Chain Substituted Cannabilactones with Selectivity for the CB2 Cannabinoid Receptor

In earlier work, we reported a novel class of CB2 selective ligands namely cannabilactones. These compounds carry a dimethylheptyl substituent at C3, which is typical for synthetic cannabinoids. In the current study with the focus on the pharmacophoric side chain at C3 we explored the effect of replacing the C1′-gem-dimethyl group with the bulkier cyclopentyl ring, and, we also probed the chain’s length and terminal carbon substitution with bromo or cyano groups. One of the analogs synthesized namely 6-[1-(1,9-dihydroxy-6-oxo-6H-benzo[c]chromen-3-yl) cyclopentyl] hexanenitrile (AM4346) has very high affinity (Ki = 4.9 nM) for the mouse CB2 receptor (mCB2) and 131-fold selectivity for that target over the rat CB1 (rCB1). The species difference in the affinities of AM4346 between the mouse (m) and the human (h) CB2 receptors is reduced when compared to our first-generation cannabilactones. In the cyclase assay, our lead compound was found to be a highly potent and efficacious hCB2 receptor agonist (EC50 = 3.7 ± 1.5 nM, E(max) = 89%). We have also extended our structure-activity relationship (SAR) studies to include biphenyl synthetic intermediates that mimic the structure of the phytocannabinoid cannabinodiol.


Introduction
The two G-protein coupled receptors (GPCRs) termed cannabinoid receptor 1 (CB1) and cannabinoid receptor 2 (CB2) are principal components of the endocannabinoid biosignaling system and molecular targets for the psychoactive constituent of cannabis ∆ 9 -tetrahydrocannabinol (∆ 9 -THC) [1][2][3]. Both CB1 and CB2 also bind the FDA approved drug Nabilone and many other exogenous (plant-derived and synthetic) cannabinoids. The CB1 receptor retains a high (≥97%) degree of amino acid sequence identity across mouse, rat, and human [4]. In contrast, the human CB2 receptor displays only 82% and 81% amino acid identity with mouse [5] and rat [6] respectively. The CB1 receptor was found to be localized primarily in the brain and it is one of the most abundant GPCRs in the central nervous system (CNS) [4]. This receptor is also found in peripheral tissues and organs [7]. Activation of the central CB1 receptors mediates most of the cannabinoid psychotropic and behavioral effects. Conversely, the CB2 receptors are detectable at very low levels in brain and they are expressed predominantly in immune cells and in the periphery [8]. Nevertheless, the CB2 receptors may be induced in the CNS under pathologic conditions. Collaborative efforts, including our laboratory, have led to the first crystal structures of the agonist and antagonist bound human CB1, while recently we published the first crystal structure of the human CB2 receptor in a complex with the antagonist AM10257. This breakthrough work provides a molecular basis for predicting the binding modes of cannabinoids with their target proteins, and offers invaluable information for target-based drug design [9][10][11][12].
Selective CB2 receptor activation is a very promising method of modulating the endocannabinoid system because the pharmacologic effects are devoid of centrally mediated liabilities that are correlated with CB1 receptor modulation either directly by exogenous cannabinoids or indirectly through functional interactions with other receptor systems [13]. Selective CB2 activation would decrease CB1 associated ataxia, hypothermia, mood and memory disturbances, and abuse potential [14][15][16][17][18][19]. Currently, the CB2 receptor has emerged as an attractive therapeutic target for the treatment of inflammatory and neuropathic pain, as well as neurodegenerative disorders [20][21][22][23], and cancer [24][25][26].
Our group has identified the cannabilactones (Figure 1), as a new class of CB2 agonists with structural similarities to the 6,6-dimethyl counterpart and phytocannabinoid cannabinol ( Figure 2). Two first generation analogs within this class of compounds namely AM1714 and AM1710 ( Figure 1) have become important pharmacological tools in establishing proof of concept for the usefulness of the CB2 receptor activation approach [27][28][29][30][31][32][33]. Both these CB2 selective agonists are found to possess potent peripheral analgesic activity in several animal models of inflammatory and neuropathic pain. Moreover, functional studies with a lead compound have highlighted the potential of the cannabilactone based CB2 agonists to behave as neutral antagonists/low potency inverse agonists at CB1, a unique property which imparts very high functional selectivity for CB2 over CB1 [30,31,33]. Thus, from a medicinal chemistry perspective, the class of cannabilactone compounds remains an attractive structural motif for further development of CB2 selective and efficacious agonists.

Chemistry
We conjectured that cannabilactones would be assessible from biphenyl compounds 7 via lactonization (Scheme 1). The bromo-or cyano-group on the C3-side chain of cannabilactones would be introduced to either the biphenyl or the cannabilactone template via substitution. Disconnection of biphenyl compounds 7 through a Suzuki coupling led to boronic acid 8 and aryl bromides 9.

