NMDA Receptors in Health and Disease

Abstract

NMDA receptors (NMDARs) are a subtype of ionotropic glutamate receptors that mediate excitatory neurotransmission and synaptic plasticity in the brain. NMDARs play important roles in various normal brain functions such as learning, memory, and cognition, but also contribute to the pathogenesis of several developmental, neurological, and psychiatric disorders. Alterations in NMDARs can result in either hypo- or hyperfunction of NMDARs, which can impair neuronal viability, synaptic efficacy, and network oscillations. In this review, we summarize the current knowledge on the involvement of NMDA receptors in Alzheimer’s disease, autism spectrum disorder, epilepsy, and schizophrenia. We also highlight the potential therapeutic strategies that target NMDAR modulation and dysfunction in these disorders.

1. Introduction

Glutamate receptors are a diverse group of proteins that play a crucial role in the nervous system. They are responsible for mediating the majority of excitatory neurotransmission in the brain. Glutamate receptors can be broadly classified into two main types: ionotropic and metabotropic [1, 2]. Ionotropic glutamate receptors (iGluRs) are ligand-gated ion channels that allow ions to flow across the cell membrane in response to the binding of glutamate [2]. They can be further divided into three main subtypes: N-methyl-D-aspartate receptors (NMDARs), α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors (AMPARs), and kainate receptors [2]. Each subtype has distinct properties and roles in synaptic transmission and plasticity. Metabotropic glutamate receptors (mGluRs), on the other hand, are G protein-coupled receptors that activate intracellular signaling pathways in response to glutamate [3]. Both iGluRs and mGluRs are involved in various physiological processes, such as neuronal development, synaptic plasticity, learning, and memory [2, 3].

2. NMDARs in health

NMDARs are involved in various processes of the synapse, such as its formation, modification, learning, memory, and cognitive functions [2]. NMDARs are different from AMPARs and kainate receptors in that they need both glycine (or D-serine in the body) and glutamate to be activated. They allow calcium to enter the cell and are blocked by magnesium when the cell is at rest [2]. NMDARs can detect the timing of pre-synaptic glutamate release and post-synaptic membrane depolarization, which makes them unique among the receptors.

NMDARs are composed of four subunits, each of which belongs to one of three families: GluN1, GluN2, and GluN3. GluN1 is the obligatory subunit that forms the core of the receptor and binds glycine as a co-agonist. GluN2 subunits (A–D) determine the pharmacological and biophysical properties of the receptor and bind glutamate as the main agonist. GluN3 subunits (A and B) modulate the receptor function and can also bind glycine. The subunit composition of NMDARs varies depending on the brain region, cell type, and developmental stage, resulting in a diversity of receptor subtypes and complexes that have distinct roles in synaptic plasticity, learning, memory, and neurological disorders [2].

Each subunit of NMDARs has a similar structure that consists of four main domains: the amino-terminal domain (ATD/NTD), the ligand-binding domain (LBD), the transmembrane domain (TMD), and the carboxyl-terminal domain (CTD) [2]. The ATD/NTD is located at the extracellular side of the receptor and is involved in subunit assembly, allosteric modulation, and receptor trafficking. The LBD is also extracellular and contains the binding sites for glutamate and glycine, as well as for various modulators such as zinc, magnesium, polyamines, protons, and neurosteroids. The TMD spans the membrane four times (TM1, TM2, TM3, and TM4) and forms the ion channel pore that allows the flux of sodium, potassium, and calcium ions upon receptor activation. The CTD is intracellular and interacts with various signaling molecules and scaffolding proteins that regulate the receptor localization, function, and coupling to downstream pathways.

Each NMDAR is composed of two GluN1 subunits and two GluN2 or GluN3 subunits, which can be either identical (diheteromers) or different (triheteromers). The subunit composition of NMDARs affects their biophysical and pharmacological properties, such as their affinity to co-agonists (glycine and glutamate), sensitivity to modulators (zinc and magnesium), and interaction with other drugs. NMDARs also have different expression patterns and functions in different brain regions and developmental stages. NMDARs are not static but rather undergo dynamic changes in their trafficking, organization, diffusion, and surface expression. These processes are regulated by various factors, such as protein-protein interactions, post-translational modifications, extracellular matrix components, and synaptic activity [4, 5, 6, 7].

The term diheteromer or triheteromer refers to the number of distinct subunits in an NMDAR, not the total number of subunits, which is always four. The most common triheteromer in the adult forebrain is GluN1/GluN2A/GluN2B, which constitutes a major synaptic NMDAR population. Other triheteromers, such as GluN1/GluN2A/GluN2C, GluN1/GluN2A/GluN2D, and GluN1/GluN2B/GluN2D, have also been reported to function in specific brain regions. NMDARs that contain GluN3 subunits are considered atypical and unconventional. GluN3A can form triheteromers with GluN1 and GluN2A or GluN2B in neurons, or with GluN1 and GluN2C in oligodendrocytes. These receptors are thought to play a role in synapse pruning and destabilization [2, 5].

NMDARs are crucial for long-term synaptic plasticity induction at different synapses and for cognitive functions. Animal models showed that non-competitive NMDAR antagonists (such as PCP, ketamine, and MK-801) induced cognitive impairment similar to schizophrenia [8]. Likewise, in healthy humans, NMDAR inhibition caused cognitive and behavioral dysfunction [9]. For instance, ketamine induced both positive and negative schizophrenia symptoms [10]. Genetically, GluN1 gene global knock-out caused neonatal death [11], confirming the importance of NMDARs. Specific GluN1 knock-out in the hippocampal CA1 region showed that NMDARs are essential for memory formation consolidation [12]. GluN1 ablation in the hippocampal CA3 region impaired spatial memory retrieval [13]. Moreover, GluN1 deletion in parvalbumin-positive interneurons disrupted hippocampal synchrony, spatial representations, and working memory [14]. Therefore, the evidence supports the roles of NMDARs in synaptic plasticity and cognition.

