Research Findings - Basic Behavioral Research

Exercise Decreases the Motivation for Cocaine in an Animal Model of Self-Administration

Much is now known about the neurobiological effects of exercise, at the systems and cellular level. Exercise has positive 'psychological' effects, produces interoceptive effects that are phenomenologically similar to those produced by addictive drugs and activates the same central dopamine neurotransmitter systems that are stimulated by drugs of abuse. Adolescents who participate in physical fitness programs have a lower incidence of tobacco and substance abuse, suggesting that exercise should be incorporated in prevention and treatment programs where it might serve as a form of alternative reinforcement. However, until recently, the causal relationship between aerobic exercise and the vulnerability for engaging in drug abuse behaviors had not been investigated. Dr. Mark Smith recently gave female rats 24 h access to running wheels and compared them to sedentary counterparts on measures of drug reward and motivation. Exercise was available for six weeks prior to training for i.v. self-administration of cocaine. All animals were first trained on a fixed ratio schedule of reinforcement and then assessed for their "willingness to work" for cocaine using a progressive ratio (PR) schedule to generate individual break-points (i.e., the maximum number of responses an animal is willing to make for cocaine). Animals in the exercise group continued to have access to running wheels during the self-administration training and testing in this study. Group comparisons revealed no difference between exercising and sedentary rats on the rate of acquisition for self-administration, but PR break-points were greater in sedentary rats. Interestingly, the investigator found significant differences for both low, and high, doses of i.v. cocaine (0.3 and 1.0 mg/kg/infusion). He also reports a positive correlation (r2=.462) between the number of wheel revolutions per day (i.e., exercise output) in the day proceeding access to cocaine and breakpoints for 1.0 mg/kg cocaine, suggesting a protective effect of exercise with high dose availability. These observations suggest that prior, and concurrent, exercise may reduce the motivation for drugs of abuse without changing the vulnerability to acquire self-administration. Smith, M.A., Schmidt, K.T., Iordanou, J.C. and Mustroph, M.L. Aerobic Exercise Decreases the Positive-reinforcing Effects of Cocaine. Drug and Alc. Dep., 98, pp. 129-135, 2008.

Initial Behavioral Response to Cocaine Predicts Sensitization of Reward

Initial "high" or "low" locomotor response to acute cocaine in the rat predicts the development of sensitization produced by repeated psychostimulant administration - that is, low responders (LRs), but not high responders (HRs), develop cocaine-induced behavioral sensitization. LRs are also more likely to develop cocaine conditioned place preference. In a recent study, male rats were habituated to a test environment and then treated with 10 mg/kg cocaine to assess acute locomotor response. They were divided into LR and HR groups on the basis of a median split. Half of the LR and half of the HR groups then received 10 mg/kg cocaine for four days (repeated exposure group, RE), while the other half received only one cocaine injection (single exposure group, SE). After acquisition training for cocaine self-administration, animals were tested on a progressive ratio (PR) schedule to assess "willingness to work" for cocaine. Break-points, determined for each rat, measured the greatest number of responses made to receive a single drug infusion. Findings revealed that initial differences in acute locomotor activation (LRs versus HRs) were eliminated by four daily injections of cocaine (the RE condition), attributed to behavioral sensitization in the LR group only. RE enhanced acquisition in both LR and HR groups. In addition to this escalated acquisition the researchers also found that RE animals took more drug during the PR schedule (measured as "break-point associated infusions") than animals who received only a single cocaine injection (SE) prior to catheter implantation. Furthermore, LR rats self-administered more cocaine, at all doses tested, than did HR animals during the PR testing (i.e., significant main effects for both exposure, and for initial response), with no significant interactions between these variables. Overall results indicate (1) initial response to the acute cocaine's locomotor effects predicts development of chronic neuroadapations (i.e., only the LR group showed sensitization, as previously reported); and (2) repeated exposure to cocaine (RE) sensitizes animals to acquire self-administration and increases the motivation to work for drug during the maintenance phase of drug intake (in both LR and HR animals). Moreover, animals with an initial insensitivity to the behaviorally stimulating effects of cocaine (LR) administer more drug on a PR schedule, suggesting that they are more motivated to work for the drug. These findings implicate a differential role for initial drug sensitivity in the initiation of drug intake versus the motivation to work for drug, once a pattern of abuse has begun. Furthermore, they also underscore powerful pharmacokinetic and/or pharmacodynamic effects of repeated drug exposure, which can overshadow individual differences in vulnerability. Mandt, B.H., Schenk, S., Zahniser, N.R. and Allen, R.M. Individual Differences in Cocaine-induced Locomotor Activity in Male Sprague-Dawley Rats and Their Acquisition of and Motivation to Self-administer Cocaine. Psychopharmacol., 201, pp. 195-202, 2008.

