Efficient Synthesis for Altering Side Chain Length on Cannabinoid Molecules and Their Effects in Chemotherapy and Chemotherapeutic Induced Neuropathic Pain

(1) Background: Recently, a number of side chain length variants for tetrahydrocannabinol and cannabidiol have been identified in cannabis; however, the precursor to these molecules would be based upon cannabigerol (CBG). Because CBG, and its side chain variants, are rapidly converted to other cannabinoids in the plant, there are typically only small amounts in plant extracts, thus prohibiting investigations related to CBG and CBG variant therapeutic effects. (2) Methods: To overcome this, we developed an efficient synthesis of corresponding resorcinol fragments using the Wittig reaction which, under acid catalyzed coupling with geraniol, produced the desired side chain variants of CBG. These compounds were then tested in an animal model of chemotherapeutic-induced neuropathic pain and to reduce colorectal cancer cell viability. (3) Results: We found that all side-chain variants were similarly capable of reducing neuropathic pain in mice at a dose of 10 mg/kg. However, the molecules with shorter side chains (i.e., CBGV and CBGB) were better at reducing colorectal cancer cell viability. (4) Conclusions: The novel synthesis method developed here will be of utility for studying other side chain derivatives of minor cannabinoids such as cannabichromene, cannabinol, and cannabielsoin.


Introduction
In the United States, 150,000 new cases of colorectal cancer will be diagnosed in 2022 with an estimated 55,000 deaths caused by the disease [1]. Cisplatin, a commonly used chemotherapeutic agent, is used to treat colorectal cancer, albeit with negative side effects, including chemotherapeutic-induced peripheral neuropathy (CIPN), for which there is a lack of treatment options [2][3][4]. The prevalence of CIPN is agent-dependent, with reported rates varying from 19% to more than 85% [5] and is the highest in the case of platinumbased drugs (70-100%), including cisplatin [6]. This outcome results in an impaired quality of life for affected patients. Therefore, there is a need to identify alternative treatments for colorectal cancer, as well as identify anti-nociceptive agents capable of treating CIPN in patients undergoing cisplatin-based chemotherapy for colorectal cancer.

Materials and Methods
Thin-layer chromatography (TLC) was performed on aluminum-supported, precoated silica gel plates (EM Industries, Gibbstown, NJ, USA). Flash column chromatography was performed using silica gel SI 60. 1 H NMR spectra were recorded on a 500 MHz Bruker mass spectrometer (Billerica, MA, USA). Proton chemical shifts are reported in parts per million (δ). The following abbreviations were used to designate chemical shift multiplicities: s = singlet, d = doublet, dd = double doublet, t = triplet, dt = doublet of triplet, m = multiplet. Mass spectrometry analysis was performed on a 4000 Q-trap hybrid triple quadrupole/linear ion trap instrument (Applied Biosystems/MDS Sciex, Waltham, MA, USA) at the proteomic facility of the Penn State College of Medicine, Hershey, PA. High Resolution Mass Spectrometry (HRMS) was performed on AB Sciex TripleTOF 5600 mass spectrometer (Farmington, MA, USA) with electrospray ionization (ESI) in positive-ion mode at the metabolomics core at Penn State University, University Park, PA. The sample was analyzed by flow infusion with a Prominence UFLC system (Shimadzu, Kyoto, Japan) at flow 300 µL/min rates of 0.1% formic acid in a mixture of methanol and water 60:40. MS1 and MS2 data were acquired using a declustering potential (DP) of 80 V. MS2 data were collected in IDA mode with collision energy (CE) of 40 V and collision energy spread (CES) 20 V. During the analysis, an ion spray voltage (IS) of 5500 V, curtain gas (CUR) of 35 psi, nebulizer gas (GS1) of 50 psi, heater gas 2 (GS2) of 55, and heater temperature of 550 C were applied. CBG was purchased from a commercial source (Cayman Chemical, Ann Arbor, MI, USA). 3,5-dimethoxybenzyltriphenylphosphonium bromide (1) was prepared as reported in the literature [26]. Briefly, 3,5-dimethoxybenzylbromide (10 g, 43.2 mmol) was heated under reflux with triphenyl phosphine (12.6 g, 47.6 mmol) in toluene (60 mL) for 6 h to give a quantitative yield of compound 1. All starting materials were obtained from Aldrich Chemical Co. (Milwaukee, WI) and used without further purification. Synthesis of 5-alkyl substituted-1,3-dihydroxybenzene precursors was conducted as described below (and illustrated in Scheme 1). mmol) in toluene (60 mL) for 6 h to give a quantitative yield of compound 1. All starting materials were obtained from Aldrich Chemical Co. (Milwaukee, WI) and used without further purification. Synthesis of 5-alkyl substituted-1,3-dihydroxybenzene precursors was conducted as described below (and illustrated in Scheme 1). Scheme 1. Synthesis of 5-alkyl substitute-1,3-dihydroxybenezene.