Chemistry
We conjectured that cannabilactones would be assessible from biphenyl compounds 7 via lactonization (Scheme 1). The bromo-or cyano-group on the C3-side chain of cannabilactones would be introduced to either the biphenyl or the cannabilactone template via substitution. Disconnection of biphenyl compounds 7 through a Suzuki coupling led to boronic acid 8 and aryl bromides 9. The present work is informed by our findings with analogs of (−)-∆ 8 -tetrahydrocannabinol (THC) and (−)-hexahydrocannabinol (HHC) with cyclic moieties at the C1 position of the side chain pharmacophore [34][35][36][37] (Figure 1). These studies have shown that analogs with six-to eight-atoms-long side chains substituted at C1 with a cyclopentyl ring exhibit remarkably high affinities for CB1 and CB2 receptors. Taken together, in this study our design replaces the 1 ,1 -gem-dimethyl group in the cannabilactone scaffold with the larger and sterically more confined cyclopentyl group ( Figure 1). Additionally, we have explored the pharmacophoric limits of side chain length, while the polar Molecules 2019, 24, 3559 3 of 21 characteristics have been enhanced by incorporating bromo-and cyano-substituents at the terminal carbon atom [37,38]. All synthesized compounds were assessed for their binding affinities at rat CB1 (rCB1) and mouse CB2 (mCB2) receptors while the most promising analogs were also tested in human CB2 (hCB2) to identify potential species differences between the mouse and human clone. One of the analogs synthesized namely AM4346 has very high affinity (K i = 4.9 nM) and 131-fold selectivity for mCB2 over rCB1 and behaves as a potent CB2 agonist in the cyclase assay. The species difference in the affinities of AM4346 between the mouse and the human CB2 receptors is reduced when compared to our first-generation compound AM1714, while AM4346 is endowed with enhanced polarity due to the presence of the cyano group. Additionally, we extended our SAR to include the biphenyl synthetic intermediates 23a-23d as they encompasses the biaryl subunit which is a privileged structure (biaryls found in 4.3% of known drugs), and also, they have structural similarities with the phytocannabinoid cannabidiol and its oxidative metabolite cannabinodiol ( Figure 2) [39][40][41][42].

Chemistry
We conjectured that cannabilactones would be assessible from biphenyl compounds 7 via lactonization (Scheme 1). The bromo-or cyano-group on the C3-side chain of cannabilactones would be introduced to either the biphenyl or the cannabilactone template via substitution. Disconnection of biphenyl compounds 7 through a Suzuki coupling led to boronic acid 8 and aryl bromides 9.
Exposure of 7a-7d to boron tribromide led to trihydroxybiphenyl intermediates 23a-23d in 89-92% yields, in which cleavage of all methyl ether groups and of the phenolic ether with introduction of the bromide in the case of 7a-7c took place in a single step. Acetic acid mediated lactonization of polyphenols 23a-23d afforded cannabilactones 3a-3d in 89-91% yields. Subsequent exposure of 3a-3c to sodium cyanide in dimethyl sulfoxide gave nitriles 3e-3g.

Affinity for Cannabinoid Receptors
The abilities of the cannabilactone analogs 3a-3g to displace the radiolabeled CB1/CB2 agonist CP-55,940 from membranes prepared from rat brain (a source of CB1) and HEK293 cells expressing mouse CB2 were determined as described earlier [38,57,58]. Inhibition constant values (Ki) from the respective competition binding curves are listed in Table 1 in which our first generation cannabilactone analog AM1714 is included for comparison. The current data of AM1714 for mCB2 are slightly different when compared to those we published earlier [27]. This is because the compound was first assayed in a different mCB2 receptor preparation, e.g., mouse spleen membrane.
It should also be noted that the rat, mouse, and human CB1 receptors have 97-99% sequence identity across species and, as shown earlier (see for example [12,37,38,57]), are not expected to exhibit variations in their Ki values. However, the CB2 receptor shows less homology (~82%) between species than does CB1 (97-99%), and that variability could cause species-related differences in affinity. Indeed, in our original work on the cannabilactone class of compounds, we have identified species-specific variation in CB2 affinity [27]. For this reason, the key compounds were also assayed using membranes from HEK293 cells expressing human CB2 (hCB2). Data from the latter preparation are listed in Table 2.
Exposure of 7a-7d to boron tribromide led to trihydroxybiphenyl intermediates 23a-23d in 89-92% yields, in which cleavage of all methyl ether groups and of the phenolic ether with introduction of the bromide in the case of 7a-7c took place in a single step. Acetic acid mediated lactonization of polyphenols 23a-23d afforded cannabilactones 3a-3d in 89-91% yields. Subsequent exposure of 3a-3c to sodium cyanide in dimethyl sulfoxide gave nitriles 3e-3g.

