Calcium-permeable AMPA receptors: an underestimated pharmacological target for the therapy of brain pathologies

Abstract

Excitotoxicity resulting from the accumulation of extracellular glutamate (the main excitatory neurotransmitter in the brain) is one of the main causes of neuronal death in various brain pathologies, including traumatic brain injury, epilepsy, stroke, and a number of neurodegenerative diseases such as Parkinson’s and Alzheimer’s diseases. An acute increase in extracellular glutamate concentration leads to neuronal dysfunction accompanied by oxidative stress, calcium homeostasis failure, a drop in energy metabolism, and loss of plasma membrane integrity. However, despite a more than 30-year history of studying glutamate excitotoxicity, there are no safe and effective drugs that prevent cell death under these pathological conditions. Numerous promising neuroprotective compounds have been rejected in clinical trials due to low efficacy and/or side effects. N-methyl-D-aspartate receptors (NMDARs) are considered the main source of Ca2+ influx during excitotoxicity, so various NMDAR antagonists have been primarily tested. However, the paradigm of NMDAR-Ca2+-mediated excitotoxicity formulated in the 1990s has outlived itself in terms of pharmacological potential. In this regard, the researchers have focused on α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors (AMPARs), the antagonists of which also demonstrate neuroprotective effects in in vitro studies. AMPARs mediate fast synaptic transmission at excitatory synapses in the central nervous system. As with other ionotropic glutamate receptors, AMPARs are tetramers assembled from a combination of GluA1–4 subunits. According to numerous studies, GluA2 is the most abundant in the mammalian brain AMPAR subunit which can be found predominantly as GluA1/GluA2 and GluA2/GluA3 heteromers. The role of GluA2 in receptor behavior is immensely significant not only due to its involvement in the regulation of the receptor trafficking and plasma membrane anchoring but also in governing ion permeability. Although a wide range of kinases is being implicated in the regulation of AMPAR-mediated synaptic plasticity in brain neuronal circuits, protein kinase C seems to be the most important in the case of GluA2 subunit surface expression. Protein kinase C-mediated phosphorylation of GluA2 causes its releasing from anchoring glutamate receptor interacting protein 1, thus promoting AMPARs internalization (Bissen et al., 2019). Regarding the role of GluA2 in AMPAR conductivity for Na+, K+, and Ca2+, GluA2-containing receptors are permeable only for K+ and Na+, while the GluA2-lacking receptor is also permeable for Ca2+ (Figure 1A). Notably, post-transcriptional editing of the GluA2 pre-mRNA also governs the calcium permeability of AMPAR. The RNA editing process (conversion of adenosine to inosine in the case of GluA2 pre-mRNA) is catalyzed by adenosine deaminases acting on RNA (ADARs). Thus, the receptors containing the unedited GluA2 subunit (GluA2(Q)) are also Ca2+-permeable.Figure 1: The differences in the structure and biophysical properties of calcium-permeable and calcium-impermeable α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors (AMPARs) underlying the approach of vital identification of neurons expressing these types of ionotropic glutamate receptors.(A) Scheme illustrating the subunit composition of calcium-permeable and calcium-impermeable AMPARs. Calcium-permeable AMPARs (CP-AMPARs) contain no GluA2 (glutamate receptor ionotropic, AMPA 2) subunit or contain unedited at the Q/R site (site of polypeptide chain where glutamine (Q) to arginine (R) change occurs due to adenosine to inosine replacement in GluA2 pre-mRNA) GluA2(Q) subunit. (B) Scheme illustrating the approach for vital identification of neurons containing CP-AMPARs using fluorescent Ca2+ imaging. Application of selective AMPAR agonists in the presence of selective antagonists of N-methyl-D-aspartate (NMDA) and kainate (KA) ionotropic glutamate receptors (NMDARs and KARs, respectively) and non-selective blockers of the voltage-gated calcium channels results in a significant increase in intracellular Na+ ([Na+]i) and Ca2+ ([Ca2+]i) concentrations. In this case, [Ca2+]i increase occurs only in neurons containing CP-AMPARs and can be detected using Ca2+-sensitive fluorescent probes such as Fluo-4 and Fura-2. The illustrations were created with Corel Draw software.Although calcium-permeable AMPARs (CP-AMPARs) are thought to play an important role in synaptogenesis and synaptic plasticity during brain development, their role in the mature brain is still poorly recognized. As studies show, most of AMPARs in the mature brain are calcium-impermeable because they contain edited GluA2(R) subunit. At the same time, there are populations of neurons that constitutively express CP-AMPARs. These neurons derive from the medial ganglionic eminence and include the basket cells, bistratified cells, ivy cells, and several classes of oriens-lacunosum moleculare interneurons (Pelkey et al., 2017). CP-AMPARs are also normally expressed