Synthetic Pathways to Non-Psychotropic Phytocannabinoids as Promising Molecules to Develop Novel Antibiotics: A Review

4.2.3. Synthesis of (−)-CBD

The first synthetic routes available to synthetize (−)-CBD [114,115,116,117] are of scarce practical value, as they lead to (−)-CBD in mediocre or even insignificant yields and the unnatural CBD isomer (Abn-CBD) (see Figure 11) was obtained in amounts considerably larger than those of (−)-CBD. Despite such poor results, the procedure proposed by Petrzilka et al. in the year 1969, but using E-(+)-p-mentha-2,8-dien-1-ol in place of Z-(+)-p-mentha-2,8-dien-1-ol and olivetol in benzene in the presence of catalytic amounts of PTSA was reproduced by Papahatjis et al. in the year 2002, achieving (−)-CBD in 31% yield [118]. The best route to (−)-CBD described is the condensation of (+)-p-mentha-diene-l-olo with olivetol in the presence of weak acids, reported by Razdan et al. in the year 1974 and Uliss et al. the next year [117,119]. In this case, the Abn-CBD obtained was converted to (−)-CBD with BF₃-etherate by a retro-Friedel-Crafts reaction, followed by recombination. However, with this reagent the reaction proceeded further causing cyclisation of (−)-CBD.

In the year 1985, it was reported that when BF₃-etherate on alumina is used as condensing agent in the reaction of Z-(+)-p-mentha-2, 8-dien-1-ol (1) with olivetol (2), (−)-CBD was obtained as the major product, in 55% yield as chromatographically pure oil, or in 41% yield as crystalline material (Scheme 14) [108].

No cyclization was observed, and side products were much more polar (14% yield) or much less polar (6% yield) than (−)-CBD.

Particularly, BF₃-etherate was added under nitrogen to a stirred suspension of basic aluminum oxide (Al₂O₃) in dry DCM, and after 15 min at room temperature and 1 min at 40–41 °C, (+)-p-mentha-2,8-dien-l-ol and olivetol dissolved in DCM were added. The reaction was quenched within 10 s with 10% aqueous solution of sodium bicarbonate (10 mL), and after evaporation of the organic extracts, (−)-CBD was obtained.

In 1988, Crombie et al. [120] reported the reaction of (1S,2S,3R,6R)-(+)-E-car-2-ene epoxide and that of p-menthadienol with olivetol in the presence of PTSA. Particularly, the starting material (1S,2S,3R,6R)-(+)-E-car-2-ene epoxide (3) was prepared starting from car-3-ene (1) by its treatment with potassium tert-butoxide to achieve the derivative 2, which after epoxidation with m-chloroperoxybenzoic acid provided the desired compound 3 (Scheme 15). Otherwise, p-menthadienol was prepared by citral with HCl in water or with PTSA in water/DCM (reaction not reported). The authors observed that in both cases, among other minor compounds, (−)-CBD, Abn-CBD, 1-THC and 6-THC were obtained. In particular, when (1S,2S,3R,6R)-(+)-E-car-2-ene epoxide was reacted with olivetol in benzene in the presence of docosane as catalyst for 45 min at 40 °C, the main product which formed was p-menthadienol with traces of (−)-CBD, Abn-CBD and THC. The further reaction of the obtained p-menthadienol with olivetol for 1h at 50 °C in benzene with docosane as well, afforded (−)-CBD in 30% yield, together with THC (18%), Δ⁸-THC (6%), and Abn-CBD (13%) (Scheme 15).

Rationally, the direct reaction of p-menthadienol (1) with olivetol in the conditions above reported afforded (−)-CBD in 30% yield (last part of Scheme 15).

Although in different conditions in terms of solvents, times, temperature, stereochemistry and catalysts, the reaction of p-menthadienol with olivetol was exploited by different research groups. Kinney et al. reacted E-(+)-p-mentha-2, 8-dien-1-ol with olivetol in toluene with PTSA for 1.5 h at 18–25 °C achieving (−)-CBD in 20% yield [121]. Also, Villano et al. in the year 2022 condensed olivetol with commercially available Z-(+)-p-mentha-2, 8-dien-1-ol in the presence of 33 mol% of wet PTSA in toluene at 0 °C for 3 h, affording a mixture of normal (−)-CBD and Abn-CBD, which were isolated in 26% and 38% yield respectively after a chromatographic column. Importantly, under these experimental conditions, no tricyclic structure was produced [122]. A different synthetic procedure was reported by Vaillancourt et al. in the year 1992 [123]. The authors described a new synthesis of (−)-CBD via the α-arylation of camphor, achieving both (−)-CBD and (−)-CBD mono methyl ether as shown in Scheme 16 [123].

