Targeting N-methyl-D-aspartate 2B receptor in painful diabetic neuropathy – mechanisms, challenges, and emerging therapeutics

Abstract

A notable number of patients with diabetes suffer from painful diabetic neuropathy (PDN), which is a debilitating complication of diabetes mellitus. Prolonged hyperglycaemia and metabolic dysregulation lead to PDN, a condition characterised by chronic pain, sensory dysfunction, and reduced quality of life. Although various treatment options are available, clinical management is challenging due to the complex and multifactorial nature of PDN pathophysiology. N-methyl-D-aspartate receptors (NMDARs), particularly the NR2B subtype (NMDAR-2B), have emerged as a key player in the pathophysiology of chronic pain states, including PDN. This review highlights the mechanistic NMDAR-2B involvement in the pathophysiology of PDN, focusing on its upregulation role in pain-processing regions, interaction with inflammatory mediators, glia-derived mediators, and oxidative stress mechanisms. Advancements in targeting NMDAR-2B as a mechanistically driven approach to PDN management also offer potential in enhancing the therapeutic efficacy of NMDAR-2B. Consequently, this review provides a novel perspective on understanding the role of NMDAR-2B in PDN for the future development of effective treatment strategies for managing the condition.

Key Words: Painful diabetic neuropathy; N-methyl-D-aspartate receptor 2B; Chronic pain; Oxidative stress; Inflammation; Nociception; Glia

Core Tip: The N-methyl-D-aspartate receptor 2B (NMDAR-2B) plays a pivotal role in the pathophysiology of painful diabetic neuropathy (DN). In this review, the contribution of NMDAR-2B to oxidative stress, neuroinflammation, and central sensitisation was underscored. Experimental and clinical data linking NMDAR-2B upregulation with chronic pain were also discussed. Furthermore, emerging therapeutics targeting the receptor were evaluated, while outlining the translational challenges. Comprehending the molecular interplay of NMDAR-2B with glial cells and signalling pathways might also provide valuable insight for developing multi-targeted or precision-based therapies to improve painful DN management.

INTRODUCTION

Diabetes-associated peripheral neuropathy (PN) is commonly referred to as diabetic PN (DPN). The condition is the most frequent long-term complication of diabetes mellitus (DM) and is detrimental to the overall health of individuals suffering from DM. Various clinical symptoms are associated with DPN, including pain and sensory loss. Different types of DPN also exhibit as motor, sensory, focal, or multifocal conditions[1]. Increased mortality risk and irreversible clinical consequences, such as diabetic foot ulcers and amputation, have been associated with individuals suffering from DPN[1,2]. The prevalence of DPN in patients with diabetes varies from 7% to 75%, with a higher incidence revealed by recent data[3]. Meanwhile, positive symptoms, including DPN, occur in one-third of the population with neuropathy, exhibiting allodynia, paraesthesia, and spontaneous pain. Painful diabetes-associated PN, or DPN, is a significantly prevalent and distressing complication of DM. The sensory abnormality syndrome develops from inadequately controlled types 1 and 2 DM. Distressing tingling, shooting, lancinating, and sharp or electric shock sensations characterise DPN. At night and sleep disturbances can also exacerbate the condition[4]. Furthermore, DPN can lead to poor wound healing of foot ulcers, which can ultimately result in amputation, which is a serious consequence of diabetic foot disease[3].

Individuals suffering from DPN have dramatically reduced physical activity and quality of life, and increased healthcare demand. The intricate and multifactorial aetiology of DPN presents obstacles to its effective treatment despite advancements in diabetes management. Moreover, patients with DPN and their families are faced with raised financial burdens due to them requiring various painkiller medications and regular health visits to healthcare professionals and diminished capacity to work or perform daily routines[3,5]. Different underlying mechanisms, including inflammation, oxidative stress, metabolic dysregulation, and neuronal hyperexcitability, cause DPN. Chronic hyperglycaemia sets off a series of metabolic modifications in the peripheral and central nervous systems, impairing neuronal function and leading to maladaptive development. Studies also suggested that central sensitisation, the process by which the brain and spinal cord amplify pain signals, is essential for maintaining chronic pain in DPN[6].

Current pharmacological DPN treatments, such as pregabalin, duloxetine, and tricyclic antidepressants, only provide pain relief and are frequently limited by their side effects including sedation, dizziness, and cognitive impairment. Moreover, the drugs primarily manage symptoms without addressing the underlying drivers of chronic pain. The mechanisms of action of the treatments also do not adequately target central sensitisation, which is a key contributor to persistent pain in DPN. Therefore, more mechanism-based interventions are necessary to bridge the significant therapeutic gap. Targeting the NR2B subunit of N-methyl-D-aspartate receptors (NMDARs), namely NMDAR-2B, directly modulates excitatory neurotransmission and maladaptive plasticity in central pain pathways, presenting a potential approach[7]. Therefore, such interventions may reduce pain perception and modify disease progression in DPN by interfering with the NMDAR-2B-driven pain signal amplification.

This review determined the molecular and functional involvement of the NMDAR-2B in the pathophysiology of DPN. The role of NMDAR-2B in the development of hyperalgesia and allodynia, and the interaction of the NMDAR subunit with oxidative stress and inflammatory mediators were discussed. Moreover, the potential of NMDAR-2B-targeted therapies for managing DPN was also evaluated.

NMDARS AND THEIR SUBUNITS: A FOCUS ON NMDAR-2B

A subtype of non-selective ionotropic glutamate receptors, NMDARs, are vital in excitatory synaptic transmission, plasticity, learning, and memory in the central nervous system (CNS). The receptors possess significant calcium (Ca²⁺) permeability and are abundantly expressed throughout the cerebral cortex, including the hippocampus[8]. Generally, the ionotropic glutamate receptor comprises five subunits: NR1, NR2, such as NMDAR-2A to NMDAR-2D subtypes, and NR3, including NR3A and NR3B[4]. As heterotetramers, NMDARs are made up of two glutamate-binding NR1 and two glycine-binding NR2 subunits[9,10]. Nonetheless, the NMDARs are typically di- (such as NR1/NMDAR-2A and NR1/NMDAR-2B) and tri-heteromers (including NR1/NR2/NR3B)[10,11]. The NR2 subunits are randomly selected from four isoforms, NMDAR-2A to NMDAR-2B, and are expressed in distinct regions of the CNS due to their special functions. Meanwhile, the NR1 subunits are extensively dispersed throughout the brain regions. NMDAR-2A is closely related to rapid pain transmission in the CNS[10]. The receptors are also reportedly linked with neuroprotective effects through cell survival mechanisms activation[12]. Conversely, NMDAR-2B is essential in maintaining and enhancing chronic pain and is frequently correlated with increased excitability and pain sensitivity[10]. Furthermore, NMDAR-2B activation is closely associated with neurodegenerative diseases, as it induces cell death signalling[12]. Glutamate binding to NMDARs activates the CNS, which opens Ca²⁺ channels, transmits impulses, and promotes neuronal plasticity. Nonetheless, NMDAR overexpression can result in neuronal hyperresponsiveness and hypersensitivity, leading to central sensitisation[10]. Primarily expressed in the postsynaptic density, NMDARs enhance the sensitivity of spinal cord dorsal horn sensory neurons to innocuous and noxious stimuli in the context of pain, facilitating hyperalgesia and allodynia development[10].