Affinity for Cannabinoid Receptors
The abilities of the cannabilactone analogs 3a-3g to displace the radiolabeled CB1/CB2 agonist CP-55,940 from membranes prepared from rat brain (a source of CB1) and HEK293 cells expressing mouse CB2 were determined as described earlier [38,57,58]. Inhibition constant values (K i ) from the respective competition binding curves are listed in Table 1 in which our first generation cannabilactone analog AM1714 is included for comparison. The current data of AM1714 for mCB2 are slightly different when compared to those we published earlier [27]. This is because the compound was first assayed in a different mCB2 receptor preparation, e.g., mouse spleen membrane.
It should also be noted that the rat, mouse, and human CB1 receptors have 97-99% sequence identity across species and, as shown earlier (see for example [12,37,38,57]), are not expected to exhibit variations in their K i values. However, the CB2 receptor shows less homology (~82%) between species than does CB1 (97-99%), and that variability could cause species-related differences in affinity. Indeed, in our original work on the cannabilactone class of compounds, we have identified species-specific variation in CB2 affinity [27]. For this reason, the key compounds were also assayed using membranes from HEK293 cells expressing human CB2 (hCB2). Data from the latter preparation are listed in Table 2.
As shown in Table 1, replacement of the C1 -gem-dimethyl group with the bulkier, more sterically confined cyclopentyl ring produces cannabilactone analogs with enhanced affinity and selectivity for Molecules 2019, 24, 3559 6 of 21 the mCB2 relative to the rCB1 receptors. We also observe that this trend for mCB2 selectivity can be optimized by varying the length of the side chain and the substituent at the terminal carbon atom. Thus, analogs carrying five-to seven-atoms long side chains terminated with a bromine atom or a cyano group (3c, 3a, 3b, 3g, and 3e) exhibit 16-to 26-fold selectivity for mCB2 over rCB1. The more lipophilic C1 -gem-dimethyl-heptyl (1b) and C1 -cyclopentyl-heptyl (3d) analogs have comparable affinity (K i = 4.7 ± 1.8 nM and 8.4 ± 2.5 nM) and selectivity (21-to 32-fold) for mCB2 over rCB1. Within these series, however, analog AM4346 with its longer chain (eight atoms) and C6 -cyano substituent has optimal properties: maximal binding affinity for mCB2 and minimal affinity for rCB1. In fact, AM4346 exhibits a remarkable 131-fold selectivity for mCB2 over rCB1.
The binding affinities for the human CB2 (hCB2) receptor of the three key analogs AM1714, AM4348, and AM4346 are listed in Table 2. We observe that although all analogs exhibit somewhat reduced affinity for hCB2 as compared to mCB2 (Table 1), the C1 -cyclopentyl-analogs show a pronounced reduction in the affinity differences between mCB2 and hCB2 as compared to C1 -gem-dimethyl analog. Thus, AM1714 has 18-fold greater affinity for mCB2 than for hCB2, but this preference is reduced to 3-fold for AM4348 and 7-fold for AM4346. With a roughly 19-fold preference, analog AM4346 has the highest selectivity for hCB2 over rCB1. The more lipophilic C1′-gem-dimethyl-heptyl (1b) and C1′-cyclopentyl-heptyl (3d) analogs have comparable affinity (Ki = 4.7 ± 1.8 nM and 8.4 ± 2.5 nM) and selectivity (21-to 32-fold) for mCB2 over rCB1. Within these series, however, analog AM4346 with its longer chain (eight atoms) and C6′-cyano substituent has optimal properties: maximal binding affinity for mCB2 and minimal affinity for rCB1. In fact, AM4346 exhibits a remarkable 131-fold selectivity for mCB2 over rCB1. The binding affinities for the human CB2 (hCB2) receptor of the three key analogs AM1714, AM4348, and AM4346 are listed in Table 2. We observe that although all analogs exhibit somewhat reduced affinity for hCB2 as compared to mCB2 (Table 1), the C1′-cyclopentyl-analogs show a pronounced reduction in the affinity differences between mCB2 and hCB2 as compared to C1′-gemdimethyl analog. Thus, AM1714 has 18-fold greater affinity for mCB2 than for hCB2, but this preference is reduced to 3-fold for AM4348 and 7-fold for AM4346. With a roughly 19-fold preference, analog AM4346 has the highest selectivity for hCB2 over rCB1. 1 Affinities were determined using rat brain (CB1) or membranes from HEK293 cells expressing mouse CB2 and [ 3 H]CP-55,940 as the radioligand following previously described procedures [38,57,58]. Data were analyzed using nonlinear regression analysis. Ki values were obtained from three independent experiments performed in triplicate and are expressed as the mean of the three values. The more lipophilic C1′-gem-dimethyl-heptyl (1b) and C1′-cyclopentyl-heptyl (3d) analogs have comparable affinity (Ki = 4.7 ± 1.8 nM and 8.4 ± 2.5 nM) and selectivity (21-to 32-fold) for mCB2 over rCB1. Within these series, however, analog AM4346 with its longer chain (eight atoms) and C6′-cyano substituent has optimal properties: maximal binding affinity for mCB2 and minimal affinity for rCB1. In fact, AM4346 exhibits a remarkable 131-fold selectivity for mCB2 over rCB1. The binding affinities for the human CB2 (hCB2) receptor of the three key analogs AM1714, AM4348, and AM4346 are listed in Table 2. We observe that although all analogs exhibit somewhat reduced affinity for hCB2 as compared to mCB2 (Table 1), the C1′-cyclopentyl-analogs show a pronounced reduction in the affinity differences between mCB2 and hCB2 as compared to C1′-gemdimethyl analog. Thus, AM1714 has 18-fold greater affinity for mCB2 than for hCB2, but this preference is reduced to 3-fold for AM4348 and 7-fold for AM4346. With a roughly 19-fold preference, analog AM4346 has the highest selectivity for hCB2 over rCB1. 1 Affinities were determined using rat brain (CB1) or membranes from HEK293 cells expressing mouse CB2 and [ 3 H]CP-55,940 as the radioligand following previously described procedures [38,57,58]. Data were analyzed using nonlinear regression analysis. Ki values were obtained from three independent experiments performed in triplicate and are expressed as the mean of the three values. The more lipophilic C1′-gem-dimethyl-heptyl (1b) and C1′-cyclopentyl-heptyl (3d) analogs have comparable affinity (Ki = 4.7 ± 1.8 nM and 8.4 ± 2.5 nM) and selectivity (21-to 32-fold) for mCB2 over rCB1. Within these series, however, analog AM4346 with its longer chain (eight atoms) and C6′-cyano substituent has optimal properties: maximal binding affinity for mCB2 and minimal affinity for rCB1. In fact, AM4346 exhibits a remarkable 131-fold selectivity for mCB2 over rCB1. The binding affinities for the human CB2 (hCB2) receptor of the three key analogs AM1714, AM4348, and AM4346 are listed in Table 2. We observe that although all analogs exhibit somewhat reduced affinity for hCB2 as compared to mCB2 (Table 1), the C1′-cyclopentyl-analogs show a pronounced reduction in the affinity differences between mCB2 and hCB2 as compared to C1′-gemdimethyl analog. Thus, AM1714 has 18-fold greater affinity for mCB2 than for hCB2, but this preference is reduced to 3-fold for AM4348 and 7-fold for AM4346. With a roughly 19-fold preference, analog AM4346 has the highest selectivity for hCB2 over rCB1. The more lipophilic C1′-gem-dimethyl-heptyl (1b) and C1′-cyclopentyl-heptyl (3d) analogs have comparable affinity (Ki = 4.7 ± 1.8 nM and 8.4 ± 2.5 nM) and selectivity (21-to 32-fold) for mCB2 