1,3-Dihydroxy-5-propylbenzene (10)
A stirring solution of 6 (200 mg, 1.11 mmol) in CH 2 Cl 2 was cooled to −10 • C for 30 min. To this cold reaction solution, BBr 3 (1 M solution in CH 2 Cl 2 , 2.78 mL, 2.78 mmol) was added dropwise. After the addition, the mixture was allowed to warm up to room temperature and stirred for an additional 24 h. The mixture was cooled to 0 • C and quenched with saturated NaHCO 3 and stirred for an additional 30 min at room temperature. The aqueous layer was extracted with CH 2 Cl 2 (2 × 25 mL), combined organic layers were dried over MgSO 4 , filtered, and evaporated. Purification on a silica gel column by using CH 2 Cl 2 :MeOH (98:2) as an eluent gave 79% yield of 10 [27]. 1

Cannabigerovarin (CBGV, 14):
To a stirred cold solution of geraniol (203 mg,1.32 mmol) in anhydrous chloroform (20 mL) was added (over a period of 20 min) a solution of 1,3-dihydroxy-5-ethylbenzene (10) (131 mg, 0.73 mmol) and p-toluenesulfonic acid (14 mg) in anhydrous chloroform (15 mL) at 0 • C and under a positive pressure of nitrogen. After stirring in the same conditions for 14 h, the reaction was quenched with saturated of NaHCO 3 (5 mL). The reaction mixture was extracted with EtOAc (3 × 15 mL) and the combined organic layers were washed with water, dried (MgSO 4 ) and evaporated. The residue was purified on silica gel column by eluting with hexane:CH 2 Cl 2 (1:1) to give 21% yield of CBGV (14) [32]. 1  Cisplatin-induced peripheral neuropathy was induced as previously described [13,15,34]. Mice were injected with 4% sodium bicarbonate solution (1 mL, s.c.) administered just prior to cisplatin (5 mg/kg, i.p.) to prevent neurotoxicity and to minimize damage to renal function. Cisplatin was administered once a week for four weeks [15,35]. Mechanical allodynia was assessed before and after cisplatin treatment to confirm neuropathic pain state as described below.
Mechanical hypersensitivity was assessed using an electronic von Frey anesthesiometer (IITC Life Sciences Inc., Woodland Hills, CA, USA) with a semi-flex tip (IITC Life Sciences Inc., Woodland Hills, CA, USA), which was applied to the plantar surface of the right hind-paw with increasing force to prompt a paw withdrawal response. Mice were placed in one of eight small acrylic chambers placed on a wire mesh table (IITC Life Sciences Inc., Woodland, CA, USA), and acclimated to the chamber for 20 min before testing. The average of three tests were calculated with each test separated by at least 3 min.
To measure the effects of test compounds, 35 neuropathic male mice were randomly assigned to 7 groups (5 mice/group). During week 1, mice were injected with vehicle (DMSO, Tween 80, saline (1:1:18), i.p.), CBGV, CBGB, CBG, CBGP, CBGN, or the nonsteroidal anti-inflammatory drug indomethacin (10 mg/kg, i.p., based upon our previous work with CBG [15]). Mice were then given a one-week washout period, and during week 2 were again randomly assigned to one of the 7 groups and injected with the test compounds. All von Frey experiments were performed by experimenters blinded to treatments.

Cell lines
The human CRC cell lines SW480, SW620, HT29, DLD-1, HCT115, LS174, and RKO were obtained from the American Type Culture Collection (ATCC, Manassas, VA) and cultured as previously described [7]. Briefly, cells were grown in Dulbecco's Modified Eagle's Media supplemented with 10% fetal bovine serum, 2 mM Glutamax, 10 U/mL penicillin, 10 µg/mL streptomycin, and 0.25 µg/mL Amphotericin B at 37 • C in 5% CO 2 . RKO cells were grown under the same conditions except that RPMI was used in place of Dulbecco's Modified Eagle's Media.
CRC cell lines were treated as previously described, except cells were seeded at a density of 10,000 cells/well [7] and 16 h later treated with vehicle (DMSO), CBGV, CBGB, CBG, CBGP, CBGN at 10 µM for 48 h. In all treatments, the DMSO was maintained at a constant 1%. Results in two cell lines (HCT116 and SW480) were confirmed by trypan blue staining. Cells were plated and treated as described above and after 48 h adherent and nonadherent cells were collected and stained with 0.2% trypan blue; cells were counted on a Countess 3 automated cell counter (ThermoFisher, Pittsburgh, PA, USA). For dose effect curves, cells were seeded as described above and then treated with Vehicle (DMSO, CBGV, CBGB, CBG, CBGP, or CBGN at concentrations of 333 nM, 1 µM, 3.3 µM, 10 µM, 18.56 µM, 33 µM, and 56µM. Cell viability for all experiments was measured using the MTT ((3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2Htetrazolium), Biovision; Milpitas, CA). MTT (0.5 mg/mL, 15µL) was added to each well and incubated for 2 h at 37 • C in 5% CO 2 . Formazan crystals were solubilized by adding stop solution (10% Triton X-100, 0.05% HCl in isopropanol) and vigorously pipetting the mixture. Absorbance was measured at 570 nm on a FlexStation 3 (Molecular Devices, San Jose, CA, USA). For each experiment, the cell line/treatment was measured from triplicate wells and the average was determined. Data are presented as the signal normalized to vehicle control.