Impulsivity as a Two-Dimensional Construct: Differences in Vulnerability Phenotypes with an Animal Model

Dr. Marilyn Carroll and her colleagues have been using rodent models to examine vulnerability phenotypes in the acquisition, maintenance, escalation and relapse to drug seeking and drug taking behaviors. As animals that prefer sucrose rich foods also self-administer more cocaine, alcohol and morphine, sweet preference may be an important vulnerability phenotype that contributes to the co-morbidity of obesity and drug addiction. Previous research by this group also reveals that animals bred for high or low saccharin preference (HiS versus LoS) differ on measures of behavioral impulsivity. Thus, sweet preference and impulsivity may be traits shared by a common intermediate phenotype for drug abuse vulnerability. Impulsivity is not a unitary construct, however, and previous studies have compared HiS versus LoS animals on measures of impulsive choice using delay discounting procedures. In a new study, the authors examined aspects of impulsivity that reflect impaired response inhibition with a Go/No-Go task to examine response inhibition. Male and female rats, selectively bred for HiS or LoS preference, were trained to administer i.v. cocaine under several different reinforcement schedules. The schedules included a signaled Go component where responding was reinforced with a drug infusion, alternating with two No-Go components where responses were without consequence. Responses made on the previously reinforced lever, during No-Go components, were taken as a measure of impaired response inhibition. Similar testing was conducted using food as the reward, instead of cocaine. Results from the Go component resemble those from previous studies, showing that: (1) female animals make more responses and infuse more drug than males, and (2) HiS rats make more responses for the self-infusion of cocaine. Of interest for the analysis of impulsivity as a vulnerability phenotype is the finding that similar profiles were seen during No-Go components, suggesting effects of sex and the ability to withhold responses contribute to greater drug intake seen by female and HiS animals. Since females and HiS animals responded significantly more during No-Go components than males and LoS animals, the authors suggested that impulsivity is an identifiable phenotype that characterizes female and HiS animals, implicating both phenotype and sex differences in impaired inhibition. By contrast, no group differences were seen on responding during No-Go periods of food reinforced operant schedules; (with the exception of males > females at high fixed ratio requirements). Therefore, sex and impulsivity selectively enhanced the vulnerability for cocaine intake in this study. In a previous experiment, these researchers found no effect for sex and sucrose preference on impulsive choice using delayed discount procedures, suggesting that different aspects of this trait may contribute selectively to vulnerability phenotypes for psychostimulant abuse. Anker, J.J., Gliddon, L.A. and Carroll, M.E. Impulsivity on a Go/No-go Task for Intravenous Cocaine or Food in Male and Female Rats Selectively Bred for High and Low Saccharin Intake. Behav. Pharmacol., 19, pp. 615-629, 2008.

Long-Access Cocaine Self-Administration Induces Sustained Cognitive Impairment in Rats

