Syntheses of Cannabinoid Metabolites: Ajulemic Acid and HU-210

Cannabinoid metabolites have been reported to be more potent than their parent compounds. Among them, ajulemic acid (AJA) is a side-chain analog of Δ9-THC-11-oic acid, which would be a good template structure for the discovery of more potent analogues. Herein, we optimized the key allylic oxidation step to introduce the C-11 hydroxy group with a high yield. A series of compounds was prepared with this condition applied including HU-210, 11-nor-Δ8-tetrahydrocannabinol (THC)-carboxylic acid and Δ9-THC-carboxylic acid.


Introduction
Phytocannabinoids and their synthetic analogues, exemplified by molecules such as ∆ 9 -THC, 1 (Figure 1), are prime candidates for pharmaceutical innovation and are known to possess potent analgesic and anti-inflammatory properties.Compound 1 was first identified in 1964 by Gaoni and Mechoulam as the principle bioactive component of marijuana (hashish), which has been used for centuries as both a therapeutic and recreational drug.More generally, cannabinoid-based chemical probes and leads are essential for the continued exploration of the endocannabinoid system [1].Prior to the 1980s, cannabinoids were hypothesized to produce their effects through nonspecific interactions with cell membranes.The absolute stereochemistry of 1 was established in 1967 [2], and more than two decades passed before it was identified as a modulator of the cannabinoid receptor [3].
The main metabolic pathways involve hydroxylations or oxygenations.It is wellknown that substituents at the C-1, C-3 and C-9 positions play critical roles in efficient binding to the cannabinoid receptors [4].SAR studies of the C-3 side chain demonstrated that a seven-carbon homologue was optimal for activity and that branched alkyl groups also led to improved binding affinity [5].For example, the dimethylheptyl analogue 3, which is an oxygenation product of 2, exhibits an approximately 50-fold improvement in activity relative to 1 [6].The relative importance of the C-1 hydroxyl differs between the two cannabinoid subtypes.Synthetic cannabinoid 3 [7] possesses two hydrogen donors and exhibits significantly enhanced affinity for both the CB2 and the CB1 receptors, producing many of the same pharmacological effects as 1 [6].
an analog, and is a 'first-in-class' chemical entity designed to have increased anti-infl matory properties and reduced psychotropic activity compared to its THC parent This compound was found to be well tolerated in a phase I clinical trial, and subseque a phase II study was completed wherein this molecule demonstrated efficacy in redu chronic neuropathic pain without any major adverse effects [11].There are few reports on the syntheses of cannabinoid metabolites, which var length from five to seven steps (Figure 2).Jiang et al. reported a synthetic route fo tricyclic hexahydrocannabinol (HHC) analogues with seven steps and 24% overall y [12].Tepper et al. developed a synthetic route to achieve the AJA within five steps but in 13% overall yields [13].Considering the rapid change in the illicit drug market, a con synthetic route (Figure 2) where the combination of a simple replacement of the ph protecting group and the optimization of the Riley oxidation condition give a better y (43% overall yield).Therefore, a synthetic route for the synthesis of a key intermed which subsequently, can be used to synthesize the cannabinoid metabolites, was d oped.The synthesis method for the intermediate is carried out on a multigram scale.Compound 4 is a synthetic analog of ∆ 9 -THC-11-oic acid, the major metabolite of the psychoactive component of marijuana, ∆ 9 -THC.∆ 9 -THC-11-oic acid [8] has no psychotropic activity and is present in the tissues of the millions of recreational cannabis users long after the mood altering effects are gone [9].Its analgesic properties suggest that it would be a good template structure for the discovery of more potent analogs.AJA is such an analog, and is a 'first-in-class' chemical entity designed to have increased anti-inflammatory properties and reduced psychotropic activity compared to its THC parent [10].This compound was found to be well tolerated in a phase I clinical trial, and subsequently, a phase II study was completed wherein this molecule demonstrated efficacy in reducing chronic neuropathic pain without any major adverse effects [11].
There are few reports on the syntheses of cannabinoid metabolites, which vary in length from five to seven steps (Figure 2).Jiang et al. reported a synthetic route for the tricyclic hexahydrocannabinol (HHC) analogues with seven steps and 24% overall yields [12].Tepper et al. developed a synthetic route to achieve the AJA within five steps but only in 13% overall yields [13].Considering the rapid change in the illicit drug market, a concise synthetic route (Figure 2) where the combination of a simple replacement of the phenol protecting group and the optimization of the Riley oxidation condition give a better yield (43% overall yield).Therefore, a synthetic route for the synthesis of a key intermediate, which subsequently, can be used to synthesize the cannabinoid metabolites, was developed.The synthesis method for the intermediate is carried out on a multigram scale.

Results and Discussion
The route to the synthesis of AJA by Tepper was optimized by changing the SeO2mediated condition to improve the yield of the key allylic oxidation step.Our synthesis started with the preparation of the tricyclic intermediate (5) [14], as shown in Scheme 1.The commercially available starting materials p-menthadienol (PMD) and 1,1-dimethylheptyl resorcinol (DMHR) were used under acid conditions to promote cyclization to obtain the tricyclic skeleton of the cannabinoid metabolites in 82% yield.Thus, the further protection of the phenol group with a TBS group afforded the key tricyclic intermediate (6) in 97% yield.To install the C11 hydroxyl group, (6) was subjected to a SeO2-mediated allylic oxidation, and product (7) bearing an aldehyde group at C11 was obtained in 65% yield.Then, the reduction of the new generated aldehyde group with NaBH4 afforded the hydroxy group in 89% yield.The resultant (8) was then treated with 1 M tetra-N-butylammonium fluoride (TBAF) in tetrahydrofuran (THF) to give product HU-210 in 93% yield.We then turned our attention to (7) for the total synthesis of AJA.To this end, (7) was first converted to acid (9) with treatment by Pinnick oxidation in 90% yield; further treatment of (9) with TBAF in THF gave product AJA in 92% yield.

