N-methyl-D-aspartate Antagonist Phencyclidine Research Articles

Alterations in High-Frequency Neuronal Oscillations in a Cynomolgus Macaque Test of Sustained Attention Following NMDA Receptor Antagonism.

A growing body of evidence indicates that neuronal oscillations in the gamma frequency range (30-80 Hz) are disturbed in schizophrenic patients during cognitive processes and may represent an endophenotype of the disease. N-methyl-D-aspartate (NMDA) receptor antagonists have been used experimentally to induce schizophrenia-like symptoms including cognitive deficits in animals and humans. Here we characterized neuronal oscillations and event-related potentials (ERPs) in Cynomolgus macaques fully trained to perform a continuous performance test (CPT) in the presence and absence of the NMDA antagonist phencyclidine (PCP). Macaques (n=8) were trained to touch 'target' stimuli and ignore 'distractor' stimuli presented randomly on a touchscreen. Subsequently, all subjects were implanted with epidural EEG electrodes over frontal (FC) and parietal cortices (PC) and later tested under vehicle (saline, i.m.) or acute PCP (0.1-0.3 mg/kg, i.m.) conditions. Compared with vehicle treatment, PCP produced a significant dose-dependent decrease in CPT performance accuracy and increased reaction times. Furthermore, PCP elevated the amplitudes of 'low' (30-50 Hz) and 'high' (51-80 Hz) gamma oscillations in FC and PC around target presentations for all correct responses. The CPT accuracy was inversely correlated with the gamma band amplitude in the presence of PCP. Additionally, PCP delayed the N100 peak latency in FC, and prolonged and suppressed the cognitively relevant P300 component of mean ERPs in FC and PC, respectively. The NMDA receptor antagonist-induced alteration in neuronal oscillations and ERPs may contribute to the observed cognitive deficits in macaques, and enhance our understanding of EEG recordings as a translatable biomarker.

Effects of novel, high affinity glycine transport inhibitors on frontostriatal dopamine release in a rodent model of schizophrenia

Dopaminergic hyperactivity within frontostriatal brain systems is a key feature of schizophrenia, and an objective neural correlate of positive schizophrenia symptoms. N-methyl-d-aspartate (NMDA) receptors are known to play a prominent role in regulation of frontostriatal dopamine release. Furthermore, disturbances in glutamatergic function are increasingly being linked to pathophysiology of both positive and negative symptoms of schizophrenia. Prior studies have demonstrated that subchronic continuous administration of the NMDA antagonist phencyclidine (PCP) induces schizophrenia-like hyper-reactivity of frontostriatal dopamine release to amphetamine (AMPH) in rodents, and that effects were reversed by glycine and the prototypic glycine transport inhibitor (GTI) NFPS. The present study investigates effectiveness of the novel, high affinity and well tolerated GTIs, R231857, R231860 and Org29335, to reverse schizophrenia-like enhancement of AMPH-induced DA release, along with effects of the partial glycine-site agonist d-cycloserine. As previously, PCP had no significant effect on basal DA levels, but significantly enhanced AMPH-induced DA release in prefrontal cortex. All GTIs tested, as well as d-cycloserine, significantly reduced PCP-induced enhancement of DA release in prefrontal cortex. Neither PCP nor GTIs significantly affected striatal DA release. Overall, these findings suggest that treatments which target the glycine modulatory site of the NMDA receptor may significantly reverse NMDA receptor antagonist-induced dysregulation of frontal DA systems, consistent with potential beneficial effects on positive-, in addition to negative-, symptoms of schizophrenia.

Analysis of licking microstructure provides no evidence for a reduction in reward value following acute or sub-chronic phencyclidine administration

The N-methyl D-aspartate antagonist phencyclidine (PCP) is purported to mimic the negative, cognitive and positive symptoms of schizophrenia. Thus, acute and sub-chronic PCP treatment in rodents might produce anhedonia, a decrease in the pleasure produced by rewards. Experiment 1 investigated whether acute PCP treatment changes the value of sucrose. A comparison was made to (+)MK-801, a drug often used interchangeably with PCP in preclinical studies. Experiment 2 assessed the effects of withdrawal from sub-chronic PCP treatment on the value of sucrose. Experiment 1 examined the dose-response effects of PCP and (+)MK-801 on licking microstructure during sucrose consumption. Experiment 2 assessed the effects of withdrawal from sub-chronic PCP treatment (5 mg/kg twice daily for 7 days), on licking microstructure during sucrose consumption. Locomotor activity testing was carried out in experiment 2 to confirm the sensitisation effect of the PCP regimen on amphetamine-induced hyperlocomotion. Low to moderate acute doses of PCP and (+)MK-801 increased the amount of sucrose consumed. Higher doses decreased consumption and the number of licks per cluster (cluster size) but also increased the average inter-lick interval, which may indicate motor impairment. There was no evidence that withdrawal from sub-chronic PCP treatment produced decreases in consumption or lick cluster size. Following acute PCP treatment, we found no evidence of reduced reward value without the presence of confounding motor deficits. Sub-chronic PCP withdrawal also produced no decrease in reward value. Therefore, the current results indicate that neither acute PCP treatment nor sub-chronic PCP withdrawal produce consummatory anhedonia.