Animal models provide an opportunity to examine cognitive effects of repeated cocaine and characterize the neurobiological processes responsible for these effects. Recently, Dr. Terry Robinson and colleagues conducted a series of experiments to study cognitive deficits induced by psychostimulant self-administration in a rodent model. Rats in this study were trained on a sustained attention task prior to i.v. cocaine self-administration. The task required animals to discriminate a (1 sec) signaled versus non-signaled trial for food reinforcement. Responses on one lever were reinforced if they followed a signal, while responses on the other lever were reinforced if made during a non-signaled trial. Illumination of the house light designated the beginning of each trial and served to focus the animal's attention toward the intelligence panel. Length of the signal was changed from 1 sec, to vary from 25 to 500 msec, randomly across trials. Rats self-administering cocaine were matched for amount of drug taken during i.v. training, and for performance on the attention task. They were divided into groups given short- (1 hr/day=SA) versus long- (6 hr/day=LA) access to cocaine for four weeks, and then were retested on the sustained attention task. In study two, additional groups of animals were tested on this task at one day, or one month, following six weeks of drug administration and post-mortem tissue collected for in situ hybridization of mRNA was collected as well. In a third study, on day 3 or day 30 after the final self-administration session, rats were sacrificed to quantify D1 and D2 receptor proteins in brain. Results from the first study reveal that LA rats showed escalated drug intake over sessions (as typically reported) and were markedly impaired on the sustained attention task 24 hr after the last self-administration session, using an overall index of performance known as VI (vigilance index). The SA rats, however, showed stable self-administration rates over the six weeks and also did not differ from sham operated controls on VI measures. The LA group was still significantly impaired on a VI measure 14 days after the last drug session. This prolonged deficit was due to increased false alarms on signal detection. Measures of mRNA expression for dopamine receptors in brain revealed decreases for D2 mRNA in the medial prefrontal cortex and the orbitofrontal cortex, at 4 and 33 days after the last drug session, for LA animals only. Interestingly, D2 receptor mRNA levels in the prefrontal region was positively correlated with VI scores for this group. In study three, analysis of D2 receptor protein in the prefrontal cortex revealed that the LA group, only, had a small, but statistically significant decrease of D2 protein. Differences in D1 mRNA or proteins were not detected in any area. These findings suggest that extended exposure to cocaine, in an animal model that mimics the progressive, compulsive escalation of drug intake seen in human addiction, can induce impairment of attention processes that are important for executive functions. Furthermore, it appears that cocaine induces this effect, in part, by perturbing D2 receptor function in cortical regions that subserve these functions. Briand, L.A., Flagel, S.B., Garcia-Fuster, M.J., Watson, S.J., Akil, H., Sarter, M. and Robinson, T.E. Persistent Alterations in Cognitive Function and Prefrontal Dopamine D2 Receptors Following Extended, but not Limited, Access to Self-administered Cocaine. Neuropyschopharmacol., 33, pp. 2269-2980, 2008.

Nicotine Taken During Adolescence Decreases Aversive Effects of Nicotine as Adults

Although adolescence has been reported to be a period of enhanced vulnerability to tobacco dependence, little is known about what explains this relationship. Dr. Laura O'Dell at the University of Texas, El Paso, used the conditioned place preference paradigm to examine rewarding and aversive effects of nicotine across different stages of development. In two studies, male rats were conditioned to associate one environmental compartment with drug (either nicotine or d-amphetamine), and to associate another environmental compartment with saline. There were 4 conditioning sessions for each environmental compartment, for a total of 8 counter-balanced conditioning sessions. On the test day, undrugged rats were able to choose between the two environments, indicating their conditioned place preference (CPP) for the drug-paired environment. Since adolescence for rats is typically between postnatal days 28 through 45, the initial conditioning session for adolescent rats took place on postnatal day 34, with the preference test taking place on postnatal day 42. For adult rats, the conditioning sessions started on postnatal day 66, with the preference test taking place on postnatal day 74. The first study compared drug-naive adolescent and adult rats who received one of several doses of nicotine or d-amphetamine during the conditioning sessions. The second study compared nicotine-induced CPP in adult rats that were exposed to nicotine during adolescence (postnatal days 28 through 42) with drug-naive adult rats. In study one, nicotine produced an inverted U-shaped dose response curve, with lower doses producing conditioned place preference, and higher doses producing conditioned place aversion. That is, the environment associated with the lower doses of nicotine was preferred on test day whereas it was avoided (e.g., the non-drug side was preferred) when the environment was associated with higher doses of nicotine. Adolescent rats displayed enhanced positive shifts in preference such that they displayed greater CPP over a wider range of nicotine doses. The high dose that produced aversion in adults still produced preference, albeit a small one, in adolescents. Conditioning with d-amphetamine produced a similar inverted U-shaped dose response curve, but there were no differences between the adolescent and adult groups. In study 2, adults who were pre-exposed to nicotine during adolescence failed to show the aversive effects produced by the highest dose of nicotine. They also showed a smaller preference for the nicotine-paired environment. That is, the dose-response curve for the pre-exposed rats was relatively flat, and lacked the preference peak and aversion drop seen in the naive adults. These findings suggest that not only do adolescents show enhanced CPP - and thus are more sensitive to nicotine's rewarding effects or less sensitive to its aversive properties, but exposure during adolescence reduces the aversion to high-dose nicotine typically seen in naive adult animals. Torres, O.V., Tejeda, H.A., Natividad, L.A., and O'Dell. L.E. Enhanced Vulnerability to the Rewarding Effects of Nicotine During the Adolescent Period of Development. Pharmacol. Biochem. Behav., 90, pp. 658-663, 2008.

