Targeting TRPM2- and TRPM4-extrasynaptic N-methyl-D-aspartate receptor coupling in ischemic stroke

Ischemic stroke represents a heavy burden on public health. Currently, recanalization is the only effective therapy for ischemic stroke in a small population of eligible patients. However, there is no effective adjunct medication for preventing neuron loss after reperfusion to mitigate long-lasting brain injury. Extrasynaptic N-methyl-D-aspartate receptors (esNMDARs) are the major cause of ischemic neuronal death. However, there is no inhibitor selectively targeting esNMDARs without compromising the function of other “beneficial” synaptic NMDARs, which is due to the limited understanding of the underlying molecular mechanisms. Recently, two members of the transient receptor potential melastatin (TRPM) channel family, TRPM2 and TRPM4, were shown to be critical in amplifying esNMDAR-mediated neurotoxic effects during ischemic stroke without influencing the functions of synaptic NMDAR. Targeting the interaction between TRPM2- and TRPM4-esNMDAR provides a novel strategy in screening effective drugs for ischemic stroke.

N-methyl-D-aspartate receptor (NMDAR): Ischemic stroke is a major contributor to global mortality and disability rates. The principal pathophysiological feature of ischemic stroke is neuronal death, which is primarily caused by NMDAR-mediated glutamate excitotoxicity (Campbell et al., 2019). However, therapies directly blocking NMDARs have all failed in clinical trials, partially due to the critical physiological functions of NMDARs in the brain. Additionally, the “double-edged sword” effects of NMDARs in both neuronal death and neuronal survival during ischemic stroke further complicate therapeutic research (Ikonomidou and Turski, 2002). Approximately two-thirds of NMDARs are synaptic NMDARs that are localized to the postsynaptic membrane. Activation of such synaptic NMDARs during ischemic stroke promotes neuronal survival. In contrast, the activation of NMDARs located outside the synapses (extrasynaptic NMDARs, esNMDARs) can cause ischemic neuronal dysfunction and death (Ikonomidou and Turski, 2002). It is unlikely that NMDAR antagonists can selectively block detrimental esNMDARs without influencing pro-survival synaptic NMDARs.

Targeting the ischemia-induced “upstream” factors that enhance the abnormal activation of esNMDAR or the “downstream” signaling cascades that magnify the cytotoxic effects of esNMDAR may represent a better strategy compared to the direct blockade of all NMDARs by NMDAR antagonists. Although it is known that NMDAR activity is mainly regulated by protein kinases and phosphatases and that esNMDAR surface expression is modulated by exocytosis, endocytosis, and local synthesis (Petit-Pedrol and Groc, 2021), the mechanisms regulating esNMDAR surface expression and activity during ischemic stroke remain elusive.

TRPM channels: TRPM subfamily of the TRP superfamily consists of eight members (TRPM1–8), among which TRPM2, TRPM4, and TRPM7 have been found to contribute to ischemic brain injury (Aarts et al., 2003; Yan et al., 2020; Zong et al., 2022c). These channels are characterized by a highly conserved N-terminal TRPM homology region (MHR domain), a six-helix transmembrane domain, a TRP domain, a C-terminal coiled-coil domain, and a variable C-terminal domain that is unique to each channel (Huang et al., 2020). TRPM7 was the first member of the TRPM family demonstrated to contribute to excitotoxicity and neuronal death caused by oxygen-glucose-deprivation in cultured neurons, but this is independent of NMDARs (Aarts et al., 2003). In recent years, two other members of the TRPM subfamily, TRPM2, and TRPM4, have been found to be actively involved in NMDAR-mediated excitotoxicity during ischemic stroke (Yan et al., 2020; Zong et al., 2022c). In addition to their activity in neurons, the activation of TRPM2 and TRPM4 in brain endothelial cells, microglia, and astrocytes also contributes to ischemic brain injury by increasing vessel leakage, promoting immune cell invasion, and amplifying inflammatory responses (Woo et al., 2020; Zong et al., 2022a). In this perspective, we focus on the interaction of NMDARs with TRPM2 and TRPM4 (NMDARs-TRPM2/4) in neuron excitotoxicity while exploring the promising translational value of targeting NMDARs-TRPM2/4 in the treatment of ischemic stroke.

TRPM2-esNMDAR coupling: The Ca²⁺-permeable, non-selective ion channel TRPM2 is characterized by its sensitivity to cellular oxidative stress (Zong et al., 2022a). TRPM2 is widely expressed in all tissues, with the highest expression in the brain. Activation of TRPM2 requires an increase of both intracellular Ca²⁺ and ADP-ribose, which binds to the unique C-terminal NUDT9-H domain of TRPM2 (Huang et al., 2020). During the early stages of oxidative stress, the mitochondrial and genomic DNA damage caused by reactive oxygen species triggers the activation of poly-ADP-ribose polymerase, which, in turn, markedly increases the production of ADP-ribose. At the same time, oxidative stress induces both Ca²⁺ influx from the extracellular environment and Ca²⁺ release from intracellular organelles such as the endoplasmic reticulum and lysosomes. The intrinsic gating property of TRPM2 by intracellular Ca²⁺ and ADP-ribose makes TRPM2 a precise sensor of oxidative stress, making it a critical player in oxidative stress-related diseases including ischemic stroke (Zong et al., 2022b).

