Characterization of the Antitumor Potential of Extracts of Cannabis sativa Strains with High CBD Content in Human Neuroblastoma

Abstract

Cannabis has been used for decades as a palliative therapy in the treatment of cancer. This is because of its beneficial effects on the pain and nausea that patients can experience as a result of chemo/radiotherapy. Tetrahydrocannabinol and cannabidiol are the main compounds present in Cannabis sativa, and both exert their actions through a receptor-mediated mechanism and through a non-receptor-mediated mechanism, which modulates the formation of reactive oxygen species. These oxidative stress conditions might trigger lipidic changes, which would compromise cell membrane stability and viability. In this sense, numerous pieces of evidence describe a potential antitumor effect of cannabinoid compounds in different types of cancer, although controversial results limit their implementation. In order to further investigate the possible mechanism involved in the antitumoral effects of cannabinoids, three extracts isolated from Cannabis sativa strains with high cannabidiol content were analyzed. Cell mortality, cytochrome c oxidase activity and the lipid composition of SH-SY5Y cells were determined in the absence and presence of specific cannabinoid ligands, with and without antioxidant pre-treatment. The cell mortality induced by the extracts in this study appeared to be related to the inhibition of the cytochrome c oxidase activity and to the THC concentration. This effect on cell viability was similar to that observed with the cannabinoid agonist WIN55,212-2. The effect was partially blocked by the selective CB1 antagonist AM281, and the antioxidant α-tocopherol. Moreover, certain membrane lipids were affected by the extracts, which demonstrated the importance of oxidative stress in the potential antitumoral effects of cannabinoids.

Keywords: antitumor; cannabis; extracts; neuroblastoma.

Figure 1

Neuroblastoma cell death upon different treatments with synthetic cannabinoids (3.5 μg/mL AM281 and 3.5 μg/mL WIN55,212-2), with or without α-tocopherol pre-treatment (1 μM). Statistical test Brown Forsythe and Welch ANOVA test was done with a Dunnett post-hoc with α set as 0.05. p-value < 0.05 (*); p-value < 0.01 (**).

Figure 2

(A) Neuroblastoma cell death upon exposure to different Cannabis plant extracts (final concentration of each plant extract was 3.5 μg/mL), in control situation and exposed to WIN55,212-2 synthetic cannabinoid (3.5 μg/mL). Each condition was analyzed per duplicate. A one-way ANOVA with Tukey’s post-hoc was performed, with α set as 0.05, two-tailed. p-value < 0.01 (**). (B) Pearson correlation between neuroblastoma cell death expressed as percentage, and quantity of tetrahydrocannabinol, cannabidiol, tetrahydrocannabinol A (THCA), and cannabidiol A (CBDA), expressed as percentage.

Figure 3

Neuroblastoma cell death at 24 h, after different treatments, both with the presence and absence of synthetic cannabinoid WIN55,212-2, and with a variety of Cannabis plant extracts (D2-7, P0-3, and D2-2), with or without antioxidant pre-treatment. The two-way ANOVA with Tukey’s post-hoc was performed with α set as 0.05, two-tailed. p-value < 0.05 (*); <0.01 (**); <0.0001 (****).

Figure 4

Inhibition of cytochrome c oxidase, mediated by different Cannabis sativa plant extracts. (A) Dose–response curve of P0-3, D2-7, and D2-2 plant extracts, expressed as percentage of inhibition. Non-linear regression analysis was performed using different curve fits for each dataset. As statistical analysis, two-way ANOVA statistical test and Tukey post-hoc analysis were performed on the whole dataset; α was set at 0.05, two-tailed. Analysis of P0-3 is expressed with an asterisk (*) and analysis of D2-7 is expressed with a hash (#) p-value < 0.05 (*); <0.01 (##). (B) Inhibition percentage of P0-3, D2-7, and D2-2 at the whole extract concentration of 100 μg/mL. The group compared with them is indicated by the color code. As statistical analysis, one-way ANOVA statistical test was performed, α was set at 0.05, two tailed. p-value < 0.05 (*). (C) Pearson correlation test was performed between cytochrome c oxidase inhibition percentage, cell viability (expressed as percentage of neuroblastoma cell death), tetrahydrocannabinol content (expressed as percentage of THC content), and cannabidiol content. α was set at 0.05, two-tailed, and correlation was expressed as R-Pearson value.

Figure 5

Changes in lipid fingerprint due to exposure to Cannabis sativa plant extracts, expressed as percentage of relative intensity of each extract with respect to the control. (A) Glycerophospholipids (PC, PC-O, PC-P, and PA). (B) Sphingolipids and glycerophospholipids (SM, CerP, PE, PE-O, PE-P, PI, and Lyso-PIP2). Normality was tested for each lipid species separately by Bartlett test; α was set at 0.05. Student t-test or Wilcoxon test were performed on the parametric or non-parametric samples, respectively. The tests were two-tailed; α was set at 0.05. p-value < 0.05 (*); <0.01 (**); <0.001 (***); and <0.0001 (****). Black arrows indicate the absence of a lipid species in this sample. Abbreviations: glycerophosphocholine, ether and plasmalogen forms—PC, PC-O, and PC-P); glycerophosphatidic acid—PA; sphingomyelin—SM; ceramide phosphate—CerP; glycerophosphoethanolamine, ether form and plasmalogen form—PE, PE-O, and PE-P; glycerophosphatidilinositol—PI; and lysophosphoinositol bisphosphate—Lyso-PIP2.

Figure 6

Changes in lipid fingerprint due to α-tocopherol pre-treatment in cell homogenates of cells exposed to Cannabis sativa plant extracts, expressed as percentage of the relative intensity of each extract, compared to the control condition. (A) Glycerophospholipids (PC, PC-O, PC-P, and PA). (B) Sphingolipids and glycerophospholipids (SM, CerP, PE, PE-O, PE-P, PG, PI, Lyso-PIP, and Lyso-PIP2). Normality was tested for each lipid species separately by Bartlett test; α was set at 0.05. Student t-test or Wilcoxon test were performed on the parametric and non-parametric samples, respectively. The tests were two-tailed and α was set at 0.05. p-value < 0.05 (*); <0.01 (**); and <0.001 (***). Black arrows indicate absence of a lipid species in this sample. Abbreviations: glycerophosphocholine, ether and plasmalogen forms—PC, PC-O, and PC-P; glycerophosphatidic acid—PA; sphingomyelin—SM; ceramide phosphate—CerP; glycerophosphoglycerol—PG; glycerophosphoethanolamine, ether form and plasmalogen form—PE, PE-O, and PE-P; glycerophosphatidilinositol—PI; lysophosphoinositol phosphate—Lyso-PIP; and lysophosphoinositol bisphosphate—Lyso-PIP2.