NMDAR-2B, also known as the NR2B subunit, is widely reported for its notable involvement in neuropathic pain development in various in vitro and in vivo models among the isoforms of NR2 receptors. Four similarly structured subunits assemble NMDAR-2B, where the extracellular region has a ligand-binding domain or agonist-binding domain. An extracellular N-terminal domain, or the amino-terminal domain, also forms the extracellular region of NMDAR-2B, which is essential for the assembly of the receptors, their trafficking, and regulation. Each transmembrane region of NMDAR-2B consists of three transmembrane domains, M1, M3, and M4, and a re-entrant loop, M2, that forms the pore of the receptor. Based on Figure 1, the intracellular C-terminal domain of the NMDAR-2B is also longer than the NR1 subunit in terms of length and amino acid sequence (e.g., 650 amino acids)[13,14]. Moreover, NMDAR-2B has a higher affinity for glutamate and a longer channel opening duration than the NR2A subunit, contributing to its swifter deactivation kinetics and potentially Ca²⁺-sensitive desensitisation attributes[11]. The unique biophysical profiles also render the NMDAR-2B more susceptible to sustained Ca²⁺ influx upon activation, which amplifies excitatory synaptic transmission and excitotoxicity susceptibility.

Figure 1 Schematic representation of an N-methyl-D-aspartate receptor-2B structure and its pharmacological regulation sites. An N-methyl-D-aspartate receptor-2B consists of four subunits with three transmembrane domains (M1, M3, and M4) and a pore (M2) that allows ion influx. The receptor is also characterised by an extracellular amino-terminal, an extracellular ligand-binding, four transmembrane, and an intracellular C-terminal domains. ATD: Amino-terminal domain; Ca²⁺: Calcium ion; CTD: C-terminal domain; LBD: Ligand-binding domain; Mg²⁺: Magnesium ion; TMD: Transmembrane domain. Adapted from Gallo et al[14] (Supplementary material).

NMDAR-2B-mediated mechanisms in PDN

Most patients suffering from DPN complain of hyperalgesia and allodynic sensations, considering that these pain responses relate to altered nociceptive signalling pathways in the spinal cord dorsal horn. The increased spinal cord neuron activities may lead to central sensitisation or hyperactivity and neuroplasticity. Several reports have also demonstrated that NMDAR activation, particularly NMDAR-2B, is involved in developing hyperalgesia in various neuropathic pain conditions. The condition is correlated with elevated glutamate release from primary afferent terminals, resulting in stronger and more frequent excitatory postsynaptic currents in lamina I-II of the spinal cord dorsal horn. Evidence also points to the role of DPN in increasing cyclic adenosine monophosphate response element-binding protein signalling, which directly controls NMDAR-2B activation. Therefore, increased glutamate release and NMDAR-2B expression can cause spinal cord hyperactivity.

Gamma-aminobutyric acid B (GABAB) receptors appear to be downregulated in the spinal cord in DN. Activation of GABAB receptors directly inhibits voltage-gated Ca²⁺, which lowers NMDAR-2B activities[15]. Considering the availability of GABAB receptors pre- and post-synaptically in the spinal cord dorsal horn, they inhibit Ca²⁺ channels or interactions with downstream release machinery and the opening of inwardly rectifying potassium ion (K⁺) channels. The processes reduce the neuronal excitability by depleting glutamate release. Furthermore, the magnesium blockade of the NMDARs is strengthened when these K⁺ channels are opened, generating slow inhibitory postsynaptic potentials. The GABAB receptors also hinder Ca²⁺ signals in dendrites and spines, downregulating the mRNA and protein expression of NMDAR-2B, which diminish their activities[1].

Increasing preclinical and clinical studies have demonstrated the significant involvement of NMDAR-2B in DPN pathogenesis[15,16]. Functional imaging reports applying functional magnetic resonance imaging in patients with DPN have also documented thalamic hyperexcitability, which is a region rich in NMDAR-2B expression, correlating with pain severity[15,16]. The thalamus has been linked to evoked and spontaneous pain in chronic conditions. For instance, a clinical trial conducted by Selvarajah et al[17] revealed that patients with type 1 DPN demonstrated significantly greater thalamic vascularity than patients with painless DN. The outcomes might be attributable to the intense pain responses felt by patients with DPN. Studies have indirectly linked NMDAR-2B activation to elevated oxidative stress markers in patients with DN[18,19], suggesting potential NMDAR-2B-mediated excitotoxicity occurrence in clinical settings. Several inflammatory signalling pathways, including kinesin family member 17, mammalian Lin-7, and amyloid beta precursor protein binding family A member 1, have also been associated with the trafficking or activation of NMDAR-2B in patients with other neurodegenerative conditions[20,21]. The data reinforce the translational potential of NMDAR-2B-targeted therapies in addressing central sensitisation and pain amplification in DPN.

The considerable increase in pain behaviour responses, such as formalin-induced pain, tactile allodynia, and thermal hyperalgesia, was notably correlated with the raised spinal expression of phosphorylated NMDAR-2B (p-NMDAR-2B) and total NMDAR-2B activation in rats with DPN[4,22]. Painless DN rats with less painful sensation also exhibited reduced expression of the NMDAR-2B markers in the spinal cord region[12]. In another study, Suo et al[23] assessed the role of spinal NMDAR-2B in modulating Fyn kinase in a pre-DN state by implementing a high-fat diet-induced mouse model. According to the findings, the high-fat diet increased NMDAR-2B phosphorylation and expression (Tyr1472) in spinal cord neurons[23]. Moreover, the condition could be related to the direct interaction of Fyn kinase (Src-family kinase) with NMDAR-2B. Fyn-knockout mice also indicated thermal hypoalgesia development. Nevertheless, tactile allodynia was not observed. The observations suggested that Fyn is an essential component for NMDAR-2B functional activation rather than expression.

