Site-specific NMDA receptor antagonists produce differential effects on cocaine self-administration in rats

https://doi.org/10.1016/S0014-2999(99)00446-XGet rights and content

Abstract

The effects of site-specific NMDA receptor antagonists on intravenous cocaine self-administration were examined in rats trained to self-administer cocaine (0.25 mg/infusion) on a fixed ratio (FR) 5 schedule with a 20-s time-out (TO) after each reinforcer. The non-competitive NMDA receptor antagonists, dizocilpine (MK-801, (+)-5-methyl-10,11-dihydro-5H-dibenzo[a,d]cyclohepten-5,10-imine hydrogen maleate) (0.05–0.2 mg/kg i.p.) and memantine (1,3-dimethyl-5-amino-adamantane hydrochloride) (2.5–20 mg/kg i.p.), dose-dependently decreased cocaine self-administration, while the competitive NMDA receptor antagonist, CGP 39551 (dl-(E)-2-amino-4-methyl-5-phosphono-3-pentanoic acid carboxyethylester) (2.5–15 mg/kg i.p.), and the NMDA/glycine receptor antagonist, L-701,324 (7-chloro-4-hydroxy-3(3-phenoxy)-phenyl-2(H)quinolone) (1.25–10 mg/kg p.o.), were without effect. Under a progressive ratio (PR) schedule, dizocilpine (0.15 mg/kg i.p.) increased the number of cocaine infusions in a manner similar to increasing the unit dose of cocaine, suggestive of potentiation of cocaine reward. Conversely, memantine (10 mg/kg i.p.) produced rate-decreasing effects on the PR schedule. These results demonstrate that NMDA receptor antagonists acting at different modulatory sites of the NMDA receptor do not share dizocilpine's cocaine reward enhancing effects although they are all known to be effective blockers of NMDA receptor activity.

Introduction

Manipulation of glutamatergic neurotransmission produced by glutamate receptor antagonists alters a wide spectrum of functional responses in rodents, including lethality, convulsions, circulatory effects and locomotor stimulation following acute administration of cocaine and other psychostimulants (see Rockhold, 1998, for review). Glutamate receptor antagonists prevent also the development of sensitization to locomotor stimulation induced by repeated administration of cocaine (Karler et al., 1989); in addition, there is now increasing evidence for glutamatergic participation in the reinforcing actions of cocaine, assessed in various self-administration paradigms (Danysz and Parsons, 1998).

Alteration of cocaine self-administration by dizocilpine (MK-801), a non-competitive antagonist of the NMDA receptor, has been demonstrated best. Systemic pretreatment with dizocilpine reduced cocaine self-administration in rats maintained on a fixed ratio (FR) schedule in a manner similar to increasing the unit dose of cocaine (Pierce et al., 1997) and increased the breaking point of cocaine responding on a progressive ratio (PR) schedule (Ranaldi et al., 1996). Taken together, these findings suggest that dizocilpine enhances the reinforcing properties of cocaine. This effect, however, can be seen only in a very narrow dose range (0.1–0.15 mg/kg), whereas higher doses of dizocilpine produce a loss of discriminative responding for cocaine and motor impairment (Schenk et al., 1993). Furthermore, dizocilpine blocks enhanced acquisition of intravenous cocaine self-administration produced by pre-exposure to amphetamine (Schenk et al., 1993) and causes reinstatement of responding for cocaine following extinction (DeVries et al., 1998).

Attempts to modulate cocaine self-administration with other NMDA receptor antagonists than dizocilpine have produced conflicting results. For example, infusion of the competitive NMDA receptor antagonist, 2-amino-5-phosphonovaleric acid (APV), into the nucleus accumbens increased cocaine self-administration in rats, indicative of antagonism of cocaine reward by APV (Pulvirenti et al., 1992). In contrast, when the non-competitive antagonist, dextrometorphan, was given systemically, it reduced cocaine responding both on a continuous reinforcement and PR schedule (Pulvirenti et al., 1997).

Although these findings suggest that manipulation of NMDA receptor activity may alter the reinforcing properties of cocaine, little is known about the underlying neuropharmacological mechanisms involved in this phenomenon. Since antagonists acting at distinct modulatory sites of the NMDA receptor complex have different pharmacological and behavioural profiles, and could thus produce differential effects on cocaine reinforcement, the purpose of the present study was to examine how cocaine self-administration is modulated by CGP 39551 (Fagg et al., 1990), an antagonist acting competitively at the endogenous agonist recognition site, dizocilpine (Wong et al., 1986) and memantine (Bormann, 1989) that bind to a non-competitive site within the NMDA receptor channel, and L-701,324 (Kulagowski et al., 1994), an antagonist at the NMDA/glycine recognition site. In this study, dizocilpine and memantine were selected as representatives for high- and low-affinity channel blockers, respectively.

