mGlu3 – New hope for pharmacotherapy of schizophrenia

Long considered promising targets for a number of neurological and psychiatric conditions, metabotropic glutamate receptors (mGlu) present as many challenges as opportunities for novel and effective therapeutic strategies. Eight subtypes of these G-protein coupled receptors have so far been identified and are classified into three groups according to their sequence homology, pharmacological properties and downstream signaling pathways. Group I mGlu1 and mGlu5 couple to Gq proteins and largely have activating or facilitatory properties. Group II mGlu2 and mGlu3, and group III mGlu4, mGlu6, mGlu7 and mGlu8 couple to Gi/o proteins and generally have inhibitory signaling effects resulting in a decrease of neurotransmitter release.

Schizophrenia is one of the disorders mGlu receptors have been explored for because it has been linked to dysfunctions of the glutamatergic system including mGlu (1). A serious mental illness schizophrenia is a psychotic disorder characterized by three symptom domains: positive or psychotic symptoms include hallucinations and delusions. Negative symptoms or deficit refer to impaired normal functions such as diminished emotional expression, avolition, anhedonia and asociality. Cognitive symptoms include impaired memory functions, attention, decision making, and social cognition. Due to the complexity of causes there is currently no cure for schizophrenia but treatments are available to manage symptoms and improve daily life functions. Anti-psychotic medications, however, have little effects on negative symptoms and cognitive impairments. Therefore, better and more effective therapeutic strategies are urgently needed.

Evidence from mouse and human genetic studies has more recently linked schizophrenia and cognitive deficits to polymorphisms in GRM3, the gene that encodes mGlu3, which can result in truncated, nonfunctional mGlu3 receptor protein (2). mGlu3 is extensively expressed pre- and post-synaptically and in glial cells throughout the nervous system (3) and recent evidence suggests that mGlu3 plays a critical role in the modulation of different forms of neuroplasticity such as long-term potentiation (LTP) and long-term depression (LTD) (4). Interestingly, postsynaptic group II mGlu can positively couple to the PLC/PKC pathway probably via mechanisms mediated by the βγ complex of the Gi/o to amplify and potentiate the typical Gq transduction signaling. It is plausible to hypothesize that mGlu3 receptors promote cognition through neuroplasticity but this has not been evaluated in preclinical mechanistic studies, in part because of a lack of subtype-specific mGlu3 ligands and other tools.

A new study (5) addresses this important knowledge gap by using novel allosteric modulators and newly developed transgenic mouse lines in combination with behavioral and electrophysiological assays to determine the role of mGlu3 in hippocampal plasticity and cognitive function in a mouse model of schizophrenia-like cognitive dysfunction induced by subchronic treatment with an NMDA antagonist (phencyclidine, PCP). These mechanistic studies found that mGlu3 activation enhanced hippocampus-dependent associate learning (trace fear conditioning) in the PCP model, reversing the cognitive deficits. Effects of mGlu3 activation were demonstrated by systemic application of an mGlu2/3 agonist (LY379268) in combination with a negative allosteric modulator (NAM) for mGlu2 (VU6001966) or for mGlu3 (VU0650786). In brain slices, mGlu3 activation restored hippocampal plasticity (CA1 neurons) by biasing synaptic plasticity from LTD to LTP, which was surprising given the commonly observed inhibitory effects of group II mGlu on synaptic transmission. However, similar to previous reports of a synergistic partnership between mGlu3 and mGlu5 in the establishment of LTD in prefrontal cortical pyramidal cells (6), the authors found that mGlu3-driven effects on synaptic plasticity required recruitment of mGlu5 in hippocampal pyramidal cells. It should be noted that different forms of LTP were explored here (electrically and chemically induced and priming). The critical role of mGlu5 was shown by using an mGlu5 NAM (MTEP) and transgenic mice (Grm5-CaMKII KO mice) with deletion of GRM5 in glutamatergic hippocampal neurons; both approaches blocked the facilitatory effects of mGlu3, suggesting that mGluR5 acts downstream of mGlu3 (Fig. 1). The effects of mGlu3 on synaptic plasticity were due to a post-synaptic action on neurons rather than astrocytes, based on paired-pulse ratio analysis and data from transgenic mice (Grm3-CaMKII KO) showing that deletion of GRM3 in glutamatergic hippocampal neurons blocked the actions of mGlu3. Considering that mGluR5-driven plasticity in the hippocampus is known to engage the activation CB1 receptors on GABAergic neurons onto pyramidal cell (7) the new study (5) showed that mGluR3-mediated synaptic plasticity involved CB1-mediated disinhibition of hippocampal neurons potentiated by mGlu5 signaling. LY379268 decreased inhibitory synaptic transmission (disinhibition), which was blocked by an mGlu3 NAM (VU0650786) and a CB1 receptor antagonist (AM251). AM251 also blocked LY379268-induced and electrically-induced LTP. Behavioral experiments validated the electrophysiological results by showing that the cognitive enhancing effects of LY379268 on associative learning (trace fear conditioning) are blocked in transgenic mouse models with mGlu3 and mGlu5 deletion in hippocampal pyramidal cells.