Daily Smokers, But Not Chippers, More Likely to be Alcohol Dependent

People with alcohol-use disorders (AUD) and anxiety disorders (ANX) are more likely to use tobacco than the general population. Dr. Sandra Morissette and colleagues therefore evaluated the contribution of tobacco use to the co-occurrence of AUD and ANX disorders by investigating differences between daily smokers, chippers (non-daily smokers), and non-smokers that might affect alcohol or anxiety symptoms. All subjects had co-occurring AUD-ANX disorders and reported their smoking status. Each subject completed several measures, including a TimeLine-Follow-Back to gather retrospective accounts of daily drinking over the past 90 days, the alcohol dependence scale, anxiety sensitivity index, depression anxiety stress scales, Albany phobia and panic questionnaire, social interaction anxiety scale and the Penn State Worry questionnaire. Some subject characteristic differences were noted, including that nonsmokers were older than chippers and that, unlike the gender breakdown in the anxiety literature that leans towards women, this sample was 80% male. Although there were no differences in emotional characteristics reported, there were significant differences in alcohol dependence, average drinks per drinking occasion, percent days alcohol-abstinent and highest and second-highest blood alcohol concentrations in a day. Nonsmokers reported lower alcohol symptoms as compared with daily smokers, with chippers falling between these two groups. It was unexpected that there were no differences with respect to percent heavy drinking days, and no difference in emotional characteristics. The authors suggest that these negative findings may be due to the small, mostly male sample. Despite these caveats, the data have potential treatment implications. Since smoking status is associated with differences in alcohol dependence, smoking status should be considered when treating patients with co-occurring AUD-ANX. Morissette, S.B., Gulliver, S.B., Kamholtz, B.W., Duade, J., Farchione, T., Devine, E., Brown, T.A., Barlow, D.H., and Circaulo, D. Differences Between Daily Smokers, Chippers, and Nonsmokers With Co-Occurring Anxiety and Alcohol-Use Disorders. Addictive Behav., 33, pp. 1425-1431, 2008.

Nutritional Changes Produce Significant Alterations in Dopamine Neurotransmission and Response to Dopaminergic Drugs

