Cannabis-based products are increasingly being used to treat refractory childhood epilepsies such as Dravet syndrome. Cannabis contains at least 140 terpenophenolic compounds known as phytocannabinoids. These include the known anticonvulsant compound cannabidiol (CBD) and several molecules showing emergent anticonvulsant properties in animal models. Cannabichromene (CBC) is a phytocannabinoid frequently detected in artisanal cannabis oils used in the community by childhood epilepsy patients. Here we examined the brain and plasma pharmacokinetic profiles of CBC, cannabichromenic acid (CBCA), cannabichromevarin (CBCV), and cannabichromevarinic acid (CBCVA) following intraperitoneal administration in mice. The anticonvulsant potential of each was then tested against hyperthermia-induced seizures in the Scn1a +/– mouse model of Dravet syndrome. All phytocannabinoids within the CBC series were readily absorbed and showed substantial brain penetration (brain–plasma ratios ranging from 0.2 to 5.8). Anticonvulsant efficacy was evident with CBC, CBCA, and CBCVA, each significantly increasing the temperature threshold at which Scn1a +/– mice had a generalized tonic–clonic seizure. We synthesized a fluorinated derivative of CBC (5-fluoro-CBC), which showed improved brain penetration relative to the parent CBC molecule but not any greater anticonvulsant effect. Since CBC and derivatives are anticonvulsant in a model of intractable pediatric epilepsy, they may constitute part of the mechanism through which artisanal cannabis oils are anticonvulsant in patients.

Objective Cannabigerolic acid (CBGA), a precursor cannabinoid in Cannabis sativa, has recently been found to have anticonvulsant properties in the Scn1a+/- mouse model of Dravet syndrome. Poor brain penetration and chemical instability of CBGA limits its potential as an anticonvulsant therapy. Here, we examined whether CBGA methyl ester, a more stable analogue of CBGA, might have superior pharmacokinetic and anticonvulsant properties. In addition, we examined whether olivetolic acid, the biosynthetic precursor to CBGA with a truncated (des-geranyl) form, might possess minimum structural requirements for anticonvulsant activity. We also examined whether olivetolic acid and CBGA methyl ester retain activity at the epilepsy-relevant drug targets of CBGA: G-protein-coupled receptor 55 (GPR55) and T-type calcium channels. Methods The brain and plasma pharmacokinetic profiles of CBGA methyl ester and olivetolic acid were examined following 10 mg/kg intraperitoneal (i.p.) administration in mice (n = 4). The anticonvulsant potential of each was examined in male and female Scn1a+/- mice (n = 17–19) against hyperthermia-induced seizures (10–100 mg/kg, i.p.). CBGA methyl ester and olivetolic acid were also screened in vitro against T-type calcium channels and GPR55 using intracellular calcium and ERK phosphorylation assays, respectively. Results CBGA methyl ester exhibited relatively limited brain penetration (13%), although somewhat superior to that of 2% for CBGA. No anticonvulsant effects were observed against thermally induced seizures in Scn1a+/- mice. Olivetolic acid also showed poor brain penetration (1%) but had a modest anticonvulsant effect in Scn1a+/- mice increasing the thermally induced seizure temperature threshold by approximately 0.4°C at a dose of 100 mg/kg. Neither CBGA methyl ester nor olivetolic acid displayed pharmacological activity at GPR55 or T-type calcium channels. Conclusions Olivetolic acid displayed modest anticonvulsant activity against hyperthermia-induced seizures in the Scn1a+/- mouse model of Dravet syndrome despite poor brain penetration. The effect was, however, comparable to the known anticonvulsant cannabinoid cannabidiol in this model. Future studies could explore the anticonvulsant mechanism(s) of action of olivetolic acid and examine whether its anticonvulsant effect extends to other seizure types.

A purified preparation of cannabidiol (CBD), a cannabis constituent, has been approved for the treatment of intractable childhood epilepsies such as Dravet syndrome. Extensive pharmacological characterization of CBD shows activity at numerous molecular targets but its anticonvulsant mechanism(s) of action is yet to be delineated. Many suggest that the anticonvulsant action of CBD is the result of G protein-coupled receptor 55 (GPR55) inhibition. Here we assessed whether Gpr55 contributes to the strain-dependent seizure phenotypes of the Scn1a+/- mouse model of Dravet syndrome. The Scn1a+/- mice on a 129S6/SvEvTac (129) genetic background have no overt phenotype, while those on a [129 x C57BL/6J] F1 background exhibit a severe phenotype that includes hyperthermia-induced seizures, spontaneous seizures and reduced survival. We observed greater Gpr55 transcript expression in the cortex and hippocampus of mice on the seizure-susceptible F1 background compared to those on the seizure-resistant 129 genetic background, suggesting that Gpr55 might be a genetic modifier of Scn1a+/- mice. We examined the effect of heterozygous genetic deletion of Gpr55 and pharmacological inhibition of GPR55 on the seizure phenotypes of F1.Scn1a+/- mice. Heterozygous Gpr55 deletion and inhibition of GPR55 with CID2921524 did not affect the temperature threshold of a thermally-induced seizure in F1.Scn1a+/- mice. Neither was there an effect of heterozygous Gpr55 deletion observed on spontaneous seizure frequency or survival of F1.Scn1a+/- mice. Our results suggest that GPR55 antagonism may not be a suitable anticonvulsant target for Dravet syndrome drug development programs, although future research is needed to provide more definitive conclusions.

ObjectiveIctal vocalizations have shown diagnostic utility in epilepsy patients. Audio recordings of seizures have also been used for seizure detection. The present study aimed to determine whether generalized tonic–clonic seizures in the Scn1a+/− mouse model of Dravet syndrome are associated with either audible mouse squeaks or ultrasonic vocalizations.MethodsAcoustic recordings were captured from group‐housed Scn1a+/− mice undergoing video‐monitoring to quantify spontaneous seizure frequency. We generated audio clips (n = 129) during a generalized tonic–clonic seizure (GTCS) that included 30 seconds immediately prior to the GTCS (preictal) and 30 seconds following the conclusion of the seizure (postictal). Nonseizure clips (n = 129) were also exported from the acoustic recordings. A blinded reviewer manually reviewed the audio clips, and vocalizations were identified as either an audible (<20 kHz) mouse squeak or ultrasonic (>20 kHz).ResultsSpontaneous GTCS in Scn1a+/− mice were associated with a significantly higher number of total vocalizations. The number of audible mouse squeaks was significantly greater with GTCS activity. Nearly all (98%) the seizure clips contained ultrasonic vocalizations, whereas ultrasonic vocalizations were present in only 57% of nonseizure clips. The ultrasonic vocalizations emitted in the seizure clips were at a significantly higher frequency and were nearly twice as long in duration as those emitted in the nonseizure clips. Audible mouse squeaks were primarily emitted during the preictal phase. The greatest number of ultrasonic vocalizations was detected during the ictal phase.SignificanceOur study shows that ictal vocalizations are exhibited by Scn1a+/− mice. Quantitative audio analysis could be developed as a seizure detection tool for the Scn1a+/− mouse model of Dravet syndrome.