NMDARs have implications for various neurological and psychiatric disorders, depending on whether they are overactivated or underactivated by different factors. When NMDARs are overactivated, they can cause excitotoxicity, which is the harmful overstimulation of neurons by glutamate. This can happen in situations such as stroke, trauma, hypoxia, ischemia, and neurological diseases such as Alzheimer’s disease (AD), epilepsy, Parkinson’s disease, Huntington’s disease, and amyotrophic lateral sclerosis (ALS). When NMDARs are underactivated, they can disrupt synaptic plasticity and cognitive functions, and lead to problems such as memory loss, schizophrenia, and autism. In the following sections, we explore the roles of NMDARs in AD, autism spectrum disorder (ASD), epilepsy, and schizophrenia. For other disease states that are not covered here, readers can refer to some excellent recent review articles [15, 16, 17, 18, 19, 20].

3. NMDARs in disease

3.1 NMDARs in AD

3.1.1 Introduction

AD is a chronic brain disorder that gradually erodes cognitive functions, memory, and behavior. It is the leading cause of dementia among older people and a major public health challenge [21].

The exact mechanisms of AD are not fully understood, but several factors contribute to its onset and progression. These include genetic variations, accumulation of amyloid-beta (Aβ) plaques, formation of tau tangles, inflammation of brain cells, oxidative damage, impaired blood flow, reduced acetylcholine levels, and environmental influences [21]. Currently, there is no cure for AD, but some therapies can help to reduce the symptoms and slow down the cognitive deterioration. These include drugs that inhibit acetylcholinesterase or block NMDARs, antioxidants, anti-inflammatory agents, and lifestyle modifications such as diet, exercise, and cognitive training [21].

3.1.2 Amyloid and tau hypotheses

Both Aβ and tau proteins are normally present in the brain, but they can undergo pathological changes that lead to their accumulation and aggregation [21, 22].

Aβ is a peptide that is derived from the cleavage of a larger protein called amyloid precursor protein (APP). Aβ can form insoluble aggregates that deposit as plaques in the brain. These plaques are believed to impair neuronal function and cause neurodegeneration [23]. Tau is a protein that binds to microtubules, which are essential for intracellular transport and neuronal structure. Tau can become hyperphosphorylated, meaning that it has too many phosphate groups attached to it. Hyperphosphorylated tau can detach from microtubules and form intracellular aggregates called neurofibrillary tangles. These tangles are believed to disrupt neuronal transport and cause cell death [24].

Both Aβ and tau have been implicated in the cognitive impairment and memory loss that are hallmarks of AD. Both Aβ and tau have been shown to interact with NMDARs, which are essential for synaptic plasticity and memory formation [24, 25].

3.1.3 NMDARs in AD

Both Aβ and tau can interact with NMDARs, which are essential for synaptic plasticity and memory formation. Aβ can interact with both synaptic and extrasynaptic NMDARs, but it preferentially impairs the trafficking and localization of NMDARs containing the GluN2B subunit, which are more- localized at extrasynaptic sites and vulnerable to excessive calcium influx and excitotoxicity. Aβ also interferes with the signaling pathways downstream of NMDARs, such as by inhibiting CaMKII activity, reducing CREB phosphorylation, activating calcineurin and caspase-3, and inducing tau hyperphosphorylation. These effects impair synaptic plasticity and promote dendritic degeneration. Moreover, Aβ can bind to other receptors that modulate NMDAR function, such as mGluR5, PrPc, and α7 nicotinic receptors, leading to further synaptic dysfunction and loss. We will now discuss in more detail how Aβ and tau proteins can interact with NMDARs, and how NMDARs can interact with Aβ and tau in return.

3.1.3.1 NMDAR trafficking

Firstly, Aβ oligomers affect synapses by altering the trafficking and localization of NMDARs, which are crucial for learning and memory. Aβ oligomers can impact on both synaptic and extrasynaptic NMDARs, but they preferentially impair NMDARs containing the GluN2B subunit, which are more vulnerable to excessive calcium influx and excitotoxicity [26]. Research has demonstrated that Aβ oligomers can suppress long-term potentiation (LTP) [27, 28, 29], enhance the induction of long-term depression (LTD) [30, 31], and trigger the internalization of AMPA and NMDA receptors [32, 33, 34, 35]. Thus, Aβ1–42 has been found to decrease the surface expression of GluN2B-containing NMDARs in cortical neurons [34, 36]. Aβ oligomers can also bind to the EphB2 receptor (Eph receptor B2) and trigger its proteasome-dependent degradation, leading to reduced surface expression and phosphorylation of GluN2B-containing NMDARs [37]. This results in reduced calcium influx into active spines, impaired induction of LTP, enhanced induction of LTD, and increased spine loss. These effects of Aβ oligomers on NMDARs contribute to cognitive impairment and neurodegeneration in AD.

3.1.3.2 NMDAR signaling

Besides impairing NMDAR expression and localization, Aβ oligomers may interfere with the signaling pathways that depend on or converge on NMDARs. Aβ oligomers can also form ion-permeable channels in cellular membranes, allowing calcium influx into neurons and affecting synaptic plasticity [38, 39, 40]. Aβ oligomers can cause overactivation of extrasynaptic NMDARs (eNMDARs), which are located outside the synaptic cleft and have different subunit composition and function from synaptic NMDARs (sNMDARs). eNMDAR activation by Aβ oligomers leads to excessive calcium influx and oxidative stress, which in turn activate various enzymes, such as calpain, caspase, Cdk5, GSK-3β, Fyn, and CaMKII, that can phosphorylate tau, cleave dynamin, disrupt cytoskeleton, and cause endocytosis of AMPARs and NMDARs [41]. These events result in synaptic depression, spine elimination, mitochondrial dysfunction, and neuronal death. On the other hand, sNMDAR activation by physiological glutamate stimulation is beneficial for neuronal survival, synaptic plasticity, and memory formation [42]. Therefore, Aβ oligomers shift the balance between eNMDARs and sNMDARs toward the former, leading to synaptic dysfunction in AD.