over rCB1. Within these series, however, analog AM4346 with its longer chain (eight atoms) and C6′-cyano substituent has optimal properties: maximal binding affinity for mCB2 and minimal affinity for rCB1. In fact, AM4346 exhibits a remarkable 131-fold selectivity for mCB2 over rCB1. The binding affinities for the human CB2 (hCB2) receptor of the three key analogs AM1714, AM4348, and AM4346 are listed in Table 2. We observe that although all analogs exhibit somewhat reduced affinity for hCB2 as compared to mCB2 (Table 1), the C1′-cyclopentyl-analogs show a pronounced reduction in the affinity differences between mCB2 and hCB2 as compared to C1′-gemdimethyl analog. Thus, AM1714 has 18-fold greater affinity for mCB2 than for hCB2, but this preference is reduced to 3-fold for AM4348 and 7-fold for AM4346. With a roughly 19-fold preference, analog AM4346 has the highest selectivity for hCB2 over rCB1. The more lipophilic C1′-gem-dimethyl-heptyl (1b) and C1′-cyclopentyl-heptyl (3d) analogs have comparable affinity (Ki = 4.7 ± 1.8 nM and 8.4 ± 2.5 nM) and selectivity (21-to 32-fold) for mCB2 over rCB1. Within these series, however, analog AM4346 with its longer chain (eight atoms) and C6′-cyano substituent has optimal properties: maximal binding affinity for mCB2 and minimal affinity for rCB1. In fact, AM4346 exhibits a remarkable 131-fold selectivity for mCB2 over rCB1. The binding affinities for the human CB2 (hCB2) receptor of the three key analogs AM1714, AM4348, and AM4346 are listed in Table 2. We observe that although all analogs exhibit somewhat reduced affinity for hCB2 as compared to mCB2 (Table 1), the C1′-cyclopentyl-analogs show a pronounced reduction in the affinity differences between mCB2 and hCB2 as compared to C1′-gemdimethyl analog. Thus, AM1714 has 18-fold greater affinity for mCB2 than for hCB2, but this preference is reduced to 3-fold for AM4348 and 7-fold for AM4346. With a roughly 19-fold preference, analog AM4346 has the highest selectivity for hCB2 over rCB1.  The more lipophilic C1′-gem-dimethyl-heptyl (1b) and C1′-cyclopentyl-heptyl (3d) analogs have comparable affinity (Ki = 4.7 ± 1.8 nM and 8.4 ± 2.5 nM) and selectivity (21-to 32-fold) for mCB2 over rCB1. Within these series, however, analog AM4346 with its longer chain (eight atoms) and C6′-cyano substituent has optimal properties: maximal binding affinity for mCB2 and minimal affinity for rCB1. In fact, AM4346 exhibits a remarkable 131-fold selectivity for mCB2 over rCB1. The binding affinities for the human CB2 (hCB2) receptor of the three key analogs AM1714, AM4348, and AM4346 are listed in Table 2. We observe that although all analogs exhibit somewhat reduced affinity for hCB2 as compared to mCB2 (Table 1), the C1′-cyclopentyl-analogs show a pronounced reduction in the affinity differences between mCB2 and hCB2 as compared to C1′-gemdimethyl analog. Thus, AM1714 has 18-fold greater affinity for mCB2 than for hCB2, but this preference is reduced to 3-fold for AM4348 and 7-fold for AM4346. With a roughly 19-fold preference, analog AM4346 has the highest selectivity for hCB2 over rCB1.  The more lipophilic C1′-gem-dimethyl-heptyl (1b) and C1′-cyclopentyl-heptyl (3d) analogs have comparable affinity (Ki = 4.7 ± 1.8 nM and 8.4 ± 2.5 nM) and selectivity (21-to 32-fold) for mCB2 over rCB1. Within these series, however, analog AM4346 with its longer chain (eight atoms) and C6′-cyano substituent has optimal properties: maximal binding affinity for mCB2 and minimal affinity for rCB1. In fact, AM4346 exhibits a remarkable 131-fold selectivity for mCB2 over rCB1. The binding affinities for the human CB2 (hCB2) receptor of the three key analogs AM1714, AM4348, and AM4346 are listed in Table 2. We observe that although all analogs exhibit somewhat reduced affinity for hCB2 as compared to mCB2 (Table 1), the C1′-cyclopentyl-analogs show a pronounced reduction in the affinity differences between mCB2 and hCB2 as compared to C1′-gemdimethyl analog. Thus, AM1714 has 18-fold greater affinity for mCB2 than for hCB2, but this preference is reduced to 3-fold for AM4348 and 7-fold for AM4346. With a roughly 19-fold preference, analog AM4346 has the highest selectivity for hCB2 over rCB1.  The more lipophilic C1′-gem-dimethyl-heptyl (1b) and C1′-cyclopentyl-heptyl (3d) analogs have comparable affinity (Ki = 4.7 ± 1.8 nM and 8.4 ± 2.5 nM) and selectivity (21-to 32-fold) for mCB2 over rCB1. Within these series, however, analog AM4346 with its longer chain (eight atoms) and C6′-cyano substituent has optimal properties: maximal binding affinity for mCB2 and minimal affinity for rCB1. In fact, AM4346 exhibits a remarkable 131-fold selectivity for mCB2 over rCB1. The binding affinities for the human CB2 (hCB2) receptor of the three key analogs AM1714, AM4348, and AM4346 are listed in Table 2. We observe that although all analogs exhibit somewhat reduced affinity for hCB2 as compared to mCB2 (Table 1), the C1′-cyclopentyl-analogs show a pronounced reduction in the affinity differences between mCB2 and hCB2 as compared to C1′-gemdimethyl analog. Thus, AM1714 has 18-fold greater affinity for mCB2 than for hCB2, but this preference is reduced to 3-fold for AM4348 and 7-fold for AM4346. With a roughly 19-fold preference, analog AM4346 has the highest selectivity for hCB2 over rCB1.  The more lipophilic C1′-gem-dimethyl-heptyl (1b) and C1′-cyclopentyl-heptyl (3d) analogs have comparable affinity (Ki = 4.7 ± 1.8 nM and 8.4 ± 2.5 nM) and selectivity (21-to 32-fold) for mCB2 over rCB1. Within these series, however, analog AM4346 with its longer chain (eight atoms) and C6′-cyano substituent has optimal properties: maximal binding affinity for mCB2 and minimal affinity for rCB1. In fact, AM4346 exhibits a remarkable 131-fold selectivity for mCB2 over rCB1. The binding affinities for the human CB2 (hCB2) receptor of the three key analogs AM1714, AM4348, and AM4346 are listed in Table 2. We observe that although all analogs exhibit somewhat reduced affinity for hCB2 as compared to mCB2 (Table 1), the C1′-cyclopentyl-analogs show a pronounced reduction in the affinity differences between mCB2 and hCB2 as compared to C1′-gemdimethyl analog. Thus, AM1714 has 18-fold greater affinity for mCB2 than for hCB2, but this preference is reduced to 3-fold for AM4348 and 7-fold for AM4346. With a roughly 19-fold preference, analog AM4346 has the highest selectivity for hCB2 over rCB1.   Table 2. Affinities of key cannabilactone analogs for hCB2 cannabinoid receptors (± 95% confidence limits). 1 Affinities were determined using rat brain (CB1) or membranes from HEK293 cells expressing mouse CB2 and [ 3 H]CP-55,940 as the radioligand following previously described procedures [38,57,58]. Data were analyzed using nonlinear regression analysis. Ki values were obtained from three independent experiments performed in triplicate and are expressed as the mean of the three values. Synthesized as precursors to cannabilactones, binding affinities of the intermediate biphenyl compounds 23a-23d for rCB1 and mCB2 were also determined ( Table 3). To our surprise, these biphenyl analogs bind mCB2 with high affinity and substantial selectivity for that receptor over rCB1. In this series, the C1′-cyclopentyl-heptyl analog AM4347 has both the greatest binding affinity (Ki = 5.7 ± 1.5 nM) for mCB2 and the highest selectivity (60-fold) for that receptor over rCB1.

