In vitro and in vivo pharmacological evaluation of the synthetic cannabinoid receptor agonist EG-018

Abstract

Synthetic cannabinoid receptor agonists (SCRAs) possess high abuse liability and complex toxicological profiles, making them serious threats to public health. EG-018 is a SCRA that has been detected in both illicit products and human samples, but it has received little attention to date. The current studies investigated EG-018 at human CB₁ and CB₂ receptors expressed in HEK293 cells in [³H]CP55,940 competition binding, [³⁵S]GTPγS binding and forskolin-stimulated cAMP production. EG-018 was also tested in vivo for its ability to produce cannabimimetic and abuse-related effects in the cannabinoid tetrad and THC drug discrimination, respectively. EG-018 exhibited high affinity at CB₁ (21 nM) and at CB₂ (7 nM), but in contrast to typical SCRAs, behaved as a weak partial agonist in [³⁵S]GTPγS binding, exhibiting lower efficacy but greater potency, than that of THC at CB₁ and similar potency and efficacy at CB₂. EG-018 inhibited forskolin-stimulated cAMP with similar efficacy but lower potency, compared to THC, which was likely due to high receptor density facilitating saturation of this signaling pathway. In mice, EG-018 (100 mg/kg, 30 min) administered intraperitoneally (i.p.) did not produce effects in the tetrad or drug discrimination nor did it shift THC’s ED₅₀ value in drug discrimination when administered before THC, suggesting EG-018 has negligible occupancy of brain CB₁ receptors following i.p. administration. Following intravenous (i.v.) administration, EG-018 (56 mg/kg) produced hypomotility, catalepsy, and hypothermia, but only catalepsy was blocked by the selective CB₁ antagonist rimonabant (3 mg/kg i.v.). Additional studies of EG-018 and its structural analogues could provide further insight into how cannabinoids exert efficacy through the cannabinoid receptors.

1. Introduction:

Originally developed for the pharmacological interrogation of the endocannabinoid system or for their potential therapeutic effects, some synthetic cannabinoid receptor agonists (SCRAs) are now being sold and used for the purpose of obtaining a cannabis-like high (Berry-Caban et al., 2012; Every-Palmer, 2011; Gunderson et al., 2012; Vandrey et al., 2012; Wiley et al., 2011). These molecules produce many effects similar to those of delta-9-tetrahydrocannabinol (THC)¹, the principle psychoactive constituent of cannabis, by activating cannabinoid type-1 receptors (CB₁) in the brain. SCRAs were originally developed to have high affinity for CB₁ and/or its sibling G-protein coupled receptor, the cannabinoid type-2 (CB₂) receptor, and incrementally modified through structure-activity relationship studies to describe the pharmacophores necessary for interaction and activation of these receptors (Wiley et al., 2011). Thus, large libraries of chemical structures are available in the scientific and patent literature which have been exploited in recent years by clandestine chemists who synthesize and sell these compounds through a variety of online and retail markets as “legal weed” and cannabis alternatives. Unfortunately, despite growing awareness of dangers associated with these products, self-described “psychonauts” and users seeking cannabis-like highs have continued to obtain and abuse these drugs, resulting in hospitalizations and deaths in numerous countries including the Australia, New Zealand and the United States (Tait et al., 2016; Trecki et al., 2015).

Among these SCRAs, EG-018 has received little attention despite being an analogue of JWH-018, which was commonly found in “Spice” products until it was scheduled (Banister et al., 2015). EG-018 is structurally differentiated from JWH-018 by its carbazole core, in place of the indole core [Figure 1]. EG-018 was first identified in the United Kingdom between February of 2013 and January of 2015 (Bijlsma et al., 2017; EMCDDA, 2015) and was subsequently detected in Japan (Uchiyama and Kikura-Hanajiri, 2016) and the United States (Worst and Sprague, 2015). Anecdotal reports of EG-018’s effects in humans suggest it to be very weak/low potency; however, the obvious caveat is there is no way to know if the compounds consumed were in fact EG-018 or if other SCRAs were also present (online forums). Despite the limited anecdotal discussion, EG-018 is sold online and has been detected in human urine samples (Mogler et al., 2018) as well as in femoral blood and brain during post-mortem of a 27 year old who fell to his death (Gaunitz et al., 2018). While the metabolism of EG-018 has been examined in a few studies (Diao et al., 2018; Gaunitz et al., 2019; Mogler et al., 2018) and EG-018 has been reported to bind the CB₁ receptor and to inhibit forskolin-stimulated cAMP production with an EC₅₀ of 40 nM (Angerer et al., 2015), there is little to no data regarding its potential in vivo effects. Specifically, it is currently unknown whether EG-018 produces cannabimimetic effects including discriminative stimulus effects which are indicative of cannabis-like psychoactivity.

Figure 1.

Structures of EG-018, JWH-018, delta-9-tetrahydrocannabinol, CP55,940, and anandamide.

The current study characterized EG-018 in terms of its molecular and behavioral pharmacology as well as its pharmacokinetics. For in vitro studies, EG-018 was examined for its binding affinity to the human CB₁ (hCB₁) and human CB₂ (hCB₂) receptors as well as its ability to stimulate [³⁵S]GTPγS binding and inhibit forskolin-stimulated cAMP production which reflect its functional activity at the level of the G protein and downstream effects on adenylyl cyclase, respectively. For in vivo studies, EG-018 was examined for its effects in the cannabinoid tetrad, including locomotor activity, catalepsy, and rectal temperature, and it was assessed in THC drug discrimination.

2. Materials and Methods:

2.1. Subjects

Drug naïve male ICR mice (31-34g), obtained from Harlan (Frederick, MD) and housed singly in polycarbonate mouse cages, were used for assessment of locomotor suppression, antinociception, hypothermia, and catalepsy. Separate mice were used for testing each dose of each compound in this battery of procedures, and mice were assigned to doses randomly. Some of these mice were later re-used to evaluate rimonabant antagonism of cannabimimetic effects of the compounds in the tetrad. These mice had free access to food when in their home cages. Singly housed male C57/B16J inbred mice (20-25g; Jackson Laboratories, Bar Harbor, ME) were used in the drug discrimination experiments. These mice were maintained at 85-90% of free-feeding body weights by restricting daily ration of standard rodent chow. At the start of this investigation, these mice had already been trained to discriminate THC from vehicle and had been tested with other SCRAs (Wiley et al., 2015). All animals were kept in a temperature-controlled (20-22°C) environment with a 12-hour light-dark cycle (lights on at 6 a.m.) and received water ad libitum. The in vivo studies reported in this manuscript were carried out in accordance with guidelines published in the Guide for the Care and Use of Laboratory Animals (National Research Council, 2011) and were approved by our Institutional Animal Care and Use Committee.

2.2. Apparatus

For the tetrad test battery in mice, measurement of spontaneous activity occurred in Plexiglas locomotor activity chambers (47 cm × 25.5 cm × 22 cm). Beam breaks (4 × 8 beam array) were recorded by San Diego Instruments Photobeam Activity System software (San Diego, CA) on a computer located in the experimental room. A standard tail flick device for rodents (Stoelting, Dale, IL) was used to assess antinociception. A digital thermometer (Physitemp Instruments, Inc., Clifton, NJ) was used to measure rectal temperature. The ring immobility device consisted of an elevated metal ring (diameter = 5.5 cm, height = 28 cm) attached to a metal stand.

