In vitro and in vivo pharmacological activity of minor cannabinoids isolated from Cannabis sativa

Abstract

The Cannabis sativa plant contains more than 120 cannabinoids. With the exceptions of ∆⁹-tetrahydrocannabinol (∆⁹-THC) and cannabidiol (CBD), comparatively little is known about the pharmacology of the less-abundant plant-derived (phyto) cannabinoids. The best-studied transducers of cannabinoid-dependent effects are type 1 and type 2 cannabinoid receptors (CB1R, CB2R). Partial agonism of CB1R by ∆⁹-THC is known to bring about the 'high' associated with Cannabis use, as well as the pain-, appetite-, and anxiety-modulating effects that are potentially therapeutic. CB2R activation by certain cannabinoids has been associated with anti-inflammatory activities. We assessed the activity of 8 phytocannabinoids at human CB1R, and CB2R in Chinese hamster ovary (CHO) cells stably expressing these receptors and in C57BL/6 mice in an attempt to better understand their pharmacodynamics. Specifically, ∆⁹-THC, ∆⁹-tetrahydrocannabinolic acid (∆⁹-THCa), ∆⁹-tetrahydrocannabivarin (THCV), CBD, cannabidiolic acid (CBDa), cannabidivarin (CBDV), cannabigerol (CBG), and cannabichromene (CBC) were evaluated. Compounds were assessed for their affinity to receptors, ability to inhibit cAMP accumulation, βarrestin2 recruitment, receptor selectivity, and ligand bias in cell culture; and cataleptic, hypothermic, anti-nociceptive, hypolocomotive, and anxiolytic effects in mice. Our data reveal partial agonist activity for many phytocannabinoids tested at CB1R and/or CB2R, as well as in vivo responses often associated with activation of CB1R. These data build on the growing body of literature showing cannabinoid receptor-dependent pharmacology for these less-abundant phytocannabinoids and are critical in understanding the complex and interactive pharmacology of Cannabis-derived molecules.

Figure 1

Cannabinoids assessed in this study. Chemical structures were drawn in Microsoft PowerPoint by the authors.

Figure 2

[³H]CP55,940 displacement from hCB1R CHO cell membranes. Compound activity was quantified for [³H]CP55,940 binding in CHO cells stably expressing hCB1R and treated with 0.1 nM–10 µM (a) THC-like compounds; (b) CBD-like compounds; or (c) CBG, or CBC. Data were fit to a variable slope (4 parameter) non-linear regression in GraphPad (v. 8). n ≥ 6 independent experiments performed in duplicate. Data are expressed as mean ± SEM. K_i and E_min are reported in Table 1.

Figure 3

hCB1R-dependent inhibition of FSK-stimulated cAMP accumulation. hCB1R-dependent inhibition of FSK-stimulated cAMP accumulation was quantified in HitHunter CHO cells stably expressing hCB1R and treated with 0.1 nM–10 µM (a) THC-like compounds; (b) CBD-like compounds; or (c) CBG, or CBC for 90 min. Data were fit to a variable slope (4 parameter) non-linear regression in GraphPad (v. 8). n ≥ 6 independent experiments performed in triplicate. E_Max refers to the top of the concentration–response curve. Data are expressed as mean ± SEM. EC₅₀ and E_max are reported in Table 2.

Figure 4

hCB1R-dependent recruitment of βarrestin2. hCB1R-dependent recruitment of βarrestin2 was quantified in PathHunter CHO cells stably expressing hCB1R and treated with 0.1 nM–10 µM (a) THC-like compounds; (b) CBD-like compounds; or (c) CBG, or CBC for 90 min. Data were fit to a variable slope (4 parameter) non-linear regression in GraphPad (v. 8). n ≥ 6 independent experiments performed in triplicate. E_Max refers to the top of the concentration–response curve. Data are expressed as mean ± SEM. EC₅₀ and E_max are reported in Table 3.

Figure 5

[³H]CP55,940 displacement from hCB2R CHO cell membranes. Compound activity was quantified for [³H]CP55,940 binding in CHO cells stably expressing hCB2R and treated with 0.1 nM–10 µM (a) THC-like compounds; (b) CBD-like compounds; or (c) CBG, or CBC. Data were fit to a variable slope (4 parameter) non-linear regression in GraphPad (v. 8). n ≥ 6 independent experiments performed in duplicate. Data are expressed as mean ± SEM. K_i and E_min are reported in Table 1.

Figure 6

hCB2R-dependent inhibition of FSK-stimulated cAMP accumulation. hCB2R-dependent inhibition of FSK-stimulated cAMP accumulation was quantified in HitHunter CHO cells stably expressing hCB2R and treated with 0.1 nM–10 µM (a) THC-like compounds; (b) CBD-like compounds; or (c) CBG, or CBC for 90 min. Data were fit to a variable slope (4 parameter) non-linear regression in GraphPad (v. 8). n ≥ 6 independent experiments performed in triplicate. E_Max refers to the top of the concentration–response curve. Data are expressed as mean ± SEM. EC₅₀ and E_max are reported in Table 2.

Figure 7

hCB2R-dependent recruitment of βarrestin2. hCB2R-dependent recruitment of βarrestin2 was quantified in PathHunter CHO cells stably expressing hCB2R and treated with 0.1 nM–10 µM (a) THC-like compounds; (b) CBD-like compounds; or (c) CBG, or CBC for 90 min. Data were fit to a variable slope (4 parameter) non-linear regression in GraphPad (v. 8). n ≥ 6 independent experiments performed in triplicate. E_Max refers to the top of the concentration–response curve. Data are expressed as mean ± SEM. EC₅₀ and E_max are reported in Table 3.

Figure 8

Analysis of ligand bias for phytocannabinoids at hCB1R (a) and hCB2R (b). Data from cAMP inhibition assays (Figs. 2, 5) and βarrestin2 recruitment assays (Figs. 3, 6) were fit to the operational model of Black and Leff (1983) using CP55,940 as the reference agonist and bias factor—reported here as ∆∆LogR—was calculated for each ligand between inhibition of FSK-stimulated cAMP accumulation and βarresitn2 recruitment. n ≥ 6 independent experiments performed in triplicate. Data are expressed as mean ± 95% CI (box) and minimum and maximum observed values (whiskers). *p < 0.05 relative to 0 as determined by non-overlapping 95% CI.

Figure 9

Acute tetrad effects in male C57BL/6 mice. Male mice aged 6–12 weeks were treated with 0.1–10 mg/kg i.p. of phytocannabinoids and assessed for catalepsy (a–c, 5 min post-injection), body temperature (d–f, 10 min post-injection), nociception in the tail flick assay (g–i, 15 min post-injection), and both locomotion (j–l) and time in the centre quadrant (m–o) in the OFT 1 h post-injection. THC-like compounds are shown in panels a, d, g, j, and m. CBD-like compounds are shown in panels b, e, h, k, and n. CBG, and CBC are shown in panels c, f, i, l, and o. (a–c) Catalepsy data are expressed as the % maximum possible effect (MPE, i.e. 60 s). (d–f) Body temperature data are expressed as change (∆) from baseline (°C). (g–i) Tail flick latency data are expressed as the % maximum possible effect (MPE, i.e. 20 s). (j–l) Locomotion data are expressed as % change from baseline total distance travelled. (m–o) Time in centre quadrant data are expressed as sec during 5 min OFT trials. n = 6/treatment. All data are expressed as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001 relative to vehicle for each assay as determined via one-way ANOVA followed by Tukey’s post-hoc analyses. Asterisk colour matches the treatment group assessed.