Since the dawn of civilization, Cannabis sativa L. has been used in a multitude of different ways for many purposes, from a recreational drug to medical uses, as well as for industrial goals including food, textile, paper, building, and energy industries, which have proven hemp to be an attractive solution for synthetic economies. (Abedi and Sahari, 2014; Alexander, 2016; Appendino, 2020; Finnan and Styles, 2013; Garcia-et al., 1998; Rehman et al., 2013). Recently, several studies have demonstrated that cannabis has positive effects on a wide range of health conditions (Campos et al., 2016). These outcomes are attributed to the active compounds found in the plant, thus prompting further research in the isolation and study of these secondary metabolites. Since the late 1960s, several compounds known to be present in cannabis have been isolated and characterized, including the psychoactive cannabinoid Δ⁹-tetrahydrocannabinol (THC or Δ⁹-THC, 1, Figure 1) and the non-euphoric cannabidiol (CBD, 2, Figure 1) (Iversen, 2018; Sholler et al., 2020). These compounds are now referred to as “major cannabinoids”, including other important ones like cannabigerol (CBG), cannabichromene (CBC), and cannabinol (CBN) (Pollastro et al., 2018; Anokwuru et al., 2022; Maioli et al., 2022). The two major phytocannabinoids THC and CBD already reached the approval by FDA for commercialization in different forms: Sativex^®, a combination of THC and CBD, is used to treat spasticity associated with multiple sclerosis, and CBD has also been developed as a single active pharmaceutical ingredient known as Epidiolex^®, drug of choice for the treatment of certain rare genetic forms of epilepsy. Additionally, the main phytocannabinoids are still being studied for their potential in treating various pathologies (mainly inflammatory diseases) both in vivo and in vitro (Baratta et al., 2022), as well as some of their derivatives, most notably the aminocannabinoquinone VCE-004.8, which has been granted orphan drug status by the FDA and EMA for the treatment of scleroderma (Caprioglio et al., 2021). More recently, Cannabis sativa showed to have a neuroprotective effect, resultant from its anti-inflammatory and antioxidant properties (Viana et al., 2022), while phytocannabinoids have been demonstrated to have potential anti-cancer properties. They induce cell death, inhibit cell migration and proliferation, decrease angiogenesis, and inhibit the invasiveness of cancer cells of, e.g., the lung, prostate, skin, breast, or brain (Hinz and Ramer, 2019; Kovalchuk and Kovalchuk, 2020; Tomko et al., 2020). Despite the very large number of phytocannabinoids isolated from cannabis (more than 150 different compounds) (Hanuš et al., 2016), the main biological activities observed during the study of this plant are to be attributed to these compounds which occur in considerably greater quantities compared to the so-called minor cannabinoids, regardless of the chemotype of the plant (Caprioglio et al., 2022).

In humans, the biological pathway regulated by cannabinoids is called endocannabinoid system (ECS). ECS is composed of two types of receptors, the CB₁R and cannabinoid 2 receptor (CB₂R), as well as enzymes that break down and synthesize their endogenous ligands (referred to as endocannabinoids) (De Petrocellis and Di Marzo, 2009). The CB₁R was identified and characterized in the rat brain in 1988 and its location was confirmed using tritiated CP 55,940 (3, Figure 1) in 1990 (Herkenham et al., 1990). In 1992, the receptor was cloned and the DNA that encodes GPCRs was found (Matsuda et al., 1990). The CB₂ receptor was instead discovered in 1993 and was found to be predominantly expressed in macrophages in the spleen (Munro et al., 1993). Once CB₁R and CB₂R have been discovered, anandamide (AEA, 4) and 2-arachidonoyl glycerol (2-AG, 5) were found to be the endogenous ligands that interact with these receptors (Figure 1) (Di Marzo et al., 2004).

