Evaluation of carboxamide-type synthetic cannabinoids as CB1/CB2 receptor agonists: difference between the enantiomers

Recently, carboxamide-type synthetic cannabinoids have been distributed globally as new psychoactive substances (NPS). Some of these compounds possess asymmetric carbon, which is derived from an amide moiety composed of amino acid derivatives (i.e., amides or esters of amino acids). In this study, we synthesized both enantiomers of synthetic cannabinoids, N-(1-amino-3-methyl-1-oxobutan-2-yl)-1-(2-fluorobenzyl)-1H-indazole-3-carboxamide (AB-FUBINACA 2-fluorobenzyl isomer), N-(1-amino-1-oxo-3-phenylpropan-2-yl)-1-(cyclohexylmethyl)-1H-indazole-3-carboxamide (APP-CHMINACA), ethyl [1-(5-fluoropentyl)-1H-indazole-3-carbonyl]valinate (5F-EMB-PINACA), ethyl [1-(4-fluorobenzyl)-1H-indazole-3-carbonyl]valinate (EMB-FUBINACA), and methyl 2-[1-(4-fluorobenzyl)-1H-indole-3-carboxamido]-3,3-dimethylbutanoate (MDMB-FUBICA), which were reported as NPS found in Europe from 2014 to 2015, to evaluate their activities as CB1/CB2 receptor agonists. With the exception of (R) MDMB-FUBICA, all of the tested enantiomers were assumed to be agonists of both CB1 and CB2 receptors, and the EC50 values of the (S)-enantiomers for the CB1 receptors were about five times lower than those of (R)-enantiomers. (R) MDMB-FUBICA was shown to function as an agonist of the CB2 receptor, but lacks CB1 receptor activity. To the best of our knowledge, this is the first report to show that the (R)-enantiomers of the carboxamide-type synthetic cannabinoids have the potency to activate CB1 and CB2 receptors. The findings presented here shed light on the pharmacological properties of these carboxamide-type synthetic cannabinoids in forensic cases.


Introduction
In recent years, several new psychoactive substances (NPS) have been newly identified from illicit drug products, one after another, and have been abused as legal alternatives to scheduled drugs [1]. One of the most popular classes of NPS is synthetic cannabinoids, which are recreationally used as substitutes for Cannabis sativa. Forensic cases of synthetic cannabinoid use have increased recently; the adverse effects of indazole-carboxamide-type synthetic cannabinoids, especially, present severe social problems [2].
In a previous study, we developed a method to separate enantiomers of synthetic cannabinoids by liquid chromatography-mass spectrometry (LC-MS) [3]. In the case of NPS classified into cathinone and phenethylamine groups, several reports have concluded that (S)-enantiomers are more potent than (R)-enantiomers [4][5][6][7]. Among the carboxamide-type synthetic cannabinoids, some compounds possess asymmetric carbon, which is derived from an amide moiety composed of amino acid derivatives (i.e., amides or esters of amino acids). These classes of compounds were first developed by Pfizer as potential therapeutic drugs; however, their patent only includes (S)-enantiomers, even though the structures of these compounds have chiral centers [8]. Some previous studies reported the cannabimimetic activities of the chiral synthetic cannabinoids, but they all investigated only the (S)-enantiomers, and thus the pharmaceutical activities of the (R)-enantiomers remain unknown [9].

Nuclear magnetic resonance spectroscopy
Nuclear magnetic resonance (NMR) spectra were recorded using an ECZS-400 spectrometer (JEOL Resonance, Tokyo, Japan) with dimethyl sulfoxide (DMSO)-d 6 as the solvent. The chemical shifts d were recorded in ppm relative to tetramethylsilane ( 1 H: d = 0 ppm, 13 C: d = 0 ppm) or the solvent ( 13 C: d = 39.5 ppm) as an internal standard. The compounds were assigned by 1 H NMR, 13 C NMR, distortionless enhancement by polarization transfer, 1 H-13 C heteronuclear multiple quantum coherence, 1 H-13 C heteronuclear multiple-bond correlation, and 1 H-1 H correlation spectroscopy.

Chiral chromatography
Chirality of the compounds was confirmed by LC-MS as described previously [3]. A CHIRALPAK AZ-3R column (3.0 lm particle size, 150 9 2.1 mm i.d.) (Daicel Corporation, Osaka, Japan) was used, and the mobile phase was composed of H 2 O/acetonitrile (55:45, v/v) under isocratic conditions. The flow rate of the mobile phase was 0.3 mL/ min, and the injection volume was 1 lL. The column temperature was 40°C. The stock standard solutions were prepared as described previously [3]. When necessary, the sample solutions and the stock standard solutions were diluted by the mobile phase for the liquid chromatographyhigh-resolution-mass spectrometry (LC-HR-MS) analysis [3].

Chemical synthesis
The synthetic pathway of these enantiomers is shown in Fig. 1. All of the target compounds were made from methyl 1H-indazole-3-carboxylate (compound 1) or methyl 1H-indole-3-carboxylate (compound 2). The enantiomers of the target compounds were synthesized by a slightly modified version of a previously described method [3,10]. Compound 1 or 2 was N-alkylated by alkyl halide under basic conditions and yielded compounds that were deprotected by hydrolysis of the methyl group. The target compounds were synthesized by chlorination of the carboxylic acids by oxalyl chloride and then amidation with the amino acid derivatives. The chemical structures of the compounds dealt with in this study are shown in Fig. 2.

