Interactions between cannabidiol and Δ9 -tetrahydrocannabinol in modulating seizure susceptibility and survival in a mouse model of Dravet syndrome

Abstract

Background and purpose: Extracts from the cannabis plant can dramatically improve the health of children suffering from refractory epilepsies such as Dravet syndrome. These extracts typically contain cannabidiol (CBD), a phytocannabinoid with well-documented anticonvulsant effects, but may also contain Δ⁹ -tetrahydrocannabinol (Δ⁹ -THC). It is unclear whether the presence of Δ⁹ -THC modulates the anticonvulsant efficacy of CBD. Here, we utilized the Scn1a^+/- mouse model of Dravet syndrome to examine this question.

Experimental approach: Scn1a^+/- mice recapitulate core features of Dravet syndrome, including hyperthermia-induced seizures, early onset spontaneous seizures and sudden death. We assessed the effects on CBD and Δ⁹ -THC alone, and in combination on hyperthermia-induced seizures, spontaneous seizures and premature mortality.

Key results: Administered alone, CBD (100 mg·kg^-1 i.p.) was anticonvulsant against hyperthermia-induced seizures as were low (0.1 and 0.3 mg·kg^-1 i.p.) but not higher doses of Δ⁹ -THC. A subthreshold dose of CBD (12 mg·kg^-1 ) enhanced the anticonvulsant effects of Δ⁹ -THC (0.1 mg·kg^-1 ). Sub-chronic oral administration of Δ⁹ -THC or CBD alone did not affect spontaneous seizure frequency or mortality while, surprisingly, their co-administration increased the severity of spontaneous seizures and overall mortality.

Conclusion and implications: Low doses of Δ⁹ -THC are anticonvulsant against hyperthermia-induced seizures in Scn1a^+/- mice, effects that are enhanced by a sub-anticonvulsant dose of CBD. However, proconvulsant effects and increased premature mortality are observed when CBD and Δ⁹ -THC are sub-chronically dosed in combination. The possible explanations and implications of this are discussed.

FIGURE 1

Schematic overview of the anticonvulsant screening pipeline in Scn1a ^+/− mice. (a) Hyperthermia‐induced seizure experiments. (b) Spontaneous seizure and survival experimental design. Figure created using BioRender.com

FIGURE 2

Δ⁹‐THC treatment in Scn1a ^+/− mice. (a) Threshold temperature of individual mice for generalized tonic–clonic seizure (GTCS) induced by hyperthermia following acute intraperitoneal treatment with vehicle (VEH) or varying doses of THC (green bars). THC (0.1 and 0.3 mg·kg⁻¹) significantly increased the temperature threshold for hyperthermia‐induced seizures. The average temperatures of seizure induction are depicted by the bars, and error bars represent SEM, with n = 13–19 per group (^* P < 0.05; log‐rank Mantel–Cox). (b) Plasma concentrations of THC from individual experimental animals. THC plasma concentrations measured following acute intraperitoneal (i.p.) administration of THC in Scn1a ^+/− mice used in hyperthermia‐induced seizure experiments or following subchronic oral (p.o.) administration of 70 mg THC·kg⁻¹ chow. The average plasma THC concentrations are depicted by the bars, and error bars represent SEM, with n = 5–9 per group. (c) GTCS frequency of individual untreated and THC‐treated mice. Drug treatment administered orally through supplementation in chow was initiated following the induction of a single thermally induced seizure. Unprovoked, spontaneous GTCSs were quantified over a 60 h recording period. THC treatment had no effect on incidence or frequency of seizures, with n = 16–27 per group (Fisher's exact text and one‐way ANOVA followed by Bonferroni's post hoc, respectively). (d) Proportion of spontaneous GTCS with (grey bars) or without (white bars) full tonic hindlimb extension is depicted. Seizure severity was not affected by THC treatment (Fisher's exact test). Total number of spontaneous GTCS was 71 (untreated) and 45 (THC). (e) Survival curves comparing untreated and THC‐treated mice. Treatment began at postnatal day 18 (P18), and survival was monitored until P30. Survival of Scn1a ^+/− mice was not affected by THC treatment, with n = 16–27 per group (log‐rank Mantel–Cox)

