Acid‐sensing ion channel 1a induces AMPA receptor plasticity: a link between acidotoxicity and excitotoxicity in hippocampal CA1 neurons

The pathological process of excitotoxicity is a well-documented phenomenon in the propagation of anoxic brain injury. Glutamate causes an over-activation of NMDA receptors (NMDARs) and AMPA receptors (AMPARs), allowing for an influx in Ca2+ and spurring a signalling cascade that ultimately results in cell death (Szydlowska & Tymianski, 2010). Interestingly, acidosis occurs with glutamate release following stroke, but little is known about the relative importance of acidotoxicity in this process. Acid sensing ion channels (ASICs) may participate in driving deleterious AMPAR plasticity, and in turn be indirectly culpable for some of the resultant cell death. As such, ASICs may even prove to be a valuable target for neuroprotection. Due to the high vulnerability of the hippocampal CA1 region to stroke and anoxic injuries, it is possible that AMPAR plasticity is a contributor to excitotoxicity and delayed degeneration. There are two forms of AMPAR plasticity that arise following ischaemia: (1) anoxic LTP (a-LTP), caused by the ongoing presence of glutamate, and (2) downregulation of GluA2 expression, resulting in an elevated expression of GluA2 lacking calcium-permeable AMPARs (CP-AMPARs). In a recent article in The Journal of Physiology, Quintana et al. (2015) described a clever procedure to identify the potential role of acid-sensing ion channel 1a (ASIC1a) in postischaemic AMPAR plasticity within CA1 hippocampal neurons, and consequently, their role in ischaemic injuries. Organotypic hippocampal slice cultures (OHSCs) from 4 to 5-day-old wild-type (WT) and ASIC1a knockout (KO) mice were exposed to oxygen–glucose deprivation (OGD) conditions for 10 min and the resultant AMPAR-mediated field excitatory postsynaptic potentials (fEPSPs) were recorded. The fEPSPs of WT OHSCs showed an increased slope, indicating the development and maintenance of a-LTP. This effect was not seen in ASIC1a KOs suggesting that ASIC1a activation may be required for the continuation of a-LTP. Interestingly, GluA2 subunit synthesis is also a requirement for the maintenance of a-LTP (Quintana et al. 2006). This earlier finding served as the rationale for examining the relative expression of GluA2 between WT and KO OHSCs. At 1 h following OGD, WT OHSCs had increased GluA2 expression, whereas ASIC1a KO OHSCs exhibited no apparent increase. WT OHSCs treated with psalmotoxin 1 (PcTx1), an ASIC1a blocker, had similar levels of GluA2 expression to the ASIC1a KO group if treated during, but not after, OGD. Importantly, 12 h following OGD, GluA2 expression was downregulated in WT OHSCs, but remained unchanged in ASIC1a KO or WT OHSCs treated with PcTx1 during and after OGD. This suggests that ASIC1a activation may be necessary for GluA2 modulation during and following anoxic injury, which could play a role in the development of GluA2-lacking CP-AMPARs. Cell death was apparent 12 h after OGD, as measured by a decrease in NeuN staining in OHSCs. Surprisingly, the neuronal cell death at 12 h following OGD did not correlate with the observed decrease in GluA2. Quintana et al. (2015) concluded that cell death alone is insufficient to explain reduced GluA2 levels. We accept this conclusion with some reservations. It is possible that cell types more susceptible to delayed death preferentially produce GluA2, potentially masking a real correlation between cell death and GluA2 expression. Ischaemic injury has an initial necrotic component and a subsequent apoptotic component. GluA2 reductions may be part of the apoptotic but not the necrotic pathway. This is further suggested by the delayed depression of GluA2 expression. Had the authors examined early and late cell death separately, they may have seen a correlation between reductions in GluA2 and NeuN staining in the later period. In addition to looking at GluA2 expression, Quintana et al. (2015) also examined evoked excitatory postsynaptic currents (EPSCs) in OHSCs from the WT and ASIC1a KO mice. Although there was no significant change from baseline 6 h after OGD, in recordings at 12 h following OGD, the WT OHSCs exhibited an increase in inwardly rectifying current–voltage (I–V) relationships, and a drop in rectification index (RI; amplitude of EPSC at +40 mV/amplitude of EPSC at −40 mV), which are both identifiers of EPSCs mediated by CP-AMPRs. These changes were not seen in ASIC1a KO OHSCs or WT OHSCs treated with PcTx1 during or after OGD, further supporting the idea that ASIC1a activation is involved in the development of GluA2-lacking CP-AMPARs. Acidic conditions, which may be brought on by protons co-released with glutamate following stroke, may also be sufficient to upregulate CP-AMPARs through the activation of ASIC1a. Quintana et al. (2015) measured the I–V relationship of glutamate-induced currents in cultured 1-day-old Wistar rat hippocampal pyramidal neurons (HPNs) 12–24 h after being