Statistics
All results are shown as mean ± standard deviation. Statistical significance was determined using GraphPad Prism Software (9.3.1, San Diego, CA, USA) using a one-way ANOVA with Dunnett's multiple comparison post hoc tests.

Results
First, to generate CBG and CBG variants, we synthesized corresponding resorcinol fragments. Several approaches have been reported in the literature for the synthesis of resorcinol derivatives [33,35]. We have developed an efficient method to generate the resorcinol fragments as shown in Scheme 1. This was accomplished in three steps involving: (1) preparation of corresponding olefins, 2-5 using 3,5-dimethoxybenzyl triphenylphosphonium bromide (1) and the corresponding aldehyde (Wittig reaction); (2) hydrogenation of the resultant E/Z-olefin mixture in Parr-hydrogenation apparatus to give compounds, 6-9; and (3) deprotection of methoxy group of compounds, 6-9 by using BBr 3 gave corresponding substituted resorcinol derivatives, 10-13. All of these operations overall gave moderate yields (40-80%).
Each of these corresponding substituted resorcinol derivatives, 10-13, were coupled with geraniol in the presence of catalytic amounts of p-toluenesulfonic acid monohydrate as shown in Scheme 2 to give the corresponding CBG analogs, 14-17, in 20-30% yields [36].
The solubility of CBGV and CBG is between 25-50 mg/mL or around 80-180 mM in pure DMSO; the solubility drops by 100-fold or more when the compounds are in 25% DMSO [37].
Given the potential of cannabinoids to evoke cancer cell death, we next investigated the effects of the CBG side-chain variants on cancer cell viability at 10 µM for 48 h. Following the 48-h timepoint, cell viability was assessed using the MTT assay. Our results show that CBGV, across all cell lines, demonstrated the greatest reduction in cell growth ( Figure 1A-G). Additionally, we found that the 4-carbon variant, CBGB, decreased cancer cell viability, although not to the same extent as CBGV. Moreover, we found that the effects of CBG were dependent upon the cell lines tested, as CBG only reduced colorectal cancer cell viability in 4 of the cell lines tested (HT-29, DLD-1, LS174, and RKO, Figure 1C,D,F,G). Even in cell lines sensitive to CBG, CBGV remained more efficacious, except in DLD-1 cells where the CBG and CBGV had a similar effect on cell viability ( Figure 1D). The longer 7 and 9 carbon variants (CBGP and CBGN) did not significantly influence cell viability in any of the cell lines tested. The results of the MTT assay were confirmed in HCT116 and SW480 cell lines by trypan blue staining (data not shown). To better examine the impact of these compounds on colorectal cancer cell growth, we performed dose effect curves. Consistent with our data at 10 μM, we found that CBGV had the lowest IC50 value across all cell lines, except DLD-1 cells (Figure 2A-G, Table 1). We also found that only in DLD-1 cells was the IC50 value of CBG similar to that found To better examine the impact of these compounds on colorectal cancer cell growth, we performed dose effect curves. Consistent with our data at 10 µM, we found that CBGV had the lowest IC 50 value across all cell lines, except DLD-1 cells (Figure 2A-G, Table 1). We also found that only in DLD-1 cells was the IC 50 value of CBG similar to that found with CBGV ( Figure 2D, Table 1). The highest IC 50 values were found for the molecules with the longer side chains (Figure 2, Table 1). In general, the IC 50 values for CBGB and CBG were found to be between those observed for CBGV and the larger chain molecules (CBGP and CBGN) (Figure 2, Table 1).   Next, we investigated the anti-nociceptive properties of CBG variants in a model of CIPN. Previously, we have demonstrated, using the von Frey test, that CBG (10 mg/kg i.p.) was effective at significantly reducing mechanical hypersensitivity in a preclinical model of CIPN [15]. Using this model, we investigated and compared the anti-nociceptive properties of the CBG side chain variants, CBGV, CBGB, CBG, CBGP, and CBGN. Neuropathic male mice were treated with vehicle control, a CBG variant, or the positive control indomethacin each at 10 mg/kg i.p. Mice then underwent von Frey testing of the hind-paw, 1 h following injections, to measure the force required to elicit a paw withdrawal response.