Lu et al[24] reported significant kalirin-7 (Kal7)/p-NMDAR-2B/postsynaptic density protein 95 (PSD-95) signalling pathway activation following DPN development in their animal model. The related mechanisms were further explored by Li et al[25], where involvement of the Janus kinase 2 (JAK2)/signal transducer of activation 3 (STAT3)/caveolin-1 (CAV-1)/NMDAR-2B signalling pathway in the pathogenesis of DPN was revealed. Similar results were documented by Shiers et al[26] and Peng et al[27] in their DN rat models. Recently, the clinical study by Shiers et al[26] reported the notable expression of peripherin-positive axons and Nageotte nodules (including clusters of non-neuronal cells that form after sensory neuron death). The report also recorded significant ligand-receptor interactions in the dorsal root ganglion neurons linking to the mechanisms of DN development. Table 1 summarises the overall NMDAR-2B-related preclinical and clinical findings[4,12,21,23-26,28-30].

Table 1 Clinical and preclinical findings of N-methyl-D-aspartate receptor-2B involvement in the pathogenesis of diabetic neuropathy.

Ref.	Research	Clinical/preclinical model type	Clinical/preclinical model duration	Sample	Key outcome
Ismail et al[4], 2019	Preclinical	STZ-induced type 1 DPN rat	23 days	Rat spinal cord tissue	Increased p-NMDAR-2B and t-NMDAR-2B protein expression in the spinal cord of DPN rats with aberrant thermal hyperalgesia and formalin-induced chemical hyperalgesia responses, which were significantly suppressed with a seven-day intrathecal injection of ifenprodil (0.5 µg/µL)
Ismail et al[12], 2022	Preclinical	STZ-induced type 1 DPN rat	23 days	Rat spinal cord tissue	Painless DN rats exhibited notably less formalin-induced flinching and licking, which was linked to reduced spinal phosphorylated and total NMDAR-2B protein levels compared to DPN and control groups
Uniyal et al[21], 2022	Preclinical	STZ-induced type 1 DPN rat	23 days	Rat spinal cord tissue	A considerable increment in formalin-induced flinching during Phase 1 (represented by peripheral nociceptive changes) and early and late Phase 2 (represented by central nociceptive changes), which were associated with increased spinal p-NMDAR-2B and t-NMDAR-2B expression; minocycline (a selective microglia OX-42 inhibitor) substantially suppressed the responses and phosphorylated and total NMDAR-2B expression; nevertheless, the substance did not exert any effects on thermal hyperalgesia
Suo et al[23], 2016	Preclinical	High-fat diet-induced prediabetic mice	24 weeks	Mouse spinal cord tissue	Tactile allodynia and thermal hypoalgesia were developed in pre-diabetic wild-type mice, which correlated with elevated NMDAR-2B expression and activation and Fyn-NMDAR-2B interaction in the spinal cord; the ro 25-6981 (selective NMDAR-2B, i.t.) also demonstrated alleviated tactile allodynia but not thermal hypoalgesia in a dose-dependent manner
Lu et al[24], 2020	Preclinical	Type 2 high-fat diet (HFD)-induced DPN rat	8 weeks HFD	Rat spinal cord tissue	Levels of kalirin-7, p-NMDAR-2B, PSD-95, and PSD-95-NMDAR-2B coupling were significantly improved in kalirin-knockout mice with type-2 DPN with diminished mechanical allodynia and thermal hyperalgesia, suggesting that spinally expressed kalirin-7 could contribute to type-2 DPN by regulating PSD-95/NMDAR-2B interaction-dependent NMDAR-2B phosphorylation in the spinal cord
Li et al[25], 2019	Preclinical	STZ-induced type 2 DPN rat (in vivo); BV2 immortalised murine microglia cell line (in-vitro)	14 days; 24 hours incubation	Rat spinal cord tissue; BV2 immortalised murine microglia cell line	The p-JAK2, p-STAT3, total-CAV-1, and p-NMDAR-2B were upregulated in the spinal cord dorsal horn of the DPN rat with enhanced mechanical and thermal hyperalgesia development; intrathecal injection of the JAK2 inhibitor significantly suppressed the expression of the markers and thereby reduced pain responses; in vitro, a high glucose environment markedly induced activation of p-STAT3 in microglia and upregulated p-CAV-1 and p-NMDAR-2B in neurons
Shiers et al[26], 2024; Nakazawa et al[28], 2001	Clinical	-	-	Human DRG tissue	Histological and spatial RNA analyses of human DRGs from DPN donors showed damaged peripherin-positive axons and Nageotte nodules (support and immune cell clusters); ligand-receptor interactions in DRG neurons linked to DN mechanisms were also identified
Shiers et al[26], 2024	Preclinical	STZ-induced diabetic neuropathy rats	3 weeks	Rat spinal cord tissue	The DN rats exhibited lower paw withdrawal threshold, displaying notably higher expressions of NMDAR-2B and p-CREB in the spinal cord dorsal horn, which were abolished by intrathecal injections of baclofen (specific GABAB receptor antagonist), suggesting that the activation of spinal GABAB receptors activation normalises NMDAR-2B expression in DN
Fajrin et al[29], 2020	Preclinical	STZ-induced DPN mice model	28 days	Mouse sciatic nerve and spinal cord tissue	Spinal NMDAR-2B and TRPV1 mRNA expression was significantly downregulated in the mice treated with 6-shogaol, with improved thermal hyperalgesia, allodynia, and pain-induced mechanical pressure
Liu et al[30], 2014	Preclinical	STZ-induced diabetic neuropathic pain rats	3 weeks	Rat spinal cord tissue	Baclofen substantially improved paw withdrawal threshold and thermal withdrawal latency, with significant downregulation of mRNA and reduced protein expression of spinal NMDAR-2B and p-CREB in DPN rats; the data postulated that GABAB activation by baclofen might attenuate diabetic neuropathic pain partly via the downregulation of p-CREB and NMDAR-2B expression

CAV-1: Caveolin-1; DN: Diabetic neuropathy; DRG: Dorsal root ganglia; GABAB: Gamma-amino butyric acid B receptor; NMDAR-2B: N-methyl-D-aspartate-2B; OX-42: CD11b protein; p-CAV-1: Phosphorylated caveolin-1; p-CREB: Phosphorylated cyclic AMP-responsive element binding protein; p-JAK2: Phosphorylated Janus kinase 2; p-NMDAR-2B: Phosphorylated N-methyl-D-aspartate-2B receptor; PDN: Painful diabetic neuropathy; PSD-95: Postsynaptic density-95; STAT3: Signal transducer and activator of transcription 3; t-NMDAR-2B: Total N-methyl-D-aspartate-2B receptor; TRPV1: Transient receptor potential vanilloid-1.

