Antidepressant Effects and Mechanisms of Group II mGlu Receptor-Specific Negative Allosteric Modulators

Abstract

In this issue of Neuron, Joffe et al. assess the antidepressant-relevant effects and underlying neural mechanisms of negative allosteric modulators selective for either metabotropic glutamate receptors 2 (mGlu₂) or 3 (mGlu₃). Negative modulation of both receptors enhanced excitatory glutamatergic input to mouse prefrontal cortex pyramidal cells, leading to antidepressant-relevant actions.

Historically, nearly all medications approved for the treatment of depression targeted monoaminergic neurotransmission, inspiring monoamine-centric views of depression’s underlying neurobiology. These conventional antidepressant medications require long-term administration for an effect to emerge, and leave a large proportion of individuals to remain treatment-resistant and continue suffering from depression. Over the last decade, intravenous infusion of (R,S)-ketamine (ketamine) has been used as an off-label treatment for depression. Ketamine acts as an antidepressant within hours following a single administration, and its effects following such an administration are typically sustained for days, and sometimes up to a week. Recently (March 2019), the United States Food and Drug Administration (FDA) approved (S)-ketamine (esketamine/Spravato^™) nasal spray for depression, constituting the first mechanistically novel antidepressant medication to be FDA-sanctioned in several decades.

A revised neurobiological understanding of depression places an emphasis on excitatory (glutamatergic) synaptic neurotransmission in specific mood- and reward-related neural circuits, including those that target or are targeted by the prefrontal cortex (PFC). Evidence indicates that synapses within these circuits are weakened by genetic and environmental factors, including exposure to chronic stress and/or circulating stress hormones (Duman et al., 2019; Gould et al., 2019), which are linked to the development of depression in susceptible individuals (Figure 1A). Clinical and preclinical studies have shown that ketamine rapidly enhances glutamatergic neurotransmission in the PFC, as well as other regions including the hippocampus, and this synaptic plasticity is currently thought to underlie its rapid antidepressant effects (Duman et al., 2019; Gould et al., 2019) (Figure 1A).

Figure 1.

(A) A combination of genetic and environmental factors contribute to decreased prefrontal glutamate signaling and depression symptoms (left). Rapid-acting antidepressant treatments, such as ketamine, are proposed to reverse this decrease in prefrontal glutamate signaling, leading to rapid and sustained relief of symptoms (right). (B) Schematic diagram of the mediodorsal thalamus to prefrontal cortex neural pathway. Activation of this pathway by metabotropic glutamate receptor 2/3 (mGlu_2/3) negative allosteric modulator (NAM) treatment is proposed to lead to a reversal of passive coping and anhedonic behaviors in mice, indicative of antidepressant-like responses. (C) Schematic diagram of the synaptic mechanisms targeted by mGlu₂- and mGlu₃-specific NAMs. Presynaptic mGlu₂ acts as an inhibitory autoreceptor to restrain vesicular glutamate release. mGlu₂-specific NAMs enhance presynaptic glutamate release by attenuating this inhibitory feedback mechanism. mGlu₃ located postsynaptically mediates long-term depression through α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor down-regulation. mGlu₃-specific NAMs enhance glutamatergic signaling by attenuating this form of postsynaptic plasticity. mGlu₃ is also found on astrocytes, and their activation enhances synaptic glutamate reuptake – however an interaction between astrocytic mGlu₃ and mGlu₃-specific NAMs is not tested for in the current study.

Abbreviations: BDNF, brain derived neurotrophic factor; EAAT1/2, excitatory amino acid transporter 1 and 2; mTOR, mechanistic target of rapamycin kinase; NMDA, N-methyl-D-aspartate

While ketamine’s (and esketamine’s) rapid onset of action represents a significant advancement in many ways over conventional antidepressants, it also exhibits side effects which limit its widespread use, namely dissociation, and ketamine is also an abused drug. One alternative target being pursued are the group II metabotropic glutamate (mGlu) receptors comprised of mGlu₂ and mGlu₃, which both mediate negative feedback at glutamatergic synapses. Activation of these receptors suppresses synaptic glutamate release (mGlu₂) or enhances glutamate reuptake and decreases postsynaptic glutamate signaling (mGlu₃) (Ionescu and Papakostas, 2017) (Figure 1). It was previously determined that the mechanism underlying antidepressant-relevant actions of ketamine converges with mGlu₂ signaling (Zanos et al., 2019) and orthosteric antagonists of group II mGlu exert rapid antidepressant-like effects in preclinical studies (Chaki, 2017). However, the relative contributions of mGlu₂ and mGlu₃ to these antidepressant actions were unknown.