N-Alkylation of methyl 1H-indazole/indole-3-carboxylate
Compound 5 was synthesized as described previously [3]. Compounds 3, 4, 6, and 7 were prepared in the same manner, except for the starting compounds. To a solution of compound 1 or 2 in tetrahydrofuran (THF), potassium tertbutoxide (t-BuOK) (1.2 equivalent) was added. To the flask containing the solution, alkyl halide (1.5 equivalent) was added and stirred at room temperature for more than 12 h. Ethyl acetate and distilled water were added to the solution, and the organic layer was separated from the aqueous layer. The organic layer was then washed with saturated aqueous NaCl and dried with anhydrous MgSO 4 . The solution was filtered, and the solvent was removed under a vacuum. The residue was purified by an Isolela Speckt System (Biotage, Tokyo, Japan) on a SNAP Ultra column (Biotage) with ethyl acetate in n-hexane as the mobile phase. The solvent of the obtained fraction was removed under a vacuum to yield compounds 3, 4, 5, 6 or 7.

Hydrolysis of methyl N-alkyl-1H-indazole-3-carboxylate
Synthesis of compound 10 was performed as described previously [3]. Compounds 8, 9, and 11 were synthesized similarly, except for the starting compounds. To a flask containing compounds 3, 4, or 6, THF, ethanol, and 1 M NaOH (aq) (approximately 2.5 mL/1 mmol of the material compounds) were added and stirred at room temperature for more than 16 h. After the reaction, the organic solvents were removed under a vacuum. Ethyl acetate and 20% Na 2 CO 3 in H 2 O were added to the solution and stirred, and the solution was allowed to separate. The organic layer was extracted by 20% Na 2 CO 3 in H 2 O again, and the aqueous layers were combined. The organic layer was extracted twice with 1 M NaOH in H 2 O. The aqueous layers were combined and neutralized with concentrated HCl, extracted with ethyl acetate twice, and then the organic layer was washed twice with brine. Anhydrous MgSO 4 was added to the organic layer and removed under a vacuum to yield compounds 8, 9, or 11.

Hydrolysis of methyl N-alkyl-1H-indole-3-carboxylate
Compound 12 was synthesized by the hydrolysis of compound 7. To a flask containing compound 7, THF, ethanol, methanol, and 10 M NaOH (aq) (approximately 2 mL/ 1 mmol of compound 7) were added and stirred at room temperature for 2.5 h. After the reaction, the organic solvents were removed under a vacuum. Ethyl acetate was added to the solution and mixed well, and the solution was allowed to separate. The organic layer was extracted twice with 1 M NaOH (aq), and the aqueous layers were then combined. The aqueous layer was neutralized with HCl and extracted with ethyl acetate twice. The organic layer was then washed twice with brine. Subsequently, anhydrous MgSO 4 was added to the combined organic layer and removed under a vacuum to yield compound 12.
To a flask containing compound 8 in dichloromethane (20 mL/1 mmol of compound 8), N,N-dimethylformamide (20 lL/1 mmol) and oxalyl chloride (2.5 equivalent) were added and stirred at room temperature for 30 min. The solvent was then removed under a vacuum to obtain 1-(2fluorobenzyl)-1H-indazole-3-carbonyl chloride as a hazy oily residue. The residue was dissolved in dichloromethane (10 mL), and L-valine amide hydrochloride (1.1 equivalent) was added. Triethylamine (1 mmol) was added to the mixture and then the resulting solution was stirred at room temperature for 90 min. The solvent was removed under reduced pressure and dissolved in ethyl acetate. The solvent was separated with water and washed with 0.1 M HCl, saturated NaHCO 3 , and saturated NaCl. Drying with MgSO 4 and removal of the solvent afforded compound 13S as a fibrous white solid. 1 Subjecting compound 8 to a procedure similar to that for 13S, but substituting D-valine amide hydrochloride for Lvaline amide, yielded 13R as a fibrous white solid. 1 R 2 NH 2 = L-valine amide R 2 NH 2 = D-valine amide R 2 NH 2 = L-valine ethyl ester R 2 NH 2 = D-valine ethyl ester R 2 NH 2 = L-valine ethyl ester R 2 NH 2 = D-valine ethyl ester R 2 NH 2 = L-phenylalanine amide R 2 NH 2 = D-phenylalanine amide R 2 NH 2 = L-tert-leucine methyl ester R 2 NH 2 = D-tert-leucine methyl ester