Functional Characterization
Functional characterization of three key cannabilactones (AM1710, AM1714, and AM4346) for the hCB2 receptor was carried out by measuring the decrease in forskolin stimulated cAMP, as detailed earlier [38,57]. Data are listed in Table 4 in which the standard cannabinoid agonist CP-55,940 is included for comparison. Our testing results show that all three compounds potently decreased the levels of cAMP, indicating that within this signaling mechanism these compounds behaved as potent agonists at the hCB2 receptor with the C1′-cyclopentyl-analog AM4346 being more potent (EC50 = 3.7  Synthesized as precursors to cannabilactones, binding affinities of the intermediate biphenyl compounds 23a-23d for rCB1 and mCB2 were also determined ( Table 3). To our surprise, these biphenyl analogs bind mCB2 with high affinity and substantial selectivity for that receptor over rCB1. In this series, the C1′-cyclopentyl-heptyl analog AM4347 has both the greatest binding affinity (Ki = 5.7 ± 1.5 nM) for mCB2 and the highest selectivity (60-fold) for that receptor over rCB1.

Functional Characterization
Functional characterization of three key cannabilactones (AM1710, AM1714, and AM4346) for the hCB2 receptor was carried out by measuring the decrease in forskolin stimulated cAMP, as detailed earlier [38,57]. Data are listed in Table 4 in which the standard cannabinoid agonist CP-55,940 is included for comparison. Our testing results show that all three compounds potently decreased the levels of cAMP, indicating that within this signaling mechanism these compounds behaved as potent agonists at the hCB2 receptor with the C1′-cyclopentyl-analog AM4346 being more potent (EC50 = 3.7  Synthesized as precursors to cannabilactones, binding affinities of the intermediate biphenyl compounds 23a-23d for rCB1 and mCB2 were also determined ( Table 3). To our surprise, these biphenyl analogs bind mCB2 with high affinity and substantial selectivity for that receptor over rCB1. In this series, the C1′-cyclopentyl-heptyl analog AM4347 has both the greatest binding affinity (Ki = 5.7 ± 1.5 nM) for mCB2 and the highest selectivity (60-fold) for that receptor over rCB1.