Results and Discussion
The route to the synthesis of AJA by Tepper was optimized by changing the SeO 2mediated condition to improve the yield of the key allylic oxidation step.Our synthesis started with the preparation of the tricyclic intermediate (5) [14], as shown in Scheme 1.The commercially available starting materials p-menthadienol (PMD) and 1,1-dimethylheptyl resorcinol (DMHR) were used under acid conditions to promote cyclization to obtain the tricyclic skeleton of the cannabinoid metabolites in 82% yield.Thus, the further protection of the phenol group with a TBS group afforded the key tricyclic intermediate (6) in 97% yield.To install the C11 hydroxyl group, (6) was subjected to a SeO 2 -mediated allylic oxidation, and product (7) bearing an aldehyde group at C11 was obtained in 65% yield.Then, the reduction of the new generated aldehyde group with NaBH 4 afforded the hydroxy group in 89% yield.The resultant (8) was then treated with 1 M tetra-N-butylammonium fluoride (TBAF) in tetrahydrofuran (THF) to give product HU-210 in 93% yield.We then turned our attention to (7) for the total synthesis of AJA.To this end, (7) was first converted to acid (9) with treatment by Pinnick oxidation in 90% yield; further treatment of (9) with TBAF in THF gave product AJA in 92% yield.
During the initial attempt, the acetyl protecting group of the tricyclic intermediate phenol 10 led to a low yield at the allylic oxidation step.We tried a variety of conditions including SeO 2 /THF/H 2 O [13], SeO 2 /AcOH/DCM [15], SeO 2 /tBuOOH/salicylic acid/DCM [16], SeO 2 /tBuOOH/DCM [17] and SeO 2 /dioxane.The results showed that there was a byproduct (12), which is a regio-selected isomer.There was also another byproduct (13), which is an aromatic product, and is shown in Table 1.Fortunately, the SeO 2 -mediated Riley oxidation in dioxane at 110 • C gave a moderate yield.When the acetyl group was replaced with the silyl-ether-protected group, the yield of the allylic oxidation was improved markedly due to the significant decrease in by-products.During the initial attempt, the acetyl protecting group of the tricyclic intermediate phenol 10 led to a low yield at the allylic oxidation step.We tried a variety of conditions including SeO2/THF/H2O [13], SeO2/AcOH/DCM [15], SeO2/tBuOOH/salicylic acid/DCM [16], SeO2/tBuOOH/DCM [17] and SeO2/dioxane.The results showed that there was a byproduct (12), which is a regio-selected isomer.There was also another byproduct (13), which is an aromatic product, and is shown in Table 1.Fortunately, the SeO2-mediated Riley oxidation in dioxane at 110 °C gave a moderate yield.When the acetyl group was replaced with the silyl-ether-protected group, the yield of the allylic oxidation was improved markedly due to the significant decrease in by-products.During the initial attempt, the acetyl protecting group of the tricyclic intermediate phenol 10 led to a low yield at the allylic oxidation step.We tried a variety of conditions including SeO2/THF/H2O [13], SeO2/AcOH/DCM [15], SeO2/tBuOOH/salicylic acid/DCM [16], SeO2/tBuOOH/DCM [17] and SeO2/dioxane.The results showed that there was a byproduct (12), which is a regio-selected isomer.There was also another byproduct (13), which is an aromatic product, and is shown in Table 1.Fortunately, the SeO2-mediated Riley oxidation in dioxane at 110 °C gave a moderate yield.When the acetyl group was replaced with the silyl-ether-protected group, the yield of the allylic oxidation was improved markedly due to the significant decrease in by-products.The optimized conditions were then applied for the syntheses of 11-nor-∆ 8 -THC-9carboxylic acid and ∆ 9 -THC-carboxylic acid (Figure 1).To this end, this method was used to successfully obtain the metabolites through a four-step strategy from ∆ 8 -THC and ∆ 9 -THC (Scheme 2).
The optimized conditions were then applied for the syntheses of 11-nor-Δ 8 -THC-9carboxylic acid and Δ 9 -THC-carboxylic acid (Figure 1).To this end, this method was used to successfully obtain the metabolites through a four-step strategy from Δ 8 -THC and Δ 9 -THC (Scheme 2).

Figure 1 .
Figure 1.General structural information for cannabinoid metabolites with oxygen at C-11 site.

Figure 1 .
Figure 1.General structural information for cannabinoid metabolites with oxygen at C-11 site.

Table 1 .
The optimization of the Riley oxidation.
a Isolated yield.b Spectrums can be found in Supplementary Materials.

Table 1 .
The optimization of the Riley oxidation.

Table 1 .
The optimization of the Riley oxidation.
a Isolated yield.bSpectrums can be found in Supplementary Materials.aIsolated yield.b Spectrums can be found in Supplementary Materials.