A New Potential Strategy for Treatment of Ischemic Stroke: Targeting TRPM2–NMDAR Association

Ischemic stroke accounts for ~85% of all strokes and generally occurs when blood flow is severely reduced due to clots or masses blocking blood vessels and cutting off blood flow to brain cells. The death or damage to neurons caused by a stroke lead to corresponding symptoms such as weakness, paralysis, and numbness in the part of the body controlled by the affected area. This is an active and prolonged process. During ischemia, the decrease in the supply of oxygen and glucose starts with loss of electrical function, a phenomenon of disordered electrophysiology of cells that cannot respond appropriately to stimuli, and progresses to membrane dysfunction due to Ca²⁺ influx, leading to Ca²⁺-dependent excitotoxicity, production of reactive oxygen species (ROS), and the eventual destruction of cell membranes and cell lysis [1, 2]. In the late 1980s, the discovery of the role of glutamate and Ca²⁺ ions in neuronal death after hypoxia and ischemia (excitotoxicity) changed the entire field of stroke research. Gradually, studies on the mechanisms of ischemic brain injury focused on glutamate release, altering transporter function, receptor expression, and the activation of downstream cell death signals. Selective compounds block these mechanisms such as tyrosine phosphatase STEP-mimetic, which targets the excitatory signaling cascade downstream of N-methyl-D-aspartate receptors (NMDARs), and microRNA-29a that reduces excitotoxicity by maintaining astrocyte glutamate transporter protein 1 levels [3]. These studies yielded promising results in animal models but failed to show efficacy in clinical translation because of their side-effects and short therapeutic windows.

During excitotoxic events, excess neurotransmitter glutamate over-activates NMDARs, and causes excessive Ca²⁺ overload, triggering a series of intracellular signaling cascades, free radical generation, and eventual cell death, which is a major cause of neuronal death in ischemic stroke [4]. Thus, these studies provide a new mechanism to treat ischemic stroke based on NMDAR-mediated excitotoxicity in the pathogenesis of stroke. NMDARs play a dual role in neuronal survival and death, and there are three main theories regarding these different roles. First, different NMDAR subtypes located in neuronal cell membranes at synaptic and extrasynaptic sites may mediate distinct physiological and pathological processes. Second, synaptic and extrasynaptic NMDARs are two functionally distinct NMDAR populations, which may control the direction of neuronal survival and death, respectively [5]. Third, NMDAR activation is associated with multiple downstream signaling pathways, including pro-survival and pro-death pathways. However, NMDARs, which regulate neural development and synaptic plasticity, play a key role in learning and memory, so individually designed NMDAR antagonists may fail to provide neuroprotection in stroke patients and may even cause worrying side-effects. In addition to NMDARs, there appear to be other newly discovered non-selective cation channels leading to neuronal cell death during stroke, including transient receptor potential (TRP) channels. Oxidative stress increases during cerebral ischemia, the balance between ROS production and elimination is disrupted, and ROS overload activates non-selective cation channels such as TRPM2. ROS-induced TRPM2-dependent neuronal cell death may involve multiple mechanisms including autophagy and apoptosis. It has been shown that ROS-activated TRPM2 inhibits autophagy through downregulation of the AMPK/mTOR pathways, leading to brain ischemia/reperfusion-induced neuronal death [6]. By controlling cytoplasmic ion homeostasis, TRPM channels may activate various downstream targets and exert different regulatory effects, which arise from cell specificity, disposition, and the subcellular localization of the channel. Activation of TRPM2 channels leads to Zn²⁺ accumulation and further induces ROS production, triggering lysosomal dysfunction and an increase in neuronal cell death [7]. In addition, TRPM2-mediated Ca²⁺ influx also activates CAMK2, which phosphorylates BECN1/Beclin1 on Ser295 to inhibit autophagy, and induces neuronal death [8]. Abnormal TRPM2 function is associated with a variety of neurological disorders such as cerebral ischemia, Alzheimer's disease, neuropathic pain, and Parkinson's disease. In ischemic stroke, among the subunits of NMDAR, GluN2A and GluN2B, the former is thought to promote neuronal survival, whereas the latter is inclined to induce neuronal death. Researchers have found that TRPM2 deletion leads to changes in the expression ratio of the NMDAR GluN2A/GluN2B subunits. TRPM2 deletion provides neuroprotection by selectively upregulating pro-survival signaling [9]. It has also been shown that in microglia-driven chronic neuroinflammation, NMDAR stimulation promotes TRPM2 channel activation as a result of Ca²⁺ and ERK1/2-dependent poly (ADP-ribose) polymerase-1 (PARP-1) recruitment [10]. However, the molecular mechanism underlying the interaction between TRPM2 and NMDARs remains unclear.

