Research Findings - Behavioral Research

Self-Administration of Heroin Leads to Alterations in Immune Function in Rats

The high incidence of bacterial, viral and fungal infections among human heroin abusers suggests that heroin use might alter infectious disease resistance. Previous work from Dr. Lysle's laboratory has demonstrated that heroin alters the induction of nitric oxide, a molecule known to regulate immune responses and resistance to infection. The present study was designed to investigate whether environmental stimuli paired with heroin administration alter the expression of inducible nitric oxide synthase (iNOS), an enzyme responsible for nitric oxide production. On three conditioning days, 48 hr apart, the experimental group received an injection of heroin (1 mg/kg) immediately prior to being placed for 1 hr in an environmental chamber that served as the conditioned stimulus. An unpaired control group was treated the same way except that the heroin injections and placement in the chamber were separated by 12 hr. A saline control group received an injection of saline before being placed in the environmental chamber. Six days after the third conditioning session, all rats were re-exposed to the conditioning chambers in the absence of heroin. Immediately after this 1 hr exposure, they were injected with lipopolysaccharide (LPS) to induce iNOS expression. Six hr later, the animals were euthanized and the production of nitric oxide was determined by measuring iNOS mRNA and protein in spleen, liver and lung. The results indicated that exposure to the conditioned stimulus on the test day in the experimental group produced a significant reduction in the expression of iNOS mRNA and protein in spleen, liver and lung. Control procedures indicated that this effect was due to conditioning processes. This study provides the first evidence that heroin-induced alteration of iNOS expression can be conditioned and suggests that environmental stimuli associated with drug abuse may modulate infection susceptibility. Lysle, D.T., and Ijames, S.G. Heroin-associated Environmental Stimuli Modulate the Expression of Inducible Nitric Oxide Synthase in the Rat. Psychopharmacology, 164, pp. 416-422, 2002.

Mechanisms of Neural Plasticity Underlying Behavioral Sensitization

Drug addiction involves persistent functional changes in the brain circuits underlying motivated behavior that are responsible for protracted craving, drug seeking and relapse. The present study investigated the role of the phosphatidylinositol 3 kinase (PI3K) signal transduction pathway in long lasting behavioral sensitization to cocaine in rats. Based upon previous research on long-term potentiation (LTP), it was hypothesized that PI3K is required for the expression but not the induction of behavioral sensitization to cocaine. Two experiments were conducted. In Experiment 1, two groups of rats were injected IP for 5 consecutive days with cocaine (15 mg/kg) or its saline vehicle. Half of each group received an ICV injection of either the reversible PI3K inhibitor LY294002 or its DMSO vehicle 20 min before each cocaine or saline injection. Sensitization to cocaine induced locomotor activity was observed in the group given cocaine/DMSO, but not in the group given cocaine/LY294002. Ten days later all animals were challenged with cocaine in the absence of LY294002. Both cocaine groups, those pretreated with DMSO as well as those pretreated with LY294002 showed a sensitized locomotor response to cocaine on this test. These results suggest that PI3K activity is not necessary for the induction of cocaine sensitization, but is necessary for its expression. Experiment 2 provided a further test of this hypothesis. As in the first experiment two groups were pretreated with either cocaine or saline, but without the ICV LY294002 or its vehicle. Then ten days after the last injection, those rats that showed behavioral sensitization were randomly assigned to one of two groups. One group was given LY294002 prior to being challenged with cocaine; the other was given DMSO before the cocaine challenge. The first group did not show locomotor sensitization to the cocaine injection, whereas the second group did. One week later, cocaine was administered again to both groups without LY294002 pretreatment and now both groups showed a sensitized locomotor response to cocaine. This result confirms the hypothesis that LY294002 reversibly blocks the expression of behavioral sensitization to cocaine and indicates that PI3K activity is necessary for the expression but not the induction and persistence of sensitization. The similarity between the role of PI3K in LTP and in cocaine-induced behavioral sensitization suggests that similar synaptic mechanisms underlie both phenomena. Izzo, E., Martin-Fardon, R., Koob, G., Weiss, F., and Sanna, P.P. Neural Plasticity and Addiction: PI3-kinase and Cocaine Behavioral Sensitization. Nature Neuroscience, 5(12), pp. 1263-1264, 2002.

