Only recently have the functional implications of the organization of the ventral striatum, amygdala, and related limbic-cortical structures, and their neuroanatomical interactions begun to be clarified. Processes of activation and reward have long been associated with the NAcc and its dopamine innervation, but the precise relationships between these constructs have remained elusive. We have sought to enrich our understanding of the special role of the ventral striatum in coordinating the contribution of different functional subsystems to confer flexibility, as well as coherence and vigor, to goal-directed behavior, through different forms of associative learning. Such appetitive behavior comprises many subcomponents, some of which we have isolated in these experiments to reveal that, not surprisingly, the mechanisms by which an animal sequences responding to reach a goal are complex. The data reveal how the different components, pavlovian approach (or sign-tracking), conditioned reinforcement (whereby pavlovian stimuli control goal-directed action), and also more general response-invigorating processes (often called "activation," "stress," or "drive") may be integrated within the ventral striatum through convergent interactions of the amygdala, other limbic cortical structures, and the mesolimbic dopamine system to produce coherent behavior. The position is probably not far different when considering aversively motivated behavior. Although it may be necessary to employ simplified, even abstract, paradigms for isolating these mechanisms, their concerted action can readily be appreciated in an adaptive, functional setting, such as the responding by rats for intravenous cocaine under a second-order schedule of reinforcement. Here, the interactions of primary reinforcement, psychomotor activation, pavlovian conditioning, and the control that drug cues exert over the integrated drug-seeking response can be seen to operate both serially and concurrently. The power of our analytic techniques for understanding complex motivated behavior has been evident for some time. However, the crucial point is that we are now able to map these components with increasing certainty onto discrete amygdaloid, and other limbic cortical-ventral striatal subsystems. The neural dissection of these mechanisms also serves an important theoretical purpose in helping to validate the various hypothetical constructs and further developing theory. Major challenges remain, not the least of which is an understanding of the operation of the ventral striatum together with its dopaminergic innervation and its interactions with the basolateral amygdala, hippocampal formation, and prefrontal cortex at a more mechanistic, neuronal level.

Acute rewarding properties are essential for the establishment of cocaine addiction, and multiple neurochemical processes participate in this complex behavior. In the present study, we used the self-administration paradigm to evaluate the role of CB1 cannabinoid receptors in several aspects of cocaine reward, including acquisition, maintenance, and motivation to seek the drug. For this purpose, both CB1 receptor knockout mice and wild-type littermates were trained to intravenously self-administer cocaine under different schedules. Several cocaine training doses (0.32, 1, and 3.2 mg/kg/infusion) were used in the acquisition studies. Only 25% of CB1 knockout mice vs 75% of their wild-type littermates acquired a reliable operant responding to self-administer the most effective dose of cocaine (1 mg/kg/ infusion), and the number of sessions required to attain this behavior was increased in knockout mice. Animals reaching the acquisition criteria were evaluated for the motivational strength of cocaine as a reinforcer under a progressive ratio schedule. The maximal effort to obtain a cocaine infusion was significantly reduced after the genetic ablation of CB1 receptors. A similar result was obtained after the pharmacological blockade of CB1 receptors with SR141716A in wild-type mice. Moreover, the cocaine dose-response curve was flattened in the knockout group, suggesting that the differences observed between genotypes were related to changes in the reinforcing efficacy of the training dose of cocaine. Self-administration for water and food was not altered in CB1 knockout mice in any of the reinforcement schedules used, which emphasizes the selective impairment of drug reinforcement in these knockout mice. Finally, cocaine effects on mesolimbic dopaminergic transmission were evaluated by in vivo microdialysis in these mice. Acute cocaine administration induced a similar enhancement in the extracellular levels of dopamine in the nucleus accumbens of both CB1 knockout and wild-type mice. This work clearly demonstrates that CB1 receptors play an important role in the consolidation of cocaine reinforcement, although are not required for its acute effects on mesolimbic dopaminergic transmission.

Activation of cannabinoid CB1 receptors (CB1R) by delta9-tetrahydrocannabinol (THC) produces a variety of negative effects with major consequences in cannabis users that constitute important drawbacks for the use of cannabinoids as therapeutic agents. For this reason, there is a tremendous medical interest in harnessing the beneficial effects of THC. Behavioral studies carried out in mice lacking 5-HT2A receptors (5-HT2AR) revealed a remarkable 5-HT2AR-dependent dissociation in the beneficial antinociceptive effects of THC and its detrimental amnesic properties. We found that specific effects of THC such as memory deficits, anxiolytic-like effects, and social interaction are under the control of 5-HT2AR, but its acute hypolocomotor, hypothermic, anxiogenic, and antinociceptive effects are not. In biochemical studies, we show that CB1R and 5-HT2AR form heteromers that are expressed and functionally active in specific brain regions involved in memory impairment. Remarkably, our functional data shows that costimulation of both receptors by agonists reduces cell signaling, antagonist binding to one receptor blocks signaling of the interacting receptor, and heteromer formation leads to a switch in G-protein coupling for 5-HT2AR from Gq to Gi proteins. Synthetic peptides with the sequence of transmembrane helices 5 and 6 of CB1R, fused to a cell-penetrating peptide, were able to disrupt receptor heteromerization in vivo, leading to a selective abrogation of memory impairments caused by exposure to THC. These data reveal a novel molecular mechanism for the functional interaction between CB1R and 5-HT2AR mediating cognitive impairment. CB1R-5-HT2AR heteromers are thus good targets to dissociate the cognitive deficits induced by THC from its beneficial antinociceptive properties.