Figure 1. mGlu3- and mGlu5-mediated hippocampal synaptic plasticity.

(A) In the absence of mGlu3 activation, mGlu5 interacts with Homer protein to engage the PI3K/Akt/GSK3β pathway, resulting in the internalization of AMPA receptors and generation of LTD. (B) In the PCP model of schizophrenia, NMDA receptor hypofunction has been linked to increased inhibitory tone onto the hippocampal pyramidal cells, resulting in impaired LTP (10). (C) Activation of mGlu3 switches mGlu5-mediated plasticity towards the generation of LTP by promoting endocannabinoid-mediated inhibition of GABAergic interneurons (disinhibition) to drive hippocampal pyramidal cells, resulting in increased NMDA receptor function. mGlu3 and mGlu5, metabotropic glutamate receptor subtypes 3 and 5; PI3K, phosphoinositide 3-kinase; Akt, known as protein kinase B (PKB); GSK3β, glycogen synthase kinase 3β; AMPA, α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor; LTD, long-term depression; NMDA, N-methyl-D-aspartate receptor; LTP, long-term potentiation; 2-AG, 2-arachidonoylglycerol; CB1, cannabinoid receptor type 1. Grey indicates hypofunction.

Taken together, this new study (5) significantly advances the field of mGlu and schizophrenia research by showing that mGlu3 activation restores neuroplasticity and mitigates cognitive deficits in a schizophrenia model through a novel mechanism that involves interaction with mGlu5 and endocannabinoid-mediated disinhibition. Thus, mGlu3 activation counteracts the changes in NMDA and GABA receptor function observed in this model (Fig. 1). The study not only employed recently discovered mGlu3 (and mGlu2) NAM but also newly developed transgenic tools that may inspire future studies necessary for the further investigation of the role of mGlu3 in schizophrenia and beyond. Still, a number of questions remain to be answered for the better understanding of mGlu3 function in neuroplasticity and schizophrenia. What is the relationship between NMDA and mGlu3 hypofunction? Is NMDA receptor hypofunction a direct outcome of mGlu3 hypofunction? Can mGlu3 or mGlu5 activation reverse mGlu3 hypofunction? Are there mGlu5-independnet beneficial effects of mGlu3 activation? Will mGlu3 agonism (mGlu3 PAM) work in conditions of nonfunctional mGlu3 receptor expression? Finally, potential sex differences remain to be explored. The study reported increased freezing behavior in female Grm3-CaMKII KO mice relative to controls.

While the new study (5) identifies mGlu3 activation as a promising therapeutic strategy for schizophrenia-associated cognitive deficits, mGlu3 still remains a challenging target for the development of selective enhancers as a pharmacological approach, which is a significant limitation for preclinical and clinical studies. In 2007, a phase II clinical trial found exciting antipsychotic effects of a group II mGlu agonist administered orally as a prodrug (LY404039, pomaglumetad methionil) in patients affected by schizophrenia, but unfortunately, these data were not confirmed in subsequent clinical studies that provided contradicting results in similar conditions, discouraging further trials (see 8). An explanation suggested for these inconsistent and disappointing findings was the recruitment of patients previously treated with atypical antipsychotic medications that were shown to epigenetically down-regulate mGlu2 in preclinical studies (see 8). However, oral treatment with an mGlu2 PAM (AZD8529) failed to ameliorate symptoms in patients affected by schizophrenia compared to the placebo group in another clinical trial (9), suggesting perhaps that the selective manipulation of mGlu2 may not be an effective strategy for psychotic patients. Based on the present study and other recent evidence from preclinical and clinical studies mGlu3 has emerged as a more promising therapeutic target for the treatment of pathological conditions such as schizophrenia and cognitive dysfunction. Several challenges remain, including the identification of subpopulations of patients with schizophrenia that could benefit from a pharmacological approach aimed at increasing mGlu3 activity in brain areas affected by the disease pathology. Moreover, a better understanding of neuronal (and perhaps non-neuronal) mechanisms and circuits that contribute to the development of schizophrenia is needed.