Functional Characterization
Functional characterization of three key cannabilactones (AM1710, AM1714, and AM4346) for the hCB2 receptor was carried out by measuring the decrease in forskolin stimulated cAMP, as detailed earlier [38,57]. Data are listed in Table 4 in which the standard cannabinoid agonist CP-55,940 is included for comparison. Our testing results show that all three compounds potently decreased the levels of cAMP, indicating that within this signaling mechanism these compounds behaved as potent 266 ± 58 28.0 ± 9.7 9.5 3.3 Synthesized as precursors to cannabilactones, binding affinities of the intermediate biphenyl compounds 23a-23d for rCB1 and mCB2 were also determined ( Table 3). To our surprise, these biphenyl analogs bind mCB2 with high affinity and substantial selectivity for that receptor over rCB1. In this series, the C1 -cyclopentyl-heptyl analog AM4347 has both the greatest binding affinity (K i = 5.7 ± 1.5 nM) for mCB2 and the highest selectivity (60-fold) for that receptor over rCB1. Synthesized as precursors to cannabilactones, binding affinities of the intermediate biphenyl compounds 23a-23d for rCB1 and mCB2 were also determined ( Table 3). To our surprise, these biphenyl analogs bind mCB2 with high affinity and substantial selectivity for that receptor over rCB1. In this series, the C1′-cyclopentyl-heptyl analog AM4347 has both the greatest binding affinity (Ki = 5.7 ± 1.5 nM) for mCB2 and the highest selectivity (60-fold) for that receptor over rCB1.

Functional Characterization
Functional characterization of three key cannabilactones (AM1710, AM1714, and AM4346) for the hCB2 receptor was carried out by measuring the decrease in forskolin stimulated cAMP, as detailed earlier [38,57]. Data are listed in Table 4 in which the standard cannabinoid agonist CP-55,940 is included for comparison. Our testing results show that all three compounds potently decreased the levels of cAMP, indicating that within this signaling mechanism these compounds behaved as potent agonists at the hCB2 receptor with the C1′-cyclopentyl-analog AM4346 being more potent (EC50 = 3.7 ± 1.5 nM, E(max) = 89%) than the C1′-gem-dimethyl-analogs AM1710 and AM1714 (EC50 = 10.5 ± 2.5 nM, Synthesized as precursors to cannabilactones, binding affinities of the intermediate biphenyl compounds 23a-23d for rCB1 and mCB2 were also determined ( Table 3). To our surprise, these biphenyl analogs bind mCB2 with high affinity and substantial selectivity for that receptor over rCB1. In this series, the C1′-cyclopentyl-heptyl analog AM4347 has both the greatest binding affinity (Ki = 5.7 ± 1.5 nM) for mCB2 and the highest selectivity (60-fold) for that receptor over rCB1.

Functional Characterization
Functional characterization of three key cannabilactones (AM1710, AM1714, and AM4346) for the hCB2 receptor was carried out by measuring the decrease in forskolin stimulated cAMP, as detailed earlier [38,57]. Data are listed in Table 4 in which the standard cannabinoid agonist CP-55,940 is included for comparison. Our testing results show that all three compounds potently decreased the levels of cAMP, indicating that within this signaling mechanism these compounds behaved as potent agonists at the hCB2 receptor with the C1′-cyclopentyl-analog AM4346 being more potent (EC50 = 3.7 ± 1.5 nM, E(max) = 89%) than the C1′-gem-dimethyl-analogs AM1710 and AM1714 (EC50 = 10.5 ± 2.5 nM, E(max) = 73% and EC50 = 36.9 ± 6.8 nM, E(max) = 77% respectively). Synthesized as precursors to cannabilactones, binding affinities of the intermediate biphenyl compounds 23a-23d for rCB1 and mCB2 were also determined ( Table 3). To our surprise, these biphenyl analogs bind mCB2 with high affinity and substantial selectivity for that receptor over rCB1. In this series, the C1′-cyclopentyl-heptyl analog AM4347 has both the greatest binding affinity (Ki = 5.7 ± 1.5 nM) for mCB2 and the highest selectivity (60-fold) for that receptor over rCB1.