Prescription Opioid Non-Medical Use and Abuse: Task Force Position Statement

A College on Problems of Drug Dependence Task Force, chaired by NIDA grantee James Zacny has issued a position paper on prescription opioid non-medical use and abuse. Non-medical use and abuse of prescription opioids are on the rise in the United States, illicit use of several widely prescribed opioids has increased disproportionately more than licit use, and the prevalence of prescription opioid abuse appears to be similar to that of heroin and cocaine abuse. The task force noted that policy decisions about prescription opioid non-medical use and abuse must take a balanced approach so that strategies developed to prevent and reduce diversion of prescription opioids do not deter physicians from prescribing high efficacy opioids when necessary for pain management. While acknowledging that the science related to the abuse of prescription opioids is limited, the task force recommended a number of actions that should be taken by scientists working with appropriate governmental, non-governmental and industry representatives: 1] Initiate a national forum to review the extant taxonomy of terms related to the non-medical use and abuse of prescription opioids and their impact on policy; 2] Initiate a comprehensive assessment of non-medical use and abuse of prescription opioids, including an assessment of its magnitude and demographics, and its impact on public health and safety; 3] Initiate a comprehensive assessment of opioid use and risk of abuse in patients with pain; 4] Initiate research programs to assess the abuse liability of prescription opioids, developing comparative data fur currently used agents; 5] Initiate the development of targeted prevention programs after identifying those sub-populations that are at risk for prescription opioid abuse; 6] Initiate the development of controlled clinical trials to assess treatments for prescription opioid abuse in different sub-populations; 7] Initiate a broad educational program on the proper prescribing of opioids for pain coordinated with pain experts and substance abuse experts. Zacny, J., Bigelow, G., Compton, P., Foley, K., Iguchi, M., and Sannerud, C. College on Problems of Drug Dependence Taskforce on Prescription Opioid Non-Medical Use and Abuse: Position Statement. Drug and Alcohol Dependence, 69, pp. 215-232, 2003.

Predicting Choice Behavior Between Cocaine and Food in Monkeys

Drug abuse can be thought of as a maladaptive, habitual choice of a drug over non-drug reinforcers. An understanding of the factors that determine the choice to self-administer a drug, therefore, will help us understand drug abuse and perhaps how to treat and prevent it. The matching law, a mathematical expression that accurately predicts response choice between alternative reinforcers, has successfully modeled choice behavior. Basically, the matching law predicts that one response is chosen over an alternative response in direct proportion to the reinforcement available for the two responses. Fro example, if food is twice as likely to occur for choosing response A over response B, the matching law correctly predicts that response A will occur twice as often as response B. The generalized matching law has provided a good fit to experimental data across a variety of species, behaviors and reinforcers. However, there have been few tests of the matching law that have pitted a drug against a non-drug alternative. Dr. Woolverton and his colleagues investigated whether the matching law provides a good fit to experimental data involving a choice between cocaine and food. Thus, four male rhesus monkeys pressed one lever for food and another lever for cocaine (0.025 and 0.05 mg/kg per injection) over a number of sessions that varied the frequency of reinforcement. Results indicated that the matching law provided a good account of the data. As with studies involving food only, monkeys showed undermatching, i.e., they responded less for the alternative with the greater reinforcement frequency than predicted by the matching law. In addition, there was a bias toward food when it was pitted against the lower dose of cocaine but a bias away from food when it was pitted against the higher dose of cocaine. These results suggest that the matching law can reliably describe the relationship between relative rate of reinforcement and behavior in the choice between a drug and a non-drug reinforcer. A quantitative model of choice like the matching law may eventually be useful for predictions of drug choice as typified in drug-abuse situations. Anderson, K.G., Velkey, A.J., and Woolverton, W.L. The Generalized Matching Law as a Predictor of Choice Between Cocaine and Food in Rhesus Monkeys. Psychopharmacology, 163, pp. 319-326, 2002.

