A complex in psychosis

Article metrics

The molecular basis of psychoses such as schizophrenia remains largely mysterious. The interaction between two of the brain receptors involved adds to evidence that will help in the search for explanations.

This is a story that involves three types of receptor in the brain that influence human perception and behaviour (those for the neurotransmitters dopamine, serotonin and glutamate), and the drugs that block or enhance their activity. Such drugs are used by researchers to investigate the causes of psychotic disorders such as schizophrenia, and by clinicians to treat patients. Classical antipsychotic drugs, designated 'typical neuroleptics', act predominantly by blocking dopamine D2 receptors. A new generation of more effective 'atypical neuroleptics' also blocks a subtype of serotonin receptor known as 5-HT2A, or more simply as 2AR. And last year we had the promising report1 of a drug that mimics the effect of glutamate at a subtype of one of its receptors, metabotropic glutamate receptor 2 (mGluR2), and that seems to be as effective as atypical neuroleptics.

This is the background against which the paper by González-Maeso et al.2 appears (page 93 of this issue): these authors show that there are direct molecular interactions between 2AR and mGluR2, and that the interactions have functional effects. Besides helping to elucidate drug actions, these findings may shed light on fundamental aspects of the causes and consequences of schizophrenia.

In the absence of definitive evidence for the molecular causes of psychosis, biological psychiatrists have long relied largely on drugs as probes. If a drug is psychotomimetic (that is, elicits a psychotic state that resembles schizophrenia), then knowing how it acts might clarify aberrations in the brains of patients with schizophrenia. Conversely, if a drug selectively alleviates symptoms of schizophrenia, understanding its mode of action will also be illuminating.

The structure and action of diverse psychotomimetic drugs have led to different models of psychosis. For example, hallucinogens such as LSD (lysergic acid diethylamide), psilocybin and dimethyltryptamine resemble the indoleamine structure of serotonin, leading to a 'serotonin concept' of psychosis, which holds that perturbation of serotonin signalling is the main cause. Although phenethylamine hallucinogens such as mescaline and methoxyamphetamines are not indoleamines, their native conformation resembles that of serotonin and they elicit psychotic states much like that produced by LSD (ref. 3). Hallucinogens are agonists (stimulating agents) at 2ARs, which are blocked by atypical neuroleptics.

High doses of amphetamines, which act by releasing dopamine, evoke a psychosis that often resembles acute paranoid schizophrenia, and low doses selectively exacerbate schizophrenic symptoms. The typical neuroleptics block dopamine D2 receptors with potencies closely paralleling the effectiveness of their antipsychotic actions, leading to the 'dopamine hypothesis' of schizophrenia4. The effects of agents that block the NMDA subtype of glutamate receptor, such as phencyclidine (PCP, also known as angel dust), usually mimic schizophrenic symptoms better than any other drug-induced psychosis. The most recent development1 in this pot-pourri of drug effects involves the striking antischizophrenic actions of LY2140023, an mGluR2 agonist.

González-Maeso et al.2 wondered whether the similar therapeutic actions of 2AR antagonists and mGluR2 agonists reflect a specific molecular interaction between the two receptors. Like many neurotransmitter receptors, mGluR2 and 2AR are coupled to G proteins, which sit inside the cell membrane and act as molecular switches that regulate intracellular signalling pathways. During the past decade, Devi5 and others have shown that different G-protein-coupled receptors can form functional complexes.

The advance made by González-Maeso et al. is the direct demonstration, using several techniques, of the existence of complexes between 2AR and mGluR2 (Fig. 1). These authors conclude that the complex is functionally relevant by showing that mGluR2 agonists increase the affinity of hallucinogens for 2AR binding, whereas 2AR agonists decrease the affinity of mGluR2 agonists for glutamate receptor binding. Moreover, activation of G proteins by 2AR was altered by co-expression of mGluR2. Induction of a gene (egr-2) that is selectively stimulated by hallucinogenic 2AR agonists was blocked by an mGluR2 agonist, whereas induction of c-fos, which responds both to hallucinogenic and non-hallucinogenic 2AR agonists, was unaffected by the same treatment.

Figure 1: Receptor interaction.

González-Maeso et al.2 find that the metabotropic glutamate receptor 2 (mGluR2) and serotonin 5-HT2A receptor (2AR) physiologically bind each other, leading to reciprocal regulation of their functions. Agonists that stimulate mGluR2 are antipsychotic, whereas 2AR agonists, such as hallucinogens, have the opposite effect. It is conceivable that the clinically significant anti-schizophrenic effects of LY2140023, an mGluR2 agonist1, derive from reducing the excessive — and hence hallucinogen-like — activity of 2AR. DOI, 2,5-dimethoxy-4-iodoamphetamine.

The authors' observations of behavioural responses in mice also support the idea of mGluR2–2AR interaction, as the effects of an mGluR2 antagonist on movement were diminished in mice with targeted deletion of 2AR. Finally, post-mortem analysis of the brains of humans with schizophrenia revealed increased numbers of 2ARs and decreased numbers of mGluR2s.

What do we learn from these findings? Using various approaches, González-Maeso et al.2 have provided a compelling case that 2AR and mGluR2 interact physically and physiologically, and that drugs influencing one of the receptors will alter the behaviour of the other. Neuronal wiring links between various transmitter circuits, and crosstalk between receptors regulated by agents called scaffolding proteins, are well known. González-Maeso and colleagues provide a new concept — that direct physical coupling of two receptors mediates complex pharmacological and behavioural responses.

Given this conclusion, researchers will be scurrying to explore what effect drugs that target 2ARs have on glutamate signalling and, conversely, to ascertain whether alterations in mGluR2s influence events mediated by serotonin. From a behavioural and clinical perspective, one would predict that drugs acting at these two different receptors would have similar influences. This possibility fits with the finding1 that the antipsychotic actions of the mGluR2 agonist LY2140023 resemble those of the atypical neuroleptics that target 2ARs. Because drugs blocking dopamine D2 receptors also act similarly, they may have some link to the receptor complex. Finally, the new results may provoke a reassessment of the effects of hallucinogens as useful models of the brain disturbances that characterize schizophrenia.


  1. 1

    Patil, S. T. et al. Nature Med. 13, 1102–1107 (2007).

  2. 2

    González-Maeso, J. et al. Nature 452, 93–97 (2008).

  3. 3

    Aghajanian, G. K. & Marek, G. J. Brain Res. Rev. 31, 302–312 (2000).

  4. 4

    Snyder, S. H., Banerjee, S. P., Yamamura, H. I. & Greenberg, D. Science 184, 1243–1253 (1974).

  5. 5

    Devi, L. A. Trends Pharmacol. Sci. 22, 532–537 (2001).

Download references

Author information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Snyder, S. A complex in psychosis. Nature 452, 38–39 (2008) doi:10.1038/452038a

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.