FIGURE 3

CBD treatment in Scn1a ^+/− mice. (a) Threshold temperature of individual mice for generalized tonic–clonic seizure (GTCS) induced by hyperthermia following acute intraperitoneal treatment with varying doses of CBD (blue bars). CBD (100 mg·kg⁻¹) significantly increased the temperature threshold for hyperthermia‐induced seizures. The average temperatures of seizure induction are depicted by the bars, and error bars represent SEM, with n = 16–17 per group (^* P < 0.05; log‐rank Mantel–Cox). (b) Plasma concentrations of CBD from individual experimental animals. CBD plasma concentrations measured following acute intraperitoneal (i.p.) administration of CBD in Scn1a ^+/− mice used in hyperthermia‐induced seizure experiments or following subchronic oral (p.o.) administration of 3,500 mg CBD or 7,000 mg CBD·kg⁻¹ chow. The average plasma CBD concentrations are depicted by the bars and error bars represent SEM, with n = 5–11 per group. (c) Generalized tonic–clonic seizure (GTCS) frequency of individual untreated and CBD‐treated mice. Drug treatment administered orally through supplementation in chow was initiated following the induction of a single thermally induced seizure. Unprovoked, spontaneous GTCSs were quantified over a 60 h recording period. Neither CBD treatment had any effect on incidence or frequency of seizures, with n = 16–27 per group (Fisher's exact text and one‐way ANOVA followed by Bonferroni's post hoc, respectively). (d) Proportion of spontaneous GTCS with (grey bars) or without (white bars) full tonic hindlimb extension is depicted. Seizure severity was not affected by either CBD treatment (Fisher's exact test). Total number of spontaneous GTCS was 71 (untreated), 28 (3,500 mg CBD·kg⁻¹ chow), and 63 (7,000 mg CBD·kg⁻¹ chow). (e) Survival curves comparing untreated and CBD‐treated mice. Treatment began at postnatal day 18 (P18), and survival was monitored until P30. Survival of Scn1a ^+/− mice was not affected by CBD treatments, with n = 16–27 per group (log‐rank Mantel–Cox) (Note that untreated mice are replotted from Figure 2 for clarity.)

FIGURE 4

Combination Δ⁹‐THC and CBD treatment in Scn1a ^+/− mice. (a) Threshold temperature of individual mice for generalized tonic–clonic seizure (GTCS) induced by hyperthermia following acute treatment with CBD (blue bar) and THC (green bar) administered individually or in combination (salmon bar). Cannabinoids were administered as intraperitoneal injections. Co‐treatment with CBD and THC (12 and 0.1 mg·kg⁻¹, respectively) and THC (0.1 mg·kg⁻¹) resulted in a significantly improved response to thermal seizure induction compared to vehicle. Combination CBD and THC treatment was significantly more effective than either treatment alone, with n = 15–36 per group (^* P < 0.05, log‐rank Mantel–Cox). (Note that vehicle, CBD, and THC are replotted from Figures 2 and 3 for clarity.) (b) Plasma concentrations of THC (left panel) and CBD (right panel) from individual hyperthermia‐induced seizure experimental animals. Combination treatment with CBD and THC (12 + 0.1 mg·kg⁻¹) resulted in significantly higher plasma concentrations of CBD and THC (^* P < 0.05, Student's t test). Error bars represent SEM, with n = 5–10 per group. (c) Spontaneous GTCS frequency of individual untreated and CBD and THC co‐treated mice. Drug treatment was administered orally through supplementation in chow following the induction of a single thermally induced seizure. Unprovoked, spontaneous GTCSs were quantified. Co‐treatment with CBD and THC had no effect on incidence or frequency of seizures, with n = 17–27 per group (Fisher's exact text and one‐way ANOVA followed by Bonferroni's post hoc, respectively). (d) Proportion of spontaneous GTCS with or without full tonic hindlimb extension is depicted. Combination treatment with CBD and THC increased the severity of GTCS in Scn1a ^+/− mice. The proportion of spontaneous GTCS with tonic hindlimb extension was significantly greater in combination‐treated mice (salmon bar) compared to untreated controls (^* P < 0.05, Fisher's exact test). Total number of spontaneous GTCS was 71 (untreated) and 54 (CBD + THC). (e) Survival curves comparing untreated and cannabinoid‐treated mice. Co‐treatment with CBD and THC resulted in significantly worse survival (salmon line) of Scn1a ^+/− mice (^* P < 0.05, log‐rank Mantel–Cox). (f) Plasma concentrations of THC (left panel) and CBD (right panel) from individual experimental animals following sub‐chronic oral administration of cannabinoids in Scn1a ^+/− mice used in spontaneous seizure and survival experiments. (Note that CBD + THC [3,500 + 70 mg] concentrations were assayed in a separate cohort of treated wild‐type mice since there was no survival of experimental animals.) Combination treatment with CBD and THC (3,500 + 70 mg) resulted in significantly higher plasma levels of CBD and THC (^* P < 0.05, one‐way ANOVA followed by Tukey's post hoc). Error bars represent SEM, with n = 5–9 per group (Note that untreated, CBD [3,500 mg·kg⁻¹ chow], and THC [70 mg·kg⁻¹ chow] are replotted from Figures 2 and 3 for clarity.)

