ASIC1a as an integrating node in ischemic neurotoxicity: crosstalk with NMDAR and TRPM signaling

Abstract

Ischemic stroke triggers acidosis, glutamate excitotoxicity and oxidative stress, which converge to disturb ionic homeostasis through several cation channels, including acid-sensing ion channel 1a (ASIC1a), N-methyl-D-aspartate receptors (NMDARs) and transient receptor potential melastatin (TRPM) channels. Emerging evidence suggests that ASIC1a functions as a context-dependent integrating node: protons directly gate ASIC1a, glutamate and bilirubin-associated signaling can modulate ASIC1a-dependent injury pathways. ASIC1a directly couples with NMDAR-dependent signaling and indirectly engages TRPM-associated pathways. Here, we discuss the mechanistic basis, therapeutic potential and unresolved questions regarding ASIC1a-associated signaling in ischemic brain injury.

Introduction

Ischemic stroke, which accounts for the majority of strokes and remains a leading cause of death and permanent disability worldwide, most commonly results from cerebral arterial occlusion and produces profound alterations in tissue pH homeostasis [24]. Reduced oxygen and glucose delivery impairs oxidative phosphorylation, depletes ATP and increases reliance on anaerobic glycolysis [70, 72]. Together with impaired clearance of lactate and H⁺ and failure of cellular pH-regulatory mechanisms, these changes produce intracellular and extracellular acidification whose severity and duration vary with residual perfusion, metabolic demand and reperfusion [11, 12, 17]. Importantly, ischemic acidosis is not merely a consequence of metabolic failure: changes in extracellular and intracellular pH influence membrane excitability, ion channels and transporter activity and signaling, thereby influencing neuronal survival [11, 88, 94]. Proton-sensitive signaling therefore provides a direct molecular interface between metabolic disturbance and ischemic brain injury [38, 94].

ASIC1a is one component of a broader pH-sensing system. Other proton-responsive molecules relevant to the ischemic brain include the proton-activated chloride channel (PAC; also known as transmembrane protein 206 [TMEM206]), two-pore-domain K⁺ channels (K2P) such as extracellular pH-sensitive TWIK-related acid-sensitive K⁺ (TASK) channels and intracellular acidosis-sensitive TWIK-related K⁺ (TREK) family members, and proton-sensing G protein-coupled receptors (GPCR) including GPR4, GPR65, and GPR68 [16, 27, 28, 97, 99]. These systems contribute to neuronal excitability, pH homeostasis, vascular responses, and inflammatory signaling [99]. However, the present review focuses on ASIC1a because it has a comparatively well-developed mechanistic and preclinical literature linking proton sensing to excitotoxic and ionic injury [19, 94].

The rationale for considering ASIC1a as an integrating node arises from its position within interacting excitotoxic, acidotic and oxidative injury pathways, although the strength of evidence is not equivalent for all proposed interactions. Direct proton gating of ASIC1a [87, 94] and functional coupling between NMDAR/ Ca²⁺/calmodulin-dependent protein kinase II (CaMKII) signaling and ASIC1a are comparatively well established [19]. By contrast, glutamate-dependent allosteric potentiation of ASIC1a is an emerging mechanism that requires independent validation [36], and bilirubin, proposed to be a putative volume transmitter for TRPM2 channels to exacerbate neurotoxicity, should be regarded as an emerging and potentially context-dependent mechanism [35, 40]. Likewise, direct physical coupling between ASIC1a and TRPM channels has not been demonstrated; ASIC1a-TRPM relationships are therefore better interpreted as functional convergence within a shared injury network. These distinctions are important because ischemic acidosis is differentially decoded across diverse ion channel systems.

We therefore use the term “integrating node” operationally. In this review, ASIC1a is considered an integrating node on the basis of three features: direct responsiveness to a major ischemic cue, functional communication with other injury pathways, and convergence onto shared downstream effectors. Current evidence most strongly supports direct proton sensing and convergence onto Ca²⁺-dependent injury mechanisms, whereas reciprocal communication is best characterized for ASIC1a-NMDAR signaling and remains less defined for TRPM-associated pathways. Accordingly, this model does not imply that ASIC1a is the first molecular substrate in every ischemic microdomain or that it physically associates with every component of the injury network. We first examine how ischemic acidosis is differentially decoded by NMDARs, TRPM channels and ASIC1a, and how emerging extracellular modulators influence ASIC1a activity; we then discuss intracellular pathway convergence and cell-type specific signaling within the neurovascular unit, before evaluating the therapeutic potential and translational limitations of targeting ASIC1a in ischemic stroke.

