Impairment of working memory is one of the most important deleterious effects of marijuana intoxication in humans, but its underlying mechanisms are presently unknown. Here, we demonstrate that the impairment of spatial working memory (SWM) and in vivo long-term depression (LTD) of synaptic strength at hippocampal CA3-CA1 synapses, induced by an acute exposure of exogenous cannabinoids, is fully abolished in conditional mutant mice lacking type-1 cannabinoid receptors (CB(1)R) in brain astroglial cells but is conserved in mice lacking CB(1)R in glutamatergic or GABAergic neurons. Blockade of neuronal glutamate N-methyl-D-aspartate receptors (NMDAR) and of synaptic trafficking of glutamate α-amino-3-hydroxy-5-methyl-isoxazole propionic acid receptors (AMPAR) also abolishes cannabinoid effects on SWM and LTD induction and expression. We conclude that the impairment of working memory by marijuana and cannabinoids is due to the activation of astroglial CB(1)R and is associated with astroglia-dependent hippocampal LTD in vivo.

Astroglial type-1 cannabinoid (CB ) receptors are involved in synaptic transmission, plasticity and behavior by interfering with the so-called tripartite synapse formed by pre- and post-synaptic neuronal elements and surrounding astrocyte processes. However, little is known concerning the subcellular distribution of astroglial CB receptors. In particular, brain CB receptors are mostly localized at cells' plasmalemma, but recent evidence indicates their functional presence in mitochondrial membranes. Whether CB receptors are present in astroglial mitochondria has remained unknown. To investigate this issue, we included conditional knock-out mice lacking astroglial CB receptor expression specifically in glial fibrillary acidic protein (GFAP)-containing astrocytes (GFAP-CB -KO mice) and also generated genetic rescue mice to re-express CB receptors exclusively in astrocytes (GFAP-CB -RS). To better identify astroglial structures by immunoelectron microscopy, global CB knock-out (CB -KO) mice and wild-type (CB -WT) littermates were intra-hippocampally injected with an adeno-associated virus expressing humanized renilla green fluorescent protein (hrGFP) under the control of human GFAP promoter to generate GFAPhrGFP-CB -KO and -WT mice, respectively. Furthermore, double immunogold (for CB ) and immunoperoxidase (for GFAP or hrGFP) revealed that CB receptors are present in astroglial mitochondria from different hippocampal regions of CB -WT, GFAP-CB -RS and GFAPhrGFP-CB -WT mice. Only non-specific gold particles were detected in mouse hippocampi lacking CB receptors. Altogether, we demonstrated the existence of a precise molecular architecture of the CB receptor in astrocytes that will have to be taken into account in evaluating the functional activity of cannabinergic signaling at the tripartite synapse.

The cannabinoid type 1 (CB1) receptor is widely distributed in the brain and peripheral organs where it regulates cellular functions and metabolism. In the brain, CB1 is mainly localized on presynaptic axon terminals but is also found on mitochondria (mtCB1), where it regulates cellular respiration and energy production. Likewise, CB1 is localized on muscle mitochondria, but very little is known about it. The aim of this study was to further investigate in detail the distribution and functional role of mtCB1 in three different striated muscles. Immunoelectron microscopy for CB1 was used in skeletal muscles (gastrocnemius and rectus abdominis) and myocardium from wild-type and CB1-KO mice. Functional assessments were performed in mitochondria purified from the heart of the mice and the mitochondrial oxygen consumption upon application of different acute delta-9-tetrahydrocannabinol (Δ9-THC) concentrations (100 nM or 200 nM) was monitored. About 26% of the mitochondrial profiles in gastrocnemius, 22% in the rectus abdominis and 17% in the myocardium expressed CB1. Furthermore, the proportion of mtCB1 versus total CB1 immunoparticles was about 60% in the gastrocnemius, 55% in the rectus abdominis and 78% in the myocardium. Importantly, the CB1 immunolabeling pattern disappeared in muscles of CB1-KO mice. Functionally, acute 100 nM or 200 nM THC treatment specifically decreased mitochondria coupled respiration between 12 and 15% in wild-type isolated mitochondria of myocardial muscles but no significant difference was noticed between THC treated and vehicle in mitochondria isolated from CB1-KO heart. Furthermore, gene expression of key enzymes involved in pyruvate synthesis, tricarboxylic acid (TCA) cycle and mitochondrial respiratory chain was evaluated in the striated muscle of CB1-WT and CB1-KO. CB1-KO showed an increase in the gene expression of Eno3, Pkm2, and Pdha1, suggesting an increased production of pyruvate. In contrast, no significant difference was observed in the Sdha and Cox4i1 expression, between CB1-WT and CB1-KO. In conclusion, CB1 receptors in skeletal and myocardial muscles are predominantly localized in mitochondria. The activation of mtCB1 receptors may participate in the mitochondrial regulation of the oxidative activity probably through the relevant enzymes implicated in the pyruvate metabolism, a main substrate for TCA activity.

Astrocytes can directly influence neuronal activity through the release of various transmitters acting on membrane receptors expressed by neurons. However, in contrast to glutamate and ATP for instance, the release of GABA (γ-amino-butyric acid) by astrocytes is still poorly documented. Here, we used whole-cell recordings in rat acute brain slices and electron microscopy to test whether hippocampal astrocytes release the inhibitory transmitter GABA. We observed that slow transient inhibitory currents due to the activation of GABAA receptors occur spontaneously in principal neurons of the three main hippocampal fields (CA1, CA3, and dentate gyrus). These currents share characteristics with the slow NMDA receptor-mediated currents previously shown to result from astrocytic glutamate release: they occur in the absence of synaptic transmission and have variable kinetics and amplitudes as well as low frequencies. Osmotic pressure reduction, known to enhance transmitter release from astrocytes, similarly increased the frequency of non-synaptic GABA and glutamate currents. Simultaneous occurrence of slow inhibitory and excitatory currents was extremely rare. Yet, electron microscopy examination of immunostained hippocampal sections shows that about 80% of hippocampal astrocytes [positive for glial fibrillary acidic protein (GFAP)] were immunostained for GABA. Our results provide quantitative characteristics of the astrocyte-to-neuron GABAergic signaling. They also suggest that all principal neurons of the hippocampal network are under a dual, excitatory and inhibitory, influence of astrocytes. The relevance of the astrocytic release of GABA, and glutamate, on the physiopathology of the hippocampus remains to be established.