Functional Characterization
Functional characterization of three key cannabilactones (AM1710, AM1714, and AM4346) for the hCB2 receptor was carried out by measuring the decrease in forskolin stimulated cAMP, as detailed earlier [38,57]. Data are listed in Table 4 in which the standard cannabinoid agonist CP-55,940 is included for comparison. Our testing results show that all three compounds potently decreased the levels of cAMP, indicating that within this signaling mechanism these compounds behaved as potent agonists at the hCB2 receptor with the C1′-cyclopentyl-analog AM4346 being more potent (EC50 = 3.7 ± 1.5 nM, E(max) = 89%) than the C1′-gem-dimethyl-analogs AM1710 and AM1714 (EC50 = 10.5 ± 2.5 nM, Synthesized as precursors to cannabilactones, binding affinities of the intermediate biphenyl compounds 23a-23d for rCB1 and mCB2 were also determined ( Table 3). To our surprise, these biphenyl analogs bind mCB2 with high affinity and substantial selectivity for that receptor over rCB1. In this series, the C1′-cyclopentyl-heptyl analog AM4347 has both the greatest binding affinity (Ki = 5.7 ± 1.5 nM) for mCB2 and the highest selectivity (60-fold) for that receptor over rCB1.

Functional Characterization
Functional characterization of three key cannabilactones (AM1710, AM1714, and AM4346) for the hCB2 receptor was carried out by measuring the decrease in forskolin stimulated cAMP, as detailed earlier [38,57]. Data are listed in Table 4 in which the standard cannabinoid agonist CP-55,940 is included for comparison. Our testing results show that all three compounds potently decreased the levels of cAMP, indicating that within this signaling mechanism these compounds behaved as potent agonists at the hCB2 receptor with the C1′-cyclopentyl-analog AM4346 being more potent (EC50 = 3.7 ± 1.5 nM, E(max) = 89%) than the C1′-gem-dimethyl-analogs AM1710 and AM1714 (EC50 = 10.5 ± 2.5 nM, 266 ± 58 28.0 ± 9.7 9.5 3.3 Synthesized as precursors to cannabilactones, binding affinities of the intermediate biphenyl compounds 23a-23d for rCB1 and mCB2 were also determined ( Table 3). To our surprise, these biphenyl analogs bind mCB2 with high affinity and substantial selectivity for that receptor over rCB1. In this series, the C1′-cyclopentyl-heptyl analog AM4347 has both the greatest binding affinity (Ki = 5.7 ± 1.5 nM) for mCB2 and the highest selectivity (60-fold) for that receptor over rCB1.

Functional Characterization
Functional characterization of three key cannabilactones (AM1710, AM1714, and AM4346) for the hCB2 receptor was carried out by measuring the decrease in forskolin stimulated cAMP, as detailed earlier [38,57]. Data are listed in Table 4 in which the standard cannabinoid agonist CP-55,940 is included for comparison. Our testing results show that all three compounds potently decreased the levels of cAMP, indicating that within this signaling mechanism these compounds behaved as potent agonists at the hCB2 receptor with the C1′-cyclopentyl-analog AM4346 being more potent (EC50 = 3.7 341 ± 100 5.7 ± 1.5 59.8

Functional Characterization
Functional characterization of three key cannabilactones (AM1710, AM1714, and AM4346) for the hCB2 receptor was carried out by measuring the decrease in forskolin stimulated cAMP, as detailed earlier [38,57]. Data are listed in Table 4 in which the standard cannabinoid agonist CP-55,940 is included for comparison. Our testing results show that all three compounds potently decreased the levels of cAMP, indicating that within this signaling mechanism these compounds behaved as potent agonists at the hCB2 receptor with the C1 -cyclopentyl-analog AM4346 being more potent (EC 50 = 3.7 ± 1.5 nM, E (max) = 89%) than the C1 -gem-dimethyl-analogs AM1710 and AM1714 (EC 50 = 10.5 ± 2.5 nM, E (max) = 73% and EC 50 = 36.9 ± 6.8 nM, E (max) = 77% respectively).
In the same assay, one of the biphenyl analogs, compound 23d (AM4347), failed to show any responses in concentrations up to 2 µM. 36.9 ± 6.8 (agonist) 77