Gonadectomy Has Opposite Effects on the Antinociceptive Effects of Opioids in Male and Female Rats

Greater sensitivity of males than females to the antinociceptive effects of mu opioids has been reported from clinical research, and is paralleled by findings from animal research with non-human primates and rodents. The basis for this sex difference is not fully accounted for by either pharmacokinetic factors, mu opioid binding affinity or receptor density. In an effort to examine the role of gonadal hormones, researchers at the University of North Carolina at Chapel Hill directly manipulated hormone levels via gonadectomy in F344 rats, which typically display relatively large sex differences in opioid antinociception, and in Sprague Dawley rats, which typically exhibit small differences. Given evidence that the magnitude of sex difference in opioid nociception is inversely related to the relative effectiveness of the opioid, the researchers examined the effects of both high-efficacy mu opioids (morphine and etorphine) and low-efficacy mu opioids (buprenorphine and dezocine) as well as mixed-action opioids with low efficacy at both mu and kappa opioid receptors (butorphanol and nalbuphine). Consistent with prior research, the researchers found greater opioid antinociception in intact males than intact females, with larger sex differences occurring with the less-effective opioids and in the F344 strain. In both strains, gonadectomy resulted in a decrease in opioid antinociception in males and an increase in females. The smallest effects were seen with high efficacy mu opioids, i.e., the magnitude of the decrease in males and the increase in females was inversely related to the relative effectiveness of the opioid. NIDA-grantee Theodore Cicero and colleagues (Cicero et al., 2002; May 2003 Director's Report) have shown that shortly after birth gonadectomy in males decreases morphine's antinociception effect, while in females the combination of gonadectomy-plus-testosterone increases morphine's antinociception effect -- reflecting the organizational effects of sex hormones in the early development of the rat brain. Taken together, these studies provide evidence that sex differences in opioid antinociception are due in part to the dual role played by gonadal hormones in sexual differentiation of the brain and in the acute response to opioids in adulthood. Terner, J.M., Barrett, A.C., Grossell, E., and Picker, M.J. Influence of Gonadectomy on the Antinociceptive Effects of Opioids in Male and Female Rats. Psychopharmacology, 163, pp.183-193, 2002.

Differential Subjective Effects of D-amphetamine by Gender, Hormone Level and Menstrual Cycle Phase

Prior research by Dr. Harriet de Wit and colleagues at the University of Chicago has shown that women report higher positive subjective effects of d-amphetamine in the follicular phase of the menstrual cycle when estrogen levels are high relative to progesterone than in the luteal phase when both hormones are high. She has also shown that exogenous administration of progesterone can produce moderate decreases in mood states. Thus, to further understand the role of estrogen and progesterone in producing the menstrual cycle modulation of the subjective effects of amphetamine and to examine possible sex differences, Dr. de Wit and her colleagues compared the subjective responses to d-amphetamine in males and females and also examined in women the correlation between these responses and endogenous levels of estrogen and progesterone during the follicular phase. Consistent with her prior study, the positive subjective responses, e.g. euphoria, feeling "high," and wanting more drug, were higher in the follicular phase than the luteal. Further, subjective responses of women during the follicular phase did not differ from men, but were lower than men during the luteal phase. In the follicular phase, these stimulant effects were inversely related to salivary levels of progesterone, but were only weakly positively correlated with salivary estradiol. These data point to the role of the menstrual cycle in understanding sex differences in the subjective response to d-amphetamine and point to the role of progesterone in understanding variations in amphetamine responsivity among women. These findings have important implications for understanding the role the menstrual cycle in the use and escalation of d-amphetamine abuse as well as its role in treatment. White, T.L., Justice, A.J., and de Wit, Harriet. Differential Subjective Effects of D-amphetamine by Gender, Hormone Level and Menstrual Cycle Phase. Pharmacology, Biochemistry and Behavior, 73, pp. 729-741, 2002.