Mice in the drug discrimination experiment were trained and tested in mouse operant chambers (Coulbourn Instruments, Whitehall, PA), housed within light- and sound-attenuating cubicles. Each chamber contained nose poke apertures, with stimulus lights located over each aperture, and a separate house light. A food dispenser delivered 20-mg food pellets (Bioserv Inc., Frenchtown, NJ) into a food cup (with a light) centered between the two levers/apertures. Illumination of lights, delivery of food pellets, and recording of nose pokes were controlled by a computer-based system (Coulbourn Instruments, Graphic State Software, v 3.03, Whitehall, PA).

2.3. Drugs and Chemicals

For the in vitro studies, [³H]CP55,940 (NIDA Drug Supply Program, NDSP; Bethesda, MD), CP55,940 (NDSP), THC (Cayman Chemical, Ann Arbor, MI) and anandamide (Cayman Chemical) were dissolved in absolute ethanol, stored at −80°C in silanized glass dram vials under a blanket of nitrogen and sealed with Teflon tape. EG-018 (Cayman Chemical) was dissolved in DMSO and stored at −80°C. Guanosine 5’ diphosphate (GDP; Sigma Aldrich, St. Louis, MO), unlabeled GTPγS (Sigma Aldrich, St. Louis, MO), and [³⁵S]GTPγS (1150-1300 Ci/mmol; Perkin Elmer Life Sciences, Boston, MA) were dissolved in distilled water, aliquoted and stored at −80°C. Scintillation fluid (MicroScint-20 and Ultima Gold) were obtained from Perkin Elmer Life Sciences. HPLC grade acetonitrile and water were purchased from Fisher Scientific (Fairlawn, NJ). Reference standards and metabolites reference standards for all compounds were obtained from Cayman Chemical (Ann Arbor, MI).

For in vivo studies, THC and rimonabant (the prototypic CB₁ receptor antagonist/inverse agonist) were obtained from the National Institute on Drug Abuse (NIDA, Bethesda, MD) through the NIDA Drug Supply Program. EG-018 [Naphthalen-1-yl(9-pentyl-9H-carbazol-3-yl)methanone] was provided by the U.S. Drug Enforcement Administration or was purchased from Cayman Chemical (Ann Arbor, MI). For the in vivo tests, the vehicle for all compounds was 7.8% Polysorbate 80 N.F.(VWR, Marietta, GA), and 92.2% sterile saline USP (Butler Schein, Dublin, OH). All compounds were injected at a volume of 10 ml/kg.

2.4. Experimental Procedures

2.4.1. Receptor and agonist-stimulated [³⁵S]GTPγS binding

Human embryonic kidney (HEK293) cells stably expressing either the hCB₁ or hCB₂ receptor (Perkin Elmer Life Sciences, Boston, MA), were grown in Dulbecco’s Modified Eagle’s Media/F12 (10-092-CV; Corning Cellgro, Manassas, VA) with 10% fetal bovine serum (FBS-BBT; Rocky Mountain Biological Laboratory, CO, USA), 50 unit/ml penicillin/streptomycin (Thermo Fisher Scientific; Waltham, MA, USA), 500 μg/mL G418 (ant-gn-5; Invivogen, San Diego, CA), and 2-4 μg/mL puromycin (ant-pr-1; Invivogen) in multilayer flasks to 90% confluence under 5% CO₂ at 37°C. Cells were detached using 1 mM EDTA in phosphate buffered saline (PBS; Sigma Aldrich, St. Louis, MO, USA), pelleted in PBS at 200 x g for 6 min, then suspended in membrane buffer (50 mM Tris Base, 1 mM EGTA, 3mM MgCl₂, pH 7.4) containing protease inhibitor cocktail (Pierce Protease Inhibitor mini-tab let, Thermo Scientific, A32953) and homogenized by Brinkmann Polytron 3000 with Kinematica 95/ PT-DA 3012/2 S generator probe for 15 s at 22,000 RPM. Cell homogenates were centrifuged at 1600 × g for 10 min at 4°C, the supernatant was collected, and the pellet was homogenized again and centrifuged at 1600 × g for 10 min at 4°C. The supernatants from the two spins were pooled and spun at 40,000 × g for 1 h at 4°C resulting in a P2 pellet. The P2 pellet was resuspended in membrane buffer and the protein amount was quantified by the Bradford method (Thermo Scientific Pierce Coomassie Plus, PI23238) using absorbance mode in a Clariostar plate reader (BMG Labtech; Cary, NC, USA). The membrane preparations were diluted to 1 mg/mL, aliquoted at 1.1 mL, snap frozen in liquid nitrogen, and stored at −80°C until the day of the experiment. Membrane preparations contain 2.1 pmol/mg of hCB₁ receptors and 1.2 pmol/mg of hCB₂ receptors from respective cell lines as determined by saturation binding. For receptor binding, reactions were carried out in assay buffer containing 50 mM Tris Base, 100 mM NaCl, 3 mM MgCl₂, 0.2 mM EGTA and 5 mg/ml BSA and 10 μg total protein per well. Reactions were carried out for 90 min at 30°C with ~1 nM [³H]CP55,940 and concentrations of compounds on a logarithmic scale. Amount of radioligand added for each experiment was determined by pipetting 50 μL of each nominal concentration stock and adding 20 mL of Ultima Gold scintillation cocktail and analysis on a Packard TriCarb 2300TR scintillation counter. Non-specific binding was determined by addition of excess unlabeled SR141716 ligand (10 μM). For receptor signaling, membranes (10 μg protein determined by Bradford method) were pre-equilibrated for 30 min at 30°C with 30 μM GDP and drugs followed by addition of 0.1 nM [³⁵S]GTPγS in a volume of 10 μL (5% final volume) to ensure compounds were pre-equilibrated close to the final concentration. Nonspecific binding was determined in presence of 30 μM unlabeled GTPγS. Binding for all reactions was terminated by vacuum filtration through a Perkin Elmer GF/C filter plate using a Perkin Elmer FilterMate followed by rinsing with 4-6 mL of cold rinse buffer (50 mM Tris, 0.1% BSA, pH 7.4) per well. Plates were dried, 35 μL of Microscint-20 (Perkin Elmer) was added to each well, sealed with TopSeal-A, and counted on a Packard TopCount NXT.