by cerebellar stellate cells (Cull-Candy and Farrant, 2021). It can be suggested that due to Ca2+ permeability and voltage-dependence CP-AMPARs mediate some forms of non-NMDAR Ca2+-dependent synaptic plasticity. However, the presence of CP-AMPARs possibly makes these neurons more vulnerable to pathological exposures accompanied by an increase in the extracellular glutamate concentration. The selective death of interneurons is observed after traumatic brain injury and epileptic seizures (Tóth and Maglóczky, 2014; Frankowski et al., 2019), but the exact reasons for this phenomenon have not been established yet. Probably, the presence of CP-AMPARs along with previously demonstrated insufficient GABA(A) receptor-mediated inhibition (Gaidin et al., 2023), is one of the reasons for the selective death of GABAergic neurons (inhibitory interneurons releasing γ-aminobutyric acid (GABA)). Additionally, some subtypes of interneurons have also been shown to contain calcium-permeable kainate receptors (CP-KARs). Notably, the populations of GABAergic neurons expressing CP-AMPARs and CP-KARs for the most part intersect (Gaidin et al., 2023). It may be possible that neurons expressing both subtypes of calcium-permeable ionotropic glutamate receptors are more vulnerable to excessive extracellular glutamate accumulation due to the presence of additional pathways for Ca2+ inflow. It is critically important to prevent or reduce damage to inhibitory interneurons, since their primary death provokes a shift in the excitation/inhibition balance towards excitation. New circuits organized in the absence of interneurons promote the formation of more excitable (epileptogenic) neuronal networks. In this regard, the use of CP-AMPARs antagonists seems like a promising approach to prevent neuronal loss and minimize the consequences of brain damage. Interestingly, one of the CP-AMPARs antagonists, 1-naphthylacetyl spermine (NASPM, a synthetic analog of Joro spider toxin), is also able to block CP-KARs selectively expressed by GABAergic neurons. This bidirectional action of NASPM towards the calcium-permeable receptors may contribute to its neuroprotective efficacy (Maiorov et al., 2021). Unfortunately, despite the above facts, to date, there is no direct evidence that exactly CP-AMPARs cause vulnerability of the neurons expressing them to glutamate excitotoxicity. Although the conclusions seem convincing, they have been drawn based on indirect data. Taking into consideration that CP-AMPARs are expressed by interneurons and involved in the control of network activity, it cannot be unequivocally stated that the neuroprotective effects of the antagonists are mediated only by reducing the calcium load in the neurons expressing them. The use of antagonists or a decrease in the number of CP-AMPARs improves the survival of neurons in general. To solve this question, a way of identifying neurons containing CP-AMPARs is necessary. Immunostaining and patch-clamp technique are traditionally used to detect the presence of CP-AMPARs, but both methods have some limitations. As we mentioned above, the Ca2+ permeability of AMPARs in addition to subunit composition also depends on the editing of GluA2 pre-mRNA. Hence, Ca2+-permeable GluA2(Q)-containing AMPARs cannot be distinguished from calcium-impermeable GluA2(R)-containing receptors using antibodies. Moreover, immunostaining is a non-vital method since most of the immunolabeling protocols require tissue fixation with paraformaldehyde or another fixative agent. Electrophysiological methods seem to be more reliable approaches providing unambiguous data about the presence of CP-AMPARs and other electrophysiological properties of vital neurons. However, the rule “one experiment – one neuron” and methodological difficulties limit the use of these methods. In light of the mentioned limitations, we have suggested a simple and effective way to identify neurons containing CP-AMPARs (Gaidin et al., 2023). We have demonstrated that neurons containing CP-AMPARs and CP-KARs can be identified with fluorescent calcium imaging using a specific cocktail of agonists and antagonists (Figure 1B). In contrast to immunostaining and electrophysiological recordings, the proposed approach makes possible real-time analysis of dozens or hundreds of living neurons in vitro and, as we expect, in vivo. Moreover, we suppose that calcium imaging allows identifying the neurons containing not only GluA2-lacking, but also GluA2(Q)-containing receptors. This fact is very important in terms of the evaluation of dynamic changes in CP-AMPARs membrane expression upon different exposures. It should be noted that if the proportion of CP-AMPARs among all AMPARs in a neuron is extremely small, the detection of such a neuron by our method will most likely be impossible due to an insignificant, difficult-to-detect change in [Ca2+]i. However, despite the highlighted limitation, we believe that this approach will allow us in the future to unequivocally answer the question of the correlation between the presence of CP-AMPARs/CP-KARs and the increased sensitivity of neurons to pathological