Influence of Estrus Cycle and Estradiol on Behavioral Sensitization to Cocaine in Female Rats

Several studies have shown that sensitization of locomotor response to cocaine and amphetamine is greater in females that males. Prior research by Dr. Kathryn Cunningham and colleagues at the University of Texas Medical Branch has shown that the acute locomotor response to cocaine in female rats is highest during proestrus when estrogen levels are highest. In a recent study, she examined the role of the estrus cycle in cocaine sensitization of the locomotor response. Following 5 days of twice daily injections of cocaine or saline and a three-day withdrawal period, the expression of locomotor sensitization was assessed via a challenge injection of cocaine. Compared to the saline group, sensitization of the locomotor response in the cocaine group was observed only in the diestrus phase when estrogen levels are lowest. In a second study, the researchers examined whether the presence or absence of estradiol would affect the development and expression of cocaine-induced locomotor sensitization. An ovariectomized group and an ovariectomized group given estradiol replacement were each given twice daily injections of cocaine for five days, and then were given a challenge injection of cocaine 3, 13, and 34 days after this treatment. During the five days of twice daily cocaine injections, the two groups developed similar levels of sensitization. After 3 days of withdrawal, neither group exhibited sensitization to the cocaine challenge, but after 13 and 34 days, both groups exhibited progressive sensitization. Again the estradiol-treated rats did not exhibit greater sensitization than the non-estradiol-treated rats. These data point to a role of the estrus cycle in the acute locomotor response to cocaine in females, but do not implicate estradiol in cocaine sensitization of the locomotor response. Sell, S.L., Thomas, M.L., and Cunningham, K.A. Influence of Estrous Cycle and Estradiol on Behavioral Sensitization to Cocaine in Female Rats. Drug and Alcohol Dependence, 67, pp. 281-290, 2002.

Synapses in Different Brain Areas are Modified by Self-administered Morphine as Compared to Experimenter-administered Drug

From studies on animal models of drug addiction it is becoming increasingly clear that drugs of abuse can have different behavioral and neurobiological effects depending on whether they are administered passively, by the experimenter, or self administered. Further understanding of these differences will help uncover causal connections between particular neurobiological alterations and the behavioral effects of drugs of abuse. Terry Robinson and his colleagues have been investigating the ability of drugs to remodel connections in the brain by altering the density of synaptic spines on neuronal dendrites. In a recent study, they compared the long-term effects of experimenter- versus self-administered morphine on dendritic spine morphology. In most brain regions (medial frontal cortex, occipital sensory cortex, and nucleus accumbens), spine density decreased irrespective of mode of administration. However, in the CA1 and dentate gyrus regions of the hippocampus, only self-administered morphine decreased spine density, while in parietal sensory cortex only experimenter-administered morphine was effective. In orbital frontal cortex, by contrast, both treatments increased spine density, but self-administration had a much greater effect. The results indicate that morphine has a persistent (at least 1 month) effect on spine density in many brain regions and on many different cell types, but the effect is specific and varies as a function of both brain region and mode of administration. The ability of morphine to remodel synaptic inputs in a regionally specific manner may account for the many different long-term sequelae of opioid use, including impairments of cognitive function. The different effects of experimenter- versus self-administered drug suggest that some neuronal changes do not result simply from the drug's molecular actions, and they help identify brain areas and mechanisms that may be particularly involved in linking external cues and motivational states with drug taking. Robinson, T.E., Gorny, G., Savage, V.R., and Kolb, B. Widespread but Regionally Specific Effects of Experimenter- versus Self-administered Morphine on Dendritic Spines in the Nucleus Accumbens, Hippocampus, and Neocortex of Adult Rats. Synapse, 46, pp. 271-279, 2002.

A Variety of Pharmacological Classes of Drugs of Abuse, or Stress, Trigger a Common Synaptic Adaptation that May Contribute to Cross Sensitization between Drugs and Stress

