suggesting the potential presence of contaminants with high cannabinoid activity, although this was not detected via NMR, GC–MS, HPLC‐DAD, and FTIR analyses. Additional HRMS analyses led to putative identifications of present impurities with structures bearing a resemblance to that of CH‐FUBBMPDORA. However, it is unclear whether these findings explain the observed activity profiles or whether other contaminants in the samples are still overlooked. This illustrates that, despite extensive analytical characterization, a combined approach—also involving activity‐based testing—might aid in providing the full picture. Furthermore, it is clear that different preparations may exhibit a much greater than anticipated cannabinoid activity, leading to more pronounced cannabinoid‐related toxicity in users. In addition, four oxidative metabolites were identified, corresponding to the hydroxylation of either the CH head group or the methyl moiety on the core of CH‐FUBBMPDORA. Broader knowledge of metabolites may contribute to more efficient monitoring of the prevalence and distribution of this newly emerging SCRA. Overall, the findings presented in this paper emphasize the importance of an in‐depth characterization of new substances. By providing insight into the pharmacology and metabolite profiles of newly emerging SCRAs, this type of research aims to provide tools to different fields to minimize SCRA‐related harms.
Disclosure
The opinions, findings, conclusions, and/or recommendations expressed in this publication are those of the authors and do not necessarily represent the official position or policies of the U.S. Department of Justice.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Acknowledgments
M.H.D. acknowledges the Research Foundation Flanders (FWO) for supporting this research (Grant 1S54521N). The lab technicians of the Laboratory of Toxicology (Ghent University) (Valerie De Muyt, Goedele Van Nuffel, and Ann Houvenaghel) are acknowledged for performing part of the experiments for the analytical characterization of the seized powder. Professor Clementina Manera is acknowledged for providing the C2 sample. The Center for Forensic Science Research and Education received funding for its portion of work from the National Institute of Justice, Office of Justice Programs, U.S. Department of Justice (award number 15PNIJ‐22‐GG‐04434‐MUMU). Part of this research was also funded by the European Commission Directorate General for Taxation and Customs Unions (DG TAXUD) for the administrative and financial support provided to the JRC through the Administrative Arrangement CLEN2SAND IV (JRC‐Nr 36204‐2021‐DG TAXUD‐Nr TAXUD/2021/DE/319). The authors would like to thank the colleagues from the EMCDDA, the French National Focal Point, the French Customs, DG TAXUD, and the Police laboratory of Milan. We also wish to express our gratitude to H. Schepers from DG TAXUD, R. Christie and A. Gallegos from EMCDDA, and the members of the Customs Laboratories European Network (CLEN) and of the Drug Working Group of the European Network of Forensic Sciences Institutes (ENFSI) for stimulating exchange of views and ideas and for their continuous support for the project CLEN2SAND.
Data Availability Statement
The data that support the findings of this study are available from the corresponding authors upon reasonable request.