Particularly, the authors prepared the endo-3-(2,5-dimethoxy-4-n-pentylphenyl)camphor (2), by reacting first olivetol dimethyl ether (2) dissolved in dry THF under nitrogen at −10 °C with tert-butyllithium (tert-BuLi) 1.7 M in hexanes for 3 h under stirring. Then the solution obtained was transferred to a solution of CuI in dry THF at 0 °C and the mixture was stirred for 20 min. Upon dilution with DMSO, the obtained solution was transferred dropwise to a solution of 3,9-dibromocamphor dissolved in dry THF/DMSO at 0 °C, and the reaction was then allowed to warm to room temperature and was stirred overnight. After the proper work-up and removal of the solvent in vacuo, the crude product was chromatographed and recrystallized from EtOH to achieve 2 in 71% yield. Compound 2 was transformed into the vinyl phosphate derivative 3, by dissolving it in dry THF under nitrogen and titrating the obtained solution cooled to −78 °C with a freshly prepared 0.4 M Na-naphthalenide/0.4 M tetra-ethyleneglycol dimethyl ether solution (see later) in THF until a deep green color persisted. The green mixture was then added with diethyl chlorophosphate and hexamethylfosforamide (HMPA), was allowed to warm to −20 °C, and opportunely treated to provide the crude product which was subjected to a short silica column obtaining the desired enol phosphate 3 as a colorless oil (89%). (−)-CBD and (−)-CBD monomethyl ether (4), were finally obtained by adding the vinyl phosphate 3 dissolved in dry THF and tert-butanol (t-BuOH) to an excess of lithium foil in methylamine (MeNH₂) at −78 °C. When addition was complete, the reaction was allowed to stir at −10 °C for 1 h and treated by acidification with HCl 1M. After extraction and removal of organic solvent the crude reaction mixture was chromatographed on neutral alumina, achieving (−)-CBD monomethyl ether in 43% yield. Further elution afforded (−)-CBD in 35% yield. The 0.4 M Na-naphthalenide/0.4 M tetra-ethylene glycol dimethyl ether solution in THF was prepared adding naphthalene in dry THF with sodium metal. The mixture was allowed to stir for 2 h, and then 2.41 mL of tetra-ethylene glycol dimethyl ether was added. The mixture was allowed to stir an additional hour at room temperature before use (Scheme not reported).

Later in the year 2002, Malkov et al. described the synthesis of (−)-CBD in 14–22% yield using Z-(+)-p-mentha-2, 8-dien-1-ol (1) or its acetate derivative 2 and olivetol in dichloromethane (DCM) and molybdenum catalysts (Scheme 17) [124].

According to the reported results, starting from Z-(+)-p-mentha-2, 8-dien-1-ol (1) and olivetol dissolved in DCM and using the molybdenum Mo (IV) triflate complex as catalyst at −20 °C for 3 h (−)-CBD was obtained in 20% yield. Similar results (22% yield) were obtained starting from Z-(+)-p-mentha-2, 8-dien-1-ol acetate (2), olivetol and the same catalyst, and stirring the reaction mixture at −10 °C for 30 min. On the contrary, with the bimetallic Mo (II) catalyst V, and stirring 2 and olivetol at 20 °C for 4 h (−)-CBD was obtained in a lower yield (14%). Kobayashi et al. in the year 2001 reported the BF₃-promoted 1,4-addition of bulky aryl groups, including dimethoxy olivetol, to an α-iodo enone (2), prepared from the parent enone (1), thus affording a β-aryl-α-iodo ketone derivative (3). Its subsequent reaction with EtMgBr furnished the magnesium enolate (4), which upon reactions with ClP(O)(OEt)₂ gave an enol phosphate (5), which was applied successfully to the synthesis of (−)-CBD (Scheme 18) [125].

Particularly, the enone 1 was converted to the α-iodocyclohexenone (2) with I₂ and pyridine in CCl₄ with good yield. The 1,4-addition of Ar₂Cu(CN)Li₂ to 2 promoted by BF₃·OEt₂ furnished ketone 3 in 67% yield, after aqueous workup and purification by chromatography. Then, EtMgBr was successful used to generate the corresponding enolate 4, which provided the enol phosphate 5 by reaction with (EtO)₂P(O)Cl in 70% yield (Scheme 18). Methylation of 5 with MeMgBr in the presence of nickel (II) acetylacetonate (Ni(acac)₂) afforded 6, and its subsequent exposure to sodium ethyl thiolate (EtSNa) in DMF resulted in the deprotection of the triethylsilil (TES) group and of one of the MeO groups to furnish 7. Attempted one-step deprotection of the two MeO groups under more vigorous conditions was unsuccessful, therefore 7 was converted in (−)-CBD by further exposure to EtSNa in DMF.