Cannabinoid CB1 receptor dependent effects of the NMDA antagonist phencyclidine in the social withdrawal model of schizophrenia

Clinical and laboratory findings suggest that cannabinoid signalling is implicated in schizophrenia. However, the interaction remains poorly understood, as data are often contradictory. Here we investigated wild-type (WT) and cannabinoid CB1 receptor-knockout (CB1-KO) mice in the phencyclidine-induced social withdrawal model of schizophrenia. N-methyl-D-aspartate (NMDA) antagonists (including phencyclidine) induce psychotic symptoms in humans, and are used to model schizophrenia in a variety of experimental conditions. In WTs, 5 mg/kg phencyclidine increased locomotion and stereotyped behaviours, and decreased social interactions. These changes are consistent with a schizophrenia-like effect. In CB1-KOs, phencyclidine decreased locomotion, enhanced ataxia and stereotypy more markedly than in WTs, but did not affect social interactions. Locomotion showed a significant negative correlation with both ataxia and stereotypy, suggesting that in CB1-KOs, the locomotor suppressive effect of phencyclidine was secondary to changes in these variables. Our findings demonstrate that CB1 gene disruption dramatically alters the behavioural effects of the NMDA antagonist phencyclidine, suggesting that the CB1 receptor is involved in schizophrenia. As social disruption and stereotypy respectively are believed to model negative and positive symptoms of schizophrenia, our findings tentatively suggest that cannabinoids are differentially involved in these two symptom categories. These findings require verification by experiments involving CB1 receptor blockers, as the genetic and pharmacological blockade of receptors may not always provide similar results.

Substitution profiles of N-methyl- d-aspartate antagonists in ethanol-discriminating inbred mice

C57BL/6J (B6) and DBA/2J (D2) inbred mice show pronounced differences in ethanol-induced behaviors, such as loss of righting reflex and locomotor activation, among others. They also differ in measures of conditioned place preference and oral self-administration of ethanol. In the current study, I examined whether B6 and D2 mice differed in their expression of the N-methyl- d-aspartate (NMDA) receptor–mediated component of the discriminative stimulus effects of ethanol. B6 and D2 mice were trained to discriminate ethanol (1.5 g/kg, i.p.) from saline in a two-choice, milk-reinforced operant procedure. After training was completed, substitution and response rate dose-effect curves were generated for ethanol; the uncompetitive NMDA antagonists phencyclidine and ketamine; and the competitive NMDA antagonist d-CPPene. Dose-effect curves were also generated for midazolam, cocaine, m-chlorophenylpiperazine (mCPP), morphine, and gamma-hydroxybutyric acid (GHB). B6 and D2 mice learned the ethanol-versus-saline discrimination. Phencyclidine produced near full substitution for ethanol in both strains, whereas ketamine fully substituted for ethanol only in B6 mice. d-CPPene partially substituted for ethanol in both strains. Moderate doses of phencyclidine produced greater response rate–increasing effects in B6 mice than in D2 mice, and high doses of phencyclidine were more potent for suppressing response rates in D2 mice. In contrast, d-CPPene had similar response rate–increasing effects in both strains, but high doses produced more potent response rate–decreasing effects in B6 mice. Among the other drugs tested, only midazolam produced substantial substitution for ethanol. Taken together, these findings seem to indicate that the behavioral effects of NMDA antagonists differ between strains, but that the NMDA-mediated component of the discriminative stimulus effects of ethanol is similar in B6 and D2 mice.

