Synthesis and Preliminary Biological Evaluation of Indol-3-yl-oxoacetamides as Potent Cannabinoid Receptor Type 2 Ligands

A small series of indol-3-yl-oxoacetamides was synthesized starting from the literature known N-(adamantan-1-yl)-2-(5-(furan-2-yl)-1-pentyl-1H-indol-3-yl)-2-oxoacetamide (5) by substituting the 1-pentyl-1H-indole subunit. Our preliminary biological evaluation showed that the fluorinated derivative 8 is a potent and selective CB2 ligand with Ki = 6.2 nM.


Introduction
For centuries, Cannabis sativa was used as medicinal plant for the treatment of pain and inflammatory processes.The discovery by Gaoni and Mechoulam [1] of ∆ 9 -tetrahydrocannabinol (∆ 9 -THC)-the primary psychoactive ingredient in Cannabis sativa-set the stage for the identification of the endogenous cannabinoid (endocannabinoid) transmitter system in the brain.The endocannabinoid system consists of cannabinoid (CB) receptors, endogenous ligands (e.g., anandamide, 2-O-arachidonoylglycerol) that activate the CB receptors, and the enzymes, which are responsible for the biosynthesis (e.g., N-acyltransferase, diacylglycerol lipase) and deactivation (e.g., fatty acid amide hydrolases, monoacylglycerol lipases) of the endogenous ligands [2,3].
Two types of specific G i/o -protein-coupled CB receptors were cloned in the 1990s, termed CB 1 and CB 2 receptor [4,5].CB 1 receptors are abundantly expressed in the central nervous system, and their function has been thoroughly investigated [6].Recently, the crystal structure of the CB 1 receptors has been reported, providing a tool for a more accurate pharmacological investigation of this receptor subtype [7,8].In contrast to CB 1 receptors, the CB 2 receptor was originally regarded as a peripheral receptor predominantly expressed in cells of the immune system [5,9,10].However, more recent investigations proved the presence of CB 2 receptors in glial cells (microglia and astrocytes) [11,12], and oligodendroglial [13] progenitors in vitro, as well as in microglia [14] and neuronal progenitors [12] in normal mouse brain in vivo [15].
The predominant expression of the CB 2 receptor in cells of the immune system suggests a modulation of diverse immune functions, including cytokine production, lymphocyte proliferation, and humoral and cell-mediated immune responses [16,17].With respect to neuroinflammation, microglia adopt a key function.Under healthy conditions, the expression of CB 2 receptors in the brain is rather low [18].However, inflammatory effects result in considerably increased expression levels of CB 2 receptors [19,20].It was found that neurodegenerative and neuroinflammatory processes in the brain related to, for example, depression, Alzheimer's disease, multiple sclerosis, amyotrophic lateral sclerosis, or brain tumors such as glioblastoma are associated with an upregulation of CB 2 receptor expression [18,[21][22][23].CB 2 receptor agonists lead to a reduction of neuroinflammation and stimulation of neurogenesis.Therefore, the therapeutic potential of CB 2 agonists is related to neurodegenerative and neuroinflammatory processes [19,24].Moreover, the availability of CB 2 receptors as measured with positron emission tomography (PET) [25] could potentially be used as a biomarker for neurodegenerative and neuroinflammatory processes in the brain [26][27][28][29].
Indole has been a very popular key building block for the medicinal chemistry of cannabinoid receptor ligands, and a large number of indole-derived CB 2 selective ligands have been developed [30][31][32][33].However, most of these ligands do not comply with the most important needs of a ligand suitable for the development of a tracer for CB 2 brain imaging with PET; namely, high affinity towards CB 2 (K i (CB 2 ) < 1 nM) and high selectivity over CB 1 (K i (CB 2 )/(CB 1 ) > 500) [26,34].

Results and Discussion
The structure-affinity relationship study which lead to the identification of compound 5 [32,36] (Scheme 1) and the work performed on the structurally related quinolinone-3-carboxamides as CB 2 ligands [37] has proven the importance of the substitution pattern at the 5-indole position, and also the need of the lipophilic adamantane subunit.Herein, we investigate the possibility of introducing a fluorine atom by modifying the alkyl chain at the indole 1-position.
Indole has been a very popular key building block for the medicinal chemistry of cannabinoid receptor ligands, and a large number of indole-derived CB2 selective ligands have been developed [30][31][32][33].However, most of these ligands do not comply with the most important needs of a ligand suitable for the development of a tracer for CB2 brain imaging with PET; namely, high affinity towards CB2 (Ki(CB2) < 1 nM) and high selectivity over CB1 (Ki(CB2)/(CB1) > 500) [26,34].Recently, we reported the development of a highly affine and selective 18 F-labeled CB2 radiotracer (Ki(CB2) = 0.4 nM, Ki(CB1) = 380 nM), and proved its applicability in a mouse model of neuroinflammation [35].However, this radioligand suffers from low metabolic stability in vivo, and therefore we redirected our focus on the structure of the literature known, highly affine 5, Ki(CB2) = 0.37 nM and Ki(CB1) = 345 nM) [32] for the development of a CB2 radiotracer.In the present work, we show the synthesis and preliminary biological evaluation of fluorine-containing indol-3-yl-oxoacetamide derivatives.