Recently, we reported that there is a physical and functional interaction between TRPM2 and esNMDAR (Figure 1A; Zong et al., 2022c). In brief, the EE₃ motif that we identified in the N-terminal domain of TRPM2 directly binds to the KKR motif in the C-terminal domain of NMDAR. The binding of TRPM2 to NMDAR appears to be preferentially present at extrasynaptic sites, as TRPM2 was not detected in synaptosome extracts, and TRPM2 knockout selectively inhibited the esNMDAR’s activity (Zong et al., 2022c). The presence of TRPM2 in the TRPM2-NMDAR complex augmented NMDAR surface expression and enhanced NMDAR currents, which is referred to as “functional coupling”. Meanwhile, TRPM2-NMDAR coupling also enhanced TRPM2 channel activity. Functional TRPM2-NMDAR coupling depends on their physical association, as disruption of the physical association by an interfering peptide, TAT-EE₃, abolished the potentiation of NMDAR activity by TRPM2. TAT-EE₃ effectively inhibited ischemia-induced neuronal death caused by oxygen-glucose-deprivation in vitro, similar to the effects produced by NMDAR antagonists MK801 and AP5, highlighting a crucial role of TRPM2-NMDAR coupling in enhancing NMDAR-mediated excitotoxicity (Zong et al., 2022c). Importantly, by using the transient middle cerebral artery occlusion model, we found that post-stroke treatment with TAT-EE₃ protected mice against ischemic brain injury and promoted neuro-behavior recovery of mice seven days after ischemic stroke (Zong et al., 2022c). In pursuing the mechanisms by which TRPM2 enhances NMDAR activity, we found that TRPM2 physically associates with protein kinase Cγ (PKCγ), the neuron-specific PKC. The binding of PKCγ to TRPM2 can be induced by oxidative stress in vitro and post-ischemic stroke in vivo (Zong et al., 2022c). PKCγ is a known regulator of NMDARs, which can increase the surface trafficking of NMDARs as well as the channel activity of NMDARs (Lan et al., 2001). The functional coupling between TRPM2 and NMDARs appears to be PKCγ-dependent, as PKCγ inhibitors significantly influenced the potentiation of NMDARs by TRPM2. Although we do not know the underlying mechanisms yet, it is plausible that TRPM2-mediated Ca²⁺ influx may serve as a PKC activator since activation of PKC requires Ca²⁺. Our studies illustrate a scenario in which TRPM2 recruits PKCγ to the TRPM2-NMDARs complex during ischemic conditions, subsequently enhancing NMDAR surface expression and channel activity by activating PKCγ in the proximity of the interacting complex. Given that the TRPM2-NMDAR association occurs under oxidative stress or ischemic stroke conditions, our findings indicate that disrupting TRPM2-esNMDAR coupling represents a new method of inhibiting NMDAR-induced excitotoxicity, thus protecting neurons in ischemic stroke.

Figure 1.

TRPM2/4-NMDAR coupling during ischemic stroke.

(A) The EE₃ motif of TRPM2 physically associates with the KKR motif of NMDAR in the neurons. Ischemia promotes the production of ROS, which further activates TRPM2 and induces the binding of PKCγ to TRPM2. The physical binding between TRPM2-PKCγ brings PKCγ within close proximity of NMDARs, which leads to the increase of NMDAR surface expression and channel activity. The subsequent Ca²⁺ overload as well as the activation of cell-death signaling promotes neuronal death. (B) The TwinF motif of TRPM4 physically associates with the I₄ motif of NMDAR in the neurons. The TRPM4-NMDAR physical coupling promotes the activation of death-related signaling downstream to extrasynaptic NMDAR during ischemic stroke. (C) TRPM2 and TRPM4 directly associate with NMDAR and enhance cytotoxic effects by inhibiting the activation of the pro-survival ERK1/2 and CREB signaling. TRPM2 potentiates NMDAR by recruiting PKCγ. CREB: cAMP-response element binding protein; ERK1/2: extracellular signal-regulated kinase 1/2; esNMDAR: extrasynaptic N-methyl-D-aspartate receptor; ROS: reactive oxygen species. Created with BioRender. com.