Adaptation of monoaminergic responses to phencyclidine in nucleus accumbens and prefrontal cortex following repeated treatment with fluoxetine or imipramine

The adaptive neuronal changes that follow chronic administration of antidepressant drugs are thought to underlie clinical improvement in patient populations. Recent evidence suggests that alterations specific to N-methyl- d-aspartate (NMDA) receptors may be a final common pathway to antidepressant action. To investigate this possibility, we sought to establish the effects of chronic fluoxetine or imipramine treatment on the monoamine stimulating effect of the non-competitive NMDA antagonist phencyclidine. Male, Sprague–Dawley rats ( n=9/group) were treated with saline (1 ml/kg, i.p.), imipramine (10 mg/kg, i.p.) or fluoxetine (10 mg/kg, i.p.) once daily for 14 consecutive days. After a 7-day drug-free period, animals given an acute challenge of either saline or phencyclidine (5 mg/kg, i.p.). One hour later, animals were killed, brains were removed, and the prefrontal cortex, striatum, and nucleus accumbens were dissected. Samples were assayed for the monoamines and their primary metabolites by HPLC. Repeated treatment with fluoxetine or imipramine did not alter baseline dopamine or serotonin turnover. Acute phencyclidine treatment increased prefrontal cortex and nucleus accumbens dopamine turnover in saline-treated animals ( P<0.01); however, the effect in the nucleus accumbens was prevented in animals pretreated with imipramine or fluoxetine. Acute phencyclidine challenge also increased serotonin turnover in prefrontal cortex of saline- or imipramine-pretreated rats ( P<0.01), though this effect was attenuated in animals pretreated with fluoxetine. Overall, the data suggest that repeated antidepressant treatment alters monoamine turnover in specific brain regions in response to blockade of NMDA receptors. The data highlight the importance of adaptive responses to NMDA receptors resulting from chronic antidepressant treatment.

Enhancement of the discriminative stimulus effects of phencyclidine by the tetracycline antibiotics doxycycline and minocycline in rats.

Tetracycline antibiotics share some neuroprotective and CNS effects with N-methyl- D-aspartate (NMDA) receptor antagonists. The acute effects of two tetracycline antibiotics were compared to those of the prototypic NMDA antagonist phencyclidine (PCP). The effects of minocycline (10-56 mg/kg) and doxycycline (10-56 mg/kg) were evaluated in Sprague-Dawley rats trained to discriminate 2.0 mg/kg IP of PCP from saline under a fixed ratio schedule of food presentation. Even though minocycline and doxycycline did not substitute for PCP, pretreatment with 32 mg/kg of either drug produced leftward shifts in the PCP dose-response curve. The 32 mg/kg dose of minocycline also produced a leftward shift in the dose-response curve for dizocilpine (MK-801), another NMDA channel blocker, in the same subjects. Tetracycline antibiotics may interact either directly or indirectly with NMDA receptors. This suggests that they might be utilized in treatment of brain disorders in which NMDA receptor over-activation has been implicated.

Continuous Phencyclidine Treatment Induces Schizophrenia-Like Hyperreactivity of Striatal Dopamine Release

Functional dopaminergic hyperactivity is a key feature of schizophrenia. Recent in vivo imaging studies have demonstrated greater striatal dopamine release in response to amphetamine challenge in schizophrenia subjects than in normal controls. N-methyl-d-aspartate (NMDA) receptors are known to play a prominent role in regulation of striatal dopamine release. In humans, NMDA antagonists induce a psychotic state that closely resembles schizophrenia. The present study investigates the degree to which chronic continuous administration of the NMDA antagonist phencyclidine (PCP) induces schizophrenia-like hyperreactivity of striatal dopamine release to amphetamine in rodents. Rats were treated with 10 or 15 mg/kg/d PCP for two weeks by osmotic minipump, and striatal dopamine release to amphetamine challenge (1 mg/kg) was monitored by microdialysis. PCP-treated rats showed significant enhancement in amphetamine-induced dopamine release, along with significantly enhanced locomotor activity. These findings support the concept that NMDA receptor dysfunction may contribute to dopaminergic dysfunction in schizophrenia.