2.4.2. cAMP assay

Forskolin (FSK)-stimulated cyclic adenosine monophosphate (cAMP) production was measured in real-time using HEK293 cells stably expressing hCB₁ (Perkin Elmer Life Sciences) or Flp-In™HEK293 cells stably expressing HA-3TCS-hCB₂ (Cawston et al., 2015; “3TCS” refers to 3x N-terminal thrombin cleavage sites which are not relevant for this study), and a transiently transfected bioluminescence resonance energy transfer (BRET) biosensor containing a cAMP binding domain (Epac1) flanked by yellow fluorescent protein (YFP) and Renilla Luciferase (RLuc) assay (CAMYEL; Cawston et al., 2013; Jiang et al., 2007). Cells were seeded in 100 mM dishes for transfection. The next day, cells were given fresh growth media and transfected with 5 μg of pcDNA3L-His-CAMYEL using linear polyethyleneimine (25 kDa, Polysciences, Warrington, PA) in 1:6 DNA:PEI ratio. The following day, cells were lifted using 1 mM EDTA in PBS and centrifuged at 200 x g for 5 min. The supernatant was removed, and cells were resuspended in growth media and plated on poly-D-lysine (Sigma Aldrich, St. Louis, MO) coated white 96 well plates (Perkin Elmer, Waltham, MA) at 60,000 cells per well, filling 2 columns of 8 wells each per plate, i.e. 8 samples in duplicate per plate. The following day, the media was removed and cells were rinsed with PBS. For hCB₁, cells were equilibrated with HBSS containing 0.5% BSA for 15 min followed by addition of coelenterazine h (5 μM) and an additional 10 min equilibration. For hCB₂, cells were equilibrated in phenol-free DMEM containing 0.1% BSA for 30 min followed by addition of coelenterazine h (5 μM) and an additional 5 min equilibration. Immediately following co-addition of forskolin (10 μM) and the respective agonist, luminescence was measured at 460 nm and 535 nm simultaneously for 1 s per well using a BMG Clariostar plate reader (hCB₁) or for 0.5 s per well using a BMG LumiSTAR Omega (hCB₂) for ~22 min at 37°C. The luminescence ratio at 460nm/535 nm (inverse BRET ratio) was calculated to quantify cAMP levels, where increases in the ratio indicate increases in cAMP.

2.4.3. Mouse Tetrad

Each mouse was tested in a battery of four tests, in which cannabinoid agonists produce a characteristic profile of in vivo effects (Martin et al., 1991): suppression of locomotor activity, antinociception, decreased rectal temperature and ring immobility. Prior to injection, rectal temperature and baseline latency in the tail flick test were measured in the mice. The latter procedure involved placing the mouse’s tail under an intense fight (radiant heat) in a tail flick apparatus and recording latency (s) for tail removal. A maximal latency of 10 s was used to minimize injury. After measurement of temperature and baseline tail flick latency, mice were injected intraperitoneally (i.p.) with vehicle or drug. Thirty min later they were placed into individual activity chambers for 10 min. Tail-flick latency and rectal temperature were measured again immediately after removal from the locomotor chambers. At 50 min post-injection, the mice were placed on the elevated ring apparatus, and the amount of time the animals remained motionless during a 5 min period was recorded. If a mouse fell off the ring during the catalepsy test, it was immediately placed back on and timing was continued for up to 9 falls. After the 10th fall, the test was terminated for the mouse.

After completion of all agonism tests, rimonabant antagonism was evaluated in a subset of the same mice, when appropriate, following a minimum of 1 week to allow for drug washout. A dose of each compound that was active in the first test (56 mg/kg THC) was re-tested in combination with vehicle or 3 mg/kg rimonabant. Procedural details were identical to those described above for agonist tests, with the exception that mice received an i.p. injection of vehicle or rimonabant 10 min prior to i.p. injection of THC.

Because EG-018 was not active at doses up to 100 mg/kg, i.p., a probe dose of 56 mg/kg was administered i.v. to separate group of mice. Evaluation in the tetrad tests proceeded as described above, with the exception that mice were placed into the locomotor chambers and on the ring apparatus 5 and 25 min after injection, respectively. Subsequently, the effect of 3 mg/kg rimonabant (i.v.) in combination with the 56 mg/kg dose (i.v.) EG-018 was assessed. Rimonabant was injected 10 min prior to each compound.

2.4.5. Drug Discrimination

Adult male mice were trained to discriminate THC prior to this study and had previously been tested with other SCRAs (Wiley et al., 2015). Briefly, each mouse was placed in a standard operant conditioning chamber with two nose poke apertures. Mice were trained to respond on one of the two apertures following i.p. administration of 5.6 mg/kg THC and to respond on the other aperture following i.p. vehicle injection, according to a fixed ratio 10 (FR10) schedule of food reinforcement under which 10 consecutive responses on the correct (injection-appropriate) aperture resulted in delivery of a food pellet. Responses on the incorrect aperture reset the ratio requirement on the correct lever. Daily injections were administered on a double alternation sequence of THC and vehicle (e.g., drug, drug, vehicle, vehicle). Daily 15 min training sessions were held Monday-Friday. When the criteria were met, acquisition of the discrimination was established, and substitution testing began.

Stimulus substitution tests were typically conducted on Tuesdays and Fridays during 15 min test sessions, with maintenance of training continuing on intervening days. During test sessions, 10 consecutive responses on either aperture delivered reinforcement. To be tested in the experiment, mice must have completed the first FR on the injection-appropriate aperture and must have responded ≥ 80% on the injection-appropriate aperture on the preceding day and during the previous training session with the alternate training compound (training drug or vehicle). A dose-effect curve was determined with THC prior to testing other compounds. For further details regarding the training of these mice see our previous paper (Wiley et al., 2015).

2.4.6. Metabolite Analysis in Mice

Mice (n=4) were given i.p. injections of 30 mg/kg EG-018. Immediately following drug administration, the mice were placed into metabolism cages and urine was collected over a 24 h period. Urine from mice dosed with the same compound was pooled for analysis. Samples were extracted using a salting out liquid-liquid extraction (SALLE) method prior to analysis. Acetonitrile (200 μL) was added to 100 μL of urine, then the samples were vortexed and 50 μL of 5 M ammonium acetate was added as a salting out agent. Samples were vortexed and centrifuged at 10,000 χ g for 5 min. The top aqueous layer was removed and dried down at 40 °C and reconstituted with 50 μL of mobile phase A.

Samples were analyzed on a Waters Acquity ultra performance liquid chromatography (UPLC) system coupled to a Waters Synapt G2 HDMS quadrupole time-of-flight (Q-TOF) mass spectrometer (Waters, Milford, MA). The mass spectrometer was operated under resolution mode, positive electrospray ionization, source temperature of 150 °C, desolvation temperature of 500 °C , desolvation gas at 1,000 L/hr, capillary voltage at 2.99 kV, sampling cone at 35 V, and extraction cone at 4.3 V. The mass spectrometer was externally calibrated from 50 - 1000 m/z using a sodium formate solution. Leucine enkephalin was used as a lockmass to correct for mass shifts during acquisition. Full scan data was collected in both low (4 eV) and high (15 to 40 eV ramp) collision energies nearly simultaneously for every m/z using MS^E acquisition mode (Bateman et al., 2002).

Samples were separated on an Acquity BEH C18 column (1.7 μm 2.1 x 50 mm) connected to a Vanguard BEH C18 pre-column (1.7 μm x 2.1 X 5 mm) and held at 30 °C. Injection volume was 10 μL. A gradient elution with a flow rate of 500 μL/min was used with mobile phase A consisting of water with 0.1% formic acid and mobile phase B consisting of acetonitrile with 0.1% formic acid. The mobile phase composition for EG-18 was held at 90% A for 0.5 min, decreased to 65% A over 1 min and 35% A over 13 min, then decreased to 5% A over 5 min and held at 90% A for 2.9 min for column re-equilibration. All mobile phase composition changes were done linearly.

