Endocannabinoids (eCBs) are a family of molecules derived from membrane phospholipids which exert biological effects through specific receptors. In the brain, eCBs are perhaps the most ubiquitous and potent neurotransmitters known to act in a retrograde manner. Within the field of endocannabinoid research, great effort has been directed towards dissecting metabolic pathways that regulate the production and degradation of eCBs. While much of this has been driven by the desire to produce better and more specific therapeutic targets, a happy by-product is, for basic neuroscientists, a better understanding of how eCBs function as signalling entities at discrete synapses in the brain. Zhong and colleagues, in a recent issue of The Journal of Physiology, provide such evidence using a recently created mouse harbouring genetic deletion of monacylglycerol lipase (MAGL), thought to be the primary degradation enzyme for the neurally abundant endogenous cannabinoid 2-arachidonoyl glycerol (2-AG) (Zhong et al. 2011). Postsynaptic activity (firing or depolarization) or G-protein-coupled receptor activation results in the production and liberation of 2-AG which targets CB1 receptors (CB1Rs) on pre-synaptic terminals to inhibit the release of neurotransmitter (Fig. 1). This results in a number of temporally distinct forms of synaptic plasticity induced and expressed in a variety of ways. Additionally, eCB signalling is also remarkably plastic following experience, at both a synaptic and whole organism level. This fantastic variety of signalling outcomes suggests that many factors may regulate the how's and why's of endocannabinoid signalling at specific synapses. Figure 1 Overview of 2-arachidonoyl glycerol modulation of cerebellar glutamatergic synapses and changes induced by genetic deletion of monoacylglycerol lipase In 2007, Blankman and colleagues showed that 2-AG is degraded by a number of serine hydrolases in the brain; a lion's share of this, 85%, is attributed to MAGL (Blankman et al. 2007). Zhong et al., using a recently developed MAGL−/− mouse Schlosburg et al. 2010 provide direct evidence that 2-AG degradation is required for normal eCB signalling at glutamate synapses in the cerebellum. This is consistent with earlier work using pharmacological inhibitors and MAGL−/− mice to show a dominant role for MAGL in vitro and in vivo (Chanda et al. 2010; Hashimotodani et al. 2007; Pan et al. 2009; Straiker et al. 2009). Here, Zhong et al. use depolarisation-induced suppression of excitation (DSE), a broadly applied electrophysiological protocol, to assay the nature and integrity of eCB signalling at glutamate synapses. They show that two key features of DSE at climbing fibre and parallel fibre synapses onto Purkinje cells in the cerebellum are altered in MAGL−/− mice. First, DSE lasts nearly three times longer in slices from MAGL−/−. Second, the maximal magnitude of DSE is blunted in older, but not younger mice. The acute addition of MAGL inhibitor JZL184 to slices prolonged DSE in MAGL+/+ mice, but this effect was occluded in MAGL−/− mice (Fig. 1). Interestingly, an inhibitor of ABDH6, another serine hydrolase, was ineffective. The authors also report prolongation of two forms of metabotropic glutamate receptor-driven 2-AG production at parallel fibre synapses. They then address two scenarios which may exist under conditions of 2-AG excess: increased tonic activation and/or desensitization of CB1Rs. Consistent with tonic CB1R activation, Zhong et al. report that basal synaptic properties are altered in MAGL−/− mice. Specifically, MAGL−/− parallel fibre synapses are less likely to release glutamate as shown by an increase in paired-pulse facilitation and a reduction in the input−output relationship. The reduced glutamate release probability at MAGL−/− synapses is reversed by addition of a CB1R antagonist, suggesting tonic activation of the receptor. In parallel, the response of MAGL−/− synapses to saturating doses of a CB1R agonist is reduced, consistent with partial desensitization of CB1Rs. This evidence indicates that deficient clearance of 2-AG from the synapse results in persistent activation of CB1Rs and subsequent desensitization. These results validate earlier studies examining the effects of acute and chronic MAGL deficiency. Clearly, 2-AG degradation by MAGL controls the duration of stimulus-driven eCB signalling without major acute influence on the magnitude of signalling at these synapses. They also indicate that MAGL is a dominant enzymatic determinant of 2-AG degradation, and that the pharmacological inhibitor JZL184 reliably exerts its effects through MAGL. This will be useful in testing the dynamics of eCB signalling at other synapses of the brain which exhibit lower CB1R densities, and may rely on alternate enzymatic strategies. Finally, the long-term absence of 2-AG degradation in MAGL−/− mice results in a synaptic ‘phenotype’ similar to chronic pharmacological MAGL inhibition or chronic cannabinoid exposure. Eloquently termed ‘endocannabinoid overload’, this results in a situation where tonically engaged CB1Rs exhibit desensitisation (Lichtman et al. 2010). These observations open the door to investigations into a topic about which little is currently known – the endogenous regulation of MAGL expression and activity and the consequences for synapses and behaviour.