Extracellular crosstalk: proton, glutamate and bilirubin signaling

During cerebral ischemia, the extracellular environment contains multiple injury-associated signals, including protons, glutamate and, under some pathological conditions, bilirubin, which influence membrane cation channels involved in ischemic injury [40, 75, 98]. Extracellular acidosis is differentially decoded across channel families, producing distinct effects on excitability, ion permeability and cell survival. The following sections examine these channel-specific responses and their convergence with glutamatergic and oxidative-metabolic signaling.

NMDARs: proton-mediated suppression within an excitotoxic environment

Extracellular protons strongly inhibit NMDARs, which display pronounced sensitivity to extracellular acidification [82, 85]. Proton-mediated inhibition reduces NMDAR channel opening probability and limits receptor-mediated Ca²⁺ entry as extracellular pH falls, with substantial inhibition occurring within the physiological-to-mildly acidic range [82, 85]. However, this suppressive effect does not eliminate NMDAR-dependent excitotoxicity during ischemia. Extracellular glutamate accumulates markedly during ischemia [5], while neuronal depolarization relieves the voltage-dependent Mg²⁺ block of NMDAR channels [56]. Moreover, ischemic tissue exhibits substantial spatial and temporal heterogeneity in extracellular pH [52], and transient local alkaline shifts can relieve proton inhibition of NMDARs [83, 84]. Thus, acidosis reshapes rather than abolishes NMDAR excitotoxicity through dynamic interactions among extracellular pH, glutamate availability and membrane potential [5, 52, 56, 83–85]. The same extracellular acidification directly activates ASIC1a, potentially altering the relative contribution of NMDAR-mediated Ca²⁺ entry and ASIC1a-mediated proton-gated cation influx.

Divergent proton regulation of TRPM channels

Unlike NMDARs that extracellular protons generally inhibit their activity, TRPM channels exhibit heterogeneous responses to ischemic pH changes. Multiple TRPM channels have been shown to aggravate ischemic neurotoxicity: TRPM2 is inhibited by extracellular protons, whereas TRPM7 inward currents are potentiated under acidic conditions; direct extracellular proton regulation of TRPM4 is less well established, highlighting that the same proton signal can produce distinct functional outcomes depending on channel identity and cellular context. These divergent responses position TRPM channels as important modulators of ischemic ionic imbalance.

TRPM2

Extracellular protons inhibit TRPM2 activity [76, 96]. In heterologous expression systems, lowering extracellular pH inhibits TRPM2-mediated currents, with half-maximal inhibition reported around pH 5.3 [15]. Mechanistically, extracellular acidification inhibits TRPM2 through protonation of residues within the outer pore region, thereby reducing channel activity despite ongoing oxidative stress signaling [15]. Whether these pore-dependent mechanisms operate similarly in native neuronal TRPM2 channels during cerebral ischemia remains to be fully established.

TRPM7

In contrast to the inhibitory effect of acidic pH on TRPM2, extracellular acidification potentiates inward TRPM7 currents [30]. Mechanistically, acidic pH appears to potentiate inward TRPM7 currents primarily by relieving divalent-cation block within the channel pore, but not by simple de novo channel activation [30]. These biophysical findings suggest that extracellular acidification could enhance TRPM7-mediated inward current; the extent to which this mechanism operates in native neurons during ischemia remains to be established.

TRPM4

TRPM4 differs fundamentally from TRPM2 and TRPM7. Although activated by elevations in intracellular Ca²⁺, TRPM4 is a monovalent-selective cation channel that is essentially impermeable to Ca²⁺ [37, 53, 54]. Accordingly, TRPM4 does not contribute directly to Ca²⁺ influx; instead, TRPM4-mediated monovalent-cation conductance [23], particularly Na⁺ influx at negative membrane potentials, can promote membrane depolarization and alter the driving force for Ca²⁺ entry through other channels [37]. In neurons, TRPM4 physically associates with extrasynaptic NMDARs to form a pro-death signaling complex, and disruption of this interface reduces excitotoxicity, mitochondrial dysfunction, and neuronal loss in experimental models [95]. In contrast to TRPM2 and TRPM7, whose extracellular proton sensitivity has been directly characterized, the influence of extracellular pH on TRPM4 remains less well defined in ischemia-relevant preparations [55].

Together, these TRPM subtype-specific responses show that extracellular acidosis is differentially decoded. ASIC1a is distinguished by direct proton gating and modulation by additional ischemic signals, including glutamate, providing the basis for considering it as an integrating node.

Glutamate as an allosteric modulator of ASIC1a

Glutamate, the major excitatory neurotransmitter accumulating during ischemia, has recently been reported to act as a positive allosteric modulator (PAM) of ASIC1a-mediated currents [36]. Site-directed mutagenesis, structure-based modeling, binding measurements and functional assays in that study support an extracellular glutamate-binding site distinct from the principal proton-sensing regions of ASIC1a [36]. In that study, glutamate increased ASIC current amplitude and altered channel gating properties, including enhanced proton sensitivity and reduced desensitization, particularly under mildly acidic conditions relevant to the ischemic penumbra [36]. Thus, elevated extracellular glutamate may amplify ASIC1a-dependent cation influx in parallel with NMDAR-mediated excitotoxicity, providing a potential molecular link between glutamatergic and acidotoxic signaling. However, because this mechanism is currently supported predominantly by a single study, independent validation across laboratories, preparations, species and clinically relevant stroke models is required.

Bilirubin-associated crosstalk with TRPM2, NMDARs and ASIC1a

Bilirubin is a context-dependent oxidative-metabolic signal that may contribute to ischemic injury. Heme catabolism generates biliverdin and subsequently bilirubin, and ischemic injury can induce endogenous bilirubin release [14, 40]. Despite being a physiological antioxidant [71, 77], bilirubin at high concentration can be neurotoxic [45]. Clinically, elevated bilirubin levels after acute ischemic stroke have been associated with greater stroke severity and poorer outcomes [42, 59, 65]. But these associations are heterogeneous and observational and do not establish causality [90]. A recent study reported that post-stroke, but not pre-stroke, bilirubin levels correlated with infarct volume [40].

At the molecular level, bilirubin may amplify ischemic signaling through interactions with oxidative stress-sensitive ion channels. Bilirubin has been proposed to work through volume neurotransmission, that is, a form of extracellular signaling by which neurotransmitters or chemical messengers released from cells can diffuse through the extracellular space and influence nearby target cells without requiring a conventional synaptic contact with pre-and postsynaptic specializations [2]. In this context, ischemic injury can induce endogenous bilirubin release from brain tissue to extracellular space where bilirubin directly binds to and activates TRPM2 channels in neurons. This cation channel is canonically known to be activated by intracellular reactive oxygen species (ROS)- and ADP-ribose associated with mitochondria reaction to ischemic challenge, but new evidence indicates bilirubin can directly gate the opening of TRPM2, thereby promoting Ca²⁺ influx and downstream neurotoxic signaling [40]. In addition, bilirubin-associated neuronal injury has been linked to NMDAR-dependent mechanisms, as pharmacological inhibition of NMDARs reduces bilirubin-induced neurotoxicity in experimental models [22, 47]. In a neonatal hyperbilirubinemia model, bilirubin was reported to potentiate ASIC1a activity under acidic conditions [35]. Whether the same mechanism contributes materially to cerebral ischemic injury remains to be established.

These extracellular signals therefore converge on channels with distinct proton, glutamate and redox sensitivities, setting the stage for intracellular pathway convergence.

Intracellular crosstalk: functional convergence of ASIC1a, NMDARs and TRPM channels

Although ASIC1a, NMDARs and TRPM channels participate in overlapping ischemic injury pathways, their molecular relationships are not equivalent. Direct physical coupling has been demonstrated for TRPM-NMDAR interactions, whereas evidence for a stable ASIC1a-NMDAR complex remains inconclusive [19, 44, 95, 100]. Accordingly, this section considers ASIC1a-related integration primarily in terms of functional coupling and pathway convergence.

Functional coupling between ASIC1a and NMDAR-dependent signaling

The strongest evidence supporting an integrative role for ASIC1a in neurotoxicity comes from reciprocal functional interactions with NMDAR-dependent signaling, although the mechanistic evidence supporting each direction is not equivalent. Ischemia enhanced ASIC1a phosphorylation at Ser478 and Ser479 through activation of GluN2B-containing NMDARs and CaMKII; blockade of GluN2B, inhibition of CaMKII, or mutation of these phosphorylation sites prevented the ischemia-induced increase in ASIC currents, cytosolic Ca²⁺ elevation, and neuronal death, thereby establishing a functional NMDAR-CaMKII-ASIC1a signaling axis in acidotoxic injury [19]. Furthermore, ASIC1a activation enhances NMDAR-mediated currents and NMDAR-dependent Ca²⁺ overload in hippocampal neurons, whereas ASIC1a inhibition suppresses NMDAR-mediated excitatory postsynaptic currents (EPSCs) and oxygen-glucose deprivation-induced overexcitation; co-expression experiments further implicated the GluN1/GluN2A/GluN2B tri-heteromer as the NMDAR subtype most strongly facilitated by ASIC1a [44]. Importantly, direct physical association between ASIC1a and NMDARs has not been conclusively demonstrated; rather, the persistence of ASIC1a-dependent enhancement of NMDAR currents under low-Mg²⁺, Ca²⁺-free, and kinase-inhibited conditions suggests that membrane-level proximity or scaffold-based interactions may allow ASIC1a conformational changes to influence adjacent NMDAR gating. Extracellular protons can also trigger neuronal necroptosis via ASIC1a independently of its ion-conducting function by recruiting receptor-interacting serine/threonine-protein kinase 1 (RIPK1; also termed RIP1) to the ASIC1a carboxyl-terminal region near the CaMKII phosphorylation sites [88, 91], supporting a non-conducting signaling role for ASIC1a in cell death.

TRPM Channels and NMDAR-associated injury signaling

TRPM channels participate in NMDAR-associated intracellular injury signaling, although their relationships with NMDARs differ mechanistically from the ASIC1a-NMDAR relationship. For example, TRPM2 contains an N-terminal interaction motif that directly interacts with C-terminal domains of extrasynaptic NMDARs, promoting surface expression and activity of GluN2A/GluN2B-containing receptors and thereby amplifying excitotoxic Ca²⁺ influx [100]. In turn, NMDAR-mediated Ca²⁺ entry may facilitate TRPM2 activation by increasing intracellular Ca²⁺/ADP-ribose and oxidative signaling, potentially creating a positive-feedback loop that links excitotoxic and oxidative injury signaling [100].

TRPM4 should be distinguished from Ca²⁺-permeable channels such as ASIC1a, TRPM2 and NMDARs [37, 53, 54]. Although activated by intracellular Ca²⁺, TRPM4 is largely impermeable to Ca²⁺ and contributes to pro-death signaling mainly through monovalent cation influx, membrane depolarization and interaction with NMDAR-associated signaling complexes. Through its N-terminal TwinF motif, TRPM4 interacts with intracellular domains of GluN2A and GluN2B; this interaction does not substantially alter NMDAR currents or Ca²⁺ influx but enables downstream pro-death signaling after extrasynaptic NMDAR activation. Disruption of the TRPM4-NMDAR interface reduces mitochondrial dysfunction and neuronal death in experimental stroke models [95].

By contrast, TRPM7 appears to contribute to ischemic neuronal injury mainly through a more parallel, glutamate-independent pathway involving delayed Ca²⁺ and Zn²⁺ overload, without a clearly identified NMDAR-scaffolding interaction [1, 78, 79].

TRPM-NMDAR interactions therefore range from direct molecular association to pathway convergence. Because direct ASIC1a-TRPM coