Identification of a new cannabidiol n-hexyl homolog in a medicinal cannabis variety with an antinociceptive activity in mice: cannabidihexol

The two most important and studied phytocannabinoids present in Cannabis sativa L. are undoubtedly cannabidiol (CBD), a non-psychotropic compound, but with other pharmacological properties, and Δ9-tetrahydrocannabinol (Δ9-THC), which instead possesses psychotropic activity and is responsible for the recreative use of hemp. Recently, the homolog series of both CBDs and THCs has been expanded by the isolation in a medicinal cannabis variety of four new phytocannabinoids possessing on the resorcinyl moiety a butyl-(in CBDB and Δ9-THCB) and a heptyl-(in CBDP and Δ9-THCP) aliphatic chain. In this work we report a new series of phytocannabinoids that fills the gap between the pentyl and heptyl homologs of CBD and Δ9-THC, bearing a n-hexyl side chain on the resorcinyl moiety that we named cannabidihexol (CBDH) and Δ9-tetrahydrocannabihexol (Δ9-THCH), respectively. However, some cannabinoids with the same molecular formula and molecular weight of CBDH and Δ9-THCH have been already identified and reported as monomethyl ether derivatives of the canonical phytocannabinoids, namely cannabigerol monomethyl ether (CBGM), cannabidiol monomethyl ether (CBDM) and Δ9-tetrahydrocannabinol monomethyl ether (Δ9-THCM). The unambiguously identification in cannabis extract of the n-hexyl homologues of CBD and Δ9-THC different from the corresponding methylated isomers (CBDM, CBGM and Δ9-THCM) was achieved by comparison of the retention time, molecular ion, and fragmentation spectra with those of the authentic standards obtained via stereoselective synthesis, and a semi-quantification of these cannabinoids in the FM2 medical cannabis variety was provided. Conversely, no trace of Δ9-THCM was detected. Moreover, CBDH was isolated by semipreparative HPLC and its identity was confirmed by comparison with the spectroscopic data of the corresponding synthetic standard. Thus, the proper recognition of CBDH, CBDM and Δ9-THCH closes the loop and might serve in the future for researchers to distinguish between these phytocannabinoids isomers that show a very similar analytical behaviour. Lastly, CBDH was assessed for biological tests in vivo showing interesting analgesic activity at low doses in mice.

To further complicate the already intricate scenario, we report a new series of phytocannabinoids that fills the gap between the pentyl and heptyl homologs of CBD and Δ 9 -THC, bearing a n-hexyl side chain on the resorcinyl moiety. At the best of our knowledge and according to the literature, no case of hexyl derivatives of cannabinoid has been reported so far. Conversely, cannabinoids with the same molecular formula and molecular weight have been classified as monomethyl ether derivatives of canonical phytocannabinoids, namely cannabigerol monomethyl ether (CBGM), cannabidiol monomethyl ether (CBDM) and Δ 9 -tetrahydrocannabinol monomethyl ether (Δ 9 -THCM) 9 . Whilst CBGM and CBDM have been already isolated and characterized 10,11 , Δ 9 -THCM has been detected in cannabis smoke 12 and some authors reported that it is present in the plant, but they were not able to isolate it due to chromatographic issues 13 . Our findings on the presence of the hexyl homologs of CBD and Δ 9 -THC, which we named cannabidihexol (CBDH) and Δ 9 -tetrahydrocannabihexol (Δ 9 -THCH) respectively, were supported by the stereoselective synthesis of the corresponding pure standards that are found in the plant prior to decarboxylation.