Results and Discussion
The structure-affinity relationship study which lead to the identification of compound 5 [32,36] (Scheme 1) and the work performed on the structurally related quinolinone-3-carboxamides as CB2 ligands [37] has proven the importance of the substitution pattern at the 5-indole position, and also the need of the lipophilic adamantane subunit.Herein, we investigate the possibility of introducing a fluorine atom by modifying the alkyl chain at the indole 1-position.The synthesis of the key building block 3 was performed as described in the literature [32] and depicted in Scheme 1. Treatment of 5-bromoindole (1) with oxalyl chloride delivered the corresponding indole-3-oxoacetyl chloride in ~60% yield, which was further reacted with amantadine (adamantan-1-amine) in presence of triethylamine (Et 3 N) to give compound 2. Next, Suzuki coupling was performed using the commercially available 2-furanboronic acid in presence of Pd(PPh 3 ) 4 and Na 2 CO 3 to give 3.The lead molecule 5 [32] was synthesized by treating 3 with 1-bromobutane in basic reaction condition in ~25% yield over four steps starting from 1, as a light yellow solid.To overcome the high lipophilicity of the indole-N-alkyl chain [38], ether groups were introduced at this site of the molecule, and simultaneously, the influence of its length on the CB 2 binding affinity was investigated.Thus, the glycol ether compound 6 was synthesized from 3 by using 2-(2-(2-fluoroethoxy)ethoxy)ethyl-4-methylbenzenesulfonate [39,40] as alkylating reagent in 72% yield.Similarly, alcohol 4 was synthesized by reacting compound 3 with 2-bromethanol.Alcohol 4 was further etherified via Williamson ether synthesis [41] by using methyl iodide and 2-fluoro-1-bromoethane to give compounds 7 and 8, respectively (for 1 H-NMR of compounds 3, 5, 6, 7 and 8 see Supplementary Materials).
The binding affinity towards CB 2 receptors was determined in vitro by radioligand inhibition binding assays according to a recently published protocol [42,43] using cell membranes from CHO cells stably transfected with the human CB 2 (Prof Paul L. Prather, University Arkansas for Medical Sciences, Little Rock, AR, USA), [ 3 H]WIN55.212-2as competitive radioligand (K D = 2.1 nM) and increasing concentrations (100 pM to 10 µM) of compounds 5, 6, 7, and 8 added in triplicate for each experiment.Non-specific binding was determined using 10 µM WIN55.212-2.The individual IC 50 values were determined by non-linear curve fitting using GraphPad Prism software (version 3.0, GraphPhad, San Diego, CA, USA), and the corresponding K i values calculated using the Cheng-Prusoff equation [44].As shown by the K i values in Figure 1b, we could not reproduce the reported sub-nanomolar CB 2 affinity of the lead molecule (compound 5) [32], but recorded a K i value of 8.5 nM instead.In addition, these values and the individual binding curves in Figure 1a illustrate that a similar low-nanomolar affinity was observed for the fluorinated derivative 8 (K i = 6.2 nM), while slightly lower affinities were determined for compounds 6 and 7 (K i = 27.3 nM, and K i = 16.1 nM, respectively).Additionally, the binding affinities towards the CB 1 receptor were investigated for compounds 6 and 8 by using [ 3 H]CP55.40as competitive radioligand.None of the two derivatives could displace [ 3 H]CP55.40from the human CB 1 receptors up to a concentration of 100 µM.Thus, within this series of fluoro-substituted compounds, the N-alkyl chain of compound 8 is most suitable to obtain high affinity binding at the CB 2 receptor, combined with an excellent selectivity against the CB 1 receptor subtype.
Molecules 2017, 22, 77 3 of 8 The synthesis of the key building block 3 was performed as described in the literature [32] and depicted in Scheme 1. Treatment of 5-bromoindole (1) with oxalyl chloride delivered the corresponding indole-3-oxoacetyl chloride in ~60% yield, which was further reacted with amantadine (adamantan-1-amine) in presence of triethylamine (Et3N) to give compound 2. Next, Suzuki coupling was performed using the commercially available 2-furanboronic acid in presence of Pd(PPh3)4 and Na2CO3 to give 3.The lead molecule 5 [32] was synthesized by treating 3 with 1-bromobutane in basic reaction condition in ~25% yield over four steps starting from 1, as a light yellow solid.To overcome the high lipophilicity of the indole-N-alkyl chain [38], ether groups were introduced at this site of the molecule, and simultaneously, the influence of its length on the CB2 binding affinity was investigated.Thus, the glycol ether compound 6 was synthesized from 3 by using 2-(2-(2-fluoroethoxy)ethoxy)ethyl-4methylbenzenesulfonate [39,40] as alkylating reagent in 72% yield.Similarly, alcohol 4 was synthesized by reacting compound 3 with 2-bromethanol.Alcohol 4 was further etherified via Williamson ether synthesis [41] by using methyl iodide and 2-fluoro-1-bromoethane to give compounds 7 and 8, respectively (for 1 H-NMR of compounds 3, 5, 6, 7 and 8 see Supplementary Materials).
The binding affinity towards CB2 receptors was determined in vitro by radioligand inhibition binding assays according to a recently published protocol [42,43] using cell membranes from CHO cells stably transfected with the human CB2 (Prof Paul L. Prather, University Arkansas for Medical Sciences, Little Rock, AR, USA), [ 3 H]WIN55.212-2as competitive radioligand (KD = 2.1 nM) and increasing concentrations (100 pM to 10 µM) of compounds 5, 6, 7, and 8 added in triplicate for each experiment.Non-specific binding was determined using 10 µM WIN55.212-2.The individual IC50 values were determined by non-linear curve fitting using GraphPad Prism software (version 3.0, GraphPhad, San Diego, CA, USA), and the corresponding Ki values calculated using the Cheng-Prusoff equation [44].As shown by the Ki values in Figure 1b, we could not reproduce the reported sub-nanomolar CB2 affinity of the lead molecule (compound 5) [32], but recorded a Ki value of 8.5 nM instead.In addition, these values and the individual binding curves in Figure 1a illustrate that a similar low-nanomolar affinity was observed for the fluorinated derivative 8 (Ki = 6.2 nM), while slightly lower affinities were determined for compounds 6 and 7 (Ki = 27.3 nM, and Ki = 16.1 nM, respectively).Additionally, the binding affinities towards the CB1 receptor were investigated for compounds 6 and 8 by using [ 3 H]CP55.40as competitive radioligand.None of the two derivatives could displace [ 3 H]CP55.40from the human CB1 receptors up to a concentration of 100 µM.Thus, within this series of fluoro-substituted compounds, the N-alkyl chain of compound 8 is most suitable to obtain high affinity binding at the CB2 receptor, combined with an excellent selectivity against the CB1 receptor subtype.