As a regulator of a wide range of physiological processes, the ECS plays an important role in many disorders (Bisogno and Di Marzo, 2010). To gain a broader view of the ECS, it would be extremely valuable to develop high-sensitivity and high-throughput analytical tools. The use of small molecule probes might provide information regarding the spatial and temporal dynamics of cannabinoid receptor expression, as it has been demonstrated in recent years for some specific classes of enzymes (Chang et al., 2009; 2012; Tully and Cravatt, 2010). As a consequence, the development of small molecule probes that can recognize cannabinoid receptors is currently in the spotlight, since they may complement, and in some cases eliminate, some of the limitations of available antibodies (Grimsey et al., 2008). The purpose of this review is to provide an overview of the recent development of CB₁R-specific probes: we will examine the various types of probes currently in use, describe how the chemistry of these molecules affects the effectiveness of these drugs, and examine where new probes and drugs may be developed in the future.

Covalent chemical probes have found widespread use as research tools and clinical agents. Their use ranges from probes that are metabolically incorporated into proteins, to probes for photoaffinity labelling of targets and electrophilic probes for activity-based protein profiling. In the contest of the CB₁Rs, covalent probes have been designed by incorporating a reactive tag into a CB₁R ligand such as 1 or 4 (Figure 2). One type of reactive tag is a chemically inert group (e.g., azides, benzophenones, or diazirine) which, upon UV irradiation, is converted into a highly reactive species able to react with any amino acid residue situated in its immediate environment. Alternatively, the reactive tag can be an electrophilic group (e.g., isothiocyanates, Michael acceptors, haloacetamides, or nitrogen mustards) able to target nucleophilic amino acids (e.g., lysine, histidine, and cysteine) situated at or near the binding site. When two reactive tags are attached to the ligands, then the corresponding bifunctionalized probes can potentially interact at two distinct sites within the CB₁R-binding domains. In this section, we reviewed the covalent probes for the CB₁Rs, classifying them as photoactivatable, electrophilic, and bifunctional probes.

Fluorescently tagged small molecules have been widely used in the past decade as biological imaging tools as they offer the advantage to allow real-time monitoring of ligand-receptor interactions with high spatio-temporal precision (Briddon et al., 2011). Conjugation of a pharmacophore with a non-peptidic fluorescent tag can detrimentally change ligand-receptor affinity and therefore, a linker is generally required to separate the two entities (Stoddart et al., 2015). The composition and length of this linker (generally PEG or methylene chain), as well as the choice of the fluorophore, can also affect the physicochemical and photophysical properties of the resulting fluorescent conjugate; therefore, the development of fluorescently labelled ligands is particularly challenging. Within recent years, a plethora of fluorescent ligands targeting CB₁R has been reported, mainly using biotin as target for the fluorophore. The addition of biotin to a ligand via a linker can, provide fluorescence, once the compound has docked with the target receptor, using fluorescent avidin conjugates. The biotinylated 2-AGE analogue 28 (Figure 10), showed moderate affinity to both human CB₁R and was selected for in vitro imaging of CB₁R (Martín-Couce et al., 2011). Biotinylated probes 29 and 30 (Figure 10), where the synthetic cannabinoid agonists HU210 and HU308, respectively, were conjugated to biotin via the free hydroxyl group, have also been successfully used for the visualization of CB₁R in neurons and in different immune cell (Martín-Couce et al., 2012). However, biotin probes require a two-step labelling process and additional steps to block endogenous biotin. This does not make them suitable for flow cytometry clinical routine or tissue staining.

To avoid these drawbacks, the cannabinoid agonist HU210 was coupled to the fluorescent tag Alexa Fluor 488 via a hexyl amide linker, generating the first fluorescent probe (31, Figure 11) with high affinity for CB₁R (Ki CB₁R = 27 nM) and selectivity over CB₂R (Martín-Fontecha et al., 2018). The use of this ligand as a chemical tool for the identification of functional CB₁R in human monocytes, T cells, and B cells was validated by multiplexed flow cytometry. This probe showed to be also suitable for the direct visualization of CB₁R in tonsil tissues, allowing the in vivo identification of tonsil CB₁R-expressing T and B cells.

