Credit: Zoonar GmbH/Alamy

Current drugs for epilepsy mainly target voltage-gated ion channels to reduce the aberrant neuronal excitability that underlies seizures, but almost one-third of patients are refractory to treatment. Now, reporting in Neuron, Nephi Stella and colleagues show that targeting the serine hydrolase α/β-hydrolase domain 6 (ABHD6), which is involved in endocannabinoid (eCB) metabolism, can protect against seizures in mouse models of epilepsy.

The eCB system in the brain controls presynaptic neurotransmitter release. The two main eCBs are anandamide and 2-arachidonoylglycerol (2-AG), which are produced 'on-demand' by specific lipases in response to synaptic activity, and are inactivated by specific hydrolases. In a process called presynaptic inhibition, eCBs produced by postsynaptic neurons can act retrogradely on CB1 receptors on presynaptic terminals, leading to a reduction of neurotransmitter release. Synthetic agonists of CB1 receptors have been shown to protect against seizures in rodent models. However, these agents can cause motor and cognitive impairment, precluding further development.

Based on their recent discovery that ABHD6 fine-tunes the production of 2-AG in the postsynaptic neuron, the authors investigated whether inhibitors of ABHD6 may boost presynaptic inhibition and exert anticonvulsant effects.

The small molecule WWL123 was identified in a screen for serine hydrolase inhibitors as a specific ABHD6 inhibitor, and found to be blood–brain barrier penetrant. In a mouse-model of pentylenetetrazole (PTZ)-induced seizures, pre-treatment with WWL123 by intraperitoneal injection blocked seizure-related mortality and reduced the severity and frequency of seizures. To test whether this effect was mediated via enhanced activity of 2-AG on CB1 receptors, the experiments were repeated in CB1-knockout mice. Surprisingly, WWL123 retained its anti-seizure effect in these mice, and also in mice that were deficient for CB2 (the other, predominantly non-neuronal, eCB receptor) — leading the authors to investigate which other receptor might reduce seizure risk in response to 2-AG.

As recent studies had shown that 2-AG can also act as a positive allosteric modulator of GABAA (γ-aminobutyric acid, type A) receptors, which are expressed postsynaptically and mediate inhibitory neurotransmission, WWL123 was tested in combination with the GABAA receptor antagonist picrotoxin in the PTZ-induced seizure model. Picrotoxin abolished the anti-seizure effects of WWL123, indicating a role for GABAA receptors in the mechanism of action of WWL123.

WWL123 was also tested in R6/2 mice, which develop spontaneous seizures. Mice were injected daily with WWL123, which reduced the incidence of spontaneous seizures. Combination treatment with the CB1 inhibitor SR1 did not antagonize the effect of WWL123, again demonstrating that there is no CB1 receptor involvement in the protective effect of WWL123.

Together, these data indicate that WWL123 exerts its anti-epileptic activity by increasing the levels of 2-AG through inhibition of its negative regulator ABHD6, leading to allosteric activation of GABAA receptors and increasing inhibitory neurotransmission, thereby reducing hyperexcitability. Importantly, unlike CB1 receptor agonists, the compound did not induce cognitive or motor impairment. Also, there was no evidence of tolerance induced by chronic treatment. This suggests that ABHD6 inhibitors may be amenable to long-term use, and have potential as a novel anti-epileptic strategy.