Section snippets

Subjects

Forty-two male Wistar rats (Wistar:Han/HY, Department of Laboratory Animals, University of Helsinki) weighing 140–160 g upon arrival were used. Animals were group housed (two per cage) in a room with controlled temperature (20±2°C) and humidity (55±5%) on a reverse 12-h light–dark cycle (lights on 6 p.m.–6 a.m.) with free access to tap water and RM1(E) pellet food (SDS, Witham, UK), except during initial food training (see Section 2.5). All behavioural testing occurred during the dark phase of

Results

Data presented in Fig. 1 show that both non-competitive NMDA receptor antagonists, dizocilpine and memantine, dose-dependently decreased cocaine self-administration during the 2-h sessions [main effects, dizocilpine: F(4,24)=18.16, P<0.0001; memantine: F(4,36)=12.06, P<0.0001]. In contrast, the competitive NMDA receptor antagonist CGP 39551 and the NMDA/glycine antagonist, L-701,324, failed to reliably alter the number of cocaine infusions [CGP 39551: F(4,28)=0.89, P>0.05; L-701,324: F

Discussion

In the present study, we investigated the modulation of intravenous cocaine self-administration by NMDA receptor antagonists acting at three different modulatory sites on the NMDA receptor: the site for the endogenous transmitter glutamate (CGP 39551), the non-competitive binding site within the NMDA receptor channel (dizocilpine, memantine) and the NMDA/glycine-specific binding site (L-701,324). The cocaine unit dose (0.25 mg/infusion) and the reinforcement schedule (FR5 TO 20-s) used in the

Acknowledgements

Supported by the Swedish Medical Research Council (Project No. 7688), and funds from the Karolinska Institutet (Stockholm) and the National Public Health Institute (Helsinki).

References (36)

Cited by (58)

  • The strength of reward-related learning depends on the degree of activation of ventral tegmental area dopamine neurons

    2018, Behavioural Brain Research
    Citation Excerpt :

    Although others have shown that scopolamine can disrupt food consumption [18] or responding for food [19], in our study scopolamine had no effect on food consumption/reward. Second, we and others previously showed that treatment with MK-801 in fact enhances the rewarding effects of cocaine [17,20–22] and food [23,24]. Therefore, it is unlikely that reduced difference scores in the CS-only test derives from scopolamine- or MK-801-induced reductions in food reward.

  • Blockade of NMDA receptors blocks the acquisition of cocaine conditioned approach in rats

    2018, European Journal of Pharmacology
    Citation Excerpt :

    The fact that MK-801 treatment decreased cocaine intake under a FR1 schedule of reinforcement (without disrupting the pattern of self-administration) suggests that MK-801, in fact, increased the rewarding effects of cocaine. Such findings are congruent with other reports (Allen et al., 2005; Hyytia et al., 1999; Ranaldi et al., 1996) and argue that increases in CS latencies observed during the CS-only test in rats previously treated with MK-801 (Experiment 1) are best understood as MK-801-induced impairment in the acquisition of conditioned approach. An alternative explanation for the significant increase in CS latencies during the CS-only test in Experiment 1 is that rats under MK-801 treatment were motorically incapacitated to approach the nose-poke port in a timely manner, and therefore missed a number of cocaine infusions (USs).

  • Psychostimulants: Basic and clinical pharmacology

    2015, International Review of Neurobiology
    Citation Excerpt :

    The study designed to assess ketamine treatment on mindfulness-based cocaine relapse prevention and abstinence rates is still running, and no data are reported. The voltage-dependent noncompetitive NMDA receptor antagonist memantine, which has received marketing approval for the treatment of Alzheimer's disease, reduces cocaine self-administration in rats and the reinstatement of cocaine-seeking behavior in rodents (Bespalov, Zvartau, Balster, & Beardsley, 2000; Hyytiä, Bäckström, & Liljequist, 1999), cocaine conditioned locomotion (Bespalov, Dravolina, Zvartau, Beardsley, & Balster, 2000), place preference (Kotlińska & Biała, 2000), and sensitization (Li, White, & Wolf, 2000). However, human studies have not supported a role for memantine in reducing cocaine self-administration in humans (Collins, Vosberg, Ward, Haney, & Foltin, 2007); however, the stratification of these patients may not have been optimal (Bisaga et al., 2010).

  • Continuous exposure to dizocilpine facilitates escalation of cocaine consumption in male Sprague-Dawley rats

    2014, Drug and Alcohol Dependence
    Citation Excerpt :

    Thus, it is not known whether this effect is specific to the NMDAR or extends to inhibition of glutamate function more generally. Rats injected with a bolus dose of dizocilpine self-administer cocaine at a lower rate under an FR1 schedule of reinforcement (with longer, but regular inter-infusion intervals) and attain a higher break point when responding under a progressive ratio (PR) schedule of cocaine reinforcement when compared with vehicle-injected controls (Allen et al., 2005, 2007b; Pierce et al., 1997; Hyytiä et al., 1999). Thus, it was unexpected that dizocilpine treatment did not lead to increases in break point when rats responded under a progressive ratio schedule after the escalation phase of the experiment, an effect observed after continuous treatment with LY235959, also (Allen et al., 2007b).

  • Nonhuman Primate Models of Drug and Alcohol Addiction

    2012, Nonhuman Primates in Biomedical Research: Diseases: Second Edition
View all citing articles on Scopus
View full text