Chromenopyrazole compounds, containing a similar THC structure, were also used as scaffolds for the conjugation with different fluorophores (BODIPY-630/650, BODIPY-FL, and Cy5) generating fluorescent ligands that have been shown to have high affinity to cannabinoid reports (Figure 11). However, these compounds (i.e., Cy-chromenylpyrazole 32) showed higher affinity to CB₂R over CB₁R (pKi hCB₁R = 5.26; pKi hCB2R = 7.83) (Singh et al., 2019).

T117, a novel diarylpyrazole fluorescent ligand (33, Figure 12), was first reported by Daly et al. and derived by conjugation of the potent CB₁R inverse agonist AM251, with a fluorescent tetramethylrhodamine group (5-TAMA) (Davenport and Daly, 2010). In ligand binding studies, AM251 competed with [³H]CP 55,940-labelled membranes to give a Ki of 0.8 nM. The addition of the 5-TAMA group significantly reduced the binding affinity of T117, providing only 10% displacement of [³H]CP 55,940 at 1 µM. However, at a lower concentration (0.3 µM), binding of T1117 to wild-type (WT) mouse mesenteric artery was observed, using 543 nm excitation (590 nm emission). This ligand also displayed binding to cannabinoid-like (GPR55) receptors through Ca⁺⁺ response in HEK-293 cells. Nevertheless, a conflicting study proved that T1117 binds endogenous and recombinant CB₁Rs with nanomolar affinity (Kd = 460 nM). Moreover, T1117 binding to CB₁R is sensitive to the allosteric ligand ORG27569 and thus it is applicable to the discovery of new allosteric drugs (Bruno et al., 2014). In 2009 a study carried out by Grant and co-workers focused on the CB₁R inverse agonist SR141716A, established the C5 position of its central pyrazole ring as the optimal site for fluorescent moiety linkage (Grant et al., 2019). Thus, CB₁R fluorescent probes (34-35, Figure 12) based on C5 conjugation of two SR141716A analogues with fluorescein isothiocyanate (FITC), were prepared. The affinity of the fluorescent probes 34 and 35 was then determined through radioligand competition binding assays at hCB₁R. The affinity of the 17-atom linker congener 35 (Ki = 2.1 μM) was modest; however, compound 34 bearing a 12-atom linker was found to display a useful level of affinity for CB₁R (K_i = 260 nM).

Positron emission tomography (PET) is a nuclear imaging technique that employs gamma rays to provide three-dimensional images that give information about the functioning of specific organs. PET is based on the detection of picomolar amounts of biological substances labeled with a short-lived positron-emitting radionuclide (tracer) sparing the biological system. This technique shows the advantage of being non-invasive, functional, and extremely sensitive (Li and Conti, 2010). Moreover, the PET probes have the same chemical structure as biomolecules and drugs, without altering their biological activity. ¹⁸F has a short half-life (109.8 min), making it the ideal radionuclide for routine PET imaging (Damont et al., 2013). Thanks to its exceptional sensitivity, PET is well suitable to measure relatively low concentrations of enzymes and receptors also in vivo considering that most neuroreceptor populations in the human brain are expressed in a range between 10^–8 and 10^–12 M (Lopresti et al., 2023). The first selective CB₁R antagonist was rimonabant (36) (Rinaldi-Carmona et al., 1994), approved in Europe in 2006, to treat obesity but withdrawn from sale 2 years later by the European Medicines Agency (EMA) due to its manifest secondary effects and not being approved by the Food and Drug Administration (FDA). However, some of its analogs have been synthesized and labeled with [¹⁸F] for PET imaging. Analogs of rimonabant showed activities on different and important biological targets, which makes them attractive for the development of new PET tracers (Gomes et al., 2020). In particular, two radiotracers [¹⁸F]SR144385 (37) and [¹⁸F]SR147963 (38, Figure 13) showed an appropriate regional brain distribution for cannabinoid receptors with a target ratio of 1.7 for [¹⁸F]SR147963 and 2.5 for [¹⁸F] SR144385 at 60 and 90 min post-injection, respectively (Mathews et al., 2000). Similar structures have been reported by Horti et al. in which the synthesis of two radiolabeled compounds named [¹⁸F]NIDA-42033 (39) and its ethyl ester derivative 40 have been described (Figure 13) (Katoch-Rouse and Horti, 2003). The radiochemical yields were in the range of 1%–6% and sufficient quantities with specific radioactivity greater than 2,500 mCi = mmol and radiochemical purity >95%.

