One of the leading hypotheses of depression suggests that its pathomechanism is associated with over-activation of N-methyl-D-aspartic acid receptors (NMDARs), excessive release of glutamate (Glu), and the induction of a depressive phenotype (Mardsen, 2021). Released Glu acts on glutamate ionotropic (NMDARs; α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid AMPARs; or Kainate) and metabotropic (mGluRs) receptors. The mGluRs form a heterogeneous group of receptors that are linked to the secondary messenger pathways via the G protein, and their activation influences behavioral changes (Mardsen, 2021). Interestingly, mGluR5 and NMDA receptors were found in proximity at the postsynaptic cleft (Szewczyk et al., 2012), and the Homer family of proteins can functionally link mGlu5 receptors with inositol trisphosphate receptors (IP3Rs) and with Shank proteins functionally existing as part of NMDA receptor-associated PSD-95 complex; thus, one can hypothesize mutual regulation (Stachowicz, 2019). NMDAR activation (e.g., by stress) opens a non-selective ion channel for Ca2+ and Na+ ions that can enter the postsynaptic neuron (Mardsen, 2021). Secondly, NMDARs can be activated by voltage-dependent mechanisms, depolarization, Mg2+ release, and channel opening (Mardsen, 2021). The synaptically or extrasynaptically located NMDAR subunits are influenced by physiological or pathological stress (Stark and Bazan, 2011). Synaptic NMDARs have a bidirectional effect on the functionality of neurons, with low-level activation that induces long-term depression (LTD) and high-level activation that induces long-term potentiation (LTP) (Mardsen, 2021). Extrasynaptic NMDARs are associated with LTD-inducing neuronal failure (Mardsen, 2021). Xu et al. (2009) found prior activity history to be an inducer of bidirectional changes in the NMDAR (GluN2A/GluN2B) subunits ratio. However, the mechanism of reciprocal regulation between the NMDA receptor and mGluR5 in the behavioral response is not known. It is reasonable to assume that there is an extrinsic factor that regulates the reciprocal interaction between these receptors in stress. Given the research, this factor could be an enzyme that controls lipid metabolism in cells by affecting arachidonic acid (AA) levels, that is, cyclooxygenase-2 (COX-2) (Stachowicz, 2021). Stark and Bazan (2011) documented lipid peroxidation of AA operating as a modulator of network activity history. Synaptic NMDARs caused enhancement of neuronal COX-2 expression, but prolonged synaptic activity suppressed its activity, acting through AA. At the same time, decreased COX-2 expression mediated by extrasynaptic NMDAR challenge was found (Stark and Bazan, 2011). Xu et al. (2009) observed that pharmacologically manipulated GluN2A/GluN2B ratio can change the subsequent LTP/LTD threshold. All of these modifications have the potential to change depressive behavior or cognition (Adell, 2020).

NMDARs consist of various combinations of GluN1, GluN2 (A-D), and GluN3 (A/B) subunits (Mardsen, 2021). Their number, localization and subunit composition are strictly regulated and differ in a cell- and synapse-specific manner (Sanz-Clemente et al., 2013a). Studies on depression and cognition have shown that GluN1 and GluN2 are present in neurons as a heterotetrameric combination and are linked to depression (Mardsen, 2021). GluN2A and GluN2B subunits are highly expressed in the cortex and hippocampus (HC), modulating the properties of endogenous NMDARs (Sanz-Clemente et al., 2013a). NMDARs mainly occur postsynaptically but are also observed extrasynaptically (Adell, 2020). Activation of synaptic NMDARs promotes survival of synapses and cells, while overactivation of extrasynaptic NMDARs is associated with excessive Glu release and cell death (Adell, 2020). Histopathological data have shown that GluN2A subunits are primarily located at the synapse, while GluN2B are primarily localized extrasynaptically (Adell, 2020). Receptors mainly containing the GluN2A subunit may contribute to improved synaptic plasticity, while selective GluN2B antagonists may exhibit neuroprotective properties (Adell, 2020). The composition of NMDA receptor synaptic subunits changes from dominant GluN2B to GluN2A during synaptic maturation and in response to activity and stress (Sanz-Clemente et al., 2013a). It was shown that the inhibition of NMDARs was connected with a decrease in extrasynaptically localized GluN2B subunits when using selective serotonin reuptake inhibitors (SSRIs) (Mardsen, 2021). Memantine and ketamine (NMDAR antagonists) act via desensitization (Glasgow et al., 2017), while ligands of mGluRs induce changes in NMDAR subunit composition (Sanz-Clemente et al., 2013a). The proportion and composition of NMDAR subunits are important factors influencing mental health, cognition, memory, and anxiety (Mardsen, 2021). Human postmortem studies have found reduced levels of GluN2A and GluN2B subunits in the prefrontal cortex (PFC), perirhinal cortex, and increased levels of GluN2A subunits in the lateral amygdala in major depression (Adell, 2020). Furthermore, higher expression levels of the NMDAR subunit 2B (Grin2b) gene in the locus coeruleus of depressed patients was detected (Adell, 2020). Grin2b has been postulated as a biomarker of suicide, and polymorphism of Grin2b has been postulated to predict treatment-resistant depression (Adell, 2020). Chronic corticosterone administration was potent in increasing Grin2a and Grin2b mRNAs in the HC (Adell, 2020). Inactivation of GluN2A subunit has been shown to possess antidepressant-like effects in mice (Adell, 2020). Furthermore, GluN2B-NMDARs regulate the activation state of the protein degradation-modulating memory liability, while GluN2A-NMDARs promote cyclic AMP response element-binding protein (CREB) phosphorylation and LTP involved in memory re-encoding and maintenance (Radiske et al., 2021). However, the way in which altering the composition of NMDAR subunits affects mood parameters by interfering with memory has not been clarified, nor have the intracellular mechanisms that induce these changes been elucidated.

Here, we used MTEP, imipramine, and NS398, separately or in combination for seven and fourteen days, and verified the levels of the NMDAR subunits using PCR (Grin2a/Grin2b) and Western blot (GluN2A/GluN2B) techniques. We found a significant effect of COX-2 inhibition on changes in NMDAR subunit composition induced by the mGluR5 antagonist or imipramine. This discovery could be crucial in the search for new treatments for depression without the side effects of clinically used antidepressants.