Identification of CBD and Δ 9 -THC hexyl homologs by UHPLC-HESI-Orbitrap.
In the attempt to provide a complete characterization of the FM2 cannabis variety, we noticed the presence of two major peaks at 18.13 and 20.21, and a minor one at 21.46 min, corresponding to the molecular formula C 23 H 32 O 4 , suggesting the presence of the carboxylic group. The analysis of the fragmentation spectra in negative ionization mode confirmed this hypothesis but showed three different fragmentation patterns (Fig. 1A). The two major peaks A and B presented very similar spectra differing only for the relative intensity of the fragments, whereas the minor peak C showed very poor fragmentation. Peak A at 18.13 min could be associated to a CBDA-like molecule (Fig. 1B), while peak B at 20.21 min had lower intensity for the fragment corresponding to [M-H 2 O] − at m/z 353 and presented a new fragment at m/z 178, not found in the other spectra (Fig. 1C). Peak C at 21.46 min showed a THCA-like fragmentation characterized by the very low intensity of the fragment [M-H 2 O] − and the absence of other major fragments besides the one corresponding to [M-CO 2 ] − at m/z 327 (Fig. 1D).
In order to identify these compounds, but being unable to isolate acidic species, we moved to work on the decarboxylated forms of such cannabinoids. Therefore, the ethanolic extract of FM2 was heated and the new mixture was analysed employing the same conditions by UHPLC-HESI-Orbitrap. As expected, in place of the previously detected peaks, three new peaks appeared at different retention times, 18.62, 20.62, and 20.77 min, with the molecular formula C 22 H 32 O 2 corresponding exactly to the loss of a CO 2 molecule. Figure 1E shows a second chromatogram with the decarboxylated compounds A d , B d , and C d . Surprisingly, peaks B d and C d had inverted elution order and peaks A d and C d presented superimposable fragmentation spectra in positive ionization mode with the same pattern as CBD and THC. Moreover, peaks A d and B d were very similar and differed for the relative intensity of the molecular ion [M+H] + at m/z 329 and the base peak at m/z 207. We concluded that peak A and B could be acidic CBD-type cannabinoids, whereas peak C could be an acidic THC-type cannabinoid ( Fig. 1F-H).
According to the literature, cannabinoids with such molecular formula and molecular ions are reported as monomethyl ethers of CBDA and THCA, named cannabidiolic acid monomethyl ether (CBDMA) and tetrahydrocannabinolic acid monomethyl ether (THCMA). Similarly, cannabinoids with molecular formula C 22 H 32 O 2 could be the corresponding decarboxylated derivatives, the already known cannabidiol monomethyl ether (CBDM) and the putative tetrahydrocannabinol monomethyl ether (THCM). However, we found three peaks corresponding to the same formula but different MS 2 spectra. By a comparison with other CBD and THC homologs present in our spectral library, such as cannabidivarin (CBDV), cannabidibutol (CBDB), cannabidiphorol (CBDP), Δ 9 -tetrahydrocannabivarin (Δ 9 -THCV), Δ 9 -tetrahydrocannabutol (Δ 9 -THCB), and Δ 9tetrahydrocannabiphorol (Δ 9 -THCP), we were able to putatively identify two new homologs of CBD and THC with a hexyl side chain. As shown in Fig. 2, the new compounds differ exactly by a -CH 2 unit (14 amu) from the corresponding pentyl (CBD and Δ 9 -THC) and the recently identified heptyl homologs (CBDP and THCP), not only for the molecular ion [M+H] + but also for all fragments. For both CBD ( Fig. 2A) and THC (Fig. 2B) homologs, it was evident that the molecular ion [M+H] + and the base peak inverted their relative intensity as the length of the side chain increased from the propyl to the heptyl homologs, most likely due to the increasing stability of the molecular ion.