3f AM4346
3.7 ± 1.5 (agonist) 89 1 Functional potencies at hCB2 receptor were determined by measuring the decrease in forskolin-stimulated cAMP levels [38,57]. EC 50 values were calculated using nonlinear regression analysis. Data are the average of two independent experiments run in triplicate. 2 Forskolin stimulated cAMP levels were normalized to 100%. E(max) is the maximum inhibition of forskolin stimulated cAMP levels and is presented as the percentage of CP-55,940 response at 500 nM.
1-(4-Bromo-3,5-dimethoxyphenyl)cyclopentanecarbonitrile (15). To a solution of 14 (3.42 g, 13.36 mmol) in anhydrous THF (120 mL), at 0 • C under an argon atmosphere, was added potassium bis(trimethylsilyl)amide (8 g, 40.2 mmol). The resulting slurry was stirred at the same temperature for 10 min, and then a solution of 1,4-dibromobutane (3.46 g, 16 mmol) in anhydrous THF (20 mL) was added dropwise. The reaction was stirred for an additional 10 min at 0 • C and then quenched by adding saturated aqueous NH 4 Cl solution (80 mL). The mixture was warmed to room temperature and diluted with diethyl ether (200 mL). The organic layer was separated and the aqueous phase extracted with diethyl ether. The combined organic layer was washed with brine and dried over MgSO 4 and the solvent evaporated under reduced pressure. The residue was purified by flash column chromatography on silica gel (25% diethyl ether in hexane) to give 15 as a white solid in 70% yield (2.9 g). mp 135-138 • C; 1  1-(4-Bromo-3,5-dimethoxyphenyl)cyclopentanecarboxaldehyde (16). To a stirred solution of 15 (1.40 g, 4.52 mmol) in dry CH 2 Cl 2 (50 mL), at −78 • C, under an argon atmosphere, was added diisobutylaluminum hydride (1 M solution in hexanes, 12 mL) dropwise. The reaction mixture was stirred at the same temperature for 1 h, and then quenched by the dropwise addition of potassium sodium tartrate (10% solution in water). The mixture was warmed to room temperature and stirred vigorously for 1 h. The organic layer was separated, and the aqueous phase extracted with diethyl ether. The combined organic layer was washed with brine, dried over MgSO 4 and concentrated in vacuo. The crude product was purified by flash column chromatography on silica gel (25% diethyl ether in hexane) to give 16 as a white solid in 85% yield (1. 2-Bromo-1,3-dimethoxy-5-[1-(4-phenoxybutyl)cyclopentyl]benzene (9a). A mixture of 17a (360 mg, 0.84 mmol) and 10% Pd/C (54 mg) in ethyl acetate (20 mL) was stirred vigorously under hydrogen atmosphere (room temperature overnight). The catalyst was removed by filtration through celite and the filtrate was evaporated under reduced pressure to give the product 9a as a viscous liquid (346 mg, 95% yield) which was used in the next step without further purification. 1  trans-3-[1-(4-Bromo-3,5-dimethoxyphenyl)cyclopentyl]acrylic acid ethyl ester (18). To a solution of triethyl phosphonoacetate (2.78 g, 12.42 mmol) in dry THF (50 mL), at 0 • C, under an argon atmosphere, was added sodium hydride (497 mg, 12.42 mmol, 60% dispersion in mineral oil). The mixture was stirred for 15 min at the same temperature and a solution of 16 (1.11 g, 3.55 mmol) in dry THF (10 mL) was added dropwise. Stirring was continued for an additional 10 min, and the reaction was quenched by adding saturated aqueous NH 4 Cl (20 mL). The mixture was warmed to room temperature and diluted with diethyl ether (100 mL). The organic phase was separated and the aqueous phase extracted with diethyl ether. The combined organic layer was washed with brine, dried over MgSO 4 and evaporated under reduced pressure. The crude product was purified by flash column chromatography on silica gel (35% diethyl ether in hexane) to give 18 as a white solid in 92% yield ( 3-[1-(4-Bromo-3,5-dimethoxyphenyl)cyclopentyl]propionic acid methyl ester (19b). A mixture of 18 (0.7 g, 1.83 mmol) and magnesium turnings (132 mg, 5.5 mmol), in dry methanol (20 mL) was stirred at 0 • C for 2 h and at room temperature for an additional 12 h. The solvent was evaporated under reduced pressure and the residue diluted with water (20 mL) and diethyl ether (50 mL). To this mixture was added 5% aqueous HCl (10 mL), the organic layer was separated, and the aqueous layer extracted with diethyl ether. The combined ethereal layer was successively washed with NaHCO 3 and brine, dried over MgSO 4 and evaporated under reduced pressure to give the product 19b as a white solid (543 mg, 80% yield) which was used in the next step without further purification. mp 71-74 • C; 1  3-[1-(4-Bromo-3,5-dimethoxyphenyl)cyclopentyl]propan-1-ol (20). To a stirred solution of 19b (520 mg, 1.4 mmol) in dry THF (20 mL), at room temperature, under an argon atmosphere, was added diisobutylaluminum hydride (3.7 mL, 1M solution in hexanes) over a period of 15 min. The mixture was stirred at the same temperature for 1 h, and then cooled to 0 • C, and the reaction was quenched by dropwise addition of aqueous potassium sodium tartrate (10% solution in water, 10 mL). The resulting mixture was warmed to room temperature, diluted with ethyl acetate (10 mL) and stirred vigorously for 1h. The organic layer was separated, and the aqueous phase extracted with ethyl acetate. The combined organic layer was washed with brine, dried over MgSO 4 and concentrated in vacuo. The residue was purified by flash column chromatography on silica gel (40% ethyl acetate in hexane) to give 20 as a colorless liquid in 85% yield (408 mg). 1  2-Bromo-1,3-dimethoxy-5-[1-(3-bromopropyl)cyclopentyl]benzene (21). To a stirred solution of 20 (400 mg, 1.17 mmol) and carbon tetrabromide (465 mg, 1.40 mmol) in anhydrous CH 2 Cl 2 (10 mL) at 0 • C, under a nitrogen atmosphere, was added triphenylphosphine (368 mg, 1.40 mmol) portionwise.
After the addition was completed, the reaction mixture was stirred for an additional 20 min, whereupon the solvent was removed in vacuo. The residue was purified by flash column chromatography on silica gel (10% diethyl ether in hexane) to give 21 as a colorless liquid in 85% yield (403 mg). 1  2-Bromo-1,3-dimethoxy-5-[1-(3-phenoxypropyl)cyclopentyl]benzene (9c). To a stirred solution of 21 (250 mg, 0.62 mmol) in DMSO (6 mL) at room temperature, under an argon atmosphere, was added sodium phenoxide trihydrate (530 mg, 3.10 mmol). The reaction mixture was stirred vigorously for 24 h and then diluted by adding ice water (5 mL) and diethyl ether. The organic layer was separated, and the aqueous phase was extracted with diethyl ether. The combined organic layer was washed with brine, dried over MgSO 4 and evaporated under reduced pressure. Purification by flash column chromatography on silica gel (10% diethyl ether in hexane) gave 9c as a colorless liquid in 50% yield (130 mg). 1