2.4.7. Pharmacokinetic Analysis in Rat and Human Liver Microsomes and Rats

Microsomal incubations were performed as detailed previously (Kevin et al., 2017a). EG-018 was incubated at 1 μM with rat (pooled IGS Sprague-Dawley) or human liver microsomes (pooled) at 37°C in triplicate. An assay mixture containing microsomes (1 mg protein/mL final concentration), NADPH (1 mM final) and a buffer consisting of 50 mM potassium phosphate buffer, pH 7.4 with 3 mM MgCl₂ was prepared and preincubated at 37 °C for 5 min. 10 μL of a 0.1 mM drug solution in acetonitrile was added to 990 μL of assay mixture in a glass test tube in a 37 °C water bath. At 0, 5, 10, 30, and 60 min timepoints, 100 μL samples were removed and added to 100 μL acetonitrile. Samples were centrifuged (11,000 χ g for 1 min) and stored at −80 °C until analysis.

Four 49-day-old male Long-Evans rats with jugular vein catheters were obtained from Charles River Laboratories (Raleigh, NC, USA) and housed in a temperature-controlled environment (20-22 °C) with ad libitum access to food and water. On the day of testing, the animals were placed individually into glass metabolite cages (Prism Research Glass, Raleigh, NC, USA), and 3 mg/kg EG-018 was administered intraperitoneally (3 mg/mL EG-018 in 7.8% polysorbate 80 and 92.2% saline). 200 μL blood samples were drawn 0.25 h prior to injection and 0.25, 0.5, 1, 2, 4, 8, and 24 h post-injection. Samples were drawn into chilled K₃EDTA collection tubes and centrifuged at 2800 χ g for 10 min at 4 °C. Plasma was decanted and stored at −80 °C until further analysis.

Extractions of EG-018 from microsome samples and plasma were performed as previously described (Kevin et al., 2017a). Briefly, acetonitrile was added in a 3:1 ratio to the sample volume (microsomes sample volume 50 μL; plasma 25 μL) and centrifuged at 4000 χ g for 15 min at 4 °C. Supernatants were transferred to vials for immediate analysis via liquid chromatography-tandem mass spectroscopy (LC-MS/MS).

2.5. Data analysis

2.5.1. In vitro data analysis

EC₅₀ values for cAMP and [³⁵S]GTPγS were calculated by non-linear regression fitting a three-parameter logistic curve using Prism 6 (Graphpad Software, San Diego, CA). For [³⁵S]GTPγS experiments, data were expressed as % net stimulation [e.g. value – basal / basal] and E_max was calculated as the top of the curve-fit. For cAMP data, the ratio of 460/535 was calculated for each time point and plotted across time and area under the curve analysis was conducted for each replicate. The cAMP data were calculated as %FSK using the formula [(sample – basal) / (forskolin – basal) x 100] and E_max was calculated as the span of the curve-fit. Data are presented as the mean ± SEM of at least n=3 independent experiments. CP55,940 binding data were fitted to either one-site homologous competition binding model for determination of K_d, whereas THC and EG-018 binding data were fitted to one-site heterologous competition binding model using concurrently determined K_d values for CP55,940 binding for curve fitting. Curve-fit parameters obtained for each n were analyzed by one-way ANOVA followed by Tukey’s post-hoc test (α = 0.05) when significant effect of treatment was determined.

2.5.2. Mouse Tetrad

Spontaneous activity was measured as total number of photocell beam interruptions during the 10-min session. Antinociception was expressed as the percent maximum possible effect (MPE) using a 10-s maximum test latency as follows: [(test-control)/(10-control)]x100. Rectal temperature values were expressed as the difference between control temperature (before injection) and temperature following drug administration (Δ°C). For catalepsy, the total amount of time (s) that the mouse remained motionless on the ring apparatus (except for breathing and whisker movement) was used as an indication of catalepsy-like behavior. This value was divided by 300 s and multiplied by 100 to obtain a percent immobility. Separate between-subjects ANOVAs were also used to analyze the four measures for THC and EG-018. Significant differences from control (vehicle) were further analyzed with Dunnett’s post hoc tests (α = 0.05) as necessary. Factorial ANOVAs (rimonabant dose X EG-018 dose) were used to analyze results of antagonist tests with rimonabant and EG-018. Significant main effects and interactions were further analyzed with Tukey post hoc tests (α = 0.05) as necessary.

2.5.3. Drug Discrimination

For each session, percentage of responses on the drug-associated aperture and response rate (responses/s) were calculated. Full substitution was defined as ≥ 80% responding on the drug-associated aperture (Vann et al., 2009). ED₅₀ values were calculated by logarithmic non-linear regression of drug aperture selection data in GraphPad Prism with the curve bottom constrained to 0 and the top constrained to 100. Since mice that responded less than 10 times during a test session did not respond on either aperture a sufficient number of times to earn a reinforcer, their data were excluded from analysis of drug aperture selection, but their response rate data were included. Response-rate data were analyzed using repeated-measures ANOVA across dose. Significant ANOVAs were further analyzed with Dunnett’s post hoc tests (α = 0.05) to specify differences compared to vehicle. For experiments that administered EG-018 prior to THC, response rates were analyzed by two-way repeated measures ANOVA. Significant ANOVAs were further analyzed by Bonferroni’s post hoc tests (α = 0.05) to compare means at each dose of THC following vehicle or EG-018 treatment.

2.5.4. Metabolite Identification

LC-MS data was analyzed using Waters MassLynx 4.1 with the aid of the MetaboLynx application manager. Automated data processing with MetaboLynx was supplemented by manual interrogation of the data using mass defect filtering, precursor ion and fragment ion searching techniques (Grabenauer et al., 2012). Presence of potential metabolites was determined by exact mass match to predicted elemental compositions in the low energy data function. Further refinement of the site of modification was determined by presence of characteristic fragment ions at the same retention time. Metabolites were provisionally identified by their molecular weight, retention time, and fragment ions. Metabolites were compared to reference standards as available.

3. Results

3.1. Receptor Binding and Function

EG-018 was assessed for affinity at cannabinoids receptors in [³H]CP55,940 competition binding assays [Figure 2]. Fitting of a homologous competition binding model to CP55,940 curves yielded pK_d values of 9.07±0.0208 (9.01 – 9.14; 0.849 nM) at hCB₁ and 9.09±0.165 (8.37 – 9.80; 0.822 nM) at hCB₂ receptors expressed in HEK293 cell membranes. THC and EG-018 both exhibited significantly lower affinity (10- to 40-fold respectively) than CP55,940 at both receptors [hCB₁: F (2, 9) = 379.2, p<0.0001; and hCB₂: F (2, 6) = 39.02, p<0.00l]. In addition, THC exhibited significantly greater affinity for hCB₁ receptors compared to EG-018 whereas EG-018 showed greater affinity for hCB₂ receptors (Table 1).

Figure 2.

[³H]CP55,940 (~1 nM) competition binding curves for CP55,940 (filled black circles), THC (filled green squares) and EG-018 (filled red triangles) using P2 membrane preparations from HEK293 cells stably expressing the (A) hCB₁ or (B) hCB₂ receptors. Compounds were incubated for 90 min at 30°C. Each data point represents the mean ± standard error of at least n=3 experiments performed in duplicate.

Table 1.

Binding affinities at cannabinoid receptors in membranes from HEK293 cells stably expressing the human CB₁ or human CB₂ receptor using [³H]CP55,940. Values reflect mean ± standard error of at least n=3 experiments performed in duplicate.