Since there is high co-morbidity between eating disorders and substance abuse, NIDA Grantee Dr. Charles France and colleagues investigated whether nutritional status impacts vulnerability to drug abuse as measured by dopamine neurotransmission. Rats were food restricted by being placed on a 10 g/day diet. This resulted in significant weight loss, but no change in blood glucose levels. Dopamine neurotransmission and behavior of food-restricted rats were compared to free-feeding rats. The first experiment examined the effects of food restriction on dopamine agonist (amphetamine) induced locomotion and reinforcing properties measured using conditioned place preference (CPP), and on dopamine clearance as measured by in vivo chronoamperometric recordings. Food-restriction did not affect amphetamine induced locomotion or CPP. However, it did reduce dopamine clearance, suggesting a reduction in dopamine transporter activity. Reductions in dopamine clearance were reversed by amphetamine treatment and by a return to free-feeding schedules. The second experiment examined effects of food restriction on dopamine receptor sensitivity by measuring raclopride (a dopamine antagonist) induced cataplexy and quinpirole (a dopamine agonist) induced yawning behavior. Food restriction reduced quinpirole-induced yawning and raclopride-induced catalepsy. Thus, the food-restricted rats were less sensitive to the behavioral effects of both drugs. Although animals remained hyposensitive to these drug challenges when amphetamine was administered, normal sensitivity was restored once food-restricted rats were again provided with unrestricted access. Taken together, food restriction reduced dopamine neurotransmission, accompanied by relatively modest changes in behavioral responses to dopaminergic drugs. While the mechanisms responsible for these changes remain to be uncovered, understanding the effects of food restriction on central dopaminergic systems may identify biological risk factors for co-morbid drug abuse and eating disorders. Sevak, R.J., Koek, W., Owens, W.A., Galli, A., Daws, L.C., and France, C.P. Feeding Conditions Differentially Affect the Neurochemical and Behavioral Effects of Dopaminergic Drugs in Male Rats. Eur. J. Pharm. 592, pp. 109-115. 2008.

Deletion of the Type 1 Cannabinoid Receptor (CB1) Enhances Preference for Cocaine in Mice

NIDA-grantee Dr. Linda Dykstra and colleagues (University of North Carolina, Chapel Hill) have been examining the role of cannabinoid signaling via the CB1 receptor in the behavioral response to stressors and to drugs of abuse. In a recent report, they examined the effects of exposure to chronic unpredictable stress (CUS) in CB1 receptor knockout (CB1 KO) mice and their wild-type (WT) littermates, on cocaine conditioned place preference (CPP). Mice were either untreated or exposed to two weeks of CUS. After this period, the acquisition of cocaine CPP was examined. Untreated CB1 KO and WT mice both acquired cocaine CPP; however, exposure to CUS enhanced acquisition of cocaine CPP in the CB1 KO mice. These findings support a role for CB1 receptors in the responses to stress as well as in the subjective effects of cocaine. Miller, L.L., Ward, S. J., and Dykstra, L.A., Behav. Pharm., 19, pp. 575-581, 2008.

Peripheral Activation of Either Cannabinoid 1 and 2 Receptors (CB1 and CB2) Attenuate Mechanical Hyperalgesia in a Rat Model of Bone Cancer Pain

NIDA-grantee Dr. Donald Simone and colleagues (University of Minnesota) have been examining the ability of peripherally administered cannabinoids to attenuate tumor-evoked mechanical hyperalgesia. Unilateral injection of osteolytic fibrosarcoma cells into and around the calcaneus bone of the hind paw of rats resulted in tumor formation and mechanical hyperalgesia. WIN 55, 212-2 injected subcutaneously into the tumor-bearing hind paw produced a dose-dependent decrease in paw withdrawal to painful mechanical stimulation without measurable side effects. Administration into the contralateral paws was ineffective. Co-administration of WIN 55,212-2 with either CB1 or CB2 receptor antagonists attenuated the antihyperalgesic effects of WIN 55, 212-2. Thus, peripherally administered WIN 55,212-2 attenuated tumor-evoked mechanical hyperalgesia by activation of both peripheral CB1 and CB2 receptors, suggesting that peripherally administering cannabinoids may be an effective approach in attenuating cancer pain in humans. Potenzieri, C., Harding-Rose, C., and Simone, D.A., Brain Res., 1215, pp. 69 -75, 2008.