exposed to acidic environments (pH 6.0 or 6.5) or a physiological control environment (pH 7.4) for 15 min. Only HPNs exposed to the most acidic environment examined (pH 6.0) displayed reduced RIs, inwardly rectifying I–V relationships and an increase in single-channel conductance. It should be noted that the most acidic condition tested is considerably less acidic than levels typically reported in severe ischaemia. These findings collectively indicate the presence of CP-AMPARs. Interestingly, GluA2 was downregulated in HPNs at 12 h following acid treatment, paralleling the downregulation seen in WT OHSCs 12 h after OGD. Changes in I–V relationships, single channel conductance, and GluA2 expression were suppressed in cultures that received PcTx1 during the acid (pH 6.0) treatment further confirming the role of ASIC1a. Voltage-gated calcium channels (VGCCs) and NMDARs are known to influence AMPAR plasticity resulting in GluA2-lacking CP-AMPARs. Quintana et al. (2015) used selective ion blockers in acidic conditions (pH 6.0) to determine where ASICs fit into this paradigm. Individual cadmium blockage of VGCCs or d-2-amino-5-phosphonovalerate (d-AP5) blockade of NMDARs resulted in partial suppression of observed AMPAR properties. However, when given simultaneously, inhibition of VGCCs and NMDARs almost entirely prevented the ASIC1a-induced AMPAR subunit switch. The modulation of both NMDARs and VGCCs by ASIC1a plays an important role in the compositional change in AMPARs, which may in turn regulate VGCC and NMDAR activity (Fig. 1). Blocking sodium channels did not inhibit subunit switch, which also suggests that Ca2+ influx is essential for a downregulation of GluA2. Excessive calcium influx can lead to excitotoxicity and cell death. It is known that blocking ASIC1a or CP-AMPARs is neuroprotective by diminishing Ca2+ influx. Quintana et al. (2015) explored the role of ASIC1a in cell death and whether a combination of AMPAR and ASIC1a blockers would produce additional neuroprotective effects. Blocking ASIC1a with PcTx1 provided neuroprotection 12 and 24 h following OGD in OHSCs, and 24 h after exposure to acidic (pH 6.0) environments. The GluA2-lacking CP-AMPAR blocker, 1-naphthyl acetyl spermine (NASPM) provided minimal protection at 12 h, but equivalent protection to PcTx1 at 24 h following OGD. Similarly, the AMPAR blocker 2,3-dihydroxy-6-nitro-7-sulfamoyl-benzo[f]quinoxaline-2,3-dione (NBQX) provided protection 24 h following exposure to acidic (pH 6.0) environments. Co-administering ASIC1a and AMPAR blockers did not afford additional protection to that offered by PcTx1 alone. This highlights the relationship of ASIC1a and AMPARs. If ASIC1a are blocked, GluA2-lacking CP-AMPARs are not produced, so a CP-AMPAR blocker would logically have no effect. The importance of ASIC1a as a target for neuroprotection was confirmed when examining ASIC1a KO mice OHSCs, which had decreased cell death 2 and 24 h following OGD when compared with WT controls. This research enhances our understanding of stroke and the complex interactions that result from anoxic injury. It is clear that ASIC1a has a role in the development of AMPAR plasticity and cell death. This data could be valuable in the development of a model of stroke using acid-solution injections. However, it is important to note that these experiments were conducted ex vivo and may not resemble similar experiments conducted in vivo. Furthermore, the occurrence of damage and the resulting response may differ by brain region (Kreple et al. 2014) or age of subject (Luján et al. 2005). It is worth noting that all rats and mice used in this experiment were 5 days old or younger. These nuances only add to the complications of in vivo models and experimentation. Furthermore, data in the work by Quintana et al. (2015) may differ from previous reports due to variability in tissue preparation, processing, OGD times, or experimental paradigms. In summary, Quintana et al. (2015) demonstrated the role of ASIC1a in the modification of AMPARs that lead to harmful plasticity. These changes proved to play a role in the electrophysiological and biochemical alterations that occur following ischaemia. Protons released during anoxic brain injury activate ASIC1a, which in turn downregulates GluA2 expression and promotes the development of GluA2-lacking CP-AMPARs, as well as resulting in a-LTP. Acidic conditions in the hippocampus seem to ultimately result in a calcium influx, which may intensify the type of cell death typically associated with glutamatergic excitotoxicity, and result in delayed neural degeneration. Excitotoxicity and acidotoxicity may not be mutually exclusive entities, but in fact two sides of the same coin. None declared. Both authors have approved the final version of the manuscript and agree to be accountable for all aspects of the work. All persons designated as authors qualify for authorship, and all those who qualify for authorship are listed.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.