Altered receptor dimerization: a new paradigm in the pathology of schizophrenia
Understanding the pathology of complex diseases such as schizophrenia requires the use of the full arsenal at the disposal of medical research. Such an approach has been used to make an exciting new discovery that suggests that abnormal dimerization between the serotonin 2A receptor (2AR) and the metabotropic glutamate 2 receptor(mGluR2) may underlie some of the symptoms of schizophrenia (González-Maeso et al., 2008).

This discovery is based on an initial finding that 2AR is coexpressed with mGluR2 in layer 5 of the mouse somatosensory cortex (SCx) and that levels of mGluR2 were decreased in the cortex of 2AR-/- mice, suggesting a relationship between the expression of the two genes. This hypothesis was further supported by data showing that expression of mGluR2 was selectively restored in mice where 2AR expression had been re-established in layer 5 of the SCx. From these data, and data from other studies suggesting G protein-coupled receptors (GPCRs) can form heterodimers (Angers et al., 2002), the authors began to test the hypothesis that 2AR and mGluR2 could form heterodimers.

Using human cortical samples and an anti-2AR antibody, the authors showed that they could immunoprecipitate an immunogenic band with a molecular weight that matches a 2AR/mGluR2 receptor dimer complex if an anti-GluR2 antibody was used with Western blotting. Significantly, that heterodimer complex could not be visualized in Western blots using anti-mGluR3 antibody instead of an anti-mGluR2 antibody. This reinforces the notion that it is mGluR2 that dimerizes with 2AR. Finally, a close interaction between the two receptors was demonstrated using fluorescence resonance energy transfer in transfected HEK-293 cells.

The authors then used molecular chimaeras to localize the site on mGluR2 that was a requirement for heterodimerization with 2AR and showed that the transmembrane helices 4 and 5 were required for this interaction. The authors then tested the posit that the interaction between 2AR and mGluR2 served to integrate cross-talk between the serotonergic and glutamatergic pathways in the CNS. To this end they showed that activation of Gαq/11 by 2AR was reduced in cells coexpressing mGluR2 and that this effect was lessened by mGluR2 receptor agonists. Significantly, this activity was dependent on the 4 and 5 transmembrane domain of the mGluR2, the domain required to form heterodimers.

Having demonstrated an impact of receptor dimerization on G protein signaling, the authors then investigated whether dimerization affected either receptor-modulated changes in c-fos, which is a marker of the signal-transduction stimulated by non-hallucinogenic 2AR agonists, or on levels of egr-2, which is induced by hallucinogens such as lysergic acid diethylamide (González-Maeso et al., 2007). The authors showed that stimulating mGluR2 with an mGluR2/3 agonist only affected the ability of hallucinogens to induce egr-2 in mouse SCx, suggesting the 2AR/mGluR2 dimers were involved in modulating hallucinogenic pathways of the CNS. To confirm this finding might have functional consequences. The authors then showed that the mGluR2/3 agonist suppressed the induction of hallucinogen-induced head twitches in the mice. These data supported the notion that receptor heterodimers are active in appropriate pathways in the CNS that have been used to model hallucinogenic effects. To extend this behavioral data, the authors also showed that mGluR2/3 agonist-induced locomotion and vertical activity were attenuated in 2AR-/- mice.

The authors had amassed a large quantity of data to suggest that 2AR/mGluR2 dimers may be important in generating hallucinogenic activity, which raised the possibility that altered levels of such dimers may be altered in the CNS of subjects with schizophrenia. To address this issue, the authors used radioligand binding to show that expression levels of 2AR and mGluR2/3 receptors were increased and decreased, respectively, in the dorsolateral prefrontal cortex (DLPFC) from untreated subjects with schizophrenia. In addition, the authors showed that the level of mGluR2, but not mGluR3, mRNA was decreased in the same CNS regions from the subjects with schizophrenia. These differences were not apparent in the same CNS regions from subjects with schizophrenia who had been treated with antipsychotic drugs. This raised the possibility that antipsychotic drug treatment may affect levels of 2AR/mGluR2 dimerization, and therefore the authors went on to show that clozapine downregulated levels of the mRNA for the two receptors in mouse cortex. The 2AR was critical in this process as clozapine did not downregulate mGluR2 mRNA in 2AR-/- mice. Haloperidol treatment had no effect on the expression of either 2AR or mGluR2. Finally, it was shown that levels of receptor binding to both receptors were reduced with aging.