Interaction of NMDAR-2B with inflammatory and oxidative stress pathways

The interaction between neuronal and non-neuronal mechanisms occurs during the pathogenesis of DN through the activation of specific signalling pathways that connect the mechanisms. Preclinical data suggest that aberrant behaviour responses characterised by elevated tactile allodynia development but not thermal hyperalgesia could be associated with increased p-NMDAR-2B and total NMDAR-2B protein levels in the spinal cord of DPN rats[4]. Meanwhile, the expression of proteins was significantly lower in the spinal cord of the painless variant of DN and was linked with hyporesponsiveness. The phenomenon could be associated with the reduced phosphorylation and total expression of NMDAR-2B detected in the ipsilateral and contralateral regions of the spinal cord dorsal horn of the rats[12].

Various possibilities have been postulated following NMDAR-2B activation in the pathogenesis of DN. For instance, Lu et al[24] hypothesized that increased activation of Kal7 in the spinal cord may regulate PSD-95 expression, which anchors NMDAR-2B to intracellular signalling proteins at the neuronal synaptic level. Kal7 is a multifunctional guanine nucleotide exchange factor of Rho GTPase, which possesses a C-terminal PDZ-binding motif regulating NMDAR-2B and PSD-95 expression and their binding. Elevated Kal7 expression in the spinal cord 2 weeks post-streptozotocin injection in the rats is also closely linked to mechanical allodynia and thermal hyperalgesia development in rats treated with dinitrophenol (DNP)[27]. Meanwhile, in the type 2-induced DNP rat model with aberrant mechanical allodynia and thermal hyperalgesia development, significant increases of Kal7, phosphorylated NMDAR-2B, and PSD-95 levels, and PSD-95-NMDAR-2B coupling have also been detected in the spinal cord. Nonetheless, the level of markers was significantly diminished in the Kal7-knockout, which correlated with reduced behavioural responses, potentially linked to reduced central sensitisation[24].

The Kal7-NMDAR-2B activation mechanism might involve PSD-95 activation, which in turn attracts Fyn, a prominent kinase that phosphorylates Tyr1472 of NMDAR-2B, activating the receptor[28]. Moreover, the pathway facilitates the development of mature dendritic spines by binding ephrin to its receptor. The process then activates Ras-related C3 botulinum toxin substrate 1 and upregulates B cell lymphoma-2 (Bcl-2) and protects against apoptosis and mediates neuroprotection by downregulating Bcl-2-associated X protein[23]. Abnormally regulated specific signal transduction linked to NMDAR-2B activation contributes to the pathogenesis of DPN. The JAK/STAT3 pathway is a vital signalling pathway that boasts a classical intracellular signal transduction pathway involving various pathophysiological processes, including cell proliferation and differentiation, inflammation, and pain transmission[25,31]. The signal transduction pathway is frequently dysregulated and overactive during the DPN state, leading to NMDAR-2B upregulation. According to Li et al[25], JAK/STAT signal transduction activation occurs due to ion channel dysfunction, which predominantly drives the pathogenesis of painful DN in the hippocampal neurons. The activated JAK/STAT3 signalling transduction can also activate the CAV-1 pathway, an essential gene targeting STAT3, which is critical in regulating neuronal plasticity and receptor transport. The JAK2/STAT3/CAV-1 combination activation increases NMDAR-2B activation and DPN maintenance.

Prolonged hyperglycaemia leads to increased sorbitol in tissues and nicotinamide adenine dinucleotide (NAD)+/nicotinamide adenine diphosphate hydride ratio. The process suppresses glyceraldehyde 3-phosphate dehydrogenase and causes diacylglycerol accumulation, which triggers advanced glycation end products and reactive oxygen species (ROS) formation by eliciting NAD phosphate (NADPH) oxidase and protein kinase C[32]. Subsequently, the activated protein kinase C may phosphorylate NMDAR-2B, enhancing its activities that eventually result in central sensitisation and pain hyperresponsiveness. Besides signalling pathways, NMDAR-2B also interacts with other non-neuronal cells to facilitate central sensitisation and chronic pain development processes. In neuropathic pain development, microglia and astrocytes are the glial cells involved. Commonly referred to as “pathological effectors and amplifiers”, microglia are the first line of immunological defence in the CNS. The cells also regularly monitor and inspect their surrounding territory utilising significantly motile processes[33].

Prolonged hyperglycaemia and systemic inflammation cause microglia to undergo morphological transformation from a resting to an activated state. During the early phases of DN, the activated microglia polarise towards the M1 phenotype and release a range of pro-inflammatory cytokines, including interleukin 1 beta (IL-1), tumor necrosis factor-alpha, IL-6, chemokines, and ROS. A primary source of ROS secreted by microglia is NADPH oxidase, which facilitates molecular oxygen to superoxide anion conversions, initiates oxidative stress cascades that potentially damage the neurons in the peripheral nerve stimulator or CNS, and intensifies glial activation. Redox-sensitive transcription factors, such as activator protein-1 and nuclear factor kappa B, can be activated by microglia-produced ROS. The substances increase the expression of other pro-inflammatory mediators and enzymes, including inducible nitric oxide. The combined superoxide and the consequent nitric oxide actions may generate the extremely cytotoxic oxidant peroxynitrite, which is significantly linked to pain hyperresponsiveness and neuronal damage. Moreover, adenosine triphosphate binding to the P2X receptor expressed on microglia could cause the release of brain-derived neurotrophic factor. The neuromodulator interacts with two types of receptors, namely neurotrophin receptor p75 and tyrosine kinase receptor B, which are low-and high-affinity receptors, respectively, to aid signal transduction and mediate neuronal survival. Among related functional and morphological modifications are transcriptional activation, migration, microgliosis, and upregulation of inflammatory genes that eventually lead to the onset of neuropathic pain[33,34].