Efficacy and safety of glycine site agonists of the N-methyl-D-aspartate receptor in schizophrenia treatment

Dopaminergic hyperactivity within frontostriatal brain systems is a key feature of schizophrenia, and an objective neural correlate of positive schizophrenia symptoms. N-methyl-d-aspartate (NMDA) receptors are known to play a prominent role in regulation of frontostriatal dopamine release. Furthermore, disturbances in glutamatergic function are increasingly being linked to pathophysiology of both positive and negative symptoms of schizophrenia. Prior studies have demonstrated that subchronic continuous administration of the NMDA antagonist phencyclidine (PCP) induces schizophrenia-like hyper-reactivity of frontostriatal dopamine release to amphetamine (AMPH) in rodents, and that effects were reversed by glycine and the prototypic glycine transport inhibitor (GTI) NFPS. The present study investigates effectiveness of the novel, high affinity and well tolerated GTIs, R231857, R231860 and Org29335, to reverse schizophrenia-like enhancement of AMPH-induced DA release, along with effects of the partial glycine-site agonist d-cycloserine. As previously, PCP had no significant effect on basal DA levels, but significantly enhanced AMPH-induced DA release in prefrontal cortex. All GTIs tested, as well as d-cycloserine, significantly reduced PCP-induced enhancement of DA release in prefrontal cortex. Neither PCP nor GTIs significantly affected striatal DA release. Overall, these findings suggest that treatments which target the glycine modulatory site of the NMDA receptor may significantly reverse NMDA receptor antagonist-induced dysregulation of frontal DA systems, consistent with potential beneficial effects on positive-, in addition to negative-, symptoms of schizophrenia.

Modulation of the ethanol-like discriminative stimulus effects of diazepam and phencyclidine by L-type voltage-gated calcium-channel ligands in rats.

Administration of voltage-gated calcium-channel (VGCC) modulators with ethanol can result in enhancement or attenuation of some behavioral effects of ethanol, including its discriminative stimulus effects. The present study used a drug-discrimination paradigm to characterize modulation of the ethanol-like discriminative stimulus effects of a gamma-amino-butyric acid (GABA)A and N-methyl-D-aspartate (NMDA) ligand by administration of VGCC ligands. Two groups of adult male Long-Evans rats were trained to discriminate either 1.0 g/kg ethanol (n=8) or 2.0 g/kg ethanol (n=9) from water under a fixed-ratio (FR) 20 schedule of food presentation. Following training, ethanol substitution tests were conducted with cumulative doses of the GABA(A)-positive modulator diazepam (0.3-10 mg/kg, i.p.) (DZP) and the uncompetitive NMDA antagonist phencyclidine (0.3-5.6 mg/kg, i.p.) (PCP). Next, a single dose of the VGCC antagonist nimodipine, nifedipine, isradipine, or the VGCC agonist (-)-BAY k 8644 (0.3 mg/kg, i.p.) was administered prior to a cumulative DZP or PCP dose-response determination. None of the VGCC modulators produced robust or consistent alterations in the ethanol-like discriminative stimulus effects of DZP in animals trained with either 1.0 g/kg or 2.0 g/kg ethanol. However, the ethanol-like discriminative stimulus effects of PCP were significantly enhanced in the presence of the VGCC antagonists and attenuated in the presence of the agonist in animals trained with 2.0 g/kg ethanol. Overall, these data show that VGCC modulation is not a robust component of ethanol-like discriminative stimulus effects of DZP in animals trained with 1.0 g/kg or 2.0 g/kg ethanol. However, the ethanol-like effects of PCP, particularly at higher training doses, appear to be modulated by dihydropyridine-sensitive VGCCs.

Behavioral pharmacology of two novel substituted quinoxalinedione glutamate antagonists

Two novel quinoxalinedione glutamatergic antagonists, with in vitro selectivity for the glycine modulatory site on the N-methyl-d-aspartate (NMDA) receptor, were evaluated in a number of behavioral tests primarily designed to compare their effects to those of the noncompetitive NMDA antagonist phencyclidine (PCP). The compounds evaluated were 5-chloro-7-trifluoromethyl-1,4-dihydro-2,3-quinoxalinedione (ACEA-1011) and 5-nitro-6,7-dichloro-1,4-dihydro-2,3-quinoxalinedione (ACEA-1021). In rats, both ACEA-1011 and ACEA-1021 were completely devoid of PCP-like discriminative stimulus effects, although behavioral activity, in the form of response rate suppression, was seen at the higher doses tested (6-25mg/kg, i.p.). ACEA-1011 and ACEA-1021 were also ineffective as antagonists of PCP discrimination in rats. ACEA-1021 failed to substitute in rhesus monkeys trained to discriminate PCP from sham injection, although in the monkeys minimal effects were observed on rates of responding even at the highest dose tested (10.2mg/kg, i.v.). ACEA-1021 also failed to produce ethanol-like discriminative stimulus effects in rats under test conditions where PCP has been shown to produce substantial levels of substitution for ethanol. Both ACEA-1011 and ACEA-1021 were also evaluated as antagonists of NMDA discrimination in rats. ACEA-1011 produced some decreases in NMDA-lever responding, with the largest effect at one intermediate dose (3mg/kg, i.p.). ACEA-1021 was ineffective as an antagonist of NMDA discrimination. Unlike results reported for PCP-like NMDA antagonists, neither ACEA-1011 nor ACEA-1021 disrupted prepulse inhibition of the acoustic startle reflex in rats. It was not possible to establish ACEA-1021 (10 or 15.6mg/kg) as a discriminative stimulus in rats. In conclusion, the novel glutamate antagonists ACEA-1011 and ACEA-1021 did not produce a profile of behavioral effects similar to that of PCP-like noncompetitive NMDA antagonists. These results are consistent with an emerging body of evidence showing differences in the behavioral effects of different classes of glutamate antagonists.

