Drug addiction is a progressive, chronic brain disorder that ravages the health and lives of people globally. In the last century, pioneering discoveries in addiction science have begun to put to rest misconceptions concerning the nature of addiction and to yield new breakthroughs, which are poised to revolutionize its prevention and treatment. Importantly, addiction is now recognized as a disordered integration of cognitive and motivational aspects of reward-directed behavior involving higher order limbic-corticostriatal structures (Kalivas and Volkow, 2005). The medial prefrontal cortex (mPFC) and associated circuits are particularly sensitive to plasticity incurred due to repeated pairing of environmental stimuli (eg, drug paraphernalia, etc) with exposure to abused drugs. These associations between environmental cues and drug-taking are essential drug-associated memories that can trigger conditioned emotional responses in addicts and ‘craving’ (desire for drug), oft cited to explain relapse to drug-seeking and drug-taking during abstinence.

The cortical control of drug-seeking is the focus of a new study by Janet Neisewander and her colleagues reported in Neuropsychopharmacology (Pentkowski et al, 2010). These studies concentrate on cocaine dependence; cocaine abuse spirals to addiction in upwards of 15% of cocaine users and limited therapeutic options to suppress recidivism are available. Pentkowski et al (2010) have uncovered a nuanced neural circuit that selectively restrains the frontocortical circuitry that controls drug-seeking behavior in rats. In the procedure employed here, intravenous catheters were surgically implanted and rats were trained to lever press to deliver an intravenous injection of cocaine plus several associated environmental cues (ie, light and tone, sound of infusion pump). Following training of the cocaine self-administration, rats were subjected to extinction during which lever presses did not result in administration of cocaine or the associated cues. Subsequent to extinction, rats were given an injection of cocaine or were presented with the cue stimuli, previously associated with cocaine delivery, either of which then ‘reinstated’ cocaine-seeking. This cocaine-seeking in rats parallels the experience of an abstinent cocaine addict exposed to either a dose of cocaine or the environment in which cocaine use occurred. These exposures often result in relapse, which cycles into a loss of behavioral control and habitual drug use. This self-administration/reinstatement model employed in preclinical studies of cocaine abuse and addiction provides a valid method to explore the neural mechanisms and circuitry underlying cocaine-seeking and cocaine-taking.

Pentkowski et al (2010) have extended our understanding of how neurotransmission through the 5-HT2C receptor (5-HT2CR) within the mPFC controls drug-seeking. The 5-HT2CR is characterized to regulate cognition, feeding, satiety and mood, although 5-HT2CR dysregulation is thought to contribute to the neuromolecular mechanisms underlying the etiology of a variety of neuropsychiatric disorders including addiction, depression and schizophrenia. The 5-HT2CR is localized to the postsynaptic specialization of neuronal synapses in mPFC neurons (Anastasio et al, 2010) and controls the excitability of mPFC neurons to afford a dynamic influence over a variety of cortically mediated autonomic, motor and limbic functions. In the Pentkowski et al (2010) study, microgram quantities of a drug were infused into one of three mPFC different subregions to activate the 5-HT2CR signalosome before analysis of cocaine- or cue-evoked reinstatement. Pentkowski et al (2010) found that local stimulation of the 5-HT2CR in the prelimbic and infralimbic subregions of the mPFC did not alter cocaine-taking, but did dose-dependently suppress reinstatement of cocaine-seeking behavior, an effect reversed by co-infusion of a selective 5-HT2CR antagonist. Microinfusions of the 5-HT2CR agonist into the anterior cingulate cortex were without effect in either paradigm. Thus, Pentkowski et al (2010) reveal that the control afforded by the mPFC 5-HT2CR is specific to cocaine-seeking over cocaine-taking, and localized to more ventral mPFC subregions.