In order to confirm the identity of these two new cannabinoids and unambiguously identify peak B d , a stereoselective synthesis of the putatively identified cannabinoids, which for sake of simplicity and consistency were called cannabidihexol (CBDH) and Δ 9 -tetrahydrocannabihexol (Δ 9 -THCH), and the monomethyl ether derivatives of CBD and Δ 9 -THC (CBDM and Δ 9 -THCM) was performed. Given 14 and cannabigerobutol (CBGB) 9 . The match of retention time (20.98 min) and fragmentation spectrum with those of synthetic CBGM and the comparison with spectral data reported in the literature for the same cannabinoid confirmed that the additional methyl group was attached to the oxygen of the resorcinyl moiety and was not part of the alkyl side chain.
(−)-trans-CBDH and (−)-trans-Δ 9 -THCH were obtained in 17% and 20% yield, respectively. The total yield of the two phytocannabinoids was 37%, which is in line with the yield obtained for the sole synthesis of CBDB or CBDP using the same procedure, but quenching the reaction after the consumption of the starting materials and before that CBDs started to isomerize into THCs (usually 30 min-1 h). Therefore, strictly monitoring the condensation of the appropriate resorcinol with (1S,4R)-1-methyl-4-(prop-1-en-2-yl)cycloex-2-enol it is possible to prepare in one-pot reaction both (−)-trans-CBDs and (−)-trans-Δ 9 -THCs, avoiding cumbersome and longer procedure to selectively prepare the (−)-trans-Δ 9 -THCs, without the awkward formation of their Δ 8 isomers. The synthesis of the monomethyl ether derivatives of CBD, THC and CBG is reported in Fig. 3A, Scheme 2. (−)-trans-CBDM and CBGM were easily prepared by methylation of the commercially available CBD and CBG by reaction with 0.5 equivalents of dimethylsulfate, in DMF at room temperature, using K 2 CO 3 as base ( A comparison of the retention time, molecular ion and fragmentation spectra of each pure synthesized standard with those found in FM2 led us to conclude that the first peak A d could be assigned to CBDH and the second one B d to CBDM (Fig. 3B). The third peak C d could most likely be associated to Δ 9 -THCH although its very low abundance and the presence of other interferents in the fragmentation spectrum from the FM2 extract did not allow an unambiguous assignment of its chemical structure (Fig. 3B). Moreover, no trace of Δ 9 -THCM was found. Fragmentation in negative ionization mode helped us to distinguish between CBDH and Δ 9 -THCH, which were identical in positive ionization mode, whereas no ionization was obtained in negative mode for Δ 9 -THCM due to the lack of free hydroxyl groups to be deprotonated (Fig. 3B). Confirmation of the identification of CBDH was achieved by isolation of pure fractions from the FM2 extract containing the acidic precursor CBDHA by semipreparative liquid chromatography. The pure compound was decarboxylated by heat and analysed by UHPLC-HESI-Orbitrap. Unfortunately, it was not possible to isolate fractions of FM2 containing THCHA due to its very low abundance. However, the stereoselective synthesis of Δ 9 -THCH allowed to assign a certain chemical structure to the corresponding peak in the FM2 sample.
Semi-quantification of CBDH and Δ 9 -THCH in the FM2 extract. Thanks to the synthetically prepared analytical standards of CBDH, Δ 9 -THCH, CBDM and CBGM, we were able to provide a semi-quantification of these cannabinoids in the FM2 cannabis variety by building the corresponding calibration curves. The results of concentration were in the order of the µg/g, while the main cannabinoids CBD and Δ 9 -THC were in the order of the mg/g (56 and 39 mg/g respectively). In particular, the hexyl homologs of CBD and Δ 9 -THC resulted 27 µg/g and 7 µg/g, while the methyl ether derivatives CBDM and CBGM were 50 µg/g and 102 µg/g. No Δ 9 -THCM was detected in the FM2.