FIGURE 5

Higher dose of Δ⁹‐THC treatment in Scn1a ^+/− mice. (a) Plasma THC concentrations of individual mice measured following sub‐chronic treatment with a high dose of THC [200 mg·kg⁻¹ chow] compared to the toxic CBD + THC co‐treatment [3,500 + 70 mg·kg⁻¹ chow]. (n.s., non significant; unpaired Student's t‐test). (Note that CBD + THC [3,500 + 70 mg] is replotted from Figure 4 for clarity.) Error bars represent SEM, with n = 5–7 per group. (b) Spontaneous GTCS frequency of individual untreated and THC‐treated mice. THC treatment had no effect on incidence or frequency of seizures, with n = 17–27 per group (Fisher's exact test and one‐way ANOVA followed by Bonferroni's post hoc, respectively). (c) Proportion of spontaneous GTCS with or without full tonic hindlimb extension is depicted. The proportion of spontaneous GTCS with tonic hindlimb extension was not different following THC treatment compared to untreated controls (Fisher's exact test). Total number of spontaneous GTCS was 71 (untreated) and 12 (THC). (d) Survival curves comparing untreated and THC‐treated mice. THC had no effect on survival of Scn1a ^+/− mice, with n = 17–27 per group (log‐rank Mantel–Cox.)

FIGURE 6

Combined lower dose of Δ⁹‐THC and CBD treatment in Scn1a ^+/− mice. (a) Plasma concentrations of THC (left panel) and CBD (right panel) from individual experimental animals following sub‐chronic oral administration of cannabinoids in Scn1a ^+/− mice used in spontaneous seizure and survival experiments. (Note that THC [70 mg·kg⁻¹ chow] and CBD [3,500 mg·kg⁻¹ chow] are replotted from Figures 2 and 3 for clarity.) Error bars represent SEM, with n = 3–9 per group. Since only three CBD + THC (900 + 35 mg)‐treated mice survived, no statistical comparisons were made between groups. (b) Spontaneous generalized tonic–clonic seizure (GTCS) frequency of individual untreated and CBD and THC co‐treated mice. Drug treatment was administered orally through supplementation in chow following the induction of a single thermally induced seizure. Unprovoked, spontaneous GTCSs were quantified. Co‐treatment with CBD and THC had no effect on incidence or frequency of seizures, with n = 17–27 per group (Fisher's exact text and one‐way ANOVA followed by Bonferroni's post hoc, respectively). (Note that untreated mice are replotted from Figure 2 for clarity.) (c) Proportion of spontaneous GTCS with or without full tonic hindlimb extension is depicted. Combination treatment with the lower dose of CBD and THC (900 + 35 mg·kg⁻¹ chow) increased the severity of GTCS in Scn1a ^+/− mice. The proportion of spontaneous GTCS with tonic hindlimb extension was significantly greater in combination‐treated mice (salmon bar) compared to untreated controls (^* P < 0.05, Fisher's exact test). Total number of spontaneous GTCS was 71 (untreated) and 73 (CBD + THC). (d) Survival curves comparing untreated and cannabinoid‐treated mice. Co‐treatment with CBD and THC resulted in significantly worse survival (salmon dashed line) of Scn1a ^+/− mice (P < 0.05, log‐rank Mantel–Cox) (Note that untreated mice are replotted from Figure 2 for clarity.)