3.1.3.3 Tau and NMDARs

Tau can interact with NMDARs, influencing their function and vice versa. Pathological tau can lead to synaptic loss and dysfunction, as evidenced by reduced spine density and impaired LTP in tau P301L transgenic mice rTg4510 [43]. The mechanisms behind tau synaptotoxicity are still being investigated, but Fyn kinase is a potential mediator. Fyn kinase, located at postsynaptic densities, can modulate tau-dependent synaptic and cognitive dysfunction. Tau binds to Fyn and enhances its interactions and stabilizing effects with NMDARs, which is required for LTP induction [44]. Deleting tau in mice alters Fyn localization in postsynaptic compartments and reduces NMDAR-dependent excitatory toxicity in response to Aβ [45]. Inhibiting Fyn kinase also reduces tau aggregation, suggesting that tau-Fyn interactions can exacerbate tau pathology in an AD mouse model [46]. Furthermore, tau can bind to Fyn and induce Fyn phosphorylation in the brain of an AD patient. Phosphorylated Fyn enhances interactions between NMDAR and the postsynaptic scaffolding component, PSD95, which can increase excitatory glutamate sensitivity and thereby exacerbate Aβ excitotoxicity [45]. Finally, Aβ can trigger an influx of calcium through NMDARs, which then activates GSK3β. This can lead to excessive phosphorylation of tau protein, causing it to mislocalize to dendrites [47].

3.1.4 Clinical implications and summary

A better understanding of NMDARs in AD may lead to more effective therapy. NMDAR antagonists are drugs that reduce NMDAR activity and may help to treat AD symptoms and slow down disease progression. The only drug of this class approved for AD is memantine, which has a weak and fast binding to NMDARs, allowing it to prevent excessive stimulation without interfering with normal function. Memantine can enhance cognition and delay functional decline in patients with moderate to severe AD, especially when used together with cholinesterase inhibitors, another class of drugs that increase acetylcholine levels in the brain. However, memantine does not work for mild AD or for preventing AD onset, and its effects are small and inconsistent among patients.

Therefore, more research is needed to develop better and more specific NMDAR modulators that can target the different types and components of NMDARs, as well as their interactions with Aβ and other factors related to AD development. Some promising candidates are NR2B-selective antagonists, glycine site modulators, and allosteric modulators. These drugs may have better outcomes than memantine in terms of strength, selectivity, safety, and effectiveness. However, more clinical studies are required to verify their usefulness and optimal dose for AD patients.

In summary, NMDARs are involved in various aspects of AD, but their exact role is still unclear. AD is characterized by the accumulation of amyloid-beta and tau proteins, which can affect NMDAR activity and lead to neuronal death, synaptic dysfunction, and spine loss. However, the effects of amyloid-beta and tau on NMDARs are not uniform, as they depend on the oligomeric state and subcellular localization of these proteins. Further research is required to understand how NMDAR dysfunction contributes to AD and to identify more effective treatments.

3.2 NMDARs in ASD

3.2.1 Introduction

ASD is a condition related to brain development that impacts how a person perceives and socializes with others, causing problems in social interaction and communication. The term “spectrum” in ASD refers to the wide range of symptoms and severity that can vary from person to person [48].

ASD is believed to be caused by a combination of genetic and environmental factors that affect synaptic plasticity and connectivity in the brain. Many genes associated with ASD encode proteins that regulate synaptic plasticity and connectivity at different levels [48]. Mutations in these genes can disrupt the balance between synaptic excitation and inhibition, leading to abnormal brain development and function. Environmental factors, such as prenatal exposure to infections, toxins or stress, and others can also modulate the effects of genetic mutations on ASD risk [48].

3.2.2 Genetic mutations in ASD

Recent advances in genomics have enabled the identification of many mutations that are associated with ASD [49]. For example, FMR1, MECP2, and ATRX are single genes that control gene expression and chromatin structure. Other mutations affect multiple genes that interact in common pathways or networks. For instance, many genes that encode proteins involved in synaptic function and plasticity, such as SynGAP, SHANK3, NRXN1, and NLGN3, have been found to be mutated in ASD. These genes modulate the formation and strength of synapses, which are essential for learning, memory, and behavior. Still, another example is the group of genes that regulate actin dynamics, such as ACTN4, SWAP-70, and SRGAP3, which have been shown to alter the morphology and localization of dendritic spines, the sites of synapses [49].

The role of NMDARs in ASD is supported by multiple lines of evidence. Firstly, several ASD-associated genes encode proteins that interact with or regulate NMDARs. Mutations in these genes, such as SYNGAP1, FMR1, SHANK2, and SHANK3, can affect the expression, trafficking, or function of NMDAR subunits, leading to altered synaptic strength and plasticity [48]. Secondly, mutations in other genes like PTEN, MECP2, and NLGN3, which are also associated with ASD, can alter the signaling pathways downstream of NMDAR activation [48]. These pathways, including mTOR, ERK, and WNT, play crucial roles in neuronal development and connectivity. Lastly, subtle changes in NMDAR expression, trafficking, or function have been observed in animal models of ASD [50]. These findings collectively suggest that NMDARs are critical for the proper formation and function of neural circuits, and their dysregulation may contribute to the cognitive and behavioral impairments observed in ASD.