Another derivative named [¹⁸F]-O1302 (41, Figure 14) having a short carbon chain at the para position of the phenyl group ending with a [¹⁸F], showed high binding affinity (Ki 0.91 nM) and moderate lipophilicity after evaluation in mice (Mathews et al., 2000; Tobiishi et al., 2007).

The [¹⁸F] isotope-labeled CB₁R inverse agonist DBPR211 (42, Figure 14) was synthesized, radiolabeled with halex exchange reaction, and analyzed for positron emission tomography scanning studies (Chang et al., 2019). After the purification, the compound was intravenously injected in mice showing a distribution percentage over 90-min scans among five regions of interest, including brain, heart, liver, thigh muscle, and kidney, lower than 1%, justifying itself as a peripherally CB₁R antagonist.

[¹⁸F]MK-9470 (43, Figure 14) is a recent selective, high-affinity, inverse agonist (human IC₅₀, 0.7 nM) for the cannabinoid CB₁R developed for the imaging of the human brain (Burns et al., 2007). Autoradiographic studies in the rhesus monkey brain showed high specific binding in the cerebral cortex, cerebellum, caudate/putamen, hippocampus, substantia nigra, and globus pallidus. Positron emission tomography (PET) images in rhesus monkeys exhibited high brain uptake.

Donohue in 2008 (Donohue et al., 2008a) described the synthesis of [¹⁸F]FMPEP-d ₂ (44, Figure 15) having a superior performance of tracing compared with [¹¹C]MePPEP (45, Figure 15), due to greater precision and accuracy in detecting significant differences in CB₁R tracer uptake (Terry et al., 2010). 44 has been used to study abnormal levels of CB₁R binding in alcohol abuse (Hirvonen et al., 2013) or neurological disorders (e.g., schizophrenia) (Jenko et al., 2012). In preclinical and clinical studies [¹⁸F]FMPEP-d2 has been used to image CB₁R expression in a mouse model of Alzheimer’s disease (Takkinen et al., 2018).

¹¹C-labeled ligands offer certain advantages compared to ¹⁸F-labeled compounds. The shorter half-life of ¹¹C enables more syntheses to be carried out in a shorter period of time, using the same hot-cell. Additionally, the lower radiation-absorbed doses permit more PET scans to be conducted on each subject. However, the shorter half-life of ¹¹C can create challenges in accurately quantifying radioligand kinetics or concentrations in plasma and the brain, especially when they are low. Today, ¹¹C-OMAR (46, Figure 15) represents the most studied and promising CB₁R radiolabeling agent for PET (Horti et al., 2006). In 2008 Donohue discovered and labeled 3,4-diarylpyrazoline derivatives as candidate radioligands for in vivo Imaging of Cannabinoid Subtype-1 using [¹¹C]cyanide ion as labeling agent and evaluated as PET radioligands in cynomolgus monkeys (Donohue et al., 2008b). Compound 47 ((−)-3-(4-chlorophenyl)-N′-[(4-cyanophenyl)sulfonyl]-4-phenyl-4,5-dihydro-1H-pyrazole-1-carboxamidine) ¹¹C-SD5024 (Figure 15) was found to get high-affinity and selectivity for binding to CB₁R. The same compound has been studied for the kinetics in humans and evaluated in seven healthy subjects with compartmental modeling (Tsujikawa et al., 2014). The compound showed a Ki = 0.47 nM at an intermediate level among the five CB₁R ligands and a lipophilicity of 3.79, which is appropriate for brain imaging together with a peak brain uptake of 1.5–3 standardized uptake value, slightly higher than that of ¹¹C-OMAR.

