High on ‘chronic’: an unexpected route to cannabinoid-induced synaptic plasticity

In this issue of The Journal of Physiology, Kim & Li (2015) demonstrate the surprising finding that chronic, but not acute, activation of the type 2 cannabinoid receptor (CB2) enhances excitatory synaptic transmission onto hippocampal CA1 pyramidal neurons in organotypic slice cultures. This report warrants a reevaluation of the prevailing notion that central nervous system CB2 is restricted to neuroimmune function.

Cannabinoid receptors come in two flavours; type 1 (CB1) and type 2 (CB2). Though both CB1 and CB2 are G_i/o-coupled G protein-coupled receptors and have ligands in common, they have vastly different expression profiles and known functions. The prevailing view of CB1 function places this receptor on axon terminals in the central nervous system where it serves to dampen neurotransmitter release upon activation by exogenous or postsynaptically released endogenous cannabinoids. CB2, on the other hand, was originally dubbed the ‘peripheral cannabinoid receptor’ due to its high expression in the spleen and detection-evading expression level in the brain as reported by Munro and colleagues (1993). Over the past decade, it has become firmly established that CB2 is actually expressed by microglia in the brain, and several reports provide evidence for postsynaptic neuronal expression of CB2 in select brain regions, including the hippocampus (Van Sickle et al. 2005; Brusco et al. 2008; Atwood & Mackie, 2010).

While the function of CB2 in the central nervous system has largely focused on microglia-mediated neuroinflammatory processes, a role for CB2 in modulating synaptic strength has thus far taken a back seat to its sister receptor CB1. In this issue of The Journal of Physiology, however, Kim & Li (2015) now demonstrate that chronic activation of CB2 enhances excitatory synaptic transmission onto CA1 pyramidal neurons of the hippocampus. In this ongoing tale of two receptors, then, these authors have begun a new and exciting chapter.

In their study, Kim and Li exposed both rat and mouse organotypic hippocampal slice cultures to either acute (3–4 days) or chronic (7–10 days) treatment with a CB2 agonist. Upon electrophysiological recording following treatment, the authors observed an increase in the frequency of spontaneous excitatory postsynaptic current events from CA1 pyramidal neurons in the chronically treated, but not acutely treated, slices. Notably, this effect was absent in slice cultures generated from CB2 knockout mice. The authors went on to show that chronic activation of CB2 also resulted in an increase in CA1 pyramidal neuron dendritic spine density and stimulation of ERK1/2 phosphorylation. Blocking ERK1/2 phosphorylation during CB2 agonist treatment prevented the increase in excitatory transmission, as well as the increase in dendritic spine density. These data suggest that the increase in spontaneous excitatory postsynaptic current events is mediated by the formation of new synapses. Finally, the authors show that this effect is not restricted to the organotypic slice culture experimental protocol. Daily in vivo injections of a CB2 agonist resulted in an increase in excitatory synaptic transmission onto CA1 pyramidal neurons in ex vivo acute slices taken from wild-type mice, but not CB2 knockouts.

Like any important advance, this work adds an important piece to the puzzle, but also unveils new vistas of the unknown. One important issue is where CB2 might be acting to mediate the synaptic enhancement. A parsimonious explanation would be that CA1 pyramidal neurons express CB2 (Brusco et al. 2008) and their activation leads to an ERK1/2 phosphorylation-dependent pathway that ultimately increases dendritic spine density. Alternatively, it is entirely possible that CB2 activation on resident microglia induces the release of a neuroactive substance that then acts upon the pyramidal neuron directly, or indirectly, in an ERK1/2-dependent manner to mediate spine formation. Indeed, microglia-dependent synaptic plasticity mechanisms are possible (Ben Achour & Pascual, 2010). Whatever the case may be, this study opens a previously unrecognized route to de novo spine formation.

This work carries several implications. One of the most important is what chronic activation of CB2 receptors might mean for regular cannabis users. Δ⁹-THC, an active ingredient of marijuana, activates both CB1 and CB2. Thus, chronic use of marijuana may be inducing the same synaptic reorganization described by Kim and Li. This could possibly affect learning mediated by the hippocampus and potentially other brain regions as well. It will be important to determine whether chronic Δ⁹-THC administration induces the same effect on CA1 pyramidal neuron spine density and whether it does so in a CB2-dependent manner. Secondly, this work potentially opens up a variety of novel routes to synaptic plasticity that would have otherwise gone unnoticed had the authors not chosen to examine the chronic effects of CB2 activation. This point raises important mechanistic questions. What cellular/molecular switch is occurring to mediate the chronic, but not acute, CB2 activation effect? Are endogenous cannabinoids capable of signalling through CB2 to induce these same plastic changes at synapses?

It is apparent that much work is needed in order to finish the tale of these two receptors and their role in synaptic plasticity. With the addition of the fine work by Kim and Li, it seems as though the intensity of the CB1–CB2 sibling rivalry has achieved a new high.

Additional information

Competing interests

None declared.