exposures accompanied by an increase in the extracellular glutamate concentration. In addition, the vital identification of CP-AMPARs-containing neurons allows a more detailed study of their features associated with the presence of calcium-permeable receptors. These data may contribute to understanding the mechanisms of toxic effects caused by Ca2+ influx through CP-AMPARs during their excessive activation. In addition to the constitutive expression of CP-AMPARs by some particular populations of neurons, the percentage of these receptors can change in other neurons upon some pathological exposures. For instance, the number of CP-AMPARs increases hours or days after brain injury or ischemic stroke. Koszegi and coauthors demonstrated that the enhanced internalization of GluA2 subunit and its lysosomal degradation were observed 15 and 60 minutes after oxygen-glucose deprivation, respectively, whereas GluA2 mRNA level decreased 24 hours after the exposure. These changes resulted in a significant increase in the number of GluA2-lacking CP-AMPARs a day after ischemia (Koszegi et al., 2017). According to other work, forebrain ischemia in adult rats leads to a decrease in ADAR2 level (Peng et al., 2006), which entails an increase in the number of CP-AMPARs containing the unedited GluA2 subunit. Thus, the percentage of CP-AMPARs rises both due to attenuation of the GluA2 pre-mRNA editing and an increase in the number of GluA2-lacking AMPARs. The fact that CP-AMPARs are expressed after some pathological exposures by neurons that normally do not contain them favorably distinguishes these receptors from NMDARs in terms of the development of neuroprotective drugs based on CP-AMPARs antagonists. NMDAR antagonists or blockers of the voltage-gated channels significantly affect normal brain activity, since these receptors and channels are involved in the implementation of numerous important functions. In turn, high expression of CP-AMPARs is observed, as a rule, under pathological conditions. However, it should be noted that CP-AMPARs play a pivotal role in brain development when neural connections actively form. Hence, we may propose that the increase in the expression of these receptors after the damage is a response aimed at “repairing” damaged connections between neurons. However, these changes in CP-AMPARs number may be more harmful than beneficial under physiological conditions, since the additional influx of Ca2+ exacerbates excitotoxic stress, thus promoting neuronal death. The neuroprotective effect of CP-AMPARs antagonists in models of ischemic stroke and traumatic brain injury has been shown previously (Guo and Ma, 2021). It was demonstrated that CP-AMPARs antagonist, Joro spider toxin, revealed more pronounced neuroprotective effects in in vitro models of traumatic brain injury compared to AMPARs or NMDARs antagonists. According to the in vivo studies, intrahippocampal injection of NASPM 9–40 hours after ischemic insult significantly increased neuronal survival in the CA1 field (Cornu Ammonis region which is most susceptible to ischemia). The small number of studies in this direction is explained by the limited repertoire of CP-AMPARs antagonists. Nevertheless, developments in this direction are underway, including the investigation of the mechanism of interaction between the antagonists and CP-AMPARs (Twomey et al., 2018). Moreover, it has been shown that not only direct inhibition of CP-AMPARs can have a protective effect. The suppression of PKC-dependent endocytosis of GluA2 subunit or repressor element-1 silencing transcription factor-mediated suppression of Gria2 (gene encoding GluA2 subunit) expression also prevent death of neurons. It is also possible to reduce the number of CP-AMPARs on the neuronal membrane by restoring ADAR2 expression, which falls after ischemic damage (Guo and Ma, 2021). Thus, the excitotoxic effect of CP-AMPAR activation can be abolished in different ways. Close attention should be paid to these approaches, since there are currently no safe and effective drugs that can protect neurons from excitotoxic damage. In summary, although most AMPARs are known to be calcium-impermeable in the adult brain, recent studies demonstrating an increase in the number of CP-AMPARs in various pathologies and the presence of populations of interneurons that constitutively express these receptors, make us take a different look at the role of CP-AMPARs in normal and pathological conditions. Cell-specific expression profile of CP-AMPARs and the dependence of their calcium conductivity not only on the subunit composition but also on pre-mRNA editing, allow fine modulation of activity of certain neuronal groups using both selective antagonists and the exposures directly or indirectly affecting RNA editing. This work was supported by the Ministry of Science and Higher Education of the Russian Federation in the framework of state assignment of PSCBR RAS 075-01512-22-02 (122112800049-0) supervising (including funding control and distribution) and realized by SGG and AMK. C-Editors: Zhao M, Liu WJ, Li CH; T-Editor: Jia Y

Topics & Concepts