The Conn group at Vanderbilt University had previously developed specific small molecule negative allosteric modulators (NAMs) of mGlu₂ (VU6001966) and mGlu₃ (VU0650786). In the present issue of Neuron, Joffe et al. (2019) report on the rapid antidepressant effects of these NAMs and their possible mechanisms to exert antidepressant actions. The rationale was that such studies would clarify whether the antidepressant signal from non-selective group II mGlu inhibition arose primarily from mGlu₂ versus mGlu₃. The authors found that a single administration of either mGlu₂- or mGlu₃-specific NAMs acutely reduced passive coping behaviors (immobility) in a mouse test for antidepressant efficacy, the forced-swim test, and exerted anti-anhedonic effects following chronic stress that was sustained 24 hours after drug administration. These latter findings are indicative of rapid-acting antidepressant-like responses of both these compounds, similar to the sustained behavioral actions of ketamine (Duman et al., 2019; Gould et al., 2019).

Based on previous studies which demonstrated that increased glutamate transmission in the PFC is necessary for the antidepressant-like effects of an orthosteric mixed-mGlu_2/3 antagonist (Fukumoto et al., 2016), Joffe et al. examined neuronal activation in the PFC following systemic administration of the mGlu-specific NAMs. By using a mouse line that reports neuronal activity (Fos-eGFP), the authors found that both mGlu₂- and mGlu₃-specific NAMs induce activation of a subpopulation of neurons within the PFC, referred to as ‘type A’ (Anastasiades et al., 2018). A previous study found that ‘type A’ neurons in the PFC are strongly and reciprocally connected with the mediodorsal thalamus (MDT) (Collins et al., 2018) (Figure 1B). Here, the experiments of Joffe et al. revealed that bath application of the mGlu₂- specific NAM acutely augments optogenetically-evoked MDT to PFC currents via a presynaptically-targeted action to enhance glutamate release, suggesting that mGlu₂ tonically restrains glutamate release in this pathway (Figure 1C). In contrast, application of the mGlu₃-specific NAM did not acutely modify synaptic transmission, which appears to contradict the neuronal activation within the PFC observed following administration of the mGlu₃ NAM in vivo. The authors addressed this discrepancy by demonstrating that both mGlu-specific NAMs attenuate mGlu₃ activation-induced long-term depression (LTD) of synaptic strength in the MDT to PFC pathway (Figure 1C). The authors did not, however, test whether tonic mGlu₃ activation-mediated LTD occurs at these synapses in vivo, or if it is expressed following chronic stress, to further confirm that this is the mechanism underlying the observed mGlu₃ NAM actions. This is an important question to be addressed in future studies.

To establish a causal role for the actions of mGlu-specific NAMs on the MDT to PFC circuit relevant to their observed antidepressant-like behavioral effects, the authors also tested mice in the forced swim test following chemogenetic inhibition of this circuit after acute drug treatment. While this manipulation was found to occlude the effects of mGlu NAM treatment on latency to immobility, chemogenetic inhibition of the MDT-PFC pathway itself tended to increase latency to immobility, in the absence of any NAM treatment, somewhat confounding interpretation of these experimental results. In contrast with these findings, previous reports found that activation, rather than inhibition, of this circuit produces antidepressant-like effects (Miller et al., 2017). As the authors note, this contradictory finding could be explained by a net disinhibition of excitatory neurons in the PFC following chemogenetic inhibition due to the presence of extensive connections between the MDT and fast-spiking inhibitory cortical neurons. In order to avoid this confounding factor, future experiments could test the effect of MDT-PFC inhibition during the time of drug treatment, and behaviorally test mice at a later time point. Such an experiment would be predicted to reveal blockade of the persistent behavioral effects of mGlu₂ and mGlu₃ NAMs, given that transient activation of these neuronal ensembles during treatment is presumably necessary for the initiation of sustained synaptic plasticity. Additionally, since it has been established that the persistent antidepressant-like effects of mixed mGlu_2/3 antagonists require downstream factors shared with other rapid-acting antidepressants (e.g., brain derived neurotrophic factor (BDNF) and mechanistic target of rapamycin kinase (mTOR) signaling (Chaki, 2017)) the role of these pathways could be examined in future studies using mGlu-specific NAMs.

mGlu receptors represent an enticing target in the search for novel rapid-acting antidepressants, given their proximity to, and function at, synapses hypothetically altered in depression. This paper provides the first evidence suggesting that both mGlu₂ and mGlu₃ inhibition may individually have rapid-acting antidepressant effects, similar to ketamine. To date, only one drug targeting group II mGlu receptors for depression has been assessed in a clinical trial. In particular, in a phase 2 clinical trial, decoglurant, a mixed mGlu_2/3 negative allosteric modulator did not differentiate from placebo in patients suffering from depression (Gould et al., 2019). However, since no marker of target engagement was assessed, it is difficult to confirm that adequate brain exposure was achieved and thus further clinical studies are required to assess for antidepressant actions of mGlu₂ or mGlu₃ inhibition, either in combination or individually.

While there continues to be lack of clarity regarding whether inhibition of mGlu₂, mGlu₃, or both these receptors together is the most viable approach to antidepressant action, the ability to separately target mGlu₂ and mGlu₃ using specific NAMs presents the potential for a more focused treatment of depression. In summary, these group II mGlu NAMs, or molecules with similar pharmacology, represent promising candidates as novel rapid-acting antidepressants.