General Methods
All reagents were used directly as obtained commercially, unless otherwise noted.Reaction progress was monitored by thin-layer chromatography (TLC) using silica gel 60 F254 (0.040-0.063 mm) with detection by UV.All moisture-sensitive reactions were performed under an argon atmosphere using oven-dried glassware and anhydrous solvents.Column flash chromatography was carried out using E. Merck silica gel 60F (230-400 mesh) (Merck Millipore, Darmstadt, Germany).Analytical TLC was performed on aluminum sheets coated with silica gel 60 F254 (0.25 mm thickness, E. Merck, Darmstadt, Germany). 1 H-NMR spectra were recorded with a Bruker-400 NMR spectrometer (Bruker, Billerica, MA, USA) at nominal resonance frequencies of 400 MHz, in CDCl 3 or DMSO-d 6 (referenced to internal Me 4 Si at δH 0 ppm).The chemical shifts (δ) were expressed in parts per million (ppm).High resolution mass spectra were recorded utilizing electrospray ionization (ESI) at the University of Notre Dame Mass Spectrometry facility.

Conclusions
In this study, the N-alkyl chain of the high affinity and selective CB 2 ligand 5 was modified for the possibility of introducing a fluorine atom.The herein developed fluorinated compound 8 retained the high affinity of the lead compound, and will be considered for the development of an 18

Figure 1 .
Figure 1.(a) Individual competition binding curves of compounds 5, 6, 7, and 8. Inhibition of [ 3 H]WIN55.212-2binding by increasing concentrations (100 pM to 10 µM) of compounds 5, 6, 7, and 8 to membrane homogenates of CHO cells stably transfected with human CB2.Each value represents the mean ± standard deviation of a triplicate in a single experiment; (b) Binding affinity (Ki) of compounds 5, 6, 7, and 8 for the human cannabinoid receptor type2 (CB2) receptor.a Values are means ± standard deviations of two to three experiments run in triplicate; b Ki of compound 5 as reported in [23].

Figure 1 .
Figure 1.(a) Individual competition binding curves of compounds 5, 6, 7, and 8. Inhibition of [ 3 H]WIN55.212-2binding by increasing concentrations (100 pM to 10 µM) of compounds 5, 6, 7, and 8 to membrane homogenates of CHO cells stably transfected with human CB 2 .Each value represents the mean ± standard deviation of a triplicate in a single experiment; (b) Binding affinity (K i ) of compounds 5, 6, 7, and 8 for the human cannabinoid receptor type 2 (CB 2 ) receptor.a Values are means ± standard deviations of two to three experiments run in triplicate; b K i of compound 5 as reported in [23].