Diphenylhydantoin potentiates the EEG and behavioural effects induced by N-methyl-D-aspartate antagonists in rats

The N-methyl-D-aspartate (NMDA) subtype of excitatory amino acid receptors are involved in the electrical and behavioural generalization of epileptiform activity within the brain. In rats, both competitive and non-competitive NMDA antagonists induce three dose-dependent stages of EEG patterns: 1) increase in cortical desynchronization periods; 2) increase in amplitude of cortical high frequency (20-30 Hz), low voltage (30-50 microV) background activity; 3) appearance of cortical slow (2-3 Hz) wave-sharp wave complexes. These EEG changes are accompanied by stimulatory-depressive behavioural effects such as stereotypy (circling, head weaving) and ataxia. In the present study, the influence of the prototypic anticonvulsant diphenylhydantoin (DPH) has been tested on the EEG and behavioural effects induced by the non-competitive NMDA antagonists phencyclidine (PCP) and dizocilpine (MK-801) and by the competitive NMDA antagonist cis-4-phosphonomethyl-2-piperidine-carboxylic acid (CGS 19755). Even though DPH (up to 100 mg/kg IP) did not markedly affect basal cortical EEG activity, at doses of 10-100 mg/kg IP it potentiated all the EEG effects induced by the NMDA antagonists. These data support involvement of NMDA neurotransmission in the pharmacological effects of DPH.

NMDA antagonists disrupt timing behaviour in rats

The behavioural effects of six compounds with N-methyl-D-aspartate (NMDA) antagonist properties were compared on the responding of rats maintained by a differential reinforcement of low rate 15s (DRL-15s) schedule. This schedule requires that responses be temporally spaced in order to be reinforced (timing behaviour). The number of reinforcers obtained was decreased by the non-competitive NMDA antagonists phencyclidine, dizocilpine and memantine, by the competitive NMDA antagonist CGS 19755, by the antitussive drug dextromethorphan, but not by the polyamine site antagonist eliprodil (SL 82.0715). Small increases in response rates were produced by phencyclidine and memantine, and dizocilpine gave rise to very marked increases in rates. CGS 19755, eliprodil and dextromethorphan only decreased rates of responding at high doses. Analysis of drug action in terms of inter-response times (IRT) showed that, except for eliprodil, all compounds shifted the peaks of the IRT distributions to the left. The percentage of short IRTs (response bursts) showed statistically significant increases after administration of dizocilpine, phencyclidine, memantine and CGS 19755. These results show that timing behaviour in rats is disturbed by several NMDA antagonists, although eliprodil, while decreasing responding at high doses, did not disrupt efficient timing.