Later in 2006, the same group reported a new reagent system for synthesizing (−)-CBD and its analogues via alkenylation of cyclohexenyl diol monoacetate according to Scheme 19 [126].

Briefly, by a nickel-catalyzed allylation of 2-cyclohexene-1,4-diol monoacetate (1) with a new reagent consisting of (alkenyl)ZnCl/TMEDA, the SN2-type product, namely E-(+)-p-mentha-2,8-dien-1-ol (2) was achieved with 94% regioselectivity in good yield. Oxidation of 2 afforded the intermediate enone, which underwent iodination at the α position by I₂ in the presence of 2,5-di-tert-butylhydroquinone (DBHQ) as a radical scavenger to produce the α-iodo enone (3) in 63% yield (two steps). Addition of the 2,6-dimethoxy-4-pentylphenyl group of olivetol (abbreviated as Ar in the first part of the Scheme) to 3 was performed with the higher-order cyanocuprate derivative (Ar₂Cu(CN)Li₂) in turn synthesized from the Aryl (Ar) lithium anion and CuCN (not reported), obtaining compound 4, as a 1:1 stereoisomeric mixture at the α position. Compound 4 underwent reaction with EtMgBr to produce the reactive magnesium enolate 5, which was quenched with ClP(O)(OEt)₂ to furnish enol phosphate 6 in 51% yield from 3. Nickel-catalyzed coupling of 6 with MeMgCl afforded dimethyl ether 7 in good yield. Finally, (−)-CBD was obtained upon demethylation of 7 using MeMgI. Zachary et al. in the year 2018 reported a practical synthetic approach to synthetize Δ⁹-THC, and (−)-CBD. Particularly, (−)-CBD was synthesized according to Scheme 20 [127].

Briefly, olivetol and K₂CO₃ in acetone were added with dimethyl sulphate (Me₂SO₄) in 5 min at room temperature and then the mixture was heated to 80 °C for 12 h under argon, achieving the crude olivetol dimethyl ether (1) as an oil which was purified by column chromatography (98% yield). A yellow solution of 1 and TMEDA in anhydrous THF at −78 °C under argon was added with sec-butyllithium (sec-BuLi) and was stirred for 30 min at −78 °C and for 60 min at 0 °C, before being added with anhydrous DMF. The mixture was stirred at 0 °C for 30 min and for additional 60 min at room temperature. Upon proper work up and silica gel column chromatography the pure aldehyde derivative 2 was obtained as a yellow oil in 85% yield. Aldehyde 2 was converted in the enone 3 via an aldolic condensation with acetone in water using a 2.5M NaOH solution and heating the reaction mixture to 60 °C for 12 h (89% yield). The carbonylic group of 3 was reduced in toluene at −78 °C under argon, using a solution of (R)-CBS oxazaborolidine ligand (see Scheme 20) and BH₃•THF complex. The reaction mixture continued to stir for 30 min at −78 °C obtaining the crude product (−)-4 which was further purified by silica gel column chromatography achieving the pure compound (−)-4 in 94% yield (77% enantiomeric excess (e.e.)) as a clear colorless oil that solidified upon standing. An alternative to generate a product with high enantiopurity consisted of an enzymatic approach using an inexpensive and readily available enzyme. In this regard, compound 3 in the blue square was reduced with sodium borohydride (NaBH₄) affording the racemic alcohol (±)-4, which was acylated with vinyl butyrate in the presence of Savinase 12T thus providing the ester (−)-4.1 which was hydrolyzed with NaOH affording (−)-4 with >98% e.e. in 38% overall yield for the three steps (blue square in Scheme 20).

