Dysfunctional brain dopamine systems induced by psychotomimetic NMDA-receptor antagonists and the effects of antipsychotic drugs

Abstract

Clinical studies utilizing imaging techniques demonstrate that classical antipsychotic drugs, such as haloperidol, in clinically effective doses display around 75% dopamine (DA)-D₂ receptor occupancy in the brain. In contrast, the atypical antipsychotic drug clozapine is even more effective at only about 45% D₂-receptor occupancy. Yet at this D₂-receptor occupancy classical antipsychotics are not effective, raising the question of which other receptors may be involved in mediating the atypical antipsychotic profile of clozapine and other atypical antipsychotics. The present paper describes experimental work aimed at elucidating this critical question, utilizing the phencyclidine (PCP) model of schizophrenia in combination with studies of typical and atypical antipsychotics as well as various specific receptor blocking agents. Both electrophysiological methods, i.e. single cell recording from DA neurons in the ventral tegmental area (VTA), and biochemical analysis of biogenic amines such as DA following microdialysis in difference DA terminal areas in the brain, were used. In addition, behavioural measurements using the conditioned avoidance response (CAR) paradigm and assessments of locomotor activity were utilized. Experiments with functional inactivation of the medial frontal cortex (mPFC) in the rat as well as with MK-801 and other antagonists at central NMDA-receptors revealed that following systemic administration of schizophrenomimetic NMDA-receptor antagonists a profound dysregulation of the mesocorticolimbic DA system occurs, severely impairing the dynamic physiological response range of the neurons. Specifically, DA neurons which largely project to the mPFC showed a profound loss of burst firing, whereas VTA-DA neurons, which mainly project subcortically, showed an increased monotonous high-frequency firing with increased DA output from nerve terminals and concomitant behavioural activation. Significantly, drugs with a prominent 5-HT_2A-receptor blocking action could effectively restore the burst firing mode, i.e. phasic responsivity, in mesocortically projecting DA neurons, and also potentiate the CAR suppressant effect of the selective D₂/D₃-receptor antagonist raclopride without increasing catalepsy scores. The selective α₁-adrenoreceptor antagonist prazosin effectively suppressed both the stereotyped, high-frequency firing of subcortically projecting DA neurons following systemic MK-801 and the concomitant behavioural, i.e. locomotor, activation. In addition, the MK-801 evoked DA release in the nucleus accumbens was suppressed. A similar effect was seen also with AMPA-receptor antagonists when applied locally into the VTA and, in addition, systemic administration of chemically different AMPA-receptor antagonists caused a CAR-suppressant effect similar to both classical and atypical antipsychotic drugs. These results and other data showing a clearcut difference between typical and atypical antipsychotic drugs on DA output in the shell and core, respectively, of the nucleus accumbens, suggest that both the 5-HT_2A- and the α₁-adrenoreceptor blocking effects of a number of atypical antipsychotic drugs in all probability contribute to their antipsychotic effect. Moreover, our results indicate that AMPA-receptor antagonists may possess an atypical antipsychotic profile.

Introduction

Preclinical and clinical evidence indicates that all effective antipsychotic drugs have one principal common denominator as regards their mode of action, i.e. they antagonise brain dopamine (DA) mediated neurotransmission. This can be achieved by means of impairing granular storage of DA within the DA neurons, thereby inhibiting nerve impulse mediated release of DA, as exemplified by reserpine, or by means of blocking various types of DA receptors [11]. In fact, most if not all clinically approved neuroleptics currently used seem to share one specific mechanism of action, e.g. they act as DA-D₂ receptor antagonists. Yet, whereas classical neuroleptics usually are therapeutically effective at about 75% D₂-receptor occupancy in brain, the atypical antipsychotic drug clozapine seems effective already at 45–50% DA-D₂ receptor occupancy, as judged by studies using positron emission tomography (PET) 23, 70. At the same time this drug causes very few extrapyramidal side effects and actually appears to possess a significantly higher efficacy than classical neuroleptics. In addition, it exerts an advantageous therapeutic action on negative symptoms which are otherwise rather resistant to treatment with conventional antipsychotics [60]. Since clozapine displays significant affinities for a large number of other neurotransmitter receptors such as various types of α₁-adrenoceptors, 5-HT₂ receptors, muscarinic and histaminergic receptors etc, a critical question is which of these other properties of the drug that contribute to its therapeutic action. Clearly, a 45% D₂ receptor occupancy alone, e.g. caused by treatment with raclopride, a relatively selective D₂/D₃-receptor antagonist, is not sufficient for an effective antipsychotic action. The analysis of this question requires a systematic evaluation of combinations of different, specific pharmacological tools, such as various types of ligands for DA receptors, 5-HT₂ receptors, α₁-adrenoceptors etc, which have now become increasingly available. For ethical and economical reasons such studies have to be performed preferentially in animals rather than in psychotic patients. Thus, a critical issue in this regard is the availability of adequate experimental models of psychotic behaviour. Initially, the dopamine hypothesis of schizophrenia [10]suggested a hyperdopaminergic state in brain. This notion was essentially based upon indirect pharmacological evidence, i.e. the aforementioned mechanisms of action of antipsychotic drugs and the fact that directly or indirectly acting DA agonists, such as amphetamine, had been found capable of inducing psychosis, particularly paranoid symptoms. However, the so-called amphetamine model of schizophrenia 46, 75has several shortcomings. Specifically, the model selects drugs with clinical effectiveness directed towards positive, rather than negative symptoms of schizophrenia. Indeed, although positive symptoms of schizophrenia tend to worsen on amphetamine challenge, negative symptomatology even appears to improve 4, 94. In addition, the amphetamine induced psychosis usually fails to encompass several core symptoms of schizophrenia, such as formal thought disorder, auditory hallucinations, flattening of affect or anhedonia. Therefore, in recent years several other experimental approaches have been adopted to circumvent these limitations of the amphetamine model.

Studies in schizophrenic patients have not yielded unequivocal support for a hyperdopaminergic state in schizophrenia. In fact, some previous data even indicate a reduced central DA output associated with the disease [44], a phenomenon that was found to correlate with negative symptoms, such as anergia, emotional blunting, lack of drive etc. On the other hand, recent studies utilizing single photon emission computerized tomography or PET in drug-free schizophrenic subjects suggest that psychotic symptoms may, indeed, be related to augmented release of DA in brain, notably an abnormal responsiveness of DA neurons 7, 51. Consequently, both hypo- and hyperfunctioning brain DA systems in schizophrenia have been proposed and, tentatively, both types of dysfunction might even occur simultaneously, albeit in different brain regions [88]. This notion might also be of heuristic value as regards the dissociation between cognitive and emotional functions in schizophrenia: Whereas a large body of evidence implicates the subcortical, mesolimbic DA projection in reward and motivational functions, the mesocortical DA projection has rather been ascribed a role in attentional processes and cognition. At any rate, clinical evidence tends to support an abnormal regulation of brain DA systems in schizophrenia, and the purported dysfunctions of e.g. mesolimbic or mesocortical DA neurons might well be caused by an altered balance of inputs converging on the DA neurons from various sources [68], such as the prefrontal cortex (PFC). Substantial evidence indicates that an impaired function of the PFC, so-called hypofrontality, particularly a reduced capacity to activate the PFC, is a common phenomenon in schizophrenia. Consequently, the functional connectivity between the PFC and the mesocorticolimbic DA system, which originates in the ventral tegmental area (VTA), is of considerable interest.