Corticostriatal mechanisms of action learning and habit formation

National Institute On Alcohol Abuse And Alcoholism

Investigators

Linked publications, trials & patents

Abstract

Striatonigral direct pathway 2-arachidonoylglycerol contributes to ethanol effects on synaptic transmission and behavior Endocannabinoids are lipid metabolites that activate cannabinoid receptors, the same receptors that are the major targets for delta-9-tetrahydrocannabinol, the major psychoactive ingredient in cannabis drugs. Endocannabinoid activation of the CB1 cannabinoid receptor inhibits release of neurotransmitters at synapses throughout the brain, spinal cord and peripheral nervous system. Activation of this receptor has been implicated in the control of alcohol intake and other alcohol-related behaviors. However, the roles in these behaviors of endocannabinoid signaling in specific brain regions and at specific synapses are not fully understood. The striatum, a subcortical brain region involved in action learning and control, expresses high levels of endocannabinoids and CB1 receptors. The dorsolateral striatum (DLS), a part of the forebrain sensorimotor circuitry, has been implicated in excessive alcohol intake. Thus, we hypothesized that alcohol alters endocannabinoid signaling in DLS and effects on this and other striatal subregions contribute to alcohol intake. To this end, we began by measuring endocannabinoid signaling in brain slices containing DLS. To measure endocannabinoid signaling at cortical inputs to DLS neurons we expressed the genetically-encoded fluorescent G protein-coupled receptor activation-based endocannabinoid sensor (GRABeCB2.0) in mouse primary motor cortex (M1) by in vivo viral injection. The GRABeCB2.0 sensor was then trafficked to cortical terminals in DLS. We prepared brain slices and used photometric recording to examine the fluorescent signals activated by electrical stimulation in DLS. As in our previous studies, we found that short bursts of electrical stimuli produced an increase in fluorescence that persisted for 10s of seconds. This increase was not observed with a mutant GRAB construct that does not bind endocannabinoids. The fluorescence increase was also prevented by a drug that blocks endocannabinoid binding to GRABeCB2.0 and a drug that blocks diacylglycerol lipase alpha (DGLalpha) the enzyme that catalyzes production of the endocannabinoid 2-arachidonoyl glycerol (2-AG). These findings are consistent with previous studies in our group indicating that 2-AG is the main endocannabinoid released following this pattern of electrical stimulation. We next used a genetic approach to determine which neurons in DLS produce the 2-AG we are detecting. In a previous study, we found that genetic removal of DGLalpha from all striatal spiny projection neurons (SPNs) eliminates stimulus-driven 2-AG release measured with GRABeCB2.0. In the persent study we specifically removed DGLalpha from only those SPNs that project to the substantia nigra pars reticulata (the "direct" pathway SPNs). Somewhat surprisingly, we found that this removal strongly reduced 2-AG release as measured with GRABeCB2.0. This indicates that the direct pathway SPNs contribute more strongly than expected to stimulation-induced 2-AG release in striatum. We then examined the effects of acute alcohol application on 2-AG release in brain slices expressing GRABeCB2.0 in M1 cortical terminals. Alcohol inhibited the stimulus-induced 2-AG release while the drug was present in the slice, and this effect was reversed when alcohol was removed from the slice. We also found that this acute alcohol exposure inhibited the small amount of 2-AG release that remained after DGLalpha removal from direct pathway SPNs. Thus, alcohol appears to reduce 2-AG release from all SPN subtypes. We next examined the potential role of 2-AG production from direct pathway SPNs in alcohol-related behaviors, with the rationale that the majority of release of this endocannabinoid in striatum comes from these neurons. We once again used mice in which DGLalpha was genetically removed from the direct pathway SPNs. We observed that the duration of alcohol-induced loss of righting reflex was reduced in the mice lacking direct pathway SPN DGLalpha. This suggests that 2-AG release from this SPN subtype contributes to the sedative effect of alcohol. We also performed a two-bottle choice alcohol drinking paradigm in the direct pathway SPN DGLalpa knockout mice and observed decreased alcohol preference and increased water preference in male, but not female mice. Changes in preference for sweet and bitter tasting substances was not altered these mice. Thus, 2-AG release from direct pathway SPNs and endocannabinoid signaling in striatum appear to have specific roles in control of alcohol drinking. In future experiments it will be interesting to determine what cellular targets of striatal endocannabinoid signaling contribute to the effects of alcohol. It will also be important to determine if other manipulations of striatal endocannabinoid and CB1 signaling, e.g. inhibition of 2-AG degradation, alter alcohol-related behaviors.