A previous collaborative study between the laboratories of Antonello Bonci and Robert Malenka showed that a single dose of cocaine altered the response of dopaminergic neurons in the VTA to glutamatergic inputs from the prefrontal cortex. Altered synaptic transmission was evident as an enhanced synaptic current through AMPA receptors relative to current through NMDA receptors, which is an established hallmark of long-term potentiation (LTP). In a new study, they found a similar effect with other addictive substances -- amphetamine, morphine, nicotine, and ethanol -- some of which have very different initial molecular mechanisms from cocaine. They also observed an enhanced AMPA:NMDA response ratio after animals were subjected to acute stress (forced swimming). As a control, they tested two psychoactive compounds, the antidepressant fluoxetine and carbamazepine (commonly used to treat seizure disorders and bipolar disorder), neither of which affected excitatory synapses in the VTA. The stress-induced potentiation could be blocked with the glucocorticoid antagonist RU-486, which did not block the response to cocaine, indicating that the cocaine effect was not simply mediated by the stress of injection. The results of these experiments suggest that plasticity at excitatory synapses on dopamine neurons may be a key early neural adaptation contributing to addiction and its interactions with stress, and thus may be an attractive therapeutic target for reducing the risk of addiction. Saal, D., Dong, Y., Bonci, A., and Malenka, R.C. Drugs of Abuse and Stress Trigger a Common Synaptic Adaptation in Dopamine Neurons. Neuron, 37, pp. 577-582, 2003.

A Likely Neurobiological Substrate for Increased Maternal Aggression Produced by Chronic Cocaine Administration During Gestation

Females of most mammalian species aggressively protect their offspring from potential harm. Maternal aggressive behavior by lactating female rats consists of a characteristic set of postures, threats and attacks used to protect the young from intruders in the nest area. This behavior is generally adaptive, but previous research by Josephine Johns has shown that, following chronic gestational cocaine administration, rat dams can become so aggressive that they leave their young unprotected and vulnerable, or even cause injury to their offspring during interactions with an intruder. Dr. Johns has previously observed a decrease in oxytocin levels in the amygdala of rats chronically treated with cocaine during gestation. This observation, together with other evidence, led to her hypothesis that heightened aggressiveness in these animals might be caused by the deficit of oxytocin in the amygdala. To test the hypothesis, she implanted dams with bilateral cannulae into the central nucleus of the amygdala and infused them with an oxytocin antagonist (OTA) or vehicle 4 hours before testing. Maternal aggressive behavior was compared among drug-na•ve dams infused with OTA, control animals infused with vehicle, and chronic cocaine-treated dams infused with vehicle only. Drug na•ve animals infused with OTA behaved significantly more aggressively towards intruders compared to control animals, and at a level comparable to that of the cocaine-treated animals. The results suggest that disruption of oxytocin activity in the central amygdala may be sufficient to selectively increase maternal aggressive behavior. To further test her hypothesis that this mechanism could account for the behavior of cocaine-treated animals, Dr. Johns plans to determine whether oxytocin infusions in these animals will reverse their abnormal aggressiveness. Cocaine abuse by human mothers is correlated with a high incidence of child neglect and abuse. Dr. Johns' research on the neurobiological substrates of these behaviors in an animal model may help identify possible therapeutic targets. Lubin, D.A., Elliott, J.C., Black, M.C., and Johns, J.M. An Oxytocin Antagonist Infused Into the Central Nucleus of the Amygdala Increases Maternal Aggressive Behavior. Behavioral Neuroscience, 117, pp. 195-201, 2003.

Serotonergic Receptor Substrates for the "Reinforcing" Effects of 3,4-methylenedioxy-methamphetamine (MDMA, "ecstasy")

