25 November 2011. A study published in Cell on November 23 aims to take the study of antipsychotic drugs far afield from the classic dopamine D2 receptor–blocking arena. Javier González-Maeso of Mount Sinai School of Medicine in New York and Diomedes Logothetis of Virginia Commonwealth University in Richmond and their colleagues propose that the therapeutic effects of both atypical antipsychotic drugs and the newer glutamate antipsychotic candidates depend on activation of hybrid receptors composed of the serotonin 5-HT2A receptor and the mGluR2 metabotropic glutamate receptor.
Though both components of the hybrid are G protein coupled receptors (GPCRs), they make an odd pair—not only do they have different neurotransmitter ligands, but they act through different G proteins, with mGluR2 stimulating Gi and 5-HT2A stimulating Gq. The researchers found that drugs acting at one receptor could take control of the counterpart receptor within the heteromer, and that the balance between these two pathways predicted whether the drug had antipsychotic or propsychotic effects.
“The findings…go beyond the scope of neuropsychopharmacology and clinical psychiatry, impinging on the biology of cell signaling and demonstrating the significance of GPCR heteromerization in cellular and behavioral responses,” write Mari Kondo and Akira Sawa of Johns Hopkins University School of Medicine in an accompanying perspective piece.
Another recent study, from a collaboration led by Jian Jin and Bryan Roth of the University of North Carolina, Chapel Hill published November 8 in the Proceedings of the National Academy of Sciences, reports dopamine receptor ligands that preferentially activate the therapeutic, β-arrestin-2 pathway over a side effect-associated G protein pathway. Together the studies suggest that the key to understanding antipsychotic function lies in the intricacies of intracellular signaling.
Not just your grandmother's D2
Though antipsychotic development has largely been a variation on a theme of blocking dopamine type 2 (D2) receptors, researchers have suspected the involvement of other neurotransmitter systems in how these drugs work. Atypical antipsychotics have higher affinity for 5-HT2A than for D2 receptors, and newer glutamate compounds that activate mGluR2/3 have shown some therapeutic effects in animal models of psychosis (Moghaddam et al., 1998) and in schizophrenia (Patil et al., 2007).
The role for serotonin and glutamate neurotransmission may connect in the curious hybrid receptors formed from 5-HT2A and mGluR2 components, and which have altered expression in postmortem brain in schizophrenia ( see SRF news story). The Cell study delves into the function of this heteromer complex.
First author Miguel Fribourg and colleagues began by getting the 5-HT2A receptors and mGluR2s to form homodimers or heterodimers in frog oocyctes. When probed with 5-HT, the heterodimer was less active than a pure 5-HT2A homodimer, with Gq activity reduced by 50% as measured by current flowing through a particular potassium channel that constitutes a terminus of this pathway. In contrast, when probed with glutamate, the heterodimer was more active than a pure mGluR2, with Gi activity increased by 200% as measured by current through a different potassium channel terminus. To capture this difference between homomeric and heteromeric signaling, the researchers came up with the 'balance index,' defined as the change in Gi activity (between hetero- and homodimers) minus the change in Gq activity.
If responses to endogenous ligands were sensitive to the heteromer state, would they also be influenced by drugs that bind to the counterpart receptor of a heteromer? Indeed, they were: mGluR2-Gi responses showed substantial modulation by 5-HT2A ligands, and vice versa. For example, glutamate-elicited Gi activity was reduced by DOI, a 5-HT2A agonist, and increased by clozapine, an inverse agonist (that is, a compound that inhibits constitutive activity of a receptor). Likewise, 5-HT-elicited Gq signaling was decreased by mGluR2 agonist LY37, and increased by LY34, an inverse agonist.
In both cases, agonists increased signaling through their target receptor, and clamped down on signaling through the counterpart receptor. In contrast, inverse agonists had the opposite effect, promoting instead activity through the counterpart receptor and diminishing that of their own target receptors. This means that clozapine (the 5-HT2A inverse agonist) and LY37 (the mGluR2 agonist) wrought a similar effect of Gi enhancement and Gq reduction though they bind different receptors. Could this state—favoring Gi over Gq—be the key to antipsychotic activity? Indeed, the researchers found a correlation between Gi-Gq balance index and anti- or propsychotic activity: effective antipsychotics (clozapine, risperidone, LY37) showed the highest balance indices, reflecting a strong bias for Gi over Gq, and drugs with pro-psychotic properties (DOI, LY34) had the lowest balance indices, favoring Gq over Gi signaling.