Two Studies Show That Calcium Calmodulin-Dependent Kinase II (Camkii) Plays a Critical Role in Drug Seeking Behavior

Two recent reports elucidate important roles for CaMKII in the long-term effects of previous exposure to psychostimulants. CaMKII has been proposed as a candidate molecule for the long-term storage of information because it can remain phosphorylated in the absence of calmodulin. In the first study, Dr. R. Chris Pierce and his colleagues used reinstatement as an animal model of relapse to cocaine seeking. They hypothesized that CaMKII is a biochemical bridge between stimulation of D1-like dopamine receptors and increases in glutamate transmission in the nucleus accumbens shell, known to independently promote reinstatement of cocaine-seeking. To test this hypothesis, they used numerous approaches, including focal injections of agonists, antagonists, and biochemical inhibitors; viral vectors to inhibit cell-surface expression of AMPA glutamate receptors; and cross-linking followed by immunoblotting to assess surface AMPA receptor levels. They found that stimulation of D1-like receptors in the nucleus accumbens shell reinstated cocaine seeking by activating L-type calcium channels and CaMKII. Reinstatement was associated with D1-like receptor-dependent increases in phosphorylation of CaMKII and of the glutamate receptor 1, which was phosphorylated on a known CaMKII target site. In addition, there was an increase in cell-surface expression of GluR1-containing AMPA receptors in the accumbens shell, and reinstatement was attenuated by administration of a viral vector that impairs the transport of GluR1-containing AMPA receptors. The results support their hypothesis that CaMKII serves as an essential link between accumbens shell dopamine and glutamate systems involved in neuronal plasticity underlying cocaine craving and relapse. The second, complementary study from Dr. Paul Vezina's laboratory used a different animal model -- repeated exposure to amphetamine (AMPH), which enhances the ability of AMPH to produce locomotor activation and dopamine overflow in the accumbens, and also leads to enhanced AMPH self administration. In previous work, Dr. Vezina demonstrated that this enhanced behavioral and neurochemical sensitivity to AMPH is impaired by microinjecting a CaMKII inhibitor into the accumbens shell. In the current study, he investigated the effect of this inhibitor on enhanced AMPH self-administration in rats that were trained to self-administer drug on a progressive ratio (PR) schedule to measure the number of responses an animal emits to receive AMPH. Rats received either 5 injections of AMPH or saline, remained abstinent from drug for 14 days, and then were tested for self-administration with PR. Results show that the AMPH-exposed rats worked harder and obtained significantly more drug infusions than saline-exposed rats. After 4 days of stable responding, a CaMKII inhibitor was injected bilaterally into the accumbens shell just before self-administra-tion. Inhibition of CaMKII reduced the enhanced drug intake in AMPH-exposed rats to levels no different from those of saline-exposed rats. Together, these two studies identify CaMKII, or L-type calcium channels, as potential targets for treating psychostimulant addiction. Anderson, S.M., Famous, K.R., Sadri-Vakili, G., Kumaresan, V., Schmidt, H.D., Bass, C.E., Terwilliger, E.F., Cha, J.H., and Pierce, R.C. CaMKII: A Biochemical Bridge Linking Accumbens Dopamine and Glutamate Systems in Cocaine Seeking. Nature Neuroscience, 11, pp. 344-353, 2008; Loweth, J.A., Baker, L.K., Guptaa, T., Guillory, A.M., and Vezina, P. Inhibition of CaMKII in the Nucleus Accumbens Shell Decreases Enhanced Amphetamine Intake in Sensitized Rats. Neurosci. Lett., 444, pp. 157-160, 2008.

