Cannabinoid receptors

The first studies identifying the receptor of psychoactive phytocannabinoids pinned down CB₁ and CB₂, G-protein coupled receptors of the rhodopsin family (Howlett et al.

1990; Matsuda et al. 1990; Munro et al. 1993). In the next decades, several receptors from all of the branches of the family were shown to be activated by phyto-, synthetic, and endocannabinoids (Brown 2007). However, the major behavioral effects of cannabinoids, also called the tetrad assay (hypolocomotion, hypothermia, analgesia, catalepsy), as well as the regulation of synapses by endocannabinoids are mediated by CB1 (Kawamura et al. 2006; Wilson et al. 2001; Zimmer et al. 1999). Being one of the most abundant GPCRs in the brain, CB1 is present in diverse cell- and synapse populations, and its multiple roles in the development and function of the CNS has been reviewed extensively (Alger 2012; Busquets-Garcia et al. 2015; Kano et al. 2009;

Katona and Freund 2012). The involvement of CB1 has been discovered in a number of disease conditions, such as epilepsy, addiction, mental- and neurodegenerative disorders (Di Marzo et al. 2015; Fattore 2015; Parsons and Hurd 2015; Soltesz et al. 2015; Volk and Lewis 2015). Presynaptic CB1 receptors were first described on the axon terminals of specific hippocampal interneurons, but CB1 expression is abundant throughout the brain, and presynaptic receptors are present on a number of cell types (Herkenham et al.

1990; Katona et al. 2001; Katona et al. 1999; Katona et al. 2006; Kawamura et al. 2006;

11

Uchigashima et al. 2007). While the number of receptors on an axon terminal ranges from barely detectable to several hundred between different synapse populations, CB1 -mediated synaptic plasticity is a robust physiological phenomenon throughout the brain (Heifets and Castillo 2009; Katona et al. 2006; Nyiri et al. 2005). What factors regulate the levels of presynaptic CB1 receptors? How is the subcellular distribution of the receptor shaped, and how the abundance and positioning of receptors determine endocannabinoid-mediated signaling? Despite some emerging results, these questions remained largely elusive.

While a low level of receptors is detectable in the somatodendritic compartment of neurons, and there are functional implications for the role of somatodendritic receptors, the axonal enrichment of CB₁ is striking, especially in neurons expressing the receptor at high levels (Bacci et al. 2004; Bodor et al. 2005; Katona et al. 1999; Maroso et al.

2016). What mechanisms ensure the accumulation of CB₁ in axons and axon terminals?

Synaptic proteins are often anchored to or trapped at nanodomains by specific scaffolding proteins in a highly organized manner (Choquet and Triller 2013; Südhof 2012; Tang et al. 2016). Although a cannabinoid receptor interacting protein (CRIP 1a) has been described, neither this, nor other binding partners of CB1 were found to be required for axonal targeting of the receptor, or for phasic endocannabinoid signaling (Howlett et al. 2010; Niehaus et al. 2007; Smith et al. 2015). On the other hand, agonist-induced and constitutive internalization of the receptors emerged as major factors regulating CB1 surface expression and distribution. Agonist application leads to rapid internalization of CB1, followed by recycling (Coutts et al. 2001; Hsieh et al. 1999).

However, the affinity of CB1 for endocytosis appears to be different in the somatodendritic and axonal compartments, with augmented constitutive endocytosis in the somatodendritic membrane (Leterrier et al. 2006; McDonald et al. 2007).

Interestingly, when CB1 was overexpressed in cultured neurons, the striking axonal targeting was preserved irrespectively of the type of the transfected neuron, suggesting the autonomous preferential trafficking of the receptor without the presence of cell type-specific auxiliary molecules. In axon terminals, CB₁ diffusion is somewhat confined to the area of the bouton, as opposed to the free diffusion in connecting axonal segments (Mikasova et al. 2008). Brief treatment with CB₁ agonist resulted in desensitization of the receptors, coupled with exclusion from synaptic domains and decreased mobility,

12

which suggests the existence of some synapse- or axon terminal-specific interactions. In contrast, long-term exposure to agonists lead to loss of surface receptors in axons, accumulation of intracellular labeling, and transport to the cell body for degradation (Coutts et al. 2001; Thibault et al. 2013). Endocytosis is likely to happen in the boutons and not in the preterminal axonal segments, as also suggested by the higher relative density of receptors in the latter (Leterrier et al. 2006; Nyiri et al. 2005). Altogether, these observations suggest that CB1 can diffuse freely in presynaptic membranes, but can probably interact with axon terminal-specific proteins.

An interesting property of CB1 is its tonic activity. The application of antagonists/inverse agonists of CB1 increases the release probability above baseline levels, uncovering presynaptic tonic endocannabinoid signaling (Losonczy et al. 2004;

Neu et al. 2007). The existence of the CB₁ tone has been partially attributed to ambient levels of endocannabinoids (Katona and Freund 2012), and recent discoveries have shown that the molecular background of tonic and phasic endocannabinoid signaling is different (Földy et al. 2013; Lee et al. 2015). In particular, transsynaptic complexes formed by presynaptic neurexin and postsynaptic neuroligin have been shown to specifically modulate tonic endocannabinoid signaling (Anderson et al. 2015; Földy et al. 2013). In neuroligin-3 KO mice, synaptic transmission is enhanced by abolished tonic, but not phasic endocannabinoid signaling (Földy et al. 2013).

Apart from their function in regulating neuronal activity, the role of CB1 and endocannabinoid signaling in the development of the CNS is also pivotal, regulating the proliferation, differentiation, migration and axonal growth of neurons. But these aspects are outside the scope of this chapter, and are reviewed in abundance (Gaffuri et al.

2012; Harkany et al. 2007; Harkany et al. 2008).

Among non-CB1 cannabinoid receptors, CB2 has been studied the most extensively.

CB2 is present on peripheral immune cells, and microglia and oligodendrocytes in the brain, and the therapeutic immune properties of cannabis are mediated through this close evolutionary relative of CB₁ (Dhopeshwarkar and Mackie 2014). Recently, convincing evidences for its neuronal expression and function have emerged (Li and Kim 2016; Stempel et al. 2016). Long-chain NAEs are known to be the endogenous ligands of peroxisome proliferator-activated nuclear receptors (PPAR-α), which are able

13

to regulate the excitability of neurons, and are co-expressed with NAPE-PLD in hippocampal neurons (Fu et al. 2003; Melis et al. 2008; Rivera et al. 2014). However, lipid messengers liberated in axon terminals are highly unlikely to reach nuclear receptors. Recently deorphanized GPCRs, GPR55 and GPR119 are also activated by NAEs, and GPR18 binds N-arachydonoylglycine, but, to date, very little is known about their physiological role (Godlewski et al. 2009). Endocannabinoids were shown to act on a number of targets in the brain, such as NMDA and glycine receptors, various transient receptor potential channels, T-type calcium and two-pore-domain potassium channels, but it is not known if these interactions in fact occur during physiological processes (Katona and Freund 2012).