Compound (−)-4 was converted in the carboxylate (−)-5 by its acylation with 5-methyl-5-hexencarboxylic acid in DCM in the presence of DCC and DMAP. After 1 h stirring at 0 °C and then overnight at room temperature, (−)-5 was achieved as crude material which was purified by column chromatography. Compound (−)-5 was treated with KHMDS in anhydrous toluene at −78 °C for 1 h, then a solution of anhydrous pyridine and tetramethylsilyl chloride (TMS-Cl) in anhydrous toluene was added and the mixture was stirred at −78 °C for 10 min and at room temperature for an additional 4 h. Upon the Ireland−Claisen rearrangement, compound (+)-6 was achieved as a white crystalline solid that could be recrystallized using hexanes in a 52% overall yield. Treatment of (+)-6 in ether at 0 °C with methyllithium (MeLi) and stirring overnight at room temperature led to the formation of ketone (+)-7 as colorless oil in 71.8% yield, after chromatography column. Compound (+)-7 could be cyclized and then converted into (−)-CBD, via Wittig methylenation and deprotection. Particularly, compound (+)-7 and Grubbs 2nd generation catalyst in DCM were first stirred for a total of 15 h at 40 °C, thus achieving compound (−)-8 in 69.6% yield. Then, by reaction at room temperature of (−)-8 with bromo(methyl)triphenylphosphorane in THF, followed by the addition of potassium tert-butoxide and stirring at 75 °C for 12 h, compound (−)-9 was isolated in 82% yield. After its demethylated in anhydrous ether under argon with MeMgI and heating to 160 °C for 1.5 h, (−)-CBD was obtained in 62% (35% on three steps) yield as a light-yellow oil.

In the year 2020, Gong et al. reported a novel synthetic procedure for making (−)-CBD on a 10 g scale, by a late-stage diversification method, starting from commercially available phloroglucinol. First, the key intermediate (−)-CBD-2OPiv-OTf was achieved which underwent Negishi cross-coupling with the pentyl chain to give (−)-CBD in 52% overall yield. By this approach using the symmetric phloroglucinol the generation of positional isomers (Abn-CBDs) was avoided (Scheme 21) [128].

Briefly, by a Friedel–Crafts alkylation of phloroglucinol with Z-(+)-p-mentha-2,8-dien-1-ol in a ratio of 1:10 in presence of BF₃ etherate gave the desired product 1 in an excellent 80% yield. By treatment of 1 with trifluoromethanesulfonic anhydride (Tf₂O) in the presence of 2,6-lutidine at −30 to −20 °C in DCM using 1.5 equivalent of 1, to prevent double triflation, afforded the triflate derivative 2, which was isolated by silica gel column chromatography in a 78.1% yield. Compound 2 was treated with a solution of pivaloyl chloride (Piv-Cl) in DCM and a solution of DMAP in pyridine at −10 to 0 °C. Subsequently, the mixture was stirred at 25 °C for 12 h, to obtain the O-protected derivative 3 in 95% yield after column chromatography as a yellow oil. Pentyl zinc chloride (C₅H₁₁ZnCl), was prepared in one step by the transmetalation of the correspondent Grignard C₅H₁₁MgBr. Particularly, anhydrous zinc chloride and anhydrous lithium chloride were dissolved in anhydrous THF and cooled to −10 °C, added with C₅H₁₁MgBr and stirred first at −10 °C for 15 min and then at room temperature for 1.5 h. The obtained mixture was added with a solution of 3 in THF and then with [1,1′-Bis(diphenylphosphino)ferrocene]dichloropalladium(II) (Pd(dppf)Cl₂), as cross-coupling agent, to provide 4 in 90% yield after stirring at 55–60 °C for proper time and after column chromatography. (−)-CBD was finally obtained upon deprotection with MeMgBr (3 M in Et₂O) in toluene at 110 °C for 12 h in 99% yield after silica gel column chromatography.

Chiurchiu et al. in the year 2021 reported an innovative and high yielding continuous approach for producing (−)-CBD, strongly reducing its cyclization into THC (traces), thus achieving (−)-CBD in 55% yield. Particularly, by means of flow chemistry, and following their studies concerning the use of this technology for synthesizing highly functionalized materials, the authors inserted acetyl isoperitenol and olivetol dissolved in DCM into a reservoir A, while BF₃·Et₂O dissolved in DCM into a reservoir B. Subsequently, they pumped the two solutions simultaneously into a T-connector before passing through a 3 mL PTFE coil reactor (7 min residence time). The outgoing solution was dropped into a flask containing a stirring saturated solution of NaHCO₃ from which (−)-CBD was extracted and purified by silica gel chromatography to provide pure (−)-CBD in 55% yield (Scheme 22) [129].

The main by-products were Abn-CBD and the dialkylated cannabidiol recovered in 19 % and 4 % of yield respectively, THC was observed in traces (GC < 0.4 %) and its isolation was unfeasible. In the same year, Navarro et al. followed a practical approach to prepare (−)-CBD that avoided the formation of the side abnormal regio-isomers by using protected 4,6-dihalo-olivetol in coupling reaction as shown in Scheme 23 [76].