From this large amount of data, the authors could conclude that they had shown that 2AR/mGluR2 heterodimers are important in hallucinogenic pathways of the CNS, using both cellular and animal models. They also argue that increased expression of 2AR and decreased expression of mGluR2 in the cortex of subjects with schizophrenia predispose these individuals to hallucinations. Presumably, therefore, the reduction in 2AR caused by certain antipsychotic drugs would be a mechanism by which a potential imbalance in heterodimer formation could be reversed to lessen hallucinations. Finally, the authors argue that the propensity for antipsychotic drugs and age to decrease levels of 2AR is why 2AR levels are reported as decreased in the majority of prior studies in schizophrenia (Dean, 2003), which mainly used cohorts made up of either treated or older subjects with schizophrenia.

As is often the case, the proposed link of a clear finding of 2AR/mGluR2 heterodimers in the mammalian cortex to hallucinations and then schizophrenia is dependent on data from the CNS of subjects with the disorder. Like many novel and compelling discoveries, the data from animal and cellular models appear clear-cut. However, there are some issues that leave in doubt the link between changes in receptor dimerization and schizophrenia. In particular, the authors did not demonstrate altered levels of dimerized receptors using the co-immunoprecipitation/Western blot approach; rather, they rely on inferences from the measurement of the two receptors separately using radioligand binding. In addition, the authors have not addressed the fact that the majority of imaging studies, many using young drug naïve subjects, did not find changes in levels of the 2AR in subjects with the disorder (Verhoeff et al., 2000; Lewis et al., 1999; Okubo et al., 2000; Trichard et al., 1998). The argument that findings in postmortem studies showing decreased levels of 2AR were due to studies being completed on treated or older subjects with the disorder is also not supported by neuroimaging studies showing decreased levels of 2AR in subjects with schizophrenia who were younger than 29 years of age (Ngan et al., 2000) or who were at high risk for the disorder (Hurlemann et al., 2005). These later studies suggest that low levels of 2AR may be more apparent earlier in the disease progression.

It is clear that the report of increased levels of 2AR with schizophrenia in the paper reporting the discovery of the 2AR/mGluR2 heterodimers (González-Maeso et al., 2008) is at odds with other postmortem (Dean, 2003) and neuroimaging studies (see above). This raises the possibility that the postmortem findings are in some way unique to the tissue collection used in the study. One difference in the postmortem tissue used in the study is that 85 percent of the subjects with schizophrenia had died by suicide. This would be higher than in most other studies of schizophrenia using postmortem CNS. Significantly, a number of studies have reported an increase in 2AR in the cortex of subjects that had died by suicide (Pandey et al., 2002; Mann et al., 1986; Hrdina et al., 1993; Escribá et al., 2004). This means the increased levels of 2AR reported in the study on heterodimers may be associated with suicide within schizophrenia, rather than schizophrenia per se.

In conclusion, like any novel finding, there are a number of important issues that will need addressing in future testing of the hypothesis that altered 2AR/mGluR2 heterodimerization is involved in the pathology of schizophrenia. However, the idea that changes in receptor heterodimerization could be involved in the pathology of schizophrenia is an exciting new direction arising from what is an excellent broad-based approach to understanding this complex disorder.

References:

González-Maeso J, Ang RL, Yuen T, Chan P, Weisstaub NV, López-Giménez JF, Zhou M, Okawa Y, Callado LF, Milligan G, Gingrich JA, Filizola M, Meana JJ, Sealfon SC. Identification of a serotonin/glutamate receptor complex implicated in psychosis. Nature. 2008 Mar 6;452(7183):93-7. Abstract

Angers S, Salahpour A, Bouvier M. Dimerization: an emerging concept for G protein-coupled receptor ontogeny and function. Annu Rev Pharmacol Toxicol. 2002 Jan 1;42():409-35. Abstract

González-Maeso J, Weisstaub NV, Zhou M, Chan P, Ivic L, Ang R, Lira A, Bradley-Moore M, Ge Y, Zhou Q, Sealfon SC, Gingrich JA. Hallucinogens recruit specific cortical 5-HT(2A) receptor-mediated signaling pathways to affect behavior. Neuron. 2007 Feb 1;53(3):439-52. Abstract

Dean B. The cortical serotonin2A receptor and the pathology of schizophrenia: a likely accomplice. J Neurochem. 2003 Apr 1;85(1):1-13. Abstract

Verhoeff NP, Meyer JH, Kecojevic A, Hussey D, Lewis R, Tauscher J, Zipursky RB, Kapur S. A voxel-by-voxel analysis of [18F]setoperone PET data shows no substantial serotonin 5-HT(2A) receptor changes in schizophrenia. Psychiatry Res. 2000 Oct 30;99(3):123-35. Abstract

Lewis R, Kapur S, Jones C, DaSilva J, Brown GM, Wilson AA, Houle S, Zipursky RB. Serotonin 5-HT2 receptors in schizophrenia: a PET study using [18F]setoperone in neuroleptic-naive patients and normal subjects. Am J Psychiatry. 1999 Jan 1;156(1):72-8. Abstract