3.2.3 NMDAR gene mutations in ASD

Mutations in genes that encode NMDAR subunits have now been firmly linked to ASD, which directly support the role of NMDARs in ASD. Two of the commonly mutated NMDAR genes in ASD are GRIN2B and GRIN2A, which encode the GluN2B and GluN2A subunits respectively. GluN2B and/or GluN2A is essential for the proper localization and function of NMDARs in the dendrites of cortical neurons. Dendrite morphology and dynamics are crucial for neuronal connectivity and information processing in the brain. Mutations in GRIN2B can disrupt the normal development and maintenance of dendritic arbors, leading to aberrant neuronal circuitry and impaired synaptic plasticity. Similarly, mutations in GRIN2A can also lead to altered synaptic signaling and neuronal development [51]. In other words, mutations in both GRIN2B and GRIN2A can interfere with how neurons grow, connect, and communicate with each other. This can lead to a range of neurodevelopmental problems, including ASD [52].

Some mutations cause truncation or loss-of-function of the GluN2B subunit, while others alter its amino acid sequence or affect its splicing or expression levels. These mutations can affect the assembly, trafficking, stability, or activity of NMDARs, resulting in either reduced or enhanced NMDAR signaling. The effects of these mutations on neuronal morphology and function may depend on the timing, location, and severity of the mutation, as well as on the interactions with other genetic or environmental factors [51, 52].

Some examples of mutations in GRIN2B that have been reported in ASD patients are:

• A single-base substitution (c27172-2A > G) at the canonical 3′ splice site of exon 10 causes a premature stop codon at position 724 of GluN2B (GluN2B 724 t). This mutation does not prevent the formation of functional NMDARs, but rather reduces their surface expression and calcium currents. It also reduces its trafficking to dendrites and synapses. The mutation also interferes with spine density and morphology, resulting in fewer and smaller spines [53, 54].

• A de novo missense mutation that causes a premature stop at position 373 of GluN2B (c.1119G > A p.(Trp373*). Rats with this mutation showed reduced social interaction and preference for social novelty, increased anxiety and stereotyped behaviors, impaired spatial learning and memory, and enhanced susceptibility to seizures. The mutation decreased the surface expression of GluN2B and the amplitude of NMDA receptor currents in the hippocampus, a brain region important for learning and memory. These findings suggest that the GluN2B-Trp373 mutation contributes to ASD-associated symptoms by disrupting the function of NMDA receptors [55, 56].

• A mutation in GluN2B (GluN2B R540H) increases NMDAR surface expression slightly compared to wild-type GluN2B. However, it does reduce NMDAR current density, as well as their sensitivity to glutamate and glycine. The mutation impairs LTP and LTD, but only when expressed in neurons that lack endogenous GluN2A. When GluN2A is present, the mutant subunits co-assemble with GluN2A and produce normal synaptic plasticity [57].

GRIN2A variants are also linked to ASD but to a lesser degree than seen with GRIN2B variants.

These mutations may affect the expression and activity of NMDARs in different brain regions or cell types, leading to synaptic dysfunction and behavioral abnormalities in ASD. Therefore, the NMDAR signaling network could be a central hub to integrate the functions of the genes that are mutated in ASD, as it modulates various aspects of neuronal development and plasticity that are relevant to the disorder. However, the relationship between NMDAR dysfunction and ASD is complex and context-dependent, as different mutations may have different effects on NMDAR function and expression, depending on the subunit composition, brain region, developmental stage, and synaptic state. Moreover, NMDAR signaling interacts with other signaling pathways that are also implicated in ASD, such as mTOR, ERK, GSK3β, and Wnt [49]. Therefore, more studies are needed to elucidate the molecular mechanisms of NMDAR mutations in ASD and to explore the potential therapeutic strategies based on modulating NMDAR function.

3.2.4 Anti-NMDAR encephalitis and ASD

Anti-NMDAR encephalitis is a rare autoimmune disorder that affects the brain and causes various neurological and psychiatric symptoms, such as psychosis, hallucinations, personality changes, cognitive impairment, seizures, and movement disorders. It is caused by the production of antibodies that target the NMDARs, which are involved in glutamate neurotransmission and synaptic plasticity.

There’s some initial evidence that suggests a potential connection between anti-NMDAR encephalitis and ASD [58]. Some instances of anti-NMDAR encephalitis, especially in children and teenagers, have been reported to show symptoms similar to ASD or have been mistakenly diagnosed as ASD. Additionally, some research has found increased levels of anti-NMDAR antibodies or other signs of autoimmunity in certain individuals with ASD. However, the exact meaning and specificity of these findings are not yet clear. The cause-and-effect relationship between anti-NMDAR encephalitis and ASD is still uncertain, and more studies are needed to understand the mechanisms and implications of this link.

3.2.5 Clinical implications

While there is broad consensus that autism involves some form of glutamatergic dysfunction, the specifics of this dysfunction (i.e., whether it involves an excess or deficiency of glutamate) remain a topic of ongoing research. Evidence suggests that NMDARs may be underactive in ASD. This evidence stems from three primary sources: (1) GRIN gene knockout mice: Mice with knockout mutations in GRIN genes, which encode NMDAR subunits, often display reduced sociability and other autism-like symptoms, (2) GRIN gene mutations in ASD: Many of the GRIN gene mutations identified in individuals with ASD appear to result in NMDAR hypofunction, and (3) Anti-NMDAR encephalitis studies: Investigations into anti-NMDAR encephalitis, a condition characterized by the presence of antibodies against NMDARs, have provided additional insights into the role of these receptors in ASD. In summary, while it’s clear that glutamatergic neurotransmission is somehow disrupted in ASD, the exact nature of this disruption is still being explored. Given the observed underactivity of NMDARs in ASD, one strategy is to enhance NMDAR function using NMDAR agonists, such as D-cycloserine [59].

Interestingly, there is evidence pointing toward an excess of glutamate (and a reduction in GABA) in ASD. Therapies targeting glutamate aim to normalize its neurotransmission. Several potential drugs aim to temper glutamate transmission. One approach is to inhibit the release of glutamate, as seen with the drug riluzole. Another strategy is to decrease glutamate signaling through the use of NMDAR antagonists, such as acamprosate (which also antagonizes mGluR5 and GABAB) and memantine [60, 61]. While some studies have shown promising results for glutamate therapy in improving ASD symptoms and cognitive function, more research is needed to establish its safety and efficacy. This includes determining the optimal dosage, duration of treatment, and combination of agents.