Burst generation in neocortical neurons after GABA withdrawal in the rat

1. gamma-Aminobutyric acid (GABA) withdrawal syndrome (GWS) represents a particular model of focal epilepsy consecutive to the interruption of a chronic intracortical GABA infusion and is characterized by the appearance of focal epileptic electroencephalographic (EEG) discharges and localized clinical signs on withdrawal of GABA. Effects of Ca2+ channel blockers and N-methyl-D-aspartate (NMDA) antagonists were evaluated in living rats presenting a GWS after interruption of a 5-day GABA infusion into the somatomotor cortex and in neocortical slices obtained from such rats. Bursting properties and morphology of neurons were also analyzed in slices. 2. In living rats, the noncompetitive NMDA antagonist phencyclidine [1-(1-phenylcyclohexyl)piperidine] and the Ca2+ antagonist flunarizine [E-1 (bis(4fluorophenyl)methyl)-4(3phenyl2-propenyl)-piperazine] were administered systemically to two groups of rats. Rats in the first group (n = 12) were injected with the drug 30-60 min before discontinuation of the GABA infusion. In this case, phencyclidine (10 mg/kg ip) prevented the development of GWS (n = 5), whereas flunarizine (40 mg/kg ip) had no consistent effect on the GWS appearance and characteristics (n = 7). Rats in the second group (n = 12) were injected 60-90 min after GABA discontinuation, i.e., during a fully developed GWS. In that case, neither drug suppressed GWS. 3. Neuronal activities in the epileptic focus were studied in slices with conventional intracellular recording and stimulation techniques. From the 65 neurons recorded, 29 responded with EPSPs and paroxysmal depolarization shifts (PDSs) to white matter stimulation (synaptic bursting or SB cells). Nineteen other neurons presented, in addition to synaptically induced PDSs, bursts of action potentials (APs) induced by intracellular depolarizing current injection (intrinsic bursting or IB cells). The remaining 17 neurons presented no bursting properties to either synaptic stimulation or depolarizing current injection (nonbursting or NB cells). 4. The recorded neurons were located 0.7-1.2 mm distant from the lesion because of the penetration of the GABA infusion cannula. Intracellular injection of neurons (n = 4) with biocytin or Lucifer yellow revealed that both SB and IB neurons were large, spiny pyramidal neurons localized in layer V of the sensorimotor cortex. 5. Bath application of the selective antagonist of NMDA receptors DL-2amino-5phosphonovalerate or DL-2amino-7phosphonoheptanoate (10-50 microM) reversibly reduced the amplitude (by 25-50%) and the duration (by 20-25%) of PDSs in all cases (n = 17).(ABSTRACT TRUNCATED AT 400 WORDS)

N-methyl-D-aspartate antagonists and working memory performance: comparison with the effects of scopolamine, propranolol, diazepam, and phenylisopropyladenosine.

The effects of the competitive N-methyl-D-aspartate (NMDA) antagonists CPP (5 & 10 mg/kg) and NPC 12626 (25 & 40 mg/kg) and the noncompetitive NMDA antagonists phencyclidine (1, 3, & 6.25 mg/kg) and MK 801 (0.1 & 0.2 mg/kg) on performance of rats on a nonspatial delayed matching-to-sample working memory task were evaluated. At the highest dose, each NMDA antagonist reduced choice accuracy at all retention intervals. In contrast, the reference anticholinergic agent scopolamine selectively reduced accuracy at long retention intervals, suggesting that scopolamine but not the NMDA antagonists directly interfered with time-dependent working memory retention. Propranolol, diazepam, and phenylisopropyladenosine had little or no effect on choice accuracy, suggesting that noradrenergic, gamma-aminobutyric acid-diazepam, and adenosine receptors may be relatively unimportant for working memory performance as assessed in this task. The NMDA antagonists also differed from scopolamine in that doses of NMDA antagonists that reduced response accuracy also reduced response probability, altered bias (competitive antagonists only), and increased intertrial interval responding (noncompetitive antagonists only). It was concluded that NMDA antagonists disrupt cognitive functions including, but not limited to, those required for accurate working memory performance.

Drug discrimination based on the competitive N-methyl-D-aspartate antagonist, NPC 12626.

The discriminative stimulus effects of the N-methyl-D-aspartate (NMDA) antagonists 3-([+/-]-2-carboxypiperazin-4-yl)propyl-1-phosphonic acid (CPP) and phencyclidine were assessed in a drug discrimination based on the competitive NMDA antagonist 2-amino-4,5-(1,2-cyclohexyl)-7-phosphonoheptanoic acid (NPC 12626). Adult male Sprague-Dawley rats were trained to discriminate NPC 12626 from saline using a standard two-lever fixed ratio 32 schedule of food reinforcement. NPC 12626 dose-dependently substituted for the training dose (20 mg/kg IP) with an ED50 of 9.5 mg/kg. The competitive NMDA antagonist CPP completely substituted for NPC 12626 (ED50 = 1.4 mg/kg IP). The non-competitive NMDA antagonist phencyclidine, as well as pentobarbital and NMDA, failed to substitute completely for NPC 12626, even at doses of these drugs that reduced response rates. These data indicate that the discriminative stimulus properties of NPC 12626 are selective and shared by CPP but not by phencyclidine, pentobarbital or NMDA. The emerging evidence for differences in the discriminative stimulus effects of competitive NMDA antagonists and phencyclidine suggests that competitive antagonists such as NPC 12626 and CPP may not have phencyclidine-like abuse liability.