Recently, Yue's team has revealed the role of TRPM2 in excitotoxicity mediated by NMDARs in stroke studies published in Neuron. They investigated the specific mechanism by which the ion channel TRPM2 interacts with NMDARs during stroke injury [11]. Most of the early research supported the “NMDAR location” hypothesis that synaptic NMDARs promote neuronal survival, while extrasynaptic NMDARs are activated in the brain in response to excess glutamate, and induce neuronal death [5]. They found that TRPM2 interacts with extrasynaptic NMDARs, and further designed a cell-penetrating peptide to block this interaction.

First, the researchers found that TRPM2 deletion was effective in reducing brain injury after stroke in a mouse model by whole-cell non-selective TRPM2-knockout and neuron-specific deletion of TRPM2. Most importantly, neuron-specific deletion of TRPM2 conferred a degree of protection similar to that of full knockout, suggesting that neuronal TRPM2 plays a key role in causing ischemic brain injury. A model of cerebral ischemia using the glucose-oxygen deprivation assay was further applied and showed that the deletion of TRPM2 inhibited the neuronal Ca²⁺ overload, mitochondrial stress, and death induced by hypoxia. Subsequently, TRPM2 physically coupled with GluN2A and GluN2B. In addition, co-expression of TRPM2 with NMDARs was also able to enhance the activation of NMDARs on the membrane surface, and functional coupling between them was also apparent. This was followed by the direct binding of the EE₃ motif in the amino-terminal domain of TRPM2 to the KKR motif in the carboxyl-terminal domain of GluN2A/B, as demonstrated by molecular cloning experiments. Moreover, the physical and functional coupling between TRPM2 and NMDARs was confirmed by EE₃ motif deletion or mutation and KKR motif deletion, which abolished the increased surface expression of NMDARs and failure to enhance NMDAR currents.

The authors subsequently investigated how exactly TRPM2 enhances the excitotoxicity of NMDARs. Protein kinase C (PKC), a family of protein kinases that phosphorylates serine/threonine residues within substrate protein molecules, includes 12 isoforms and plays an important role in learning and memory [12]. PKC-γ is a neuron-specific PKC and the authors found that it bound to the amino-terminus of TRPM2. PKC-γ directly activated NMDARs and promoted the cell surface transport and expression of NMDARs. Moreover, PKC-γ activators further amplified the enhanced the effect of TRPM2 on NMDAR activation, while PKC-γ inhibitors significantly inhibited the enhanced activation of NMDARs induced by TRPM2. Furthermore, the interaction between PKC-γ and TRPM2 was facilitated by the induction of oxidative stress in vitro.

Finally, the molecular binding mechanism of TRPM2–NMDAR was elucidated. They further produced a membrane-permeable interfering peptide TAT-EE3 (Transactivator-EE3), TAT, which has the ability to carry substances with different biological activities into living cells that are 100 times larger than its molecular weight. TAT-EE3 mimicked the sequence of TRPM2 and bound to NMDARs, thus disrupting the interaction between TRPM2 and NMDARs. Fully considering the potential impact of delayed neuronal death after stroke, they designed two stroke models, short-term and long-term, both of which showed that the disruptive peptide TAT-EE3 significantly reduced the area of brain damage and improved brain function in mice, indicating a potential strategy for the treatment of ischemic stroke (Fig. 1).

Fig. 1.

The TRPM2–NMDAR interaction exacerbates excitotoxicity during ischemic stroke, and uncoupling this interaction attenuates ischemic brain injury. In ischemic stroke, excessive activation of NMDAR-type ionotropic glutamate receptors by excess glutamate leads to Ca²⁺ overload in neurons, resulting in neuronal death. TRPM2–NMDAR binding promotes the surface expression of extrasynaptic NMDARs (esNMDARs), leading to enhanced NMDAR activity and increased neuronal death. Under oxidative stress conditions, TRPM2 recruits PKC-γ to the TRPM2/NMDAR complex, bringing PKC-γ into close proximity with NMDAR-interacting partners, thereby promoting the surface trafficking of NMDARs and leading to enhanced excitability. TRPM2 interacts directly with GluN2A/B through its unique EE₃ pattern in the amino terminal and the KKR pattern in the carboxyl terminal of GluN2A/B, but not GluN1A. The TRPM2-derived interfering peptide TAT-EE₃ disrupts this interaction and prevents the reduction of phosphorylated CREB and ERK1/2 levels, effectively inhibiting excitotoxicity, reducing infarct volume, and providing a therapeutic effect in ischemic stroke.