Discussion
Recently, a variant of CBD and THC has been identified in which the 5-carbon side chain is two carbons longer, these molecules were termed CBDP and THCP [22]. The authors went on to show that THCP binds to cannabinoid receptors with a higher affinity and was more effective at reducing pain than THC. In contrast, we did not see any impact of side-chain length on the ability of CBG to reduce pain in a mouse model of CIPN. However, we did find a significant anti-nociceptive effect of CBG and all CBG variants in a model of CIPN.
Our data indicate that side-chain length plays a role in the ability of CBG to reduce colorectal cancer cell growth in vitro. We found that, molecules with shorter side chains are more efficacious at reducing cell growth compared to longer side chains. Our findings with CBG are in contrast with previous work on CBDV and THCV which did not find a significant difference between these compounds and the more common 5-carbon variant (CBD and THC) [38 -40]. Furthermore, neither the 4 or 7 carbon variants of CBD was found to have any greater impact on breast cancer cell growth than the 5-carbon molecule [41]. This could be due to the unique nature of CBG, which has been found to be an agonist of α2-adrenogeric receptors [42] and peroxisome proliferator-activated receptors (PPAR) α

Discussion
Recently, a variant of CBD and THC has been identified in which the 5-carbon side chain is two carbons longer, these molecules were termed CBDP and THCP [22]. The authors went on to show that THCP binds to cannabinoid receptors with a higher affinity and was more effective at reducing pain than THC. In contrast, we did not see any impact of side-chain length on the ability of CBG to reduce pain in a mouse model of CIPN. However, we did find a significant anti-nociceptive effect of CBG and all CBG variants in a model of CIPN.
Our data indicate that side-chain length plays a role in the ability of CBG to reduce colorectal cancer cell growth in vitro. We found that, molecules with shorter side chains are more efficacious at reducing cell growth compared to longer side chains. Our findings with CBG are in contrast with previous work on CBDV and THCV which did not find a significant difference between these compounds and the more common 5-carbon variant (CBD and THC) [38][39][40]. Furthermore, neither the 4 or 7 carbon variants of CBD was found to have any greater impact on breast cancer cell growth than the 5-carbon molecule [41]. This could be due to the unique nature of CBG, which has been found to be an agonist of α2-adrenogeric receptors [42] and peroxisome proliferator-activated receptors (PPAR) α and γ [43][44][45][46], and the activation of these receptors has previously been reported to inhibit colorectal cancer cell growth.
It has been shown that side chain length can influence receptor binding for cannabinoids. For example, THCP has been shown to bind to CB1 receptors with a higher affinity than THC [22]. In contrast, THCV acts as an antagonist of the CB1 receptor, the opposite activity of THC, THCB, and THCP [24,47]. It is known that CBG binds with differing affinities and activities to a variety of receptors compared to THC and CBD [25]. One possible explanation for the differences we observe between the pain assay and cytotoxicity effects of these molecules is that different receptors mediate the analgesic and cytotoxic effects of CBG. Alternatively, the varying side chain lengths may create differing pharmacokinetics in vivo, thus normalizing their effects. Further studies on the binding of these novel molecules at known CBG receptors and additional studies on the mechanism that leads to cytotoxicity may provide useful insights into the mechanism by which CBGV and CBGB are slightly more cytotoxic but not more analgesic. These studies would also provide novel insights into how CBG interacts with known receptors.

Conclusions
We have developed a unique and adaptable process for generating cannabinoids with varying side chain lengths. Several recent reports have identified variants of CBD and THC with 3, 4, 6, and 7 carbon side chains; however, such side chain variants also likely exist for other cannabinoids such as cannabichromene (CBC) and must exist for CBG since this is the precursor molecule for the other cannabinoids. Surprisingly, we did not observe any effect of side-chain length regarding the ability to reduce neuropathic pain, which is in contrast to the data regarding THCP. However, we did find that CBG variants, such as CBG, produced significant anti-nociceptive effects in a murine model of CIPN. Importantly, we found that shorter side-chain variants of CBG were better able to reduce colorectal-cancer cell viability compared to longer-side chains.