Astrocytes notably contribute to the latter phase of neuropathic pain. Astrocytes maintain pathological pain by increasing cell hypertrophy and astrocytic marker glial fibrillary acidic protein[25]. Although typically associated with metabolic support and neurotransmitter regulation, astrocyte is also involved in DN development. Moreover, reactive astrocytes release pro-inflammatory cytokines and the neurotransmitter glutamate and upregulate glial fibrillary acidic protein, leading to a neuroinflammatory milieu. Consequently, excess extracellular glutamate may increase NMDAR activation and exacerbate neuronal excitotoxicity, especially in NMDAR-2B, which is susceptible to the redox changes. Another isoform of NADPH oxidase, NADPH oxidase 4, can be enhanced by astrocyte production. The glial cells are also insufficiently neutralised due to compromised antioxidant mechanisms, including glutathione peroxidase and superoxide dismutase, as demonstrated in patients with diabetes. Persistent oxidative stress arises from the redox imbalance, upsetting neuronal homeostasis and exacerbating chronic pain development. Therefore, inflammatory cytokines released from glial cells, such as IL-6, could activate NMDAR and trafficking via JAK/STAT3 signalling, resulting in neurodegenerative lesions and central sensitisation during DPN[25,31]. Moreover, the glial interplay could activate oxidative stress signalling pathways in pain-processing regions, such as the spinal cord dorsal horn and brain thalamus, which eventually contribute to central sensitisation development in DPN. Figures 2 and 3 demonstrate the overall mechanisms involving NMDAR-2B and various signalling pathways.

Figure 2 Possible mechanisms leading to N-methyl-D-aspartate receptor-2B activation in diabetic peripheral neuropathy. Hyperglycaemia leads to the accumulation of glycolytic precursor diacylglycerol. The glycolytic precursor stimulates advanced glycation end product and reactive oxygen species formation by activating protein kinase C (PKC) and nicotinamide adenine dinucleotide phosphate (NADPH) oxidase. The activated PKC then stimulates N-methyl-D-aspartate receptor-2B (NMDAR-2B) activation and central sensitisation. Brain-derived neurotrophic factor (BDNF) may also activate NMDAR-2B through tyrosine kinase B (TrkB) receptor binding, which activates various phospholipase C (PLC)/diacylglycerol (DAG)/PKC and calmodulin-dependent protein kinase (CaMK) signaling pathways, leading to the development of neuropathy and transcription of various pro-nociceptive mediators, enhancing NMDAR-2B activation, resulting in central sensitisation and chronic pain. BAX: B-cell lymphoma-2 associated X protein; Bcl-2: B-cell lymphoma-2; Ca²⁺: Calcium ion; CAV-1: Caveolin-1; EphB: Ephrin B; JAK: Janus kinase; PaK: P21-activated kinase; PSD-95: Postsynaptic density protein 95; ROS: Reactive oxygen species; STAT3: Signal transducer and activator of transcription 3.

Figure 3 Potential increased release of inflammatory markers from activated glial cells due to prolonged hyperglycaemia. The binding of mediators to their receptors activates Janus kinase 2 (JAK2) and phosphorylates signal transducer and activator of transcription 3 (STAT3). The processes upregulate caveolin-1 (CAV-1) expression and promote N-methyl-D-aspartate receptor-2B (NMDAR-2B) trafficking and activation, increasing calcium ion (Ca²⁺) influx, central sensitisation, and aberrant pain signaling pathway, which leads to the development of diabetic peripheral neuropathy. CaMKII: Calcium-calmodulin-dependent protein kinase II; CREB: Cyclic adenosine monophosphate response element binding protein; IL: Interleukin; SOCS3: Suppressor of cytokine signaling 3; TGF-β: Transforming growth factor beta; TNF-α: Tumor necrosis factor alpha.

Crosstalk between NMDAR-2B and comorbidities: Implications for PDN therapeutics

NMDAR-2B is expressed in pain-processing regions, such as the spinal cord dorsal horn and brain thalamus, and limbic structures, including the anterior cingulate cortex (ACC) and hippocampus[13], which are known for their critical processing of emotion and cognitive aspects of pain. Therefore, comorbid conditions, such as depression, anxiety, and cardiovascular disease, which typically coexist with DPN, can profoundly influence pain perception, disease progression, and treatment outcomes[35-38]. The comorbidities also have overlapping pathophysiological pathways with DPN, including glutamatergic dysregulation, oxidative stress, and neuroinflammation[37]. Magnetic resonance imaging scans of the brains of individuals with type 1 DM have demonstrated higher glutamate-glutamine-GABA levels in the prefrontal cortex region than healthy individuals. The elevated concentrations also correlate with mild depressive symptoms, indicating that hyperglycaemia significantly impacts cognition and mood brain function[36]. Nevertheless, animal studies have documented considerably lowered diabetes susceptibility induced by a high-fat diet after birth, utilising memantine (e.g., NMDAR antagonist)[39]. These findings indicate a notable association of NMDAR-2B involvement in diabetic-induced comorbidities despite limited related clinical evidence. The comorbidities influence NMDAR-2B activity, and dysregulated NMDAR-2B contributes to worsening comorbidity states[39]. Consequently, designing therapeutic strategies to manage DPN is challenging.

Upregulation of NMDAR-2B in limbic systems, particularly the ACC, has been associated with emotional and affective pain and mood regulation components. According to preclinical reports, inflammation-induced persistent pain promotes NMDAR-2B expression in the ACC, whereas inhibition of NMDAR-2B in this area alleviates pain behaviours. Although NMDAR-2B antagonism can decrease nociceptive responses, ACC inhibition can also impair memory[40,41]. Moreover, cardiovascular disease might influence drug pharmacokinetics and blood-brain barrier (BBB) integrity, influencing therapeutic delivery and response[39]. Therefore, better clinical outcomes with fewer side effects could be achieved through a region-specific targeting approach or the employment of peripherally restricted NMDAR-2B antagonists.

Limitations and challenges in translational research involving NMDAR-2B

Substantial preclinical evidence supports the role of NMDAR-2B in mediating central sensitisation and neuropathic pain, particularly in DPN. In streptozotocin-induced diabetic rodent models, intrathecal ifenprodil significantly diminished tactile allodynia, thermal hyperalgesia, and p-NMDAR-2B expression in the spinal cord[4,12,22]. Nevertheless, translating such compelling preclinical results to human clinical efficacy has been hindered by limitations and translational challenges. A predominant obstacle is the species-specific differences in NMDAR-2B expression, distribution, and regulatory mechanisms between rodents and humans. Extrapolating the findings from animal models to human neuropathic pain states is complicated by interspecies variations[34,42]. The NMDAR-2B is abundantly expressed in pain-processing regions, such as the spinal cord dorsal horn, thalamus, and prefrontal cortex. Upregulated NMDAR-2B in response to a peripheral nerve injury or metabolic insult has also been consistently linked to central sensitisation development and chronic pain behaviour. Nonetheless, the spatiotemporal dynamics of NMDAR-2B expression in humans are poorly characterised. The regional distribution of the NMDAR subtype may also differ significantly[43].

In studies involving rodents, sustained NMDAR-2B elevation was observed in chronic pain states, whereas human postmortem and imaging data on the NMDAR-2B regulation in neuropathic pain states are limited[42]. Moreover, intracellular signalling pathways and post-translational modifications governing NMDAR-2B functions, including phosphorylation, ubiquitination, and receptor trafficking, may exhibit species-specific profiles[42]. Therefore, the potential responsiveness and safety profiles of the therapeutic approach could differ in humans.

Although exhibiting substantial preclinical efficacy, early clinical trials involving NMDAR-2B-selective antagonists, such as ifenprodil and Traxoprodil, have reported disappointing results (Table 2)[44-52]. Taxoprodil (CP-101, 606), a selective NMDAR-2B antagonist developed by Pfizer, also advanced to Phase 2 neurological trials, including stroke and depression. Nonetheless, clinical development of the drug was terminated due to prolongated QTc and dissociative side effects, despite modest antidepressant effects observed[45]. Therefore, the feasibility of taxoprodil application in chronic pain management is limited. Clinical ifenprodil data, specifically for neuropathic pain, remain scarce, without completed Phase 2 or 3 trials for DPN. Although some trials conducted in peripheral vascular disease (such as Leriche-Fontaine stage II) demonstrated improved walking distance and good tolerability, analgesia in neuropathic pain contexts was not addressed[53]. Ifenprodil has also been associated with off-target effects due to its affinity for α1-adrenergic and sigma-1 receptors[54,55]. The interactions have resulted in adverse outcomes, including sedation, dizziness, and cardiovascular effects, which narrows the therapeutic window of ifenprodil.

Table 2 Success or failures of clinical investigations on N-methyl-D-aspartate receptor-2B antagonists from previous clinical trials.

Ref.	NMDAR-2B antagonist	Clinical indication	Study phase/type of clinical trial	Outcome summary
Preskorn et al[44], 2008; Machado-Vieira et al[45], 2017	Traxoprodil (CP-101, 106)	Treatment-refractory major depressive disorder	Randomised, double-blind study	Traxoprodil produced an exceptional antidepressant response, well-tolerated without producing a dissociative reaction in patients. Nevertheless, further development of traxoprodil was not pursued due to concern for potential cardiovascular toxicity risk via QTc prolongation
Yurkewicz et al[46], 2005	Traxoprodil (CP-101, 106)	Severe traumatic brain injury	Randomised, double-blind study	Traxoprodil failed to demonstrate any mortality rate improvement or a favourable outcome that achieved statistical significance. No definitive claim of efficacy was made for traxoprodil for the treatment of severe TBI
Merchant et al[47], 1999	Traxoprodil (CP-101, 106)	Mild or acute moderate traumatic brain injury or hemorrhagic stroke	Double blind, placebo-controlled study	Minimal adverse effects of traxoprodil were reported in healthy subjects at 3 mg/kg/hour for 72 hours. The 72-hour-infused taxoprodil exerted no psychotropic effects in mild or moderate TBI or hemorrhagic stroke, but is well-tolerated by patients
Kotajima-Murakami et al[48], 2022	Ifenprodil	Methamphetamine use disorder	Exploratory, randomised, double-blind, placebo-controlled study	Ifenprodil at 120 mg/day showed safety and efficacy in reducing emotionality problems but exerted no effect on primary or secondary outcomes (such as drug use status, relapse risk, drug craving, and methamphetamine in urine)
Sugaya et al[49], 2018	Ifenprodil	Alcohol dependence	Prospective, randomised, controlled, rater-blinded study	Ifenprodil (60 mg/day for 3 months) significantly lowers the rate of heavy drinking in patients with alcohol dependence
Sasaki et al[50], 2022	Ifenprodil	Adolescent post-traumatic stress disorder (PTSD)	Randomised, double-blind, placebo-controlled trial	Ifenprodil can be a short-term, effective, safe alternative treatment for adolescent patients with PTSD
Danton et al[51], 2004	Ifenprodil	Stroke	Phase 2 clinical trial	Although eliprodil demonstrated a safer side-effect profile with promising findings in Phase 2 clinical trials; however, no considerable effect was observed in patients with stroke
Jain et al[52], 2000	Ifenprodil	Acute ischaemic stroke	Phase 3 clinical trial	Treatment of eliprodil in 8 hours (time window) for 14 14-day duration of treatment indicated no functional outcome improvement at 3 months. This clinical trial was discontinued, thus unpublished, due to lack of efficacy in sequential analysis
Branchereau and Rouffy[53], 1995	Ifenprodil	Chronic arterial occlusive disease of the legs (Fontaine stage II)	Double-blind randomised controlled trial	Ifenprodil tartrate (60 mg a.d.) improved maximum walking distance in patients. Nonetheless, no considerable evolution of ankle/brachial systolic post detected compared to placebo, ifenprodil exhibited excellent clinical and biological tolerance in the patients

PTSD: Post-traumatic stress disorder; TBI: Traumatic brain injury.

Context-specific NMDAR-2B expression and function present another primary challenge. The NMDAR subunit contributes to maladaptive plasticity and central sensitization in chronic pain. Nevertheless, the receptor subtype is also vital in the physiological function of normal synaptic transmission, learning, and memory, particularly in the cortex and hippocampus[9]. The dual roles of NMDAR-2B raise concerns regarding off-target effects and neurocognitive side effects when applying NMDAR-2 B-selective antagonists in clinical settings. Moreover, the positive effects of various NMDAR-2B antagonists, such as ifenprodil and Traxoprodil, observed in animal models[55] failed to achieve consistent effectiveness in clinical trials due to narrow therapeutic windows and dose-limiting side effects[56]. Heterogeneity amongst patients with DPN poses a translational barrier. Several factors, including the duration of diabetes, glycaemic control, comorbidities, and individual pain phenotypes, may influence NMDAR-2B-associated mechanisms and treatment responsiveness. Functional alterations, such as enhanced excitatory neurotransmission or transient NMDAR-2B upregulation, may dominate during the early course of diabetes. Nonetheless, chronic or long-term diabetes is frequently linked to irreversible nerve damage, glial scarring, and reduced neuroplasticity, which can lead to less effective NMDAR-2B modulation. Consequently, diabetic duration plays a critical role in the pathophysiology of DPN and treatment effectiveness. Comorbidities, including depression, anxiety, cardiovascular disease, or obesity, can implicate central pain modulation systems[57]. For instance, depression and DPN reportedly altered glutamatergic transmission and NMDAR-2B function in limbic regions, such as the ACC and hippocampus. The information suggested overlapping but different involvement of NMDAR-2B in the emotional and sensory parts of pain. Consequently, predicting treatment outcomes, especially for therapeutics that act on central receptors, including NMDAR-2B, is challenging. Furthermore, patient stratification remains a barrier without reliable biomarkers to identify pain mechanisms, potentially restricting the success of targeted therapies.