Effects of CBDH on the formalin test in mice.
Formalin paw injection is a solid and widely used model of nociception with high face validity when tested with analgesic drugs. A nociceptive response to subcutaneous formalin induced an early, short-lasting first phase (0-7 min) followed by a quiescent period, and then a second, prolonged phase (15-60 min) of tonic hyperalgesia (Fig. 3C). In the tonic phase, two-way ANOVA revealed that CBDH (1, 2 mg/kg, i.p.) significantly reduced the late phase of the formalin-induced nocifensive behavior when compared to the vehicle-treated group (treatment F (4,288) = 17.32, P < 0.0001, time F ( www.nature.com/scientificreports/ interaction F (48,288) = 3.02, P < 0.0001); also, the dose of 2 mg/kg had a significant antinociceptive effect as compared to the vehicle group. The doses of 3 and 5 mg/kg had no effect on the formalin test (Fig. 3C).

Discussion
The comprehensive characterization of the chemical profile of a cannabis variety is a rather arduous task as the analytical tools in the chemist's hand are not able to cover such a broad range of compounds. However, the high sensitivity and selectivity of the high-resolution mass spectrometry, for example those achieved with the Orbitrap technology, can enable the identification of a reasonable number of molecules, even when present in very small traces. This approach allowed for the identification of new series of cannabinoids, CBD and THC homologs, with different lengths of the alkyl side chain, which were recently reported by our group [2][3][4]8 . The present work expanded the scope of cannabinoids identification completing the series of homologs with different alkyl side chain from three to seven methylene units. Up to now, only cannabinoids with an odd number of carbon atoms on the side chain have been reported and those with an even number of carbon atoms have been supposed to be artifacts derived from fungal ω-oxidation of their corresponding homologs 7 . The investigation of the origin of these species, such as those with a butyl and hexyl side chain, is beyond the scope of this work, but, although surprising, it is certain that such cannabinoids are actually present in a medicinal cannabis variety. The literature reports the existence of monomethyl ether derivatives of the canonical pentyl cannabinoids to justify the presence of compounds bearing an additional methyl group. Although on one side the structural identity of such derivatives was confirmed, our findings pointed out a new series of cannabinoids with the same molecular formula of the monomethyl ethers but with a different arrangement. Their origin, whether it is from the plant or from microorganisms, should be investigated as this might disclose new insights in the cannabis biochemistry. It is certainly important to underline that it is very easy to confuse CBDH and CBDM, as well as Δ 9 -THCH and Δ 9 -THCM. However, the match of the high-resolution fragmentation patterns with their pure synthetic standards was determinant to assign the respective chemical structure. This work might serve in the future for any researcher to distinguish between two species that show a very similar analytical behaviour. In a similar way, the methyl ether derivative of CBG was also identified (CBGM). It is worth noting that no Δ 9 -THCM was detected in the FM2 variety. On the other hand, both CBDM and CBGM showed a high peak as well as their native precursors CBDMA and CBGMA. Achieved results are in accordance with what reported by Lumır Ondrej Hanus et al., which showed that the cannabigerol monomethyl ether is always presents in greater quantities than its products, THCMA and CBDMA 7, 16 .
de Meijer et al. demonstrated that the cannabinoid acid synthases (THCAS, CBDAS) show a different affinity for CBGA alkyl homologues. This concept would explain the achieved results. CBDAS could be competitively stronger than THCAS when the substrate is CBGMA 17,18 .
The comparison of the results obtained for the concentrations of unorthodox cannabinoids in the FM2 variety suggested that there is no relationship between the different series as the same CBD to THC ratio is not respected. Considering our recent work on the heptyl derivatives of CBD and Δ 9 -THC, CBDP and Δ 9 -THCP were found in the FM2 at the concentration of 243 µg/g and 29 µg/g 4 ; whereas, in the same cannabis variety, the butyl series of CBD and THC was found at the concentration of 500 µg/g and 400 µg/g for CBDB and Δ 9 -THCB respectively 8 . In this work, we found 27 µg/g and 7 µg/g for CBDH and Δ 9 -THCH respectively. However, this data should be verified considering a larger number of different varieties in order to provide a reliable statistical significance.
The CBDH could have pleiotropic mechanisms of action through which it can exert its pharmacological effect. We found that the doses of the 1 and 2 mg/Kg significantly reduced the late phase of the formalin-induced nocifensive behavior, whereas the higher doses 3 and 5 mg/Kg were uneffective. This could be due, at least in part, assuming that at these doses CBDH can activate receptor facilitating nociception such as TRPV1 or other channels. On the other hand, we can speculate that CBDH at the higher doses could block receptors involved in antinociception such as CB1 or CB2. However, further pharmacological studies are needed to better investigate the pharmacodynamics profile of this interesting compound.
Another piece of knowledge towards understanding Cannabis Sativa L. cannabinoma has been added with this work. In particular, clarity has been made about the possible confusion between phytocannabinoids with a 6-term alkyl chain (CBDH, THCH) and those with a methylated resorcinolic hydroxyl group (CBDM, THCM). Furthermore, two new phytocannabinoids CBDH and THCH have been identified in the FM2 variety by Data are represented as means ± SEM (n = 5-6). +,+++ indicate statistically significant differences versus veh/form, p < 0.05 and p < 0.001, respectively. 2-way ANOVA followed by Bonferroni's post hoc tests was used for statistical analysis.

Scientific Reports
| (2020) 10:22019 | https://doi.org/10.1038/s41598-020-79042-2 www.nature.com/scientificreports/ comparison with their respective authentic synthesized compounds. In particular, CBDH has been isolated and its pharmacological activity has been evaluated in vivo in mice. At extremely low doses (1 mg/kg) it showed an interesting nocifensive activity. However, the CBDH concentration of 27 μg/g found in the FM2 variety is too low to exert the pharmacological effect but it is not excluded that other cannabis varieties may contain higher concentrations. More in-depth pharmacological studies are currently underway to clarify the mechanism of action of this new phytocannabinoid.

UHPLC-HESI-Orbitrap metabolomic analysis. Analyses on FM2 extracts were performed on a Thermo
Fisher Scientific Ultimate 3000 provided with a vacuum degasser, a binary pump, a thermostated autosampler, a thermostated column compartment and interfaced to a heated electrospray ionization source and a Q-Exactive Orbitrap mass spectrometer (UHPLC-HESI-Orbitrap). The HESI and Orbitrap parameters were set following our previous work 4 . Briefly, the capillary temperature was set at 320 °C, the vaporizer temperature at 280 °C, the electrospray voltage at 4. The chromatographic separation was carried out on a core shell C 18 stationary phase (Poroshell 120 SB-C18, 3.0 × 100 mm, 2.7 µm, Agilent, Milan, Italy) following the conditions employed for our previous work 4 .
A semi-quantitative analysis of Δ 9 -THC and CBD, their hexyl homologs CBDH and Δ 9 -THCH, and the methyl ether derivatives of CBD and CBG, CBDM and CBGM, was carried out using a calibration curve with the external standard method. A stock solution of CBD and Δ 9 -THC (1 mg/mL) was properly diluted to obtain five non-zero calibration points at the final concentrations of 50, 100, 250, 500 and 1000 ng/mL; a stock solution of CBDH, CBDM, Δ 9 -THCH, CBDM and CBGM was diluted to obtain the final concentrations of 5, 25, 50, 100 and 250 ng/mL. The linearity was assessed by the coefficient of determination (R 2 ), which was greater than 0.992 for each analyte.