3.2.6 Summary

In summary, NMDARs have been implicated in ASD. Two of the genes that have been associated with ASDs are GRIN2B and GRIN2A. Mutations in GRIN2B and/or GRIN2A can result in a defective GluN2 protein that does not form functional receptors resulting in abnormal synaptic connectivity and function in the brain. Another way that NMDARs could be involved in ASDs is through autoimmunity. Some cases of anti-NMDAR encephalitis have been reported to be associated with ASDs, suggesting that autoantibodies could impair glutamate neurotransmission and contribute to ASD pathophysiology. Therapeutic strategies based on a better understanding of the molecular nature of ASD to target NMDARs are undergoing, but more studies are required to confirm the efficacy and safety [61].

3.3 NMDARs in epilepsy

3.3.1 Introduction

Epilepsy is one of the most common neurological disorders characterized by recurrent seizures, which can affect the quality of life and cognitive function of patients. Epilepsy can be caused by various factors, such as brain injury, genetic mutations, or inflammation, that disrupt the balance between excitatory and inhibitory neurotransmission in the brain [62]. One of the major excitatory neurotransmitters in the brain is glutamate, which acts on different types of receptors, including NMDARs. NMDARs play a complex and dynamic role in epilepsy, as they can modulate seizure initiation, propagation, termination, and epileptogenesis. Depending on the subunit composition, location, and activation state of NMDARs, they can have pro-convulsant or anti-convulsant effects on different types of seizures and epileptic syndromes. Abnormal expression or function of NMDARs can lead to neuronal excitotoxicity, inflammation, and epileptogenesis. Therefore, NMDARs are regarded as a potential target for suppressing epileptogenesis and treating epilepsy [63]. In this section, I will review how NMDARs contribute to the pathophysiology of epilepsy and how targeting NMDARs can offer novel therapeutic opportunities for epilepsy treatment.

NMDARs are widely distributed in the central nervous system (CNS) and play critical roles in neuronal excitability in the CNS [4]. Both clinical and preclinical studies have revealed that the abnormal expression or function of these receptors can underlie the pathophysiology of seizure disorders and epilepsy. For example, genetic studies have identified mutations in NMDAR subunits, such as GRIN1, GRIN2A, GRIN2B, and GRIN2D, that can cause various forms of epilepsy, such as developmental and epileptic encephalopathies (DEEs), rolandic epilepsy, and infantile spasms. These mutations can alter the biophysical properties, trafficking, or interactions of NMDARs, leading to either gain-of-function or loss-of-function effects [63]. We now review the molecular mechanisms and clinical implications of NMDAR mutations in epilepsy.

3.3.2 NMDAR gene mutations in epilepsy

Mutations in NMDAR genes can affect the structure and function of the NMDARs in various ways, such as altering the binding affinity of glutamate, the channel opening and closing kinetics, the receptor biogenesis and trafficking, and the sensitivity to modulators and inhibitors. These changes can have different impacts on synaptic and non-synaptic NMDAR activity, which are important for brain development, plasticity, and cognition. Depending on the location and type of mutation, as well as the expression pattern and function of the affected subunit, mutations in NMDAR genes can cause various epilepsy and other neurodevelopmental disorders such as intellectual disability, ASD, and schizophrenia [64]. Indeed, epileptic phenotypes have been linked to mutations in the genes GRIN1, GRIN2A, GRIN2B, and GRIN2D. These genes are responsible for the production of the GluN1, GluN2A, GluN2B, and GluN2D subunits of NMDARs respectively [65].

Mutations in the GRIN1 gene impair the function of the NMDAR, leading to abnormal neuronal activity and seizures. GRIN1 mutations in patients with epilepsy include duplication mutation, nonsense mutation, and missense mutations, which affect different domains of the GluN1 subunit [65]. These mutations may alter the biophysical properties of the NMDAR, such as channel opening, closing, and desensitization [65]. A novel mutation in GRIN1 (c.1923G > A, p.Met641Ile) was found in a child with refractory epilepsy and early-onset epileptic encephalopathy. Laboratory tests show that NMDARs with GluN1-M641I have increased agonist affinity and decreased Mg2+ sensitivity. NMDARs with GluN1-M641I are more responsive to the NMDAR channel inhibitors memantine, ketamine, and dextromethorphan than the normal receptors. The patient’s seizure frequency was significantly reduced by adding memantine to the anti-seizure therapy [66].

One of the most common NMDAR subunit mutations found in epilepsy is the p.Arg518His substitution in the GluN2A subunit, which affects the transmembrane domain of the protein. This mutation was first reported in a patient with Landau-Kleffner syndrome (LKS), a rare form of epileptic amnesic syndrome (EAS) that manifests with acquired aphasia and focal epileptic activity [67]. The p.Arg518His mutation was shown to impair the surface expression and trafficking of GluN2A-containing NMDARs, resulting in reduced NMDAR-mediated currents and synaptic plasticity. The mutation also altered the sensitivity of NMDARs to Mg2+ block and glycine modulation, which may affect the balance between excitation and inhibition in the brain. The p.Arg518His mutation has been found in several other patients with EAS disorders, suggesting a common pathogenic mechanism for these syndromes [67].

Mutations in the GRIN2B gene have also been linked to epilepsy [68]. The pathophysiological mechanism underlying the variability of clinical phenotypes in patients with GRIN2B mutations is unclear. A recent study identified a novel GRIN2B mutation (c.3272A > C, p.K1091T) in a patient with epilepsy and intellectual disability [69]. This mutation reduces the interaction of GluN2B with PSD-95, a scaffolding protein that stabilizes NMDA receptors at the synapse. The mutation also impairs the surface expression, glutamate sensitivity, and current density of NMDA receptors in HEK 293 T cells and hippocampal neurons. Moreover, p.K1091T mutation decreases the dendritic spine density and excitatory synaptic transmission in hippocampal neurons. These findings suggest that the GRIN2B-K1091T mutation causes a loss-of-function effect on NMDA receptors, which may contribute to the patient’s phenotype [69].