	CP55,940	EG-018	THC
		hCB1
pKi±SEM	9.07±0.0208^^^^	7.66±0.0739****	8.09±0.0182####
95% C.I.	(9.01 – 9.14)	(7.34 – 7.98)	(8.05 – 8.14)
Ki (nM)	0.849 nM	21.8 nM	8.05 nM
95% C.I.	(0.724-0.977)	(10.5-45.7)	(7.24-8.91)
		hCB2
pKi ±SEM	9.09±0.165^^^	8.11±0.132**	7.5±0.0691#
95% C.I.	(8.37 – 9.8)	(7.54 – 8.68)	(7.2 – 7.79)
Ki (nM)	0.822 nM	7.76 nM	32 nM
95% C.I.	(0.158-4.27)	(2.09-28.8)	(16.2-63.1)

The number of symbols indicates the increasing level of significance, i.e. 1 = p < 0.05, 2 = p < 0.01, 3 = p < 0.001, 4 = p < 0.0001;

^*vs. CP55,940,

^#vs. EG-018,

^{^}vs. THC

CP55,940, EG-018, THC, and anandamide stimulated [³⁵S]GTPγS binding [Figure 3, top panel], varied in both potency and efficacy. At hCB₁ receptors, CP55,940 was the most potent, followed by EG-018 which was approximately 90-fold less potent than CP55,940. THC was 2.5-fold less potent than EG-018 and anandamide was 3.5-fold less potent than THC [F (3, 11) = 276.6, p<0.0001]. At hCB₁ receptors, CP55,940 had 2-fold greater efficacy than THC and 3.5-fold greater efficacy than EG-018, but CP55,940 was not significantly different than anandamide; THC exhibited approximately 2-fold less efficacy than anandamide whereas EG-018 exhibited roughly 2-fold less efficacy than THC [F (3, 12) = 60.19, p<0.0001]. There were fewer differences between E_max values and pEC₅₀ values at hCB₂ receptors. At hCB₂ receptors [Figure 3B]. CP55,940 was 12-fold more potent than EG-018, 25-fold more potent than THC, and 200-fold more potent than anandamide [F (3, 9) = 14.40, p<0.001] but no differences between the latter three compounds were observed. CP55,940 exhibited 2.5-fold greater efficacy than THC, anandamide and EG-018 [F (3, 9) = 7.744, p<0.01].

Figure 3.

Top Panel: Concentration response curves for CP55,940 (filled black circles), THC (filled green squares), EG-018 (filled red triangles), and anandamide (AEA; filled orange downward triangles) in [³⁵S]GTPγS binding using P2 membrane preparations from HEK293 cells expressing the (A) hCB₁ or (B) hCB₂ receptors. [³⁵S]GTPγS reactions were incubated for 1 h at 30°C. Bottom Panel: Concentration response curves inhibiting forskolin-stimulated cAMP production using intact HEK293 cells expressing the (C) hCB₁ or (D) hCB₂ receptors. and cAMP assay was incubated for 22 min at 37°C. Each data point represents the mean ± standard error of at least n=3 experiments performed in duplicate.

CP55,940, THC and EG-018 all inhibited forskolin-stimulated cAMP production in intact hCB₁ and hCB₂ expressing HEK293 cells [Figure 3, bottom panel]. E_max is calculated as the span of the inhibition curve, i.e. greater values reflect greater efficacy. At hCB₁ receptors [Figure 3C]. CP55,940 was 10-fold more potent than THC and 53-fold more potent than EG-018 [F (2, 8) = 19.2, p<0.001]. In agreement with the [³⁵S]GTPγS experiments, EG-018 also exhibited a significantly lower E_max than that of CP55,940 [F (2, 8) = 9.60, p<0.01] but not THC. At hCB₂ receptors [Figure 3D]. CP55,940 was 26-fold more potent than THC and 114-fold more potent than EG-018 [F (2, 6) = 30.4, p<0.001]. Further, THC and EG-018 exhibited significantly lower efficacy than CP55,940 [F (2, 6) = 16.91, p<0.01], inhibiting forskolin-stimulated cAMP production by half of the response induced by CP55,940.

3.2. Behavioral Studies

3.2.1. Drug Discrimination

EG-018 was tested in mice trained to discriminate THC (5.6 mg/kg, i.p.) from vehicle in order to assess its ability to substitute for THC [Figure 4A–B], an effect which can predict psychoactive effects similar to cannabis in humans. THC fully substituted for itself with an ED50 value of 2.2 mg/kg (95% CI: 1.4 – 3.6). In contrast, EG-018 (10 – 100 mg/kg, i.p.) did not substitute for THC at any dose. Neither THC nor EG-018 significantly affected response rates compared to vehicle. Considering the low efficacy of EG-018 in stimulating [³⁵S]GTPγS binding, we hypothesized that EG-018 may insufficiently activate CB₁ receptors in brain to produce stimulus effects; hence, we evaluated it for its ability to act as an in vivo antagonist to shift THC’s dose-effect curve. In this experiment, mice [Figure 4C–D] were pretreated with either vehicle or EG-018 (100 mg/kg, i.p.) 30 min prior to THC (0.3, 1, 3, 5.6, and 10 mg/kg, i.p.). Results showed that there was no significant difference in THC’s potency, regardless of whether mice were pretreated with vehicle [ED₅₀: 1.5 mg/kg (0.79 – 2.7)] or EG-018 [ED₅₀ of 1.2 mg/kg (0.57 – 2.6)]. Interestingly, there was a significant main effect of THC [F (4, 28) = 2.863, p<0.05] and EG-018 [F (1, 7) = 6.225], and a significant interaction effect [F (4, 28) = 4.021, p<0.05] on response rates [Figure 4D], Tukey’s post-hoc analysis indicated that EG-018 pretreatment resulted in significantly greater response rates at 5.6 and 10 mg/kg THC as compared to responding at these doses following vehicle pretreatment.

Figure 4.

Top Panel: Effects of THC (filled green squares) and EG-018 (filled red triangles) in mice trained to discriminate THC (5.6 mg/kg, i.p.) from vehicle using nose-poke response. Bottom Panel: Effects of EG-018 pretreatment (10 – 100 mg/kg, i.p.) on the discriminative stimulus effects of THC in mice trained to discriminate THC (5.6 mg/kg, i.p.) from vehicle using nose-poke response. Left Panels: percent THC-paired aperture responding. Right Panels: response-rates. Vehicle and THC training dose control tests are depicted on the left half of the X-axis. Values represent the mean ± standard error of data from 7 male C57/BL6J mice. The number of symbols indicates the increasing level of significance, i.e. 1 = p < 0.05, 2 = p < 0.01, 4 = p < 0.0001; * vs. Vehcontrol test (left x-axis), # vs. Veh+THC (PanelD)

3.2.2. Cannabinoid tetrad

To determine if EG-018 could produce typical cannabimimetic effects, it was tested in the cannabinoid tetrad, which includes measures of locomotor activity, catalepsy, nociception, and body temperature [Figure 5]. In mice treated acutely, THC (3, 10 or 56 mg/kg, i.p.) produced hypomotility [F (3, 20) = 7.15, p<0.01], anti-nociception [F (3,20)=21.7, p<0.05], hypothermia [F (3,20)=63.9, p<0.05], and catalepsy [F (3,20)=27.3, p<0.05]. In contrast, EG-018 (1, 3, 10, 30, or 100 mg/kg, i.p.) did not produce significant effects in any of the four assays [Figure 5, top panel].