Enhanced Methamphetamine Self-Administration in a Rat Model of Schizophrenia

The high prevalence of dual-diagnosis drug abuse and other mental disorders is well established, but the biological basis for this co-morbidity is not well understood. Animal models of mental disorders can be used to test, for example, the self medication hypothesis versus a hypothesis of shared vulnerability and etiology. Although animal models cannot capture the full spectrum of mental disorders, they can replicate some of the symptoms of these diseases. One such model is the neonatal ventral hippocampal lesion (NVHL), which reproduces several behavioral abnormalities observed in schizophrenia, including hypersensitivity to stimulants, hyperactivity, reduced social interactions, and impaired working memory. With respect to comorbidity, NVHL rats exhibit enhanced reinstatement of cocaine-seeking behavior and enhanced sensitization to cocaine and nicotine. The current study investigated drug-seeking behavior in the NVHL model by assessing methamphetamine (METH) self-administration. Rats received excitotoxic lesions of the ventral hippocampus, or sham lesions, when they were 7 days old and were then trained as adults to self-administer METH or to respond for natural reinforcers (water or food). NVHL rats were faster than shams to learn to respond for both drug and natural rewards, but after stable responding was acquired, both groups performed similarly for the two types of reward on a fixed ratio schedule and were equally sensitive to METH dose changes. However, when tested under a progressive-ratio schedule, the NVHL animals had significantly higher break points compared to the shams for METH, but not for food reinforcement. This result suggests that the motivation of NVHL rats to acquire the drug was enhanced (they were willing to work harder for it) specifically for the drug reinforcer. The result supports the hypothesis that schizophrenia and drug abuse arise from neuropathology that confers vulnerability for both disorders, which also appear to be characterized by similar dysfunctions in corticolimbic dopamine and glutamate systems. Brady, A.M., McCallum, S.E., Glick, S.D., and O'Donnell, P. Enhanced Methamphetamine Self-administration in a Neurodevelopmental Rat Model of Schizophrenia. Psychopharm (Berlin), 200, pp. 205-215, 2008.

Deep Brain Stimulation Successfully Attenuates Reinstatement of Cocaine Self-Administration

Deep brain stimulation (DBS), originally developed as a therapy for Parkinson's disease, has recently shown promise in treating psychiatric disorders, including a case study indicating that DBS of the nucleus accumbens reduced alcohol consumption in an alcohol-dependent patient. Thus, DBS is being considered as a potential treatment in cases of extreme cocaine addiction. This study in rats was designed to examine the effect of DBS of the accumbens shell on cocaine priming-induced reinstatement of drug seeking. Rats were trained to self-administer cocaine for 21 days and then cocaine-seeking behavior was extinguished by replacing cocaine with saline. In the reinstatement phase of the study, rats were given small priming doses of cocaine. Reinstatement was assessed by number of lever presses on the lever previously associated with drug delivery (although responses during reinstatement testing do not produce a drug infusion). They also investigated several doses of cocaine used to prime reinstatement in this paradigm, (0, 5, 10 or 20 mg/kg). DBS was delivered to assess its ability to reduce the number of presses during reinstatement at different doses of cocaine priming. DBS was administered bilaterally to the accumbens shell after the priming injection. The two higher doses of cocaine both produced robust reinstatement, and DBS significantly attenuated this reinstatement of cocaine seeking at both doses. A separate experiment on animals trained to respond for sucrose pellets showed that DBS had no effect on reinstatement of food seeking. Additionally, DBS of the dorsal striatum had no effect on cocaine reinstatement. Thus, the DBS in this study was both anatomically and reinforcer specific. These results suggest that DBS of the accumbens shell could be a therapeutic option in the treatment of severe cocaine addiction. Vassoler, F.M., Schmidt, H.D., Gerard, M.E., Famous, K.R., Ciraulo, D.A., Kornetsky, C., Knapp, C.M., and Pierce, R.C. Deep Brain Stimulation of the Nucleus Accumbens Shell Attenuates Cocaine Priming-induced Reinstatement of Drug Seeking in Rats. J. Neurosci., 28, pp. 8735-8739, 2008.

Single Neurons in the Amygdala Track the Moment-to-Moment Value of the Animal's Current State