These results are interpreted as reflecting 5-HT2CR-evoked decrements in the incentive-motivational value of the cocaine-primed and cue-elicited cocaine-seeking behavior. The mechanisms underlying suppression of cocaine-seeking by intra-mPFC infusion of a 5-HT2CR agonist are likely to occur through stimulation of the 5-HT2CR located on gamma-aminobutyric acid (GABA) interneurons or glutamate projection neurons that regulate mPFC output (Bubar and Cunningham, 2008) possibly via actions at the level of dopamine neurotransmission from the ventral tegmental area to limbic corticostriatal circuits known to be involved in goal-directed behaviors (Kalivas and Volkow, 2005). Future research is required to test this hypothesis; however, given the postulated role for hypofrontality and cognitive deficits in compulsive addictive disorders (Kalivas and Volkow, 2005), these new preclinical data from Pentkowski et al (2010) are important in identifying the 5-HT2CR as a functional rheostat for mPFC output within the circuit that links motivational aspects of cues associated with drug reward with the cognitive control over drug-seeking behavior. It is possible that the connection between the cocaine experience and the need to seek this drug after repeated self-administration has been carved into the mPFC in part via 5-HT2CR-mediated neuroplasticity developed during repeated pairings of this appetitive reward and its associated conditioned stimuli.

The mPFC is integral to the storage and maintenance of long-term memories, and the knowledge that these processes are malleably regulated by serotonergic neurotransmission through the 5-HT2CR, encourages the design of preclinical experiments to decipher the role of the mPFC 5-HT2CR in the acquisition, consolidation, retrieval and reconsolidation of cocaine-associated memories (Nic Dhonnchadha and Cunningham, 2008). Pre- and post-training administration of specific 5-HT2CR ligands into discrete mPFC subregions would allow analysis of whether this receptor is involved in the acquisition of the pairing between cocaine and associated cues, or can disrupt the consolidation or retrieval of the memory of the cocaine–cue association in the self-administration paradigm, respectively. Furthermore, evidence that consolidated memories become labile again after ‘reactivation,’ that is, soon after the memory is retrieved by reminder stimuli that apparently reactivate the original memory trace, as is the case in the cue-reinstatement model, may provide an additional window in which to manipulate long-lasting cocaine-associated memories. Thus, manipulation of 5-HT2CR signaling in the mPFC at various intervals following a retrieval session would further delineate the role of this system in reconsolidation of cocaine-associated memories. Preclinical evaluations of the role of 5-HT2CR in acquisition, consolidation, retrieval and associated memory processes (eg, reconsolidation) and their special properties vis à vis drug-associated cues will help to identify how 5-HT2CR-selective pharmacotherapic strategies may be useful to selectively minimize the strong incentive-motivational pull of cocaine-associated cues potentially imprinted as memory within mPFC circuits.

The observations reported by Pentkowski et al (2010) support the assertion that selective activation of the 5-HT2CR signalosome may prove therapeutically useful to reduce craving and extend abstinence (Bubar and Cunningham, 2008). However, the progress of selective 5-HT2CR agonists from preclinical evidence of efficacy to verification of safety and efficacy as psychotherapeutic medications in humans has been stymied until recently. Phase II and III clinical trials evaluating the efficacy of the selective 5-HT2CR agonist lorcaserin (formerly APD 356, Arena Pharmaceuticals) for weight reduction have been successfully completed (Smith et al, 2009). Lorcaserin, which exhibits ∼15-fold and ∼100-fold greater potency to stimulate the 5-HT2CR over the homologous 5-HT2AR and 5-HT2BR, respectively, was well tolerated by subjects in the weight reduction trials (Smith et al, 2009). Further, when evaluated in polydrug users, the doses shown to be therapeutic in obesity (Smith et al, 2009) were well tolerated, exhibited low abuse liability, and were not associated with notable neurocognitive or perceptual effects (Schram et al, 2009). The selective 5-HT2CR agonist vabicaserin (SCA-136; Wyeth Research/Pfizer) has been under evaluation to establish its efficacy and safety in subjects with an acute exacerbation of schizophrenia. Vabicaserin exhibits an interesting profile as a selective 5-HT2CR full agonist, a 5-HT2BR antagonist and a very weak 5-HT2AR antagonist (Sharon Rosenzweig-Lipson, personal communication). Thus, armed with the knowledge from present Pentkowski et al (2010) study, further preclinical studies to elucidate the functional regulation of meso- and allo-cortical circuitry by the 5-HT2CR involved in the learning and memory processes relevant to cocaine self-administration and reinstatement, and the potential availability of two potent and selective 5-HT2CR agonists opens the door to implementation of clinical research to evaluate these compounds as innovative pro-abstinence, anti-relapse therapeutics for stimulant addiction.