Finally, changes in the GRIN2D gene can cause different types of epilepsy. Up to now, 11 particular GRIN2D variants have been found in patients with developmental and epileptic encephalopathy (DEE). Out of them, 6 were in the M3 domain (Val667Ile, Leu670Phe, Thr674Lys, Ala675Thr, Ala678Asp, and Met681Ile). Laboratory tests of GRIN2D variants showed that Val667Ile and Leu670Phe variants greatly enhance agonist affinity and channel opening probability, and extend the closing time, leading to increased calcium entry that probably affects the clinical features seen in patients [70, 71, 72].

Mutations in the NMDAR gene result in either a gain-of-function (GoF) or loss-of-function (LoF) protein. GoF mutations in NMDAR can lead to an increase in excitatory postsynaptic currents (EPSC), which can disrupt the balance between excitatory and inhibitory discharges in the neuronal network. On the other hand, LoF mutations are thought to diminish inhibitory postsynaptic currents (IPSC) by either reducing the release of GABA from the presynaptic membrane of GABAergic neurons or through a postsynaptic mechanism [64], which can result in epilepsy. However, the exact mechanism through which GoF or LoF mutations in the NMDAR gene cause epilepsy is not fully understood.

To summarize, mutations in NMDAR subunit-encoding genes can cause various forms of epilepsy by altering the structure, function, and regulation of NMDARs. These mutations can affect the expression, trafficking, gating, modulation, and signaling of NMDARs, leading to changes in synaptic transmission, plasticity, and excitotoxicity. The functional consequences of these mutations depend on the location, type, and severity of the amino acid substitution, as well as on the genetic background and environmental factors of the patients. Understanding the molecular mechanisms of these mutations may help to identify novel therapeutic targets and strategies for epilepsy.

3.3.3 NMDAR as a therapeutic target in epilepsy

NMDAR antagonists and agonists have been investigated for their potential therapeutic effects on epilepsy, as well as their possible adverse effects. Some NMDAR antagonists, such as ketamine, memantine, and efavirenz, are FDA-approved drugs for other indications. However, there is evidence from clinical trials and case reports that these drugs can also reduce seizure frequency and severity in some patients with refractory epilepsy or GRIN mutations.

Ketamine, a noncompetitive NMDAR antagonist, has been used in several clinical studies and case reports for the treatment of refractory SE (RSE) and super-refractory SE (SRSE) in adults and children, with variable doses and durations. The reported efficacy rates range from 32 to 74%, depending on the timing of administration and the type and duration of SE. However, the evidence on ketamine is still limited by the retrospective and heterogeneous nature of the data, and more prospective and randomized trials are needed to establish its optimal dose, timing, and duration in SE treatment [73].

Memantine, a noncompetitive, open-channel NMDAR antagonist, has been used as an add-on therapy for a girl with refractory epilepsy due to a de novo GRIN2A mutation. The patient had a reduction in seizure frequency [74]. In another randomized control trail, memantine was effective for nine patients of 27 (33%), compared to only two patients (7%) in the placebo group (P < 0.02). Therefore, memantine appears to be a safe and effective treatment for children with developmental epileptic encephalopathy [75].

Efavirenz is a drug that can increase the level of 24(S)-hydroxycholesterol, a molecule that can restore the function of NMDARs with GluN2A/Grin2a-V685G mutation. This mutation causes a loss of function of NMDA receptors. Efavirenz treatment improved the seizure susceptibility, cortical EEG activity, and synaptic currents of GluN2A/Grin2a-V685G mutant mice. These results suggest that efavirenz may be a potential therapeutic option for epilepsy caused by NMDA receptor dysfunction [76].

In addition, there are other strategies to antagonize NMDARs. These include the use of drugs such as amantadine, magnesium sulfate, remacemide, and MK-801 in epilepsy [77]. These examples illustrate that some NMDAR blockers can reduce seizure frequency and severity in some patients or animal models with refractory epilepsy or GRIN mutations. However, the mechanisms of action, efficacy, safety, and optimal dosing of these drugs for epilepsy are still unclear and need further investigation [77].

In summary, NMDAR antagonists are promising pharmacological agents that can modulate NMDAR activity for epilepsy treatment. They can have beneficial effects on seizure control, neuroprotection, and neuroplasticity in some patients with refractory epilepsy or GRIN mutations. However, they can also have adverse effects on cognition, mood, and neurotoxicity in some patients or at high doses. Therefore, more research is needed to elucidate the mechanisms of action, efficacy, safety, and optimal dosing of these drugs for epilepsy. Moreover, biomarkers and predictors of response and adverse effects are needed to guide the personalized use of these drugs for epilepsy.

3.3.4 NMDAR interaction with other neurotransmitter systems

NMDARs interact with GABA, glutamate, and serotonin to modulate or mediate epileptic phenomena [63, 77]. For example, NMDARs can modulate the function of GABAergic neurons and interneurons, which affects the balance between excitation and inhibition. In epilepsy, glutamate levels are elevated, which can lead to overstimulation of NMDARs and cause neuronal hyperexcitability. NMDARs can also interact with other glutamate receptor subtypes, which can modulate their trafficking, expression, and function. In addition, serotonin can modulate the release of glutamate or GABA from presynaptic terminals, which can affect NMDAR activation and synaptic transmission. In epilepsy, serotonin levels are altered, which may influence NMDAR function and epileptic activity. In conclusion: NMDARs play a crucial role in epilepsy by interacting with other neurotransmitter systems. Therefore, understanding these interactions may provide new insights into the pathophysiology and treatment of epilepsy.