The drug ecstasy produces subjective effects that resemble both psychostimulants, like amphetamine, and hallucinogenics, like LSD. In human subjects, some of these subjective effects arise from activation of central dopamine (DA) transmitter systems, and other effects appear more closely linked with stimulation of the neurotransmitter, serotonin (5-HT). Unlike the classic hallucinogenic drugs, which act primarily at 5-HT2a receptors, MDMA is self-administered in animal models. However, it is unknown whether dopaminergic or serotonergic substrates mediate these effects. Recently, investigators from the laboratories of Drs. James Woods and Gail Winger trained four rhesus monkeys to self-administer i.v. S(+)-MDMA and R(-)-MDMA and pretreated these animals with either a 5-HT2 antagonist (ketanserin) or a 5-HT2a antagonist (MDL100907) prior to the session. Dose response determinations (range=0.001-0.3 mg/kg/injection) were made for both enantiomers and the racemic mixture of MDMA. The same animals were also tested with 0.01 mg/kg/injection cocaine and 0.0003-0.003 mg/kg/injection methamphetamine. All animals self-administered racemic MDMA and both of the stereoisomers, generating inverted U dose response curves. However, none of the doses maintained injection rates as high as seen for cocaine or the 0.003 dose of methamphetamine. When animals were pretreated with the serotonin receptor antagonists, cocaine responding was not significantly affected by 0.1 mg/kg pretreatment with either drug. And while both antagonists produced shifts in the inverted U function for (+)-MDMA these effects were also not significant. However, with (-)-MDMA, the 0.1 mg/kg dose of either antagonist significantly suppressed self-administration. Due to subject variability in self-administration response rates, this suppression was statistically significant only at 0.03(-)-MDMA for both pretreatments. Several interesting observations were made in this study. First, self-administration of MDMA was highly variable across animals - much more so than for the comparison drug, cocaine. Secondly, while further studies are needed to assess relative reinforcement (e.g., choice procedures), in the doses tested, cocaine was a more potent reinforcer of operant behavior. And lastly, in agreement with a growing body of literature to support the role of serotonergic substrates in drug reinforcement (e.g., interacting DA-5HT systems), it appears that 5-HT2a receptors may be important for ecstasy's hedonic or rewarding properties. Fantegrossi, W.E., Ullrich, T., Rice K.C., Woods, J.H. and Winger, G. 3,4-methylenedioxymethamphetamine (MDMA, "ecstasy") and its Stereoisomers as Reinforcers in Rhesus Monkeys: Serotonergic Involvement. Psychopharmacology, 161, pp. 356-364, 2002.

Sucrose and the Development of Conditioned Place Preference with Fentanyl and Amphetamine: The Role of Reinforcement History?

Animals who consume palatable foods and fluids show altered behavioral responses to drugs of abuse; (e.g., more pronounced opiate-induced antinociception and anorexia). However, providing palatable solutions either preceding, or concomitant with availability of self-administration, attenuates drug intake. Conditioned place preference (CPP) is a paradigm used to assess incentive motivational value of an environment previously paired with a drug reward. When environmental cues are paired with the positive effects of drugs, these cues take on reinforcing properties of their own by virtue of classically conditioned associations with the subjective drug experience. In CPP, these reinforcing properties are measured by time spent in proximity of the cues, or what some have called "approach responding" to the drug-associated environment. Recently, Dr. Robin Kanarek and colleagues have examined the ability of sucrose availability to alter the establishment of CPP with an opiate, fentanyl, and the psychostimulant, amphetamine. All rats in this study were fed rat chow, but some groups also had access to a 32% (w/v) sucrose solution 24 hr per day. They then underwent conditioning with 0.004 or 0.016 mg/kg fentanyl paired with the non-preferred side of a CPP apparatus and were tested in the drug free state to see if side preferences had shifted as a result of the association with drug. Collapsing across chow-only and chow+sucrose groups, there was a significant increase of time spent in the high dose-paired side compared to baseline times before the drug conditioning. However, when data were examined separately for the two groups, only the sucrose drinking animals showed a significant CPP. Interestingly, sucrose drinking rats also showed greater antinociceptive effects of fentanyl when assessed with tail-flick measures of analgesia, indicating a greater sensitivity to opiate effects. When a separate set of animals was tested for CPP to 0.33 or 1.00 mg/kg amphetamine, collapsed groups showed a significant shift in time spent on the drug-paired side at both doses. But again, only the sucrose-drinking rats showed a significant CPP when groups were analyzed separately. In fact, at the lower dose, sucrose-drinking rats spent double the amount of time in the drug-associated environment compared to chow-only animals. The investigators suggest that sucrose may enhance the rewarding effects of both drugs by increasing dopamine release in the nucleus accumbens, or perhaps via actions upon endogenous opioid systems. In light of previous observations that sucrose attenuates drug self-administration, this CPP finding suggests that: (1) less drug is needed to produce reinforcement in self-administration when sucrose is available; and (2) sucrose intake stimulates central reward systems, which sums with the reinforcing effects of these drugs in CPP. Vitale, M.A., Chen, D. and Kanarek, R.B. Chronic Access to a Sucrose Solution Enhances the Development of Conditioned Place Preferences for Fentanyl and Amphetamine in Male Long-Evans Rats. Pharmacol. Biochem. Behav., 74, pp. 529-539, 2003.