Apart from pharmacodynamic considerations, significant pharmacokinetic obstacles, including poor BBB permeability, rapid systemic metabolism, and limited bioavailability of numerous NMDAR-2B-selective antagonists, persist in developing NMDAR-2B-targeted therapies[58,59]. Achieving sufficient drug concentrations in central sites, such as the spinal cord dorsal horn and brain thalamus, is challenging, although NMDAR-2B is abundantly expressed in these pain-processing regions due to the BBB. Some compounds, such as ifenprodil, eliprodil, and 3-substituted aminocyclopentanes, can cross the BBB and exert central effects[60]. Nevertheless, a few NMDAR-2B antagonists are substrates of P-glycoprotein (P-gp), which is a key efflux transporter at the BBB that actively expels drugs from the CNS back into the circulation, limiting their intracerebral accumulation[61]. Therefore, designing NMDAR-2B-targeted drugs with optimal CNS bioavailability presents a critical barrier. Therapeutic agents also must achieve adequate penetration to modulate central sensitisation effectively and avoid off-target effects, such as sedation, dizziness, or cognitive impairment. Accordingly, formulation and delivery strategies of the potential therapeutics should be optimised to acquire effective doses at pain-relevant CNS sites without eliciting systemic toxicity. Among the strategies may involve P-gp bypassing delivery systems, prodrug designs, or allosteric modulators with improved BBB transport profiles to enhance efficacy and safety in clinical implementation. A systems-level approach is necessary to address the complex interplay among NMDAR-2B signalling, glial activation, oxidative stress, and inflammation rather than the isolated targeting of single pathways. Accordingly, future research should focus on the interactions when designing multi-targeted or combining therapies that better reflect the pathophysiological complexity of DPN. Therefore, improving animal models with better pharmacological tools, robust biomarkers, and personalised therapeutic strategies is essential to facilitate the successful translation of NMDAR-2B-targeted interventions from bench to bedside. Table 2 lists successful and failed key clinical and translational reports involving NMDAR-2B antagonists[44-53].

Therapeutic targeting of NMDAR-2B in PDN

Therapeutic DPN management remains challenging, given incomplete comprehension of its pathophysiology. Furthermore, pain relief remains insufficient in most patients despite several medications introduced. In clinical trials, a treatment is considered successful if the patients indicate a 50% decrease in pain and some other positive benefits on quality of life, fatigue, sleep, and depression[15]. Treating DPN primarily involves ruling out alternative causes of painful PN, whereas managing DPN comprises three strategies: Strict glycaemic control and risk factor management, pathogenetic mechanism-based therapies, and symptomatic pain management. Meanwhile, pharmacological therapies are typically symptomatic, inadequately focusing on pathophysiological mechanisms, and are constrained by tolerance development and side effect profiles[15]. Among the approaches, only intensive glucose control has lowered the likelihood of developing neuropathy. Although reversing or preventing DPN is notably challenging, improving glucose monitoring and other lifestyle habits, such as quitting smoking and minimising alcohol consumption, can enhance the quality of life of the patients. A singular or a combination of medications has considerably minimised neuropathic pain in various clinical studies[60,61]. Utilising medications, such as NMDAR-2B antagonists with chemicals targeting complementary pathways, offers a promising therapeutic strategy, considering the multifactorial nature of DPN. Oxidative stress is a main contributor to the pathophysiology of DPN. Nebivolol is a beta-adrenoceptor antagonist with antioxidant properties, which has demonstrated efficacy in alleviating thermal hyperalgesia in diabetic rats by enhancing nitric oxide activity[60,61]. Despite limited data on combining DPN medications with antioxidants, the rationale for this strategy is supported by the overlapping mechanisms of oxidative stress and neuronal excitotoxicity in DPN.

Preclinical studies have underscored the pivotal role of NMDAR-2B in DPN pathogenesis[32]. Increased phosphorylation of spinal NMDAR-2B has been linked to heightened nociceptive responses in various streptozotocin-induced diabetic rodent models. Several promising therapeutics have also been extended to clinical trial phases to evaluate their efficiency as NMDAR-2B antagonists in human subjects. For example, ifenprodil, a selective NMDAR-2B antagonist, has significantly attenuated tactile allodynia and thermal hyperalgesia. Reduced phosphorylated NMDAR-2B expression in the spinal cord of DPN rodents has also been observed[4]. The findings suggest that NMDAR-2B antagonism may modulate central sensitisation mechanisms underlying DPN[4]. Several safety and pharmacological challenges may impede clinical translation of NMDAR-2B antagonists exhibiting therapeutic potential. The therapeutics either require modification due to their dual activities, are halted due to inefficiency in human biological systems, or exhibit adverse effect profiles in clinical phases. For instance, ifenprodil is a promising NMDAR-2B therapeutic in preclinical studies. Nonetheless, the drug demonstrated off-target interactions with alpha 1-adrenergic and sigma receptors[54], raising concerns about potential side effects. Therefore, the dual activities of ifenprodil in NMDAR-2B antagonism and sigma-1 receptors were researched in various contexts to minimise the unwanted side effects. Similarly, traxoprodil, a selective NMDAR-2B antagonist, has exhibited neuroprotective effects on various animal models[46,62-64]. Unfortunately, the drug has been correlated with prolonged QT interval in humans, leading to its clinical development cessation[46]. Conversely, eliprodil, which is an NMDAR-2B antagonist and sigma-1 receptor agonist, failed to demonstrate notable therapeutic efficacy after undergoing clinical trials for acute ischaemic stroke and schizophrenia[64]. The outcomes highlight the complexity of translating NMDAR-2B-targeted therapies from preclinical models to clinical settings, requiring an understanding of receptor interactions in therapeutic development.