3.3.5 Summary

In conclusion, NMDARs play a crucial role in epilepsy by interacting with other neurotransmitter systems. Evidence from genetic studies, clinical trials, animal models, and pharmacological research has shown that these receptors and their subunits play crucial roles in excitatory neurotransmission, synaptic plasticity, and the balance between excitation and inhibition in the brain. Mutations in NMDAR subunits can cause various forms of epilepsy, and modulating NMDAR activity with antagonists or agonists can reduce seizure frequency and severity in some patients. However, the expression and function of NMDARs can be influenced by many genetic and environmental factors, and they interact with other neurotransmitter systems in ways that may modulate or mediate the epileptic phenomena. Future studies could focus on identifying new NMDAR subunit mutations associated with epilepsy, investigating the effects of different environmental factors on NMDAR expression and function, exploring the interactions of NMDARs with other neurotransmitter systems in more detail, and testing new NMDAR modulators in preclinical models and clinical trials [63, 77].

3.4 NMDARs in schizophrenia

3.4.1 Introduction

Schizophrenia is a severe mental disorder that affects how people think, feel, and behave. It is characterized by symptoms such as hallucinations, delusions, disorganized speech, and cognitive impairment. Schizophrenia affects about 1% of the world’s population and has a significant impact on the quality of life and functioning of the affected individuals and their families [78].

One of the most intriguing aspects of schizophrenia is the involvement of NMDARs in its pathophysiology. NMDARs have been implicated in schizophrenia by various lines of evidence, including genetic studies, pharmacological studies, animal models, and clinical trials. However, the exact role of NMDARs in schizophrenia remains unclear and controversial.

The aim of this section is to review the current state of knowledge on the role of NMDARs in schizophrenia and critically evaluate the strengths and limitations of different approaches to studying this topic. The section will cover the following aspects: (1) genetic studies linking NMDARs and schizophrenia; (2) pharmacological studies examining the effects of drugs that target NMDARs on schizophrenia symptoms; (3) animal models used to investigate the role of NMDARs in schizophrenia; and (4) clinical trials testing NMDAR-targeting treatments for schizophrenia. The section will conclude with a summary of the main points and a reflection on the potential future directions for research on NMDARs and schizophrenia.

3.4.2 Genetics studies

Genetic studies have provided evidence for the involvement of NMDARs in schizophrenia. Several genes that encode NMDAR subunits or modulate their function have been associated with schizophrenia risk or altered expression in patients. For example, GRIN2A, which codes for the GluN2A subunit of the NMDAR, has been firmly linked to schizophrenia by genome-wide association studies (GWAS) and meta-analyses [79, 80, 81]. To a lesser extent of confidence, schizophrenia has also been linked to gene mutations in GRIN2B, GRIN2C, GRIN2D, GRIN3A, and GRIN3B [82, 83, 84]. Moreover, these genes show reduced expression in the postmortem brains of schizophrenia patients, suggesting impaired NMDAR signaling. Other genes that interact with NMDARs, such as dysbindin, neuregulin, DISC1 (Disrupted-in-Schizophrenia 1), D-Amino acid oxidase, and RGS4, have also been implicated in schizophrenia by genetic and functional studies. These genes may affect NMDAR trafficking, localization, or modulation, thereby influencing synaptic plasticity and information processing [78, 85, 86].

3.4.3 Pharmacological studies

The observation that drugs that block NMDARs, such as ketamine and phencyclidine (PCP), can induce psychotic symptoms in healthy individuals and exacerbate them in patients with schizophrenia has led to the NMDAR hypofunction hypothesis of schizophrenia [87]. These drugs act as noncompetitive antagonists at the PCP binding site of the NMDAR, preventing the influx of calcium ions and reducing the synaptic transmission mediated by glutamate. The psychotomimetic effects of these drugs are dose-dependent and correlate with the degree of NMDAR blockade [88].

The pharmacological studies that support the NMDAR hypofunction hypothesis of schizophrenia have used various approaches, such as measuring the effects of NMDAR antagonists on cognitive functions, neurophysiological parameters, neurochemical markers, and brain imaging in healthy volunteers and patients with schizophrenia [89]. These studies have shown that NMDAR antagonists impair working memory, attention, executive functions, and sensory gating, as well as alter the activity of dopaminergic, serotonergic, and cholinergic systems in the brain [90]. Moreover, these drugs affect the regional cerebral blood flow and glucose metabolism, especially in the prefrontal cortex and the hippocampus, regions that are implicated in the pathophysiology of schizophrenia [91].

The pharmacological studies have also examined the effects of agents that modulate NMDAR function, such as glycine, D-serine, sarcosine, and NMDAR co-agonists, on the symptoms and cognitive deficits of schizophrenia [92]. These agents act by enhancing the activity of NMDARs by binding to the glycine site or by increasing the availability of glycine or D-serine. Some of these agents have shown beneficial effects on negative symptoms and cognitive functions in patients with schizophrenia, especially when combined with antipsychotic drugs [87]. However, the results are not consistent across studies and depend on various factors, such as the dose, duration, and type of treatment, as well as the patient population and outcome measures.

The pharmacological studies on NMDARs and schizophrenia have provided valuable insights into the role of glutamatergic neurotransmission in the etiology and treatment of this disorder. However, they also have some limitations, such as the lack of specificity of NMDAR antagonists and co-agonists, the complexity of NMDAR subtypes and interactions with other receptors and channels, and the variability of individual responses to these drugs [93].

3.4.4 Animal models

Animal models are useful tools to study the role of NMDAR hypofunction in schizophrenia, as they can mimic some of the behavioral and neurobiological features of the disorder. There are two main approaches to generate animal models of NMDAR hypofunction: pharmacological and genetic.