Figure 5.

Top Panel: Effects of THC (filled green squares) and EG-018 (filled red triangles) on (A) locomotor activity (beam breaks during 10 min session), (B) catalepsy (ring immobility), (C) antinociception (latency to tail-flick), and (D) hypothermia (rectal temperature) in mice administered drugs i.p. 30 min prior to start of the experiment. Bottom Panel: Effects of EG-018 (56 mg/kg, i.v., 5 min) in mice given a 10 min pretreatment of vehicle (white bars) or rimonabant (3 mg/kg i.v.) on (E) locomotor activity (beam breaks during 10 min session), (F) catalepsy (ring immobility), (G) antinociception (latency to tail-flick), and (H) hypothermia. Values represent the mean±standard error of 6 male ICR mice. The number of symbols indicates the increasing level of significance, i.e. 1 = p < 0.05, 2 = p < 0.01, 3 = p < 0.001, 4 = p < 0.0001; * vs. Veh (A-D) or Veh+Veh (E-H), ^ vs. vehicle+rimonabant, # vs. EG-018+Veh, $ main effect of EG-018 treatment. V = Veh

To account for potential pharmacokinetic effects that could prevent sufficient EG-018 brain levels following i.p. administration, EG-018 (56 mg/kg) was administered intravenously [Figure 5, bottom panel] to mice 10 min following vehicle or the CB₁ antagonist rimonabant (3 mg/kg, i.v.). In contrast to i.p. administration, EG-018 administered i.v. resulted in hypomotility [significant main effect EG-018: F (1, 20) = 20.4, p<0.001; Figure 5E], catalepsy [significant main effect EG-018: F (1, 20) = 36.8, p<0.0001; Figure 5F], and hypothermia [significant main effect EG-018: F (1, 20) = 97.7, p<0.0001; Figure 5H], but it did not significantly elicit antinociceptive behavior [main effect EG-018: F (1, 20) = 4.24, p=0.0527; Figure 5G]. In addition, the magnitude of each effect was substantially reduced compared to that obtained with THC [Figure 5, top panel]. Pretreatment with rimonabant attenuated catalepsy [significant main effect rimonabant: F (1, 20) = 9.90, p<0.01; significant interaction effect: F (1, 20) = 10.7, p<0.01], but did not significantly reduce hypothermia [main effect rimonabant: F (1, 20) = 1.32, p=0.265] or hypomotility [main effect rimonabant: F (1, 20) = 0.554, p=0.465].

3.3. Pharmacokinetics and Metabolism

Pharmacokinetic parameters derived from microsomal incubations and rat plasma are presented in Table 4. Rat microsome elimination half-life was shorter than for human microsomes, but broadly predicted kinetic parameters observed from our in vivo rat plasma assay. No intact parent compound or metabolites were identified in the urine from mice dosed with EG-018, despite using automated tools within the Metabolynx software package for expected and unexpected metabolites, and manual interrogation of the data for protonated and sodiated accurate masses of predicted metabolites and predicted and known fragment ions. A strong signal for EG-018 was obtained from the dosing solution when analyzed using the metabolite LC-MS method.

Table 4.

Pharmacokinetic parameters of EG-018 in rat and human liver microsomes and rat plasma

Parameter	EG-018
Rat liver microsomes
Half life (min)	4.29
CLint, micr (mL/min/mg)	0.16
CLint (mL/min/kg body wt.)	290.77
CLH (mL/min/kg body wt.)	46.39
ER	0.84
Human liver microsomes
Half life (min)	2.04
CLint, micr (mL/min/mg)	0.34
CLint (mL/min/kg body wt.)	393.09
CLH (mL/min/kg body wt.)	19.03
ER	0.95
Rat Plasma
Half life (h)	2.53
CL/F (mL/min/kg body wt.)	520.93
Cmax (ng/mL)	26.01
Tmax (h)	2.00
AUC 0-24 h (h.ng/mL)	154.06
AUC 0-∞ (h.ng/mL)	167.92

AUC area under the curve, CL/F observed apparent clearance, CL_H estimated hepatic clearance, CL_int estimated intrinsic clearance, CL_{int, micr} intrinsic microsome clearance, C_max mean maximum observed concentration, ER extraction ratio, T_max mean time of C_max

4. Discussion

The current studies evaluated the in vitro and in vivo pharmacology of the synthetic cannabinoid receptor agonist (SCRA) EG-018 [Figure 1]. Binding studies were conducted to determine EG-018’s affinity at the hCB₁ and hCB₂ receptors using [³H]CP55,940 as the probe radioligand [Figure 2] and THC and CP55,940 were included as a controls. Results showed that EG-018 bound to both cannabinoid receptors with moderate to high affinity, exhibiting approximately 3-fold greater affinity for hCB₂ (8 nM) over hCB₁ (22 nM). Determined K_i values for EG-018 were similar to nanomolar affinities reported previously (Schoeder et al., 2018). K_d values determined for CP55,940 using a homologous competition binding model were consistent with previous K_d values determined from saturation binding experiments in the same cell line (Gamage et al., 2018; Gamage et al., 2019) and with affinities reported in the literature (Mauler et al., 2002; Showalter et al., 1996). THC’s affinity at both receptors was also consistent with previous observations (De Vry et al., 2004; Gamage et al., 2018; Mauler et al., 2002).

In [³⁵S]GTPγS binding [Figure 3, top panel] using membranes from HEK293 cells stably expressing the cannabinoid receptors, CP55,940 exhibited sub nanomolar potency similar to previous studies (Gamage et al., 2018; Gamage et al., 2019; Thomas et al., 2017). EG-018 was notably more potent at hCB₂ than hCB₁ which aligns with its observed greater affinity in binding. As expected, THC exhibited markedly lower efficacy than CP55,940, consistent with its established designation as a partial agonist and in line with our previous studies (Gamage et al., 2018). While EG-018 exhibited similar low efficacy to that of THC at hCB₂ receptors, its efficacy at hCB₁ receptors was significantly lower. In fact, EG-018 appeared to be inactive in [³⁵S]GTPγS binding in initial studies where the ligands were not pre-equilibrated prior to addition of [³⁵S]GTPγS (data not shown). Recently, the low efficacy of EG-018 at hCB₁ receptors has also been reported in assays of mini-Gα_i and beta arrestin-2 recruitment (Wouters et al., 2019), wherein EG-018 exhibited an E_max of 11.9% for mini-G and 7.8% for beta arrestin-2 relative to CP55,940’s maximal response in those assays. Interestingly, EG-018 exhibited 26-fold greater potency for mini-Gα_i over beta arrestin-2, whereas CP exhibited only a four-fold shift in potency between these assays, suggesting a bias for G protein signaling over beta arrestin. Similar low efficacy was observed for the 5-fluoro analog of EG-018, EG-2201, in the same study, although the apparent G-protein bias was not observed for this compound.