Neuroimaging studies in humans and animals have shown that the amygdala is part of a distributed neural network that functions abnormally in addiction and underlies altered decision-making, the processing of emotional information, and the experience of "craving." Dr. C.D. Salzman and his colleagues are interested in understanding how the brain encodes information about the value of an organism's current state. In ordinary language, how does the brain monitor whether things are OK, or not OK, or sort of OK? They examined how the brain mediates this process by recording the activity of neurons in the amygdala while monkeys performed a trace-conditioning task during which the presentation of different stimuli induced state transitions. The stimuli were both unconditioned stimuli (USs) -- liquid rewards of two magnitudes, and a mildly aversive airpuff -- and visual conditioned stimuli (CSs) that signaled which US would follow. In each session, new arbitrary CSs were used. After the initial values of the CS-US associations were learned in each session, values for the large reward and the airpuff were reversed. In addition, they also monitored neural activity in response to the fixation stimulus that started the trial, which was considered to always induce a positive value state because the monkey chose when to start the trial by fixating and two thirds of the trials produced positive rewards. The complicated design and analysis of the neural responses was necessary to answer the question of whether amygdala neurons reflect the value of states induced by sensory events independent of the specific stimulus properties of the events. About half of the 145 neurons recorded in 3 monkeys showed differential responses indicating that they encoded value, with about two thirds of these encoding positive value and one third encoding negative value. Importantly, these neurons were not simply associated with rewarding vs. aversive USs and the CSs that predicted them. Rather, they showed graded responses indicating the relative value of the stimuli. As an example, a negative value neuron would be most excited by the CS predicting the airpuff and not fire much or at all when the large-reward CS was presented, but the neuron would fire a modest amount in response to the small-reward CS. This representation of state value could underlie how the amygdala helps coordinate cognitive, emotional, and behavioral responses depending on the value of an organism's state. Knowing the normal function of the amygdala will help us understand how specific alterations of the amygdala in drug abuse contribute to addiction. Belova, M.A., Paton, J.J., and Salzman, C.D. Moment-to-moment Tracking of State Value in the Amygdala. J. Neurosci., 28, pp. 10023-10030, 2008.

Allopregnanolone Decreases Cocaine-Primed Reinstatement of Cocaine-Seeking Behavior in Female But Not Male Rats

Recent studies by Marilyn Carroll, Justin Anker, Erin Larson and their colleagues at the University of Minnesota have highlighted the influence of gonadal hormones, and especially progesterone and its metabolite allopregnanolone (ALLO), on cocaine self-administration in rodent models. Using an escalation of cocaine self-administration procedure, which is commonly used to model the transition from moderate drug use to addiction, they found that progesterone prevented escalation, whereas estrogen facilitated escalation, in ovariectomized female rats, (Larson et al., 2007). They also found that progesterone has an inhibitory effect on reinstatement of cocaine-seeking behavior in females (Anker et al., 2007). In a recently published follow up to that study, they investigated the effects of both progesterone and its metabolite allopregnanolone (ALLO) on cocaine reinstatement in both females and males (Anker et al., 2008). They also tested progesterone in combination with finasteride which is a 5-alpha reductase inhibitor that prevents the metabolism of progesterone into ALLO. Following acquisition and extinction of cocaine self-administration and prior to testing of cocaine-primed reinstatement of responding, separate groups of rats were administered either ALLO, progesterone, progesterone plus finasteride, or saline. As the researchers had previously reported, progesterone suppressed cocaine reinstatement of responding (measured in a paradigm proposed to mimic human relapse); however, greater suppression was produced with ALLO. The combination of progesterone + finasteride (to inhibit progesterone's conversion to ALLO) failed to suppress reinstatement, thus suggesting that progesterone's suppressive effects may be mediated by its conversion to ALLO. In contrast to ALLO's strong suppression of reinstatement in females, ALLO failed to suppress reinstatement in males. The authors discuss possible neurobiological mechanisms for these suppressive effects of progesterone and ALLO including their modulation of GABA-A receptors to inhibit cocaine-induced dopamine release and their interaction with the HPA-axis to dampen the stress response. The results of this study complements recent work from other NIDA-supported investigators who have shown that cocaine cue-induced craving is inversely related to circulating plasma progesterone levels in women (Sinha et al., 2007) and that experimentally administered progesterone decreases the positive subjective effects cocaine (Evans & Foltin, 2007) in women but not men. The present results are also consistent with recent preclinical NIDA-support research showing an inverse relationship between cocaine-primed reinstatement and plasma progesterone levels in freely cycling female rats across the estrous cycle (Feltenstein and See, 2007). This line of research points to the potential clinical use of progesterone, ALLO or related compounds in the treatment of cocaine use and addiction. Anker, J.J., Holz, N.A., Zlebnik, N., and Carroll, M.E. Effects of Allopregnanolone on the Reinstatement of Cocaine-Seeking Behavior in Male and Female Rats. Psychopharm. [Epub ahead of print], 2008.