Targeting metabotropic glutamate receptors for novel treatments of schizophrenia

Abstract

Support for the N-methyl-D-aspartate receptor (NMDAR) hypofunction hypothesis of schizophrenia has led to increasing focus on restoring proper glutamatergic signaling as an approach for treatment of this devastating disease. The ability of metabotropic glutamate (mGlu) receptors to modulate glutamatergic neurotransmission has thus attracted considerable attention for the development of novel antipsychotics. Consisting of eight subtypes classified into three groups based on sequence homology, signal transduction, and pharmacology, the mGlu receptors provide a wide range of targets to modulate NMDAR function as well as glutamate release. Recently, allosteric modulators of mGlu receptors have been developed that allow unprecedented selectivity among subtypes, not just groups, facilitating the investigation of the effects of subtype-specific modulation. In preclinical animal models, positive allosteric modulators (PAMs) of the group I mGlu receptor mGlu₅ have efficacy across all three symptom domains of schizophrenia (positive, negative, and cognitive). The discovery and development of mGlu₅ PAMs that display unique signal bias suggests that efficacy can be retained while avoiding the neurotoxic effects of earlier compounds. Interestingly, mGlu₁ negative allosteric modulators (NAMs) appear efficacious in positive symptom models of the disease but are still in early preclinical development. While selective group II mGlu receptor (mGlu_2/3) agonists have reached clinical trials but were unsuccessful, specific mGlu₂ or mGlu₃ receptor targeting still hold great promise. Genetic studies implicated mGlu₂ in the antipsychotic effects of group II agonists and mGlu₂ PAMs have since entered into clinical trials. Additionally, mGlu₃ appears to play an important role in cognition, may confer neuroprotective effects, and thus is a promising target to alleviate cognitive deficits in schizophrenia. Although group III mGlu receptors (mGlu_4/6/7/8) have attracted less attention, mGlu₄ agonists and PAMs appear to have efficacy across all three symptoms domains in preclinical models. The recent discovery of heterodimers comprising mGlu₂ and mGlu₄ may explain the efficacy of mGlu₄ selective compounds but this remains to be determined. Taken together, compounds targeting mGlu receptors, specifically subtype-selective allosteric modulators, provide a compelling alternative approach to fill the unmet clinical needs for patients with schizophrenia.

Keywords: Allosteric modulator; Glutamate; Heterodimer; NAM; PAM; Schizophrenia; Signal bias; mGlu receptor.

Fig. 1

Localization of metabotropic glutamate receptor subtypes. Schematic representation of the predominant locations of mGlu receptors at the synapse. mGlu₁ (light blue) is found on postsynaptic glutamatergic neurons as well as on GABAergic neurons. mGlu₅ (yellow) can be located on the same neurons as mGlu₁ as well as on glia. mGlu₂ (purple) is found primarily presynaptically as both a homodimer as well as a heterodimer with mGlu₄ (red). mGlu₃ (dark blue) is found on both presynaptic and postsynaptic glutamatergic, GABAergic, and neuromodulatory neurons as well on glia. mGlu₄ is localized to both modulatory neurons as well as on presynaptic glutamatergic neurons as either a homodimer or heterodimer. mGlu₇ (green) is localized to presynaptically neurons as well as GABAergic neurons. Lastly, mGlu₈ (orange) is primarily localized presynaptically and mGlu₆ is not shown since it is restricted to the retina

Fig. 2

Emerging concepts in the development of mGlu receptor-targeting antipsychotic therapeutics. a mGlu₅ PAMs have recently been developed that bias mGlu₅ signaling away from NMDAR modulation but still mobilize intracellular Ca²⁺ and activate ERK1/2. The mechanism of this bias is still unclear but could involve G-protein dependent versus independent coupling of mGlu₅ to NMDAR. The biased mGlu₅ PAM VU0409551 retains antipsychotic-like efficacy in vivo but does not cause excitotoxicity or seizures observed with unbiased mGlu₅ PAMs that enhance mGlu₅-mediated modulation of NMDAR currents. This suggests that NMDAR modulation is not necessary for in vivo efficacy and that this signal bias may provide a means to overcome the NMDAR-mediated excitotoxicity that has stalled mGlu₅ PAM development. b Recently, functional mGlu_2/4 heterodimers with unique pharmacology have been identified. This suggests that actions at the mGlu_2/4 heterodimer rather than at the mGlu_4/4 homodimer might underlie the antipsychotic efficacy of mGlu₄ PAMs, such as Lu AF21934, consistent with the antipsychotic-like effects of mGlu₂-specific ligands. This remains to be tested experimentally but may provide an interesting alternative to failed mGlu₂ clinical programs