As expected based upon the [³⁵S]GTPγS results, EG-018 and both control compounds attenuated forskolin-stimulated cAMP production [Figure 3, bottom panel]. Relative to CP55,940, EG-018 exhibited greater efficacy in the forskolin-stimulated cAMP assay (73% of CP55,940 E_max at hCB₁; 48% at hCB₂) as compared to the [³⁵S]GTPγS studies (28% of CP55,940 E_max at hCB₁; 40% at hCB₂). This was also the case for THC in cAMP (91% of CP55,940 E_max at hCB₁; 45% at hCB₂) as compared to [³⁵S]GTPγS (48% of CP55,940 E_max at hCB₁; 40% at hCB₂). This would be expected since this assay is thought to have high amplification such that partial agonists can appear to be equi-efficacious to higher efficacy agonists (Finlay et al., 2017). Further, while EG-018 was more potent than THC in [³⁵S]GTPγS binding, it appeared less potent in the cAMP assay at both receptors. This would be consistent with a lower efficacy agonist that has to activate more receptors than a higher efficacy agonist to produce the same response, since the higher efficacy agonist can saturate the response by occupying a smaller fraction of the available receptors (Kenakin, 2002). The concentration response curve for the low efficacy agonist might then be right-shifted relative to a higher efficacy agonist when compared to an assay that has little or no receptor reserve, reflecting a requirement for greater fractional receptor occupancy. Since the [³⁵S]GTPγS binding assay reflects total Ga and does not discriminate between subtypes, it could be that a greater proportion of the Gα subtypes activated by THC regulate adenylyl cyclase than the subtypes activated by EG-018. For example, in N18TG2 cells, Gα_i3 was suggested to be the predominant subtype regulating adenylyl cyclase (Mukhopadhyay and Howlett, 2001) and cannabinoids can differentially activate Gα_i subtypes in a ligand-dependent manner in this cell line (Mukhopadhyay and Howlett, 2005).

Given previous reports that EG-018 was detected in products seized by drug enforcement agencies (Bijlsma et al., 2017; EMCDDA, 2015; Uchiyama et al., 2009; Worst and Sprague, 2015), EG-018 was also evaluated in THC discrimination and cannabinoid tetrad assays to determine the extent to which it would produce cannabimimetic effects despite its low in vitro efficacy. These effects are indicative of CB₁ agonism in the central nervous system, a necessary pre-requisite for psychotropic effects and abuse liability (Wiley, 1999). In particular, THC-like discriminative stimulus effects in animals are predictive of cannabis-like subjective effects, and they provide a high degree of pharmacological specificity (Balster and Prescott, 1992; Barrett et al., 1995; Wiley et al., 1995); i.e., drugs of abuse with different receptor mechanisms are unlikely to fully substitute for one another. Past studies have shown that SCRAs, based on several chemical templates (e.g., indoles, indazoles), that were abused by humans substituted for THC in drug discrimination procedures in rodents (Gamage et al., 2018; Gatch and Forster, 2014; Marusich et al., 2017; Wiley et al., 2014; Wiley et al., 2019; Wiley et al., 2015; Wiley et al., 2013). These reports are consistent with human reports of using SCRAs as a substitute for cannabis (Berry-Caban et al., 2012; Every-Palmer, 2011; Gunderson et al., 2012; Vandrey et al., 2012).

In the present study, EG-018 did not substitute for THC even at a high dose of 100 mg/kg. Further, this EG-018 dose did not substitute for the more structurally similar compound, JWH-018, in mice trained to discriminate 0.3 mg/kg JWH-018 from vehicle (data not shown). Due toEG-018’s solubility limitations, doses above 100 mg/kg were not tested in either discrimination procedure. In contrast, at doses consistent with its high affinity for the CB₁ receptor, JWH-018 has been shown to produce THC-like effect in drug discrimination in several labs (Brents et al., 2013; Marshell et al., 2014) via CB₁ receptor activation (Ginsburg et al., 2012; Wiley et al., 2014; Wiley et al., 2016), and other psychoactive SCRAs and/or their degradants have been shown to substitute for JWH-018 in JWH-018-trained rodents (Thomas et al., 2017; Wiley et al., 2016) and primates (McMahon, 2009). Consistent with these results, human use of EG-018 appears to be minimal compared to other SCRAs, despite sporadic reports of its detection in human samples (Gaunitz et al., 2018; Mogler et al., 2018) and seized products (Bijlsma et al., 2017; EMCDDA, 2015; Uchiyama et al., 2009; Worst and Sprague, 2015).

Because EG-018 did not show agonist-like activity in THC discrimination, it was evaluated using an antagonist paradigm, in which mice were preheated with EG-018 (100 mg/kg, i.p.) prior to administration of various doses of THC. Given that EG-018’s in vitro profile showed that it activated hCB₁ receptors with lower efficacy, but greater potency, than THC, a rightward shift in the THC dose-effect curve was predicted. To the contrary, however, results revealed that EG-018 also did not serve as an antagonist / partial agonist in the discrimination assay, as it did not shift THC’s dose-effect curve in either direction, and it did not decrease THC’s maximal substitution. These results suggest that EG-018’s lack of substitution when administered alone may not be due to efficacy but rather a lack of receptor occupancy in the brain. The lack of effects of EG-018 on its own or in combination with THC may partially be explained by pharmacokinetic factors. Metabolite analysis of urine collected from mice administered EG-018 i.p. did not yield any detected metabolites or the parent compound. While these results might suggest that EG-018 is not eliminated in urine in mice, structurally similar SCRAs and/or metabolites have been detected in mouse urine (Kusano et al., 2016; Öztürk et al., 2015) and metabolites of EG-018 have been detected in human urine samples (Mogler et al., 2018). Hence, it is unclear whether the lack of detection was due to excretory differences specific for EG-018 in mice or if another issue (e.g., solubility, drug precipitation, technical) precluded adequate absorption.

Indeed, when administered i.p. at doses up to 100 mg/kg, EG-018 also did not produce cannabimimetic effects in the cannabinoid tetrad, although it elicited modest hypothermia, hypermotility and catalepsy (but not antinociception) when administered i.v. at a dose of 56 mg/kg. A similar phenomenon was observed for the SCRA FUBIMINA, in that doses up to 100 mg/kg administered i.p. were largely without effect; however, with FUBIMINA, an i.v. dose 56 mg/kg produced marked tetrad effects in all four measures, all of which were reversed by pre-administration of 3 mg/kg i.v. rimonabant (Wiley et al., 2015). The ability of rimonabant to antagonize i.v. administered EG-018 was also examined. The same group of mice that were tested for agonism were given a one-week drug washout period prior to antagonism testing with rimonabant. While one week is typically long enough to ensure sufficient clearance of test drugs, the disposition of EG-018 was not determined in the present study and it is possible residual drug could have been present in the antagonism tests or that acute tolerance to the initial 100 mg/kg dose of EG-018 may have developed. When mice were pretreated with the CB₁ antagonist rimonabant (3 mg/kg), only the cataleptic effects of EG-018 were blocked, but not the hypothermic or locomotor suppressive effects. Considering the relatively high affinity (21 nM) and very low efficacy of EG-018 compared to the relatively low affinity (296.1 nM) and higher efficacy of FUBIMINA, it is possible that the rimonabant dose of 3 mg/kg was too low to sufficiently compete with the higher affinity of EG-018. The dose of rimonabant was approximately 18-fold lower than EG-018 and rimonabant has an affinity approximately 20-fold higher (Gamage et al., 2018). Depending upon their relative penetration into tissue compartments, the dose of rimonabant may simply need to be increased. The lack of antagonism by rimonabant on EG-018’s locomotor and hypothermic effects could be due to differences in receptor reserve in brain regions mediating these effects. However, a study that utilized CB₁ (+/− ) mice in order to determine receptor reserve for pharmacological effects of cannabinoids found that catalepsy, but not hypothermia, was mediated by areas with high receptor reserve (Grim et al., 2016). The current study employed ICR mice for tetrad testing whereas the CB₁ (+/−) mice were generated on a C57/BL6 background, so strain differences could be involved. Indeed, anandamide was reported to produce ataxia in C57BL/6 mice but not ICR mice (Chakrabarti et al., 1998); thus, mouse strain differences in the behavioral effects of cannabinoids have precedent.

