David W. Self
 
David Self is in the Division of Molecular Psychiatry, Yale University School of Medicine, Connecticut Mental Health Center, 34 Park St., New Haven, Connecticut 06508, USA.
e-mail: david.self@yale.edu

Endocannabinoids are endogenous substances that mimic the psychoactive effects of marijuana on cannabinoid receptors¹. The story of their discovery goes back to the last decade, when pharmacological and molecular studies^{2, 3} led to the identification of a G-protein-coupled receptor that was activated by ⁹-tetrahydrocannabinol (⁹-THC), the major psychoactive substance in marijuana. Just as the existence of opioid receptors led to the discovery of endogenous opioid neurotransmitters in the 1970s⁴, the identification of the brain cannabinoid receptor CB₁ spurred a search for naturally occurring ligands within the brain.

Several endogenous ligands for the CB₁ receptor have been discovered, but none has yet been shown to function as a neurotransmitter. In this issue of Nature Neuroscience, Giuffrida and colleagues report that local depolarization can trigger the release of anandamide, the first endocannabinoid identified, in the striatum of awake, freely moving rats⁵. They also show that anandamide release can be stimulated by dopamine receptors, and that this leads to the inhibition of dopamine-mediated locomotor behavior via cannabinoid receptors. Their findings promise to propel anandamide from candidate status to bona fide neurotransmitter, and may also open the door to novel treatments for diseases that involve dysfunction of dopamine signaling.

The name anandamide is derived from the Indian Sanskrit term ananda, meaning 'bliss and tranquillity'¹, undoubtedly in reference to psychoactive effects of cannabinoids in humans. Anandamide belongs to a class of molecules called eicosaniods, and it was first isolated based on its hydrophobic properties, by analogy with exogenous cannabinoids such as ⁹-THC⁶. It is expressed throughout the brain, and it is most prevalent in the hippocampus, striatum, cerebellum and cortex, structures that regulate learning, movement and cognition, among other behaviors. Another endocannabinoid, 2-arachidonylglycerol (2-AG), which was discovered more recently, is even more highly expressed in the brain¹. Both molecules fulfill at least some of the criteria for neurotransmitter status. They both activate the brain cannabinoid receptor CB₁, and both have putative biosynthetic pathways. (They are synthesized from arachidonic acid and phospolipids.) Anandamide also has a putative mechanism for its inactivation via re-uptake and intracellular degradation. Being hydrophobic molecules, neither anandamide nor 2-AG is packaged into synaptic vesicles (in contrast to conventional neurotransmitters); instead, they are thought to be released by phospholipase-mediated cleavage followed by passive diffusion across the plasma membrane¹.

Because of their low (micromolar) affinities for the CB₁ receptor, however, many investigators were skeptical as to whether anandamide or 2-AG ever attain sufficient concentrations to activate the receptor in the brain. Although certain memory- and anxiety-enhancing effects of the cannabinoid receptor antagonist SR 141716 suggest that endocannabinoids are tonically active, the new findings of Giuffrida and colleagues⁵ lay this concern to rest for anandamide.

The authors focused on the striatum, which expresses high levels of CB₁ receptors. They induced depolarization in the striatum of awake rats and showed that this leads to the release of anandamide. Release of 2-AG, in contrast, did not reach detectable levels either before or after stimulation. This suggests that anandamide is the main ligand for striatal cannabinoid receptors, although it remains possible that other endocannabinoids might also be involved.

The authors found that anandamide release could also be induced by pharmacological activation of the D₂ class of dopamine receptors, which are known to be important in striatal function. D₂ receptor activation causes rats to become hyperactive, so the authors asked whether anandamide release might be involved in this behavioral response. They found that blocking the CB₁ receptor with a pharmacological antagonist potentiated D₂-receptor-induced hyperactivity, whereas it had no effect on baseline activity. Their results suggest that anandamide can reach a sufficient concentration to produce functional effects, but only after stimulation of D₂ receptors. Thus, the released anandamide seems to function as a 'brake' that limits the behavioral response to D₂ receptor activation.

Although the source of the released anandamide in the striatum is not yet known, both pre- and postsynaptic elements of striatal architecture contain D₂ receptors that could trigger anandamide release (Fig. 1). Presynaptic D₂ receptors could stimulate anandamide release from dopamine terminals to provide negative feedforward regulation of postsynaptic D₂- receptor-mediated locomotor behavior, in conjunction with the presynaptic D₂ receptor's negative feedback effects on dopamine release itself. Alternatively, postsynaptic D₂ receptors could stimulate anandamide release from striatal neurons as a negative feedback mechanism on the same CB₁-receptor-containing neurons. In view of the latter possibility, it is interesting that CB₁ receptors apparently switch coupling from inhibition to activation of adenylyl cyclase when stimulated concurrently with D₂ receptors, thereby counteracting the inhibitory effects of D₂ receptors on the cyclase⁷. Yet another possibility is that anandamide release is triggered indirectly by other neurotransmitter systems within the striatum that are modulated by D₂ dopamine receptors.

The pleasant psychoactive effects of cannabinoids in humans are well known, and some studies have reported that laboratory animals will self-administer cannabinoids intravenously^{8, 9}. However, other studies suggest that systemically administered cannabinoids produce anxiety and dysphoria and oppose reward mechanisms in rodents¹. Because dysphoria and anhedonia have been associated with reduced dopamine levels in striatal subregions¹⁰, cannabinoid-induced inhibition of dopamine-mediated behavior may contribute to these aversive effects of exogenous cannabinoids.

In any event, the interaction of endogenous cannabinoids with dopaminergic systems reported by Giuffrida and colleagues may have important therapeutic implications for the development of treatments for movement disorders. For example, drugs that block endocannabinoid effects at the CB₁ receptor could potentiate or prolong the therapeutic efficacy of dopamine-based treatment strategies currently used in Parkinson's disease while having minimal effects on their own. In contrast, drugs that stimulate the CB₁ receptor could reduce dyskinesias associated with Huntington's disease or antipsychotic treatment, possibly at doses with minimal psychoactive effects. Indeed, CB₁ receptor binding is decreased in target regions of striatal neurons during the early stages of Huntington's disease^{11, 12}, suggesting that alterations in endocannabinoid signaling possibly contribute to the disease pathology itself.

Given their diffuse localization throughout the brain, it is likely that anandamide and other endocannabinoids interact with multiple neurotransmitters in ways that are yet to be discovered. In addition to reward and anxiety, behavioral studies suggest an interaction between endocannabinoid systems and appetite, pain, epilepsy and other behavioral states¹. As these complex interactions are unraveled and understood, endocannabinoid systems are likely to gain appreciation as a prominent signaling pathway in the brain, which could open the door to new treatment strategies for a variety of disorders associated with these behaviors.

REFERENCES