Pharmacological models involve the administration of NMDAR antagonists, such as ketamine, phencyclidine (PCP), or dizocilpine (MK-801), to induce transient or chronic NMDAR blockade in rodents or non-human primates [94]. These drugs can produce positive, negative, and cognitive symptoms of schizophrenia in humans, as well as neurochemical, electrophysiological, and neuroplastic changes in the brain that resemble those observed in schizophrenia patients. However, pharmacological models have some limitations, such as the lack of specificity for NMDAR subtypes, the potential involvement of other receptors or channels, and the variability in dose, route, and duration of administration.

Genetic models involve the manipulation of genes that encode NMDAR subunits or modulators, such as NR1, NR2A, NR2B, NR3A, D-serine, glycine transporter 1 (GlyT1), or serine racemase [94, 95]. These genes have been implicated in schizophrenia by human genetic studies or by their role in regulating NMDAR function. Genetic models can provide more selective and stable NMDAR hypofunction than pharmacological models, as well as insights into the developmental and cell-type specific effects of NMDAR dysfunction. However, genetic models also have some drawbacks, such as the potential compensatory mechanisms, the pleiotropic effects of gene manipulation, and the difficulty in replicating human genetic variations in animals [94, 95]. Despite these challenges, animal models of NMDAR hypofunction have contributed to the understanding of the molecular and cellular mechanisms underlying schizophrenia pathophysiology, as well as to the identification of novel therapeutic targets for this disorder.

3.4.5 Clinical trials

Several clinical trials have tested the efficacy and safety of NMDAR-targeting treatments for schizophrenia, with mixed results. Some of the treatments that have been investigated include:

• Glycine modulatory site agonists: These compounds bind to the glycine site on the NMDAR and enhance its function. Examples are glycine, d-serine, d-cycloserine, and sarcosine. These agents have shown some benefits for negative and cognitive symptoms of schizophrenia, but their effects are modest and inconsistent [96, 97, 98].

• Kynurenine pathway inhibitors: These compounds inhibit the enzymes that degrade tryptophan into kynurenic acid, a potent NMDAR antagonist. Examples are laquinimod, roquinimex, and JM6. These agents have shown some neuroprotective and anti-inflammatory effects in animal models of schizophrenia, but their clinical efficacy is unclear [99].

• Cystine-glutamate antiporter modulators: These compounds increase the extracellular levels of cystine, which stimulates the cystine-glutamate antiporter to release more glutamate into the synaptic cleft. This enhances the activation of NMDARs and other glutamate receptors. Examples are N-acetylcysteine (NAC) and l-cysteine. These agents have shown some benefits for negative and cognitive symptoms of schizophrenia, as well as reducing oxidative stress and inflammation [100].

• mGluR modulators: These compounds modulate the activity of mGluRs, which are G-protein coupled receptors that regulate glutamate release and synaptic plasticity. Examples are mGluR2/3 agonists (e.g., LY2140023), mGluR2 positive allosteric modulators (PAMs) (e.g., AZD8529), mGluR5 PAMs (e.g., basimglurant), and mGluR4 PAMs (e.g., VU0155041). These agents have shown some benefits for positive, negative, and cognitive symptoms of schizophrenia, as well as improving neuroplasticity and neurogenesis [3].

The results of these clinical trials suggest that targeting NMDARs may be a promising strategy for treating schizophrenia, but more research is needed to identify the optimal compounds, doses, combinations, and patient populations.

In conclusion, the evidence from genetics studies, pharmacological studies, animal models, and clinical trials suggests that NMDARs play a significant role in schizophrenia. A comprehensive understanding of the role of NMDARs in schizophrenia should consider the complexity and diversity of NMDAR function and regulation; the interactions between NMDARs and other neurotransmitter systems; the developmental and environmental factors that influence NMDAR function; and the individual differences among schizophrenia patients in terms of genetics, symptoms, cognition, and response to treatment.

4. Conclusions

The NMDAR, a critical component of the central nervous system, has been the subject of extensive research due to its pivotal role in neurological function and disease.

In health, the NMDAR plays a crucial role in synaptic plasticity, learning, and memory. It is a key player in the delicate balance of excitation and inhibition that maintains homeostasis in the brain. However, when this balance is disrupted, the consequences can be severe.

In disease states such as AD, autism, epilepsy, and schizophrenia aberrant NMDAR function has been implicated. One of the common themes that emerge from the review of NMDARs in health and disease is the role of genetic mutations in NMDAR subunits or genes that encode proteins that interact with NMDARs. These mutations can affect the expression, trafficking, function, or modulation of NMDARs, resulting in either hypo- or hyperfunction of NMDARs. Depending on the brain region, developmental stage, and synaptic state, these mutations can have different effects on neuronal excitability, synaptic plasticity, network oscillations, and cognitive functions. These mutations can also interact with environmental factors, such as infections, toxins, or stress, to modulate the risk and severity of neurological and psychiatric disorders. Therefore, understanding the molecular mechanisms and clinical implications of these mutations is crucial for developing novel diagnostic and therapeutic strategies for these disorders. However, genetic studies only provide a framework to link the genes and proteins in a network. It requires a concerted effort to understand the disease mechanisms at multiple levels, from molecular to cellular to network to behavior. NMDARs are central to this effort, as they are involved in various aspects of neuronal development and plasticity that are relevant to these disorders.

The development of pharmacological agents targeting the NMDAR offers promising avenues for therapeutic intervention. However, the challenge lies in selectively modulating NMDAR activity without disrupting its physiological functions.

Looking forward, our journey in understanding the NMDAR is far from over. Future research should focus on elucidating the precise mechanisms of NMDAR regulation and exploring novel therapeutic strategies. The development of drugs with improved specificity for NMDAR subtypes or co-agonist sites could potentially minimize side effects and enhance therapeutic efficacy.

In conclusion, the NMDAR remains a fascinating subject of study with significant implications for neuroscience and medicine. As we continue to unravel its mysteries, we move closer to our ultimate goal: improving human health through science.

Acknowledgments

We wish to acknowledge the financial support from the Philadelphia College of Osteopathic Medicine Chief Research and Science Officer (CRSO) Fund.