Ethyl [1-(5-fluoropentyl)-1H-indazole-3-carbonyl]-Dvalinate (15R)
Subjecting compound 10 to a procedure similar to that for 13S, but substituting D-valine ethyl ester hydrochloride for L-valine amide, yielded 15R as a clear oily residue. 1  Subjecting compound 11 to a procedure similar to that for 16S, but substituting D-phenylalanine amide hydrochloride for L-phenylalanine amide hydrochloride, yielded 16R as a crystalline white solid. 1  Subjecting compound 12 to a procedure similar to that for 17S, but substituting D-tert-leucine methyl ester hydrochloride for L-tert-leucine methyl ester hydrochloride, yielded 17R as a white powder. 1  In vitro assays to evaluate the CB 1 /CB 2 receptor activities For the evaluation of CB 1 /CB 2 cannabinoid receptor activity, [ 35 S]GTPcS binding assays were performed. These assays were performed at ADME and Tox. Research Institute, Sekisui Medical Co., Ltd. (Tokai-mura, Ibaraki, Japan). The assay conditions were as described previously [11], except for the tested concentration levels of the compounds ranging from 1 9 10 -11 to 1 9 10 -4 M. Agonistic activities (EC 50 value: concentration showing 50% response) of the test compound to the cannabinoid receptors CB 1 and CB 2 were measured.

Results and discussion
In this study, we synthesized both (S)-and (R)-enantiomers of synthetic cannabinoids to evaluate their pharmacological properties. All target compounds were reported in an EMCDDA-Europol joint report to be NPS sold and recreationally used in Europe [12,13]. NMR data for the enantiomers were similar, as shown in the section on materials and methods above, and the results are in good agreement with the values expected for their structures. Figure 3 shows the extracted ion chromatograms for the enantiomers of the target compounds (1 lg/mL solution for both enantiomers) obtained by LC-HR-MS using the method that we developed previously [3]. By the previously developed method, all of the enantiomers of the tested compounds were separated, and no enantio-impurities were found in their chromatograms, except for (S) APP-CHMINACA. Approximately less than 2% of (R) APP-CHMINACA was confirmed in the extracted ion chromatogram of (S) APP-CHMINACA as shown in Fig. 3. However, because (R)-enantiomer was much less effective than (S)-enantiomer to both CB 1 and CB 2 receptors, it is suspected that the effect of contaminated (R)-enantiomer on the EC 50 value was almost negligible. Table 1 shows the results of the [ 35 S]GTPcS binding assays performed in this study. Each compound functioned as both a CB 1 and a CB 2 receptor agonist, except for (R) MDMB-FUBICA. The EC 50 values of (S)-enantiomers for the CB 1 receptors were at least one-fifth of or less than those of (R)-enantiomers, which indicates that (S)-enantiomers are more potent CB 1 receptor agonists than (R)enantiomers. The (S)-enantiomer of APP-CHMINACA could activate CB 1 at a concentration 134 times lower than that for the (R)-enantiomer. For MDMB-FUBICA, the CB 1 receptor activity was not confirmed for the (R)-enantiomer even at the maximum tested concentration, but the EC 50 of the (S)-enantiomer was 9.72 9 10 -9 M; the difference in the EC 50 levels between the enantiomers of MDMB-FUBICA to activate the CB 1 receptor was more than 10,000-fold, which indicates that the activities for the CB 1 receptor are markedly different depending on the chirality of the compounds. In contrast, we could not observe a clear relevance between EC 50 values for the CB 2 receptor and chirality of the compounds. (R) APP-CHMINACA and (R) MDMB-FUBICA lacked the potential to be CB 1 receptor agonists, but both of their EC 50 values for the CB 2 receptor were less than 10 -6 M; all of the target compounds were strong agonists for the CB 2 receptor, irrespective of their chirality.
As for the CB 1 /CB 2 selectivity, (R)-enantiomers tended to be CB 2 -selective and were at least 20-fold more active than for CB 1 for all of the tested compounds. Interestingly, (R) MDMB-FUBICA functioned only on the CB 2 receptor as an agonist, while showing no CB 1 activity as mentioned above, although (S) MDMB-FUBICA could activate both of these receptors at nanomolar levels. In several preclinical studies, compounds such as JWH-133 and HU308 were used as CB 2 -selective ligands [14]. More recently, the synthesis and evaluation of novel CB 2 -selective agonists have been repeatedly reported. In these pharmacological or medicinal chemistry studies seeking CB 2 agonists, the most selective compound showed a selective index of CB 1 /CB 2 of only 100-to 1000-fold. In this study, (R) MDMB-FUBICA was shown to have potency to activate the CB 2 receptor signal with an EC 50 value of 3.1 9 10 -9 M, but had no [ 35 S]GTPcS binding activity for the CB 1 receptor, even at 1.0 9 10 -4 M, which corresponds to a selective index of more than 3.2 9 10 4 -fold.
The metabolism of MDMB-FUBICA enantiomers has not been reported yet; thus it remains unclear whether activities of each enantiomer presented in this study may be helpful to precisely recognize forensic cases associated with these synthetic cannabinoids when evaluated with the quantitative data of a specimen analyzed after enantiomeric separation.

Conclusions
The  EC 50 (M) S/R ratio a EC 50 (M) S/R ratio a MDMB-FUBICA) were evaluated by the [ 35 S]GTPcS binding assay. We also confirmed that these enantiomers could be clearly differentiated by chiral-LC-MS. These findings should contribute to better understanding of the forensic cases associated with these synthetic cannabinoids.