Okubo Y, Suhara T, Suzuki K, Kobayashi K, Inoue O, Terasaki O, Someya Y, Sassa T, Sudo Y, Matsushima E, Iyo M, Tateno Y, Toru M. Serotonin 5-HT2 receptors in schizophrenic patients studied by positron emission tomography. Life Sci. 2000 Jan 1;66(25):2455-64. Abstract

Trichard C, Paillère-Martinot ML, Attar-Levy D, Blin J, Feline A, Martinot JL. No serotonin 5-HT2A receptor density abnormality in the cortex of schizophrenic patients studied with PET. Schizophr Res. 1998 May 4;31(1):13-7. Abstract

Ngan ET, Yatham LN, Ruth TJ, Liddle PF. Decreased serotonin 2A receptor densities in neuroleptic-naive patients with schizophrenia: A PET study using [(18)F]setoperone. Am J Psychiatry. 2000 Jun 1;157(6):1016-8. Abstract

Hurlemann R, Boy C, Meyer PT, Scherk H, Wagner M, Herzog H, Coenen HH, Vogeley K, Falkai P, Zilles K, Maier W, Bauer A. Decreased prefrontal 5-HT2A receptor binding in subjects at enhanced risk for schizophrenia. Anat Embryol (Berl). 2005 Dec 1;210(5-6):519-23. Abstract

Pandey GN, Dwivedi Y, Rizavi HS, Ren X, Pandey SC, Pesold C, Roberts RC, Conley RR, Tamminga CA. Higher expression of serotonin 5-HT(2A) receptors in the postmortem brains of teenage suicide victims. Am J Psychiatry. 2002 Mar 1;159(3):419-29. Abstract

Mann JJ, Stanley M, McBride PA, McEwen BS. Increased serotonin2 and beta-adrenergic receptor binding in the frontal cortices of suicide victims. Arch Gen Psychiatry. 1986 Oct 1;43(10):954-9. Abstract

Hrdina PD, Demeter E, Vu TB, Sótónyi P, Palkovits M. 5-HT uptake sites and 5-HT2 receptors in brain of antidepressant-free suicide victims/depressives: increase in 5-HT2 sites in cortex and amygdala. Brain Res. 1993 Jun 18;614(1-2):37-44. Abstract

Escribá PV, Ozaita A, García-Sevilla JA. Increased mRNA expression of alpha2A-adrenoceptors, serotonin receptors and mu-opioid receptors in the brains of suicide victims. Neuropsychopharmacology. 2004 Aug 1;29(8):1512-21. Abstract

View all comments by Brian Dean

Another bicycle trip?
Ever since dopamine was first implicated in the therapeutic effects of antipsychotic drugs by Arvid Carlsson and colleagues over 50 years ago, and then dopamine D2 receptors were implicated in the Parkinsonian side effects and late-evolving movement disorders, an intense search has been underway for antipsychotic drugs that might act through other mechanisms. In parallel with this search, drugs with psychotomimetic effects in healthy volunteers or exacerbating psychosis have also been used to discover new antipsychotic drugs. With an evolving understanding of the neuropharmacology underlying ketamine or PCP, amphetamines, and serotonergic hallucinogens (LSD, mescaline, and psilocybin), glutamatergic, dopaminergic, and serotonergic theories of psychotic pathophysiology have been advanced. Converging evidence points to activation of 5-HT2A receptors as a necessary action in the psychotomimetic effects of the serotonergic “hallucinogens.” The recent description of a proof-of-concept clinical study where a prodrug for a metabotropic glutamate2/3 (mGlu2/3) receptor agonist exerted therapeutic effects in schizophrenic patients may be the most promising report for an elusive antipsychotic medication generally viewed as lacking direct effects on dopamine D2 receptors (Patil et al., 2007). More recently, a report has appeared which raises the possibility that glutamate and serotonin may be involved in the therapeutic effects of mGlu2/3 receptors by virtue of a molecular complex between mGlu2 and 5-HT2A receptors (González-Maeso et al., 2008). Beyond replication of these effects in other laboratories, several fundamental questions have been raised that should be addressed.

First, does this type of interaction occur in the prefrontal cortex, which (through cortico-thalamo-striatal loops) is more closely related to the core symptoms of schizophrenia than the somatosensory cortex? Second, are the therapeutic actions of mGlu2/3 receptors mediated through activation of postsynaptic mGlu2 receptors, rather than presynaptic mGlu2 receptors (Marek et al., 2001)? Third, do other G protein-coupled receptors similarly act through complexes with 5-HT2A receptors?