In summary, Yue's team revealed the physical and functional interactions of NMDARs and TRPM2 and further identified their interaction motifs. In addition, the mechanisms of TPRM2-NMDAR functional coupling were illuminated. Most importantly, they identified a strategy using a peptide, TAT-EE3, to treat ischemic stroke more effectively and safely by targeting uncoupling TRPM2–NMDAR interactions. Similarly, a previous study by Yan et al. demonstrated that the mechanism of excitotoxicity generation is the physical coupling of NMDARs to TRPM4, with GluN2A/B containing four regularly-distributed isoleucines (I4), which interact with the 57-amino-acid structural domain (TwinF) in the amino terminal of TRPM4 to enhance excitotoxicity [4]. In addition, researchers have found that floralozone has a neuroprotective effect after chronic cerebral ischemia, a process that involves the reduction of TRPM2 protein levels and the enhancement of NR2B protein levels of NMDARs, thereby reducing neuronal apoptosis [13]. Thus, targeting the interactions between NMDARs and the TRPM family shows great potential for the treatment of cerebral ischemia and other neurodegenerative diseases.

Yue's group has been working on the relationship between TRP channels and many diseases, and the TRPM family is the largest and most diverse subfamily of the TRP superfamily. Besides TRPM2 and TRPM4, TRPM7, which is critical for axonal development [14] and mediates hypoxic and ischemic neuronal cell death [15], has been identified as a potential non-glutamate target for hypoxic-ischemic neuronal injury [16]. Therefore, identifying the key downstream proteins involved in TRPM7 signaling under ischemic conditions in the brain seems to support another pharmacological tool for potential drug development.

In addition, this interfering peptide faces several obstacles before entering clinical studies. The role of neuronal TRPM2 in cerebral ischemia has been shown to be sexually dimorphic, TRPM2 inhibition reduces neuronal damage after middle cerebral artery occlusion in the male mouse brain or ischemia in vitro, while having no effect on the female brain [17]. Since female mice were not used in this study, we remain skeptical that disrupting the interaction between TRPM2 and NMDARs would have the same effect in female mice. Ischemic brain injury involves complex molecular signaling in neurons, glial cells, and the blood-brain barrier. The interaction between TRPM2 and NMDARs may also regulate signaling molecules associated with neuroprotection in other cells, including astrocytes, pericytes, and endothelial cells in different aspects of neuroprotection, including cell death, regeneration, perfusion injury, and events around the neurovascular unit.

Mitochondrial dysfunction in cerebral ischemia leads to neuronal death, and mitochondrial fusion and fission are key dynamic processes that maintain mitochondrial function and cell viability. The balance of mitochondrial fusion and fission is mediated by a variety of proteins, like Mfn2, the expression level and mitochondrial translocation of which have been shown to decrease across the cerebral cortex, but unchanged in the hippocampus during excitotoxicity [18]. This may be due to the expression of TRPM2 in different brain regions during excitotoxicity, which then affects the susceptibility of specific regions to stroke. Current treatment for ischemic stroke is mainly thrombolysis for embolism, but this approach has a strict time window, contraindication restrictions, and time/space limitations, and many stroke patients are unable to receive this treatment. In this case, the early application of neuroprotective agents in patients with acute ischemic stroke has a broader scope. The optimal time window and optimal dose of a drug need to be identified in clinical studies. Considering the possible disadvantages of interfering peptides such as instability and rapid clearance rates, designing new dosage forms or optimizing them may achieve better efficacy. In addition, since the majority of stroke patients are elderly, they often have hypertension, a major factor in hemorrhagic and atherothrombotic stroke, which can promote arterial spasm and arterial wall damage, followed by the progressive development of atherosclerosis and plaque formation. It has been estimated that ~54% of strokes worldwide are associated with hypertension and studies have shown that the relationship between hypertension and risk of stroke is continuous, consistent, and independent of other risk factors. Once an abnormal blood pressure plaque is dislodged, the patient may have a sudden ischemic stroke, so a stroke model for the elderly group is also worth exploring. Anyway, in the long run, this discovery has shed light on ischemic stroke and has promising clinical application.