EG-018 was eliminated quickly from rat plasma (half-life 2.5 h), although our microsomal data suggests that EG-018 may be cleared more quickly in rats than in humans. The T_max value of 2 h in rats suggests a relatively slow distribution compared to some other SCRAs such as CUMYL-PICA (Kevin et al., 2017a); however, the carboxamide analogues of JWH-018, NNEI and MN-18, exhibited divergent T_max values of approximately 0.5 and 2.5 hours, respectively, using the same dosing procedure, and these only differed by the presence of an additional nitrogen for MN-18 in the core (Kevin et al., 2017b). This might contribute to MN-18’s solubility and explain the shorter T_max. This would be consistent with enhanced lipophilicity afforded by the additional benzene ring conjugated to the indole-like moiety in EG-018, compared to indole/indazole based SCRAs (e.g. JWH-018). This could further explain the lack of behavioral effects following i.p. administration as enhanced lipophilicity could have resulted in poor absorption of EG-018 (Arnott and Planey, 2012).

In summary, EG-018 possesses interesting and peculiar pharmacology. The observed in vivo results are not those that would have been predicted by EG-018’s in vitro profile or by the results of previous in vivo studies with structurally similar molecules (e.g. JWH-018). EG-018 binds to both cannabinoid receptors with appreciable affinity, similar to that of THC. It also activates both receptors with reasonable potencies, albeit with efficacies that are similar to or even lower than the prototypic partial agonist THC. Yet, in contrast with THC, EG-018 was mostly devoid of behavioral effects except following high dose (56 mg/kg) i.v. administration where it produced modest effects in hypothermia, locomotor and catalepsy assays. The pharmacokinetics for EG-018 in mice have not been completely characterized; hence, lack of observed effects in the behavioral studies could be explained by relatively low brain levels of the compound, resulting from decreased absorption and/or distribution to the brain. A structurally similar cannabinoid agonist, CB-13, is peripherally restricted (Dziadulewicz et al., 2007) and lowers gastrointestinal motility at doses that do not produce catalepsy (Cluny et al., 2010). As such, further examination of the pharmacokinetic parameters of EG-018 in mice is warranted. While issues regarding the pharmacokinetics of EG-018 still need to be explored further to elucidate its in vivo profile, future in vitro studies of this compound and its analogs could yield promising insight into how ligands impart efficacy at CB₁ and CB₂ receptors, especially in light of the recent finding that EG-018 exhibits pronounced G-protein bias (Wouters et al., 2019). Ironically, pursuit of this goal would represent a return to one of the original purposes of synthetic cannabinoid synthesis: to investigate the functioning of the endocannabinoid system.

Table 2.

Pharmacological parameters at human cannabinoid receptors in HEK293 membrane preparations on [³⁵S]GTPγS binding. Values reflect mean ± standard error of at least n=3 experiments performed in duplicate.

	CP55,940	EG-018	THC	AEA
hCB1
pEC50±SEM	9.38±0.0375	7.43±0.137****^	7.05±0.015****††	6.48±0.0229****#
95% C.I.	(9.26 – 9.5)	(7 – 7.87)	(6.98 – 7.11)	(6.41 – 6.56)
EC50 (nM)	0.413	37.1	89.9	328
95% C.I.	(0.316-0.55)	(13.5-100)	(77.6-105)	(275-389)
Emax±SEM	265±19.3^^^^	73.9±9.12%*****	128±3.82%#†††	234±7.86%####
95% C.I.	(204 – 327)	(44.9 – 103)	(116 – 140)	(209 – 259)
hCB2
pEC50±SEM	9.08±0.154	7.95±0.215*	7.69±0.295*	6.8±0.375***
95% C.I.	(8.59 – 9.57)	(7.02 – 8.87)	(6.43 – 8.96)	(5.18 – 8.41)
EC50 (nM)	0.839	11.3 nM	20.3	160
95% C.I.	(0.269-2.57)	(1.35-95.5)	(1.1-372)	(3.89-6610)
Emax±SEM	116±19.3%	46.2±2.72%*	46.5±7.94%*	49.2±2.25%*
95% C.I.	(55 – 178)	(34.5 – 57.9)	(12.3 – 80.6)	(39.6 – 58.9)

The number of symbols indicates the increasing level of significance, i.e. 1 = p < 0.05, 2 = p < 0.01, 3 = p < 0.001, 4 = p < 0.0001;

^*vs. CP55,940,

^#vs. EG-018,

^{^}vs. THC,

^†vs. AEA

Table 3.

Pharmacological parameters at human cannabinoid receptors expressed in intact HEK293 cells on forskolin-stimulated cAMP production Values reflect mean ± standard error of at least n=3 experiments performed in duplicate.

	CP55,940	EG-018	THC
hCB1
pEC50±SEM	9.24±0.149	7.51±0.0603***	8.29±0.351*
95% C.I.	(8.82 – 9.65)	(7.25 – 7.77)	(6.78 – 9.8)
EC50 (nM)	0.581	30.9	5.15
95% C.I.	(0.224-1.51)	(17-56.2)	(0.158-166)
Emax±SEM	45.4±1.73%	32.9±2.27%**	41.5±2.3%
95% C.I.	(40.6 – 50.2)	(23.1 – 42.6)	(31.6 – 51.4)
hCB2
pEC50±SEM	8.8±0.0608	6.74±0.292***	7.39±0.144**
95% C.I.	(8.54 – 9.06)	(5.49 – 8)	(6.77 – 8.01)
EC50 (nM)	1.58	182	40.8
95% C.I.	(0.871-2.88)	(10-3240)	(9.77-170)
Emax±SEM	58.8±2.48%	28.1±6.14%**	27.3±3.68%**
95% C.I.	(48.2 – 69.5)	(1.72 – 54.5)	(11.4 – 43.1)

^*p < 0.05,

^**p < 0.01,

^***p < 0.001 vs. CP55,940

Highlights.

• EG-018 binds to human cannabinoid type-1 and type-2 receptors with high affinity

• EG-018 acts as a very weak partial agonist of both cannabinoid receptors in vitro

• Intraperitoneal EG-018 treatment did not produce cannabinoid-like effects in mice

• Intravenous EG-018 treatment caused hypomotility, catalepsy, and hypothermia in mice

• Further study of EG-018 analogues might reveal insight into cannabinoid efficacy

Abbreviations:

SCRA

synthetic cannabinoid receptor agonist

hCB₁

human cannabinoid type-1 receptor

hCB₂

human cannabinoid type-2 receptor

DMSO

dimethyl sulfoxide

BSA

bovine serum albumin

HEK293

human embryonic kidney-293 cells

PBS

phosphate buffered saline