Further research will be required to address this first question, especially since both mGlu2/3 agonists and NMDA receptor antagonists appear to have more potent or efficacious effects in the prefrontal cortex than the somatosensory cortex under either in vitro or in vivo conditions. The second question will be important to address at a fundamental level, since “simple” intra-cortical processes invoke different levels of analyses than do hypotheses that presynaptic mGlu2 receptors on long-loop afferents may play key roles as therapeutic targets. In fact, previous experiments involving rescue of 5-HT2A receptors in the cortex or thalamus appear to be compromised by confounds. Namely, the cortical rescue of 5-HT2A receptors in the htr2A-/- mice using the Emx1 promoter does not rule out an involvement of afferents to the cortex from a poorly understood region involved in integrating multi-modal associations, the claustrum. 5-HT2A receptor expression was also restored to the claustrum with this rescue strategy (Weisstaub et al., 2006). The thalamic rescue of 5-HT2A receptors, which generally fails to reprise the effects seen in the cortical rescue preparation, may be problematic in that the promoter utilized expresses SERT in thalamocortical projections from primary sensory relay neurons rather than the midline and intralaminar thalamic neurons intimately involved in arousal and stress-related biology (Lebrand et al., 1996; Van der Werf et al., 2002). The relatively dense expression of cortical 5-HT2A and mGlu2 receptor expression in layers I and Va of the prefrontal cortex is an excellent match for the laminar distribution of afferents from the midline and intralaminar thalamic nuclei (Marek et al., 2001). Further work is required to understand the magnitude of the involvement of thalamic afferents from the posterior thalamic nucleus to the somatosensory cortex vs. involvement of the afferents from midline and intralaminar thalamic nuclei throughout the prefrontal cortex. Third, do other Gi/Go-coupled GPCRs form complexes with 5-HT2A receptors? Other, much stronger candidates for such a role than mGlu3 receptors would be μ-opioid receptors and adenosine A1 receptors. The physiology of both μ-opioid receptors and adenosine A1 receptors share a striking degree of similarity with mGlu2 receptors ranging from regulating excitatory synaptic afferents to the prefrontal cortex in slice preparations to in vivo modulation of the three major classes of psychotomimetic drugs.

Both the replication of the exciting basic findings reported by the Gingerich and Sealfon laboratories and answers to these questions above should add another chapter to the story that began in earnest over 60 years ago with a bicycle ride by the Sandoz chemist Albert Hoffman following the ingestion of the twenty-fifth lysergic diethylamide that he had synthesized.

References:

Patil ST, Zhang L, Martenyi F, Lowe SL, Jackson KA, Andreev BV, Avedisova AS, Bardenstein LM, Gurovich IY, Morozova MA, Mosolov SN, Neznanov NG, Reznik AM, Smulevich AB, Tochilov VA, Johnson BG, Monn JA, Schoepp DD. Activation of mGlu2/3 receptors as a new approach to treat schizophrenia: a randomized Phase 2 clinical trial. Nat Med. 2007 Sep 1;13(9):1102-7. Abstract

González-Maeso J, Ang RL, Yuen T, Chan P, Weisstaub NV, López-Giménez JF, Zhou M, Okawa Y, Callado LF, Milligan G, Gingrich JA, Filizola M, Meana JJ, Sealfon SC. Identification of a serotonin/glutamate receptor complex implicated in psychosis. Nature. 2008 Mar 6;452(7183):93-7. Abstract

Marek GJ, Wright RA, Gewirtz JC, Schoepp DD. A major role for thalamocortical afferents in serotonergic hallucinogen receptor function in the rat neocortex. Neuroscience. 2001 Jan 1;105(2):379-92. Abstract

Weisstaub NV, Zhou M, Lira A, Lambe E, González-Maeso J, Hornung JP, Sibille E, Underwood M, Itohara S, Dauer WT, Ansorge MS, Morelli E, Mann JJ, Toth M, Aghajanian G, Sealfon SC, Hen R, Gingrich JA. Cortical 5-HT2A receptor signaling modulates anxiety-like behaviors in mice. Science. 2006 Jul 28;313(5786):536-40. Abstract

Lebrand C, Cases O, Adelbrecht C, Doye A, Alvarez C, El Mestikawy S, Seif I, Gaspar P. Transient uptake and storage of serotonin in developing thalamic neurons. Neuron. 1996 Nov 1;17(5):823-35. Abstract

Van der Werf YD, Witter MP, Groenewegen HJ. The intralaminar and midline nuclei of the thalamus. Anatomical and functional evidence for participation in processes of arousal and awareness. Brain Res Brain Res Rev. 2002 Sep 1;39(2-3):107-40. Abstract

View all comments by Gerard J. Marek