The synthesis of [¹¹C]MePPEP (45, Figure 15), a CB₁R mixed inverse agonist and antagonist, has been first reported in 2008 to ameliorate previous CB₁R ligands (Donohue et al., 2008b). The compound showed fairly high lipophilicity (LogD7.4 = 4.8) but still preserving high selectivity and affinity for the CB₁R. It readily entered the monkey brain within 20 min displaying stable measurements of distribution volume within 90 min (Yasuno et al., 2008).

PSNCBAM-1 (1-(4-Chlorophenyl)-3-(3-(6-(pyrrolidin-1-yl) pyridin-2-yl)phenyl)urea) (48, Figure 16) was developed as a potent allosteric antagonist for CB₁, able to reduce the appetite and body weight of rats. Other Several analogs were synthesized and radiolabeled using [¹¹C]COCl₂ and evaluated as PET ligands for CB₁R imaging using in vitro and in vivo techniques (49-50, Figure 16) (Yamasaki et al., 2017). In particular, compound 49 showed a strong binding affinity for peripheral CB₁R in an in vitro binding assay. PET imaging with showed considerable binding to peripheral CB₁R in the mouse brown adipose tissue (BAT), suggesting that 49 is a promising PET imaging agent for further evaluating pathophysiological and biological processes mediated by peripheral CB₁R.

Gao et al. reported an alternative synthetic route to PET CB₁R radioligands [¹¹C]OMAR analogs (51-53, Figure 17) that were prepared in high overall chemical yields (Gao et al., 2012). The radiosynthesis was employed at the oxygen position of the precursor the O-[¹¹C]methylation radiolabeling. Radiolabeling procedures incorporated efficiently [¹¹C]CH₃O with [¹¹C]CH₃OTf. The target tracers were isolated and purified in high radiochemical yields, short overall synthesis time, and high specific activity making them the potential preclinical and clinical PET agents in animals and humans.

In Table 1 are summarised the CB₁R binding of the probes reported in this review. As a result of the important repercussions in ECS modulation, targeting this system is currently one of the major trends in drug discovery. If we take in consideration CB₁R only, this receptor can be exploited to treat a variety of pathologies, including neurological disorders (i.e., Huntington’s disease, multiple sclerosis and Alzheimer’s disease) (McCaw et al., 2004; Pertwee, 2006; Pertwee, 2007; Liu et al., 2015), as well as peripheral disorders that involve energy metabolism, food intake, and obesity (DiPatrizio, 2021; Hijová, 2022). In addition to controlling liver and kidney function, it also controls bone remodeling, skeletal mass, and elongation under normal and pathophysiological conditions (Tam et al., 2018).

A significant amount of information has been accumulated in the literature concerning the in vivo and in vitro pharmacology of the CB₁R over the past decades, revealing new insights into pathways controlled and the roles of receptors, enzymes, and ligands. This knowledge has, however, made a complete transition into drug development in only a few cases (i.e., Sativex) (Namdar et al., 2020).

In order to develop novel therapeutic and diagnostic tools, it is necessary to understand the functions and molecular mechanisms associated with CB₁R modulation; to this aim, several molecular probes that utilize multiple interaction mechanisms (i.e., fluorescence, PET) have been developed in the past 30 years. These results demonstrate the great interest in this biological target: the development of new selective probes is therefore essential to obtaining new results that can lead to the introduction of new CB₁R-based drugs on the market.