Cannabis sativa extracts protect LDL from Cu2+-mediated oxidation

Abstract

Background: Multiple therapeutic properties have been attributed to Cannabis sativa. However, further research is required to unveil the medicinal potential of Cannabis and the relationship between biological activity and chemical profile.

Objectives: The primary objective of this study was to characterize the chemical profile and antioxidant properties of three varieties of Cannabis sativa available in Uruguay during progressive stages of maturation.

Methods: Fresh samples of female inflorescences from three stable Cannabis sativa phenotypes, collected at different time points during the end of the flowering period were analyzed. Chemical characterization of chloroform extracts was performed by ¹H-NMR. The antioxidant properties of the cannabis sativa extracts, and pure cannabinoids, were measured in a Cu²⁺-induced LDL oxidation assay.

Results: The main cannabinoids in the youngest inflorescences were tetrahydrocannabinolic acid (THC-A, 242 ± 62 mg/g) and tetrahydrocannabinol (THC, 7.3 ± 6.5 mg/g). Cannabinoid levels increased more than twice in two of the mature samples. A third sample showed a lower and constant concentration of THC-A and THC (177 ± 25 and 1 ± 1, respectively). The THC-A/THC rich cannabis extracts increased the latency phase of LDL oxidation by a factor of 1.2-3.5 per μg, and slowed down the propagation phase of lipoperoxidation (IC₅₀ 1.7-4.6 μg/mL). Hemp, a cannabidiol (CBD, 198 mg/g) and cannabidiolic acid (CBD-A, 92 mg/g) rich variety, also prevented the formation of conjugated dienes during LDL oxidation. In fact, 1 μg of extract was able to stretch the latency phase 3.7 times and also to significantly reduce the steepness of the propagation phase (IC₅₀ of 8 μg/mL). Synthetic THC lengthened the duration of the lag phase by a factor of 21 per μg, while for the propagation phase showed an IC₅₀ ≤ 1 μg/mL. Conversely, THC-A was unable to improve any parameter. Meanwhile, the presence of 1 μg of pure CBD and CBD-A increased the initial latency phase 4.8 and 9.4 times, respectively, but did not have an effect on the propagation phase.

Conclusion: Cannabis whole extracts acted on both phases of lipid oxidation in copper challenged LDL. Those effects were just partially related with the content of cannabinoids and partially recapitulated by isolated pure cannabinoids. Our results support the potentially beneficial effects of cannabis sativa whole extracts on the initial phase of atherosclerosis.

Keywords: Cannabis sativa; atherosclerosis; hemp; low-density lipoprotein; maturation; oxidation; phytocannabinoids.

Fig. 1

Mechanism of unsaturated fatty acid oxidation. In the first step, the unsaturated fatty acid (LH) is attacked by an oxidant (Ox^●), abstracting a bis-allylic hydrogen and giving rise to an alkyl radical (L^●), this radical can react with O₂ generating a peroxyl radical (LOO^●). The propagation phase is triggered and perpetuated by the reaction of L^● and LOO^● with reduced lipid molecules. Antioxidants (XH) can act preventing the initiation step or participating in the termination phase of lipid oxidation

Fig. 2

Temporal evolution of the glandular trichomes. Cannabis sativa inflorescences were captured with a digital microscope immediately before the collect for analysis at 15 (sample 1), 10 (sample 2) and 0 (sample 3) days before regular harvest time and trichomes dried for 30 days after harvest (sample 4)

Fig. 3

¹H-RMN spectra and a zoom in showing the hydrogen signals used for quantification. Details of the spectra from SB3 is shown as an example. The H10: 6.39 ppm for THC-A and H2: 6.14 ppm for THC are shown (a). The amount of THC-A (b) and THC (c) per gram of extract determined by ¹H-RMN from the inflorescences obtained during the first (white bars), second (grey bars), and third harvest (dashed bars) and after drying of the mature inflorescences (black bars). Statistically significant differences against the first harvest were found by one-way ANOVA followed by Sidak-Bonferroni multiple comparisons test (*p < 0.05, ** p < 0.01 **** p < 0.0001)

Fig. 4

UV-Vis spectra of Cannabis sativa extracts. a The spectrum of the SB3 extract at 0.6; 0.8; 1.25; 1.7 mg/mL in 100 mM phosphate pH 7.4 is shown as an example. The inset shows the linear regression of absorbance at 257 nm (●) and 298 nm (♦) against the concentration of extract. b Correlation between the levels of THC-A determined by ¹H-RMN and the extinction coefficients (ε) at 257 nm (○) and 298 nm (●) determined from the slope of plots as the one represented in the inset of (a)

Fig. 5

Conjugates dienes from LDL oxidation. a LDL (0.1 mg/mL) was exposed to CuSO₄ (50 μM), and the formation of conjugated dienes was followed at 234 nm in the absence (dashed line) and the presence of cannabis extracts (continuous lines). Number over the curves represent the concentration of extract in μg/mL. Representative curves obtained in the presence of extracts from SB3 (1.25–5 μg/mL) are shown. The dotted line represents control LDL assayed in the same condition but without the addition of cupper ions. b Linear regression of latency ratios and the amount of SB3. The slope of these graphs (antioxidant capacity (AC/μg)) are summarized in Table 3. c Correlation between the AC/μg and the concentration of THC (R² = 0.43). and THC-A (R² = 0.49) in each extract. d Representative plot of percentage of protection, determined as described in Materials and Methods, against the concentration of SB3. Analogous plots obtained for each extract were used to obtain the values of IC₅₀ shown in Table 3

Fig. 6

Effect of hemp extracts on LDL oxidation. a LDL was oxidized as in Fig. 5 and the time course of diene production was followed at 234 nm. Sigmoidal kinetics were obtained for the condition without (dashed line) and with hemp extracts (continuous lines). Number over the curves represent concentration of extract in μg/mL. b Linear regression of latency ratios against μg of hemp were used to calculate the AC/μg reported in Table 4. c A plot of percentage of protection against concentration was used to determine the IC₅₀ reported in Table 4