TRPM4-esNMDAR coupling: TRPM4 is another TRP channel which associates with NMDARs and enhances the death-promoting activity of NMDARs (Yan et al., 2020). An elegant study by Yan et al. (2020) reported the discovery of neuroprotectants developed based on the interaction interface between TRPM4 and NMDARs. TRPM4 is a non-selective cation channel expressed in various cells including neurons, microglia, and endothelial cells, and is also involved in various neurological diseases (Mathar et al., 2014). Activation of TRPM4 requires an elevation of intracellular Ca²⁺. A unique feature of TRPM4 that distinguishes it from other TRP channels is its impermeability to Ca²⁺. TRPM4 is one of two TRP channels (TRPM4 and TRPM5) that are monovalent cation-selective. Due to its gating by intracellular Ca²⁺, TRPM4 may also be activated under oxidative stress conditions including ischemic injury. In 2020, Yan et al. in the Bading group, reported that TRPM4 interacts with NMDARs, and the disruption of the interaction using chemical compounds inhibits ischemic neuronal death (Figure 1B). Like TRPM2, TRPM4 interacts with GluN2A and GluN2b subunits but not GluN1 subunits, thereby promoting neuronal death under ischemic injury conditions. Interestingly, the interaction of TRPM4 with NMDARs or disruption of TRPM4-NMDAR coupling neither influences NMDAR currents nor NMDAR-mediated Ca²⁺ influx (Yan et al., 2020). Instead, disruption of TRPM4-NMDAR coupling detoxifies esNMDAR signaling, thus preventing mitochondrial dysfunction and hence inhibiting neuron-death pathways. In contrast to the conventional Ca²⁺ overload theory, Yan et al. (2020) revealed that the physical coupling of NMDARs with TRPM4 at the extrasynaptic sites promotes excitotoxicity. Functionally, it does this by inhibiting the pro-survival extracellular signal-regulated kinase 1/2 and cAMP-response element binding protein signaling. It will be of a great interest to understand the exact mechanisms as to how TRPM4 and NMDAR association mediates NMDARs’ excitotoxicity (Figure 1C). In our study, we found that TRPM2 interacts with esNMDAR both physically and functionally, thereby exacerbating NMDAR’s excitotoxicity. Thus, TRPM2-NMDARs coupling-induced excitotoxicity seems to converge conventional Ca²⁺ overload and unconventional esNMDAR coupling-induced neuronal death into a conjoined mechanism. As both TRPM2 and TRPM4 interact with esNMDAR, it will be important to investigate how those two TRP channels localize NMDARs to the extrasynaptic location under ischemic conditions, and whether TRPM2 and TRPM4 may work cooperatively or competitively in coupling with esNMDAR.

Summary: The failure of NMDAR antagonists in clinical trials highlights an urgent need of finding better therapeutic targets for ischemic stroke. The inhibition of the abnormal activation of TRPM2 and TRPM4 in the TRPM2/4-esNMDAR complex induced by ischemic stroke has two major advantages compared to the direct inhibition of NMDARs. Firstly, the baseline expression of TRPM2 and TRPM4 channels in healthy neurons is low, and impairment of brain function is not observed in TRPM2 and TRPM4 knockout mice (Yan et al., 2020; Zong et al., 2022c). Thus, inhibition of TRPM2 and TRPM4 would result in fewer side effects than those induced by NMDAR antagonists. Secondly, activation of TRPM2 and TRPM4 by oxidative stress or elevated intracellular Ca²⁺ during ischemic stroke appears to always be detrimental, therefore inhibition of TRPM2 and TRPM4 after ischemic stroke would produce largely protective effects without compromising the neuroprotective effects of synaptic NMDARs, thus avoiding serious side effects such as neuron-degeneration induced by synaptic NMDAR blockade (Yan et al., 2020). Thirdly, inhibition of TRPM2 and TRPM4 selectively inhibits the detrimental activation of esNMDAR during ischemic stroke, which is especially advantageous compared to the indiscriminate inhibition of both esNMDARs and synaptic NMDARs by NMDAR antagonists.

Currently, while specific blockers for TRPM2 and TRPM4 are not yet available, inhibiting the functional coupling between TRPM2/4 and NMDAR using disrupting peptides or chemical compounds represents a promising therapeutic strategy for ischemic stroke. Such a disruption strategy has a strong advantage in comparison with the direct inhibition of TRPM2 or TRPM4. As NMDARs are not expressed in other cell types, the disruption of NMDAR-TRPM2/4 interactions would only occur in neuronal cells, substantially reducing the risk of side effects caused by a systemic inhibition of TRPM2 and TRPM4. However, there is a possibility that the binding of disrupting peptides or chemical compounds might influence NMDAR binding with other associating proteins, causing off-target effects by interfering with the physiological functions of NMDARs. The structural studies using cryo-electron microscopy in recent years have revealed the high-resolution structure of TRPM2 and TRPM4, paving the way for the discovery of specific channel inhibitors. Moreover, the rapid advancement of nano-lipid particles for drug delivery suggests a cell-type and tissue-specific drug delivery system. This would significantly increase the specificity of TRPM2 and TRPM4 inhibitors in ischemic stroke, as those channels are widely expressed in almost all cell types. In summary, inhibiting the pathological activation of TRPM2 and TRPM4 and their coupling with NMDAR in the brain represents a novel and more effective treatment for ischemic stroke.

We thank the professional proofreading and editing by grant science writers Dr. Christopher Bonin and Dr. Geneva Hargis at UConn Health.

We apologize to many peers and colleagues whose elegant work is not cited due to space limitations.

This work was partially supported by the National Institute of Health (R01-HL143750) and American Heart Association (19TPA34890022) to LY.