Briefly, the synthesis began by the Wittig reaction of butyl phosphonium bromide with commercially available 3,5-dimethoxybenzaldehyde (1), to deliver the olefin derivative 2 as mixture of Z and E isomers, which was conveniently reduced with hydrogen under pressure in presence of Pd/C as catalyst to the olivetol dimethyl ether (3). Regioselective electrophilic aromatic bromination of 3 using 2.3 equivalents of N-bromosuccinimide (NBS) in DCM at room temperature produced exclusively the 4,6-dibrominated product 4 in good yield. Then, the methyl ether-protecting groups were removed with boron tribromide to generate the key resorcinol intermediate 5 which was submitted to the Friedel–Craft alkylation with (1S,4R)-4-isopropenyl-1-methyl-2-cyclohexen-1-ol (6) in DCM, in the presence of PTSA as catalyst, thus affording the adduct 7 as single diastereomer. Finally, reductive dehalogenation using sodium sulfite in the presence of triethylamine (Et₃N) in a mixture of MeOH and H₂O at 75 °C delivered the targeted cannabinoid (−)-CBD in 43% yield. Anand et al. in the year 2022, developed a three-step concise and stereoselective synthesis route to (−)-CBD and (+)-CBD, using inexpensive and readily available starting material, such as R-(+)-limonene and S-(−)-limonene respectively. The synthesis involved the diastereoselective bi-functionalization of limonene, followed by effective elimination leading to the generation of the key chiral (+)- or (−)-p-mentha-2,8-dien-1-ols. Such dienols on coupling with olivetol under silver bis (trifluoromethanesulfonyl) imide (AgN(SO₂CF₃)₂) as catalysis provided regiospecific (−)-CBD or (+)-CBD in good yield (Scheme 24) [130].

Briefly, the present approach started with the direct generation of diastereoselective bi-functionalized 2-phenylseleninyl-p-menth-8-en-1-ol (1) from readily available and inexpensive starting material R-(+)-limonene. Particularly, electrophilic phenyl selenium bromide and H₂O₂ in a mixture acetonitrile/water at −30 °C were used, thus achieving compound 1 in 53% yield. The removal of SePh to synthesize (+)-menthadienol 2 was carried out using Selectfluor in THF at room temperature for 10 h achieving optically pure (+)-p-mentha-2,8-dien-1-ol 2 in 86% yield. Compound 2 was then coupled with olivetol in DCM using AgN(SO₂CF₃)₂ at room temperature for 10 h achieving (−)-CBD in 46% yield. Similarly, starting from S-(−)-limonene, (+)-CBD was synthesized. Briefly, the synthesis began from commercially available S-(−)-limonene, which was subjected to stereoselective bifunctionalization and elimination cascade to afford (−)-p-mentha-2,8-dien-1-ol. Pleasingly, (−)-p-mentha-2,8-dien-1-ol on reaction with olivetol in the presence of AgN(SO₂CF₃)₂ afforded the single isomer of (+)-CBD. The last synthetic procedure we report here to obtain (−)-CBD was very recently described by Grimm et al. (Scheme 25) [131].

Briefly, neral (Z-citral) was cyclized to isopiperitrol using imino-imidodiphosphates (iIDP), featuring a bifunctional inner-core system with an acidic P=NHTf moiety and a basic P=O moiety, thus combining excellent reactivity and selectivity, and furnishing (1R,6S)- E-isopiperitenol (1) in good yield (77%) and excellent diastereo- and enantioselectivity. Interestingly, such cyclization can be performed easily on a multigram scale (>4 g) without any loss of selectivity or yield, and catalyst can be recovered in excellent yield (95%) and re-used in further cyclization reactions. Then, direct access to (−)-CBD from isopiperitenol (1) and olivetol was provided in 35% yield under mild conditions using PTSA as catalyst in DCM at room temperature for 5 days. It is noteworthy that no further reaction of (−)-CBD to the corresponding THC was observed.

Following an overview scheme (Scheme 26) showing the synthetic procedures selected by us as the most convenient in terms of yields to obtain CBC (purple route) [103], CBD (pink route) [108] and CBG (amaranth route) [113]. Note that, while the synthetic paths leading to CBC and CBD are both one step processes shearing olivetol as reagent, that leading to CBG is a complicate multi-step process, using reagents completely different from those leading to CBC and CBD.