NMDA Receptors in Neurodevelopmental Disorders: Pathophysiology and Disease Models

Abstract

N-methyl-D-aspartate receptors (NMDARs) are critical components of the mammalian central nervous system, involved in synaptic transmission, plasticity, and neurodevelopment. This review focuses on the structural and functional characteristics of NMDARs, with a particular emphasis on the GRIN2 subunits (GluN2A-D). The diversity of GRIN2 subunits, driven by alternative splicing and genetic variants, significantly impacts receptor function, synaptic localization, and disease manifestation. The temporal and spatial expression of these subunits is essential for typical neural development, with each subunit supporting distinct phases of synaptic formation and plasticity. Disruptions in their developmental regulation are linked to neurodevelopmental disorders, underscoring the importance of understanding these dynamics in NDD pathophysiology. We explore the physiological properties and developmental regulation of these subunits, highlighting their roles in the pathophysiology of various NDDs, including ASD, epilepsy, and schizophrenia. By reviewing current knowledge and experimental models, including mouse models and human-induced pluripotent stem cells (hiPSCs), this article aims to elucidate different approaches through which the intricacies of NMDAR dysfunction in NDDs are currently being explored. The comprehensive understanding of NMDAR subunit composition and their mutations provides a foundation for developing targeted therapeutic strategies to address these complex disorders.

Keywords: ASD—autism spectrum disorder; ATD—amino-terminal domain; CNS—central nervous system; CTD—C-terminal domain; EPSCs—excitatory postsynaptic currents; LBD—ligand-binding domain; NDD—neurodevelopmental disorder; SCZ—schizophrenia; TMD—transmembrane domain; mEPSCs—miniature excitatory postsynaptic currents.

Figure 1

Structural composition of the NMDA receptors; (A) Graphical representation of the subunit composition of NMDA receptors. Each monomer of the receptor contains four functional domains, the amino-terminal domain (ATD), ligand-binding domain (LBD), transmembrane domain (TMD), and C-terminal domain. (B) Ion channel activity of AMPA and Kainate receptors (members of iGluR family). (C) Crystal structure displaying 3D conformation of the heterotetramer NMDA receptor containing GluN1 and GluN2B subunits (PDB: 6WHX); GluN3 subunits also form functional receptors with GluN1 subunit. (D) The proposed GluD2 ion channel activity mechanism is through interaction with pre-synaptic linker proteins; the representation is adapted from Carillo et al. SciAdv, 2021 [12].

Figure 2

Activation of the “coincidence receptor”. (A) Initial interaction of AMPAR (blue) and NMDAR (Purple) with pre-synaptic glutamate; the receptors are inactive at the resting membrane potential (−70 mV); upon interaction with the presynaptic glutamate, NMDAR remains inactive still since Mg²⁺ is blocking the ion channel pore, while AMPA receptor allows in-flow of Na⁺ ions initiating the depolarization of the postsynaptic membrane. (B) Depolarization of the postsynaptic membrane facilitates the release of Mg²⁺ ions from the NMAR ion channel pore, activating the receptor. (C) Activation of NMDAR and repetitive presynaptic glutamate release leads to increased in-flow of Ca²⁺ and Na⁺ ions into the postsynaptic neurons. The representation is adapted from Sprengel et al. 2022 [43]. This fundamental characteristic of NMDA receptors is disrupted in NDDs, where disease-associated variants are distributed across various domains of the GRIN proteins within NMDARs, impacting multiple physiological properties and leading to either receptor hypofunction or hyperfunction. Missense variants within the transmembrane helix may alter NMDAR surface expression and modify receptor sensitivity to endogenous agonists and inhibitors [2]. Likewise, variants in the extracellular ATD and LBD regions are often linked to receptor loss of function in DD/ID patients [27]. These alterations impact the neurons’ downstream calcium signaling, which affects long-term potentiation and synaptic plasticity in NDD patients.

Figure 3

Schematics of the eight functional isoforms of the GluN1 subunit resulting from alternative RNA splicing. Color coding represents the splicing sites at exons 5, 21, and 22 or 22’, denoted as cassettes N1, C1, C2, and C2’, respectively. N1 cassette site is in the amino-terminal domain, whereas C1, C2, and C2’ sites are in the C-terminal domain. The schematics are adapted from Li et al., PNAS, 2021 [45].

Figure 4

Modeling neurodevelopmental disorders using patient-derived iPSCs. The illustration depicts the general protocol for modelling human neurodevelopmental disorders (NDDs) using patient-derived induced pluripotent stem cells (iPSCs). The iPSCs can be derived from the fibroblasts, PBMCs, or lymphoblastoid cell lines (LCLs) of the patients as well as the healthy subjects (Isogenic controls). These iPSCs provide a platform to study the disease biology and explore novel therapeutic strategies through various techniques that can lead to the development of personalized and more effective treatment options for patients suffering with different NDDs. The schematics of the figure are adapted from Liu et al., Development, 2018 [148]. iPSC-derived neurons from ASD patients frequently exhibit reduced synaptic activity and altered excitatory/inhibitory signaling balance, both of which are critical to understanding synaptic dysfunctions characteristic of ASD [149,150,151,152]. With their ability to model complex synaptic processes, aberrant connectivity, and neurotransmitter imbalances, hiPSC-derived neurons serve as invaluable tools for dissecting the molecular mechanisms underlying neuropsychiatric disorders such as ASD, schizophrenia, bipolar disorder, and ID [153,154,155,156,157]. Integration-free methods for iPSC generation avoids genomic integration of vectors, thereby preserving genetic integrity and reducing tumorigenic risks [158]. Consequently, several groups have successfully adopted this method for modeling different neurological disorders. For instance, iPSCs were generated by reprogramming fibroblasts derived from a Phelan-McDermid syndrome (PMS) patient, harboring an insertion mutation in SHANK3 (C.3679insG) [159]. The iPSCs were observed to express the pluripotency markers, differentiate into the three germ layers, retain the disease-causing mutation, and display normal karyotypes. Therefore, this technology allows researchers to explore the functional properties of cellular factors involved in the pathology of NDDs, which can be translated into a patient specific therapeutic intervention.