Regarding refractory DPN, where the patients frequently fail to achieve sufficient pain relief from monotherapies, combining NMDAR-2B antagonists and existing analgesics offers a rational and potentially synergistic strategy. The NMDAR-2B antagonists can improve the effectiveness of first-line medications, such as gabapentinoids (for example, pregabalin and gabapentin) or serotonin-norepinephrine reuptake inhibitors (such as duloxetine)[65-67], which primarily alter peripheral neurotransmission and descending pain inhibition pathways by reducing excitatory glutamatergic signalling and central sensitisation. Moreover, dual treatment might facilitate addressing the tolerance development commonly observed with chronic opioid or gabapentinoid employment. NMDAR-2B antagonists may also restore or prolong the analgesic window of other drugs by reducing the threshold for pain signal amplification at glutamatergic synapses[68]. Despite scarce clinical studies reporting such combinations, the approach may be particularly relevant in patients with refractory DPN where monotherapies are insufficient and polypharmacy is inevitable. Therefore, future clinical trials are warranted to validate the synergistic effects of the combination of drugs and determine optimal combination regimens with favourable safety profiles.

Advancements in gene therapy offer novel avenues for modulating NMDAR-2B expression in DPN. For example, targeted genetic material delivery to downregulate NMDAR-2B expression, such as delivering short hairpin RNA against the glutamate ionotropic receptor NMDA type subunit 2B (Grin2b) gene encoding NMDAR-2B utilising adeno-associated virus (AAV) vectors, could mitigate central sensitisation development[8]. Although the strategy has not been specifically applied to NMDAR-2B, a related proof-of-concept study revealed that intrathecal AAV9-short hairpin RNA injection targeting transient receptor potential vanilloid-1 suppressed gene expression in the dorsal root ganglion and spinal cord, effectively attenuating neuropathic pain behaviours in mice[69]. By analogy, AAV9-shGrin2b constructs might feasibly downregulate spinal NMDAR-2B and mitigate central sensitisation in DPN. Nonetheless, future studies should develop and validate such vectors with cell-specific promoters and optimal dosing regimens as a targeted means to modulate NMDAR-2B activities. Allosteric modulators present an alternative approach, which fine tunes NMDAR-2B activity without completely inhibiting it. Compounds, such as ifenprodil and its derivatives, act on the amino-terminal domain of the NMDAR-2B[70]. Therefore, the drug offers a mechanism to modulate receptor function with potentially fewer side effects.

Traxoprodil (CP-101,106) is an example of a high-affinity, negative allosteric NMDAR-2B modulator with analgesic and neurobehavioural effects in rodent and non-human primate studies[71,72]. The chemical has also undergone Phase 1 and 2 clinical trials as an experimental antidepressant; however, its rapid efficacy was hampered by cognitive effects and QT prolongation[44,45]. Therefore, future studies could consider designing drug-like, selective NMDAR-2B allosteric modulators with optimised CNS pharmacokinetics, efflux avoidance (e.g., P-gp resistance), and minimal cardiac and cognitive liabilities. Early assessments of target engagement and pharmacodynamic action across species could also be facilitated by incorporating quantitative electroencephalogram measures as translational biomarkers. Nonetheless, the strategies still warrant further investigation to determine their efficacy and safety in DPN.

CONCLUSION

The DPN condition is affected by a multifactorial pathophysiology, including metabolic disturbances, neuroinflammation, oxidative stress, and maladaptive neuroplasticity. Substantial preclinical evidence highlighted the aberrant expression and phosphorylation of NMDAR-2B in the spinal cord and thalamus of diabetic rodent models of studies evaluating the key role of NMDAR-2B in facilitating excitatory neurotransmission and central sensitisation, correlating with hyperalgesia, allodynia, and neuroinflammatory signalling[24,27,28]. In this review, the mechanistic role of the NMDAR-2B in the pathogenesis of DPN was outlined, including its interaction with inflammatory cytokines, oxidative stress, and glial activation. Therapeutics targeting the NMDAR-2B employing selective antagonists, such as ifenprodil and Ro 25-6981, have also demonstrated potential in preclinical models, where they attenuated pain behaviours and modulated receptor activities. Combining therapeutic approaches involving antioxidants and anti-inflammatory agents may enhance therapeutic efficacy by addressing multiple pain mechanisms concurrently. For instance, synergistic analgesic effects were observed by co-administering NMDAR-2B antagonists with antioxidants, potentially mitigating the limitations of monotherapies. Antioxidants, such as astaxanthin, have also exhibited substantial analgesic effects in several preclinical models, including diminishing neuropathic pain behaviour responses and NMDAR-2B expression in the spinal cord[73,74]. Nevertheless, establishing the combinatorial mechanisms and optimising cotreatment strategies requires further investigations.

Several research avenues are essential to overcome current translational barriers. Firstly, developing next-generation allosteric modulators with improved selectivity and NMDAR-2B subunit binding precision. Such modulators may minimise off-target effects and enhance safety by fine-tuning receptor activity without complete inhibition. Moreover, integrating cutting-edge human-based platforms, such as brain and spinal cord organoids, can offer a more physiologically relevant model to evaluate NMDAR-2B-targeted therapies. The three-dimensional organoids derived from human pluripotent stem cells recapitulate complex cytoarchitecture and cellular interactions, potentially enabling superior drug efficacy, toxicity, and mechanistic pathways prediction.

Advanced translational models, including humanised NMDAR-2B-expressing rodents and patient-derived iPSC-based neuronal systems, are critical to address drug response differences among varying species. The models possess the capacity to mimic human receptor pharmacology more accurately and improve predictive validity for clinical outcomes. Considering that iPSCs are reprogrammed from patient somatic cells, they retain genetic and epigenetic individual variability profiles, including chromosomal abnormalities and disease-specific mutations[75]. Future strategies should focus on the rational design of multi-targeted therapies that simultaneously address neuroinflammation, oxidative stress, and excitatory dysfunction beyond monotherapy options. Combining gene therapy approaches targeting NMDAR-2B expression with small-molecule modulators may also offer long-term disease modification. Conclusively, NMDAR-2B remains a compelling therapeutic target in DPN. Nonetheless, realising the clinical potential of the receptor requires a multidisciplinary strategy that bridges molecular insights, human-relevant disease models, biomarker-driven patient stratification, and innovative pharmacological platforms.