The researchers also found evidence for this heteromeric cross-talk in mouse brain using membrane preparations from mouse frontal cortex, and in cultured cortical neurons. To explore a connection to behavior, the researchers induced excessive locomotion with MK-801, an antagonist of the N-methyl D-aspartate (NMDA) glutamate receptor. Drug-induced hyperlocomotion is widely used, though not uniformly accepted, as a stand-in for psychosis in rodent studies. Though LY37 (mGluR2 agonist) reduced this hyperlocomotion in wildtype mice, the drug failed to do so in mice lacking 5-HT2A receptors. Likewise, clozapine could reduce hyperlocomotion in wildtype mice, but not in mGluR2 KOs. This means that the mGluR2-dependent hyperlocomotion effect needed 5-HT2A expression, and the 5-HT2A-dependent antipsychotic effect relied on mGluR2 expression, results that implicate the mGluR2-5-HT2A heteromer.
Noting how combinations of low doses of atypical and glutamate antipsychotics have shown therapeutic effects in clinical trials (Uslaner et al., 2009), the researchers asked whether suboptimal doses of clozapine or LY37 would have the same effect in their mouse model of psychosis. Administering LY37 or clozapine in mice missing either one copy the gene encoding mGluR2 or one copy of the gene encoding 5-HT2A had no effect on MK801-induced hyperlocomotion; however, co-administering them at the same dose significantly reduced hyperlocomotion in both kinds of mice.
The authors suggest that combining the shared Gi-favoring and Gq-diminishing effects of mGluR2 agonists and 5-HT2A inverse agonists creates synergy to shift the Gi-Gq balance away from a psychosis-associated state—tilted away from Gi and toward Gq. This concept could answer the question of why some 5-HT2A drugs are antipsychotics and others are pro-psychotic: it might depend on how the drug alters the Gi-Gq balance, with agonists and inverse agonists behaving differently. Though it is still unclear where exactly these heteromers reside within the brain, this Gi-Gq balance is a testable idea that may help researchers screen for new antipsychotic compounds.
Designing dopamine drugs
The second recent paper turns to the more familiar target of antipsychotics, dopamine D2 receptors. Jin and Roth led an effort to synthesize new D2R ligands that could diminish psychosis without side effects. When bound by antipsychotics, D2Rs typically activate both a β-arrestin-2 pathway important to the therapeutic effects, and a G protein coupled pathway associated with motor side effects.
As described in an SRF meeting report earlier this year (see SRF related news story), co-first authors John Allen, Julianne Yost, and Vincent Setola and colleagues successfully developed three D2 ligands that spurred β-arrestin-2 signaling without recruiting the Gi–cAMP pathway. One of these drugs, UNC9975, decreased phencyclidine- or amphetamine-induced hyperlocomotion in mice without increasing catalepsy. This antipsychotic effect was abolished in β-arrestin-2 knockouts, and motor effects emerged. These and other results point to the β-arrestin-2 pathway as instrumental to antipsychotic efficacy, and show the way for developing functionally selective ligands that can maximize the positive while minimizing the negative in drugs for schizophrenia.—Michele Solis.
Fribourg M, Moreno JL, Holloway T, Provasi D, Baki L, Mahajan R, Park G, Adney SK, Hatcher C, Eltit JM, Ruta JD, Albizu L, Li Z, Umali A, Shim J, Fabiato A, MacKerell AD, Brezina V, Sealfon SC, Filizola M, González-Maeso J, Logothetis DE. Decoding the Signaling of a GPCR Heteromeric Complex Reveals a Unifying Mechanism of Action of Antipsychotic Drugs. Cell. 2011 Nov 23; 147: 1011-1023.
Kondo M, Sawa A. Anti-/Propsychotic Drug Signaling via Heteromeric GPCRs – A Balancing Act? Cell. 2011 Nov 23; 147: 964-965.
Allen JA, Yost JM, Setola V, Chen X, Sassano MF, Chen M, Peterson S, Yadav PN, Huang XP, Feng B, Jensen NH, Che X, Bai X, Frye SV, Wetsel WC, Caron MG, Javitch JA, Roth BL, Jin J. Discovery of β-Arrestin-Biased Dopamine D2 Ligands for Probing Signal Transduction Pathways Essential for Antipsychotic Efficacy. Proc Natl Acad Sci U S A. 2011 Nov 8; 108: 18488-18493.
Zhou X. The transcription factor SP4 is reduced in postmortem cerebellum of bipolar disorder subjects: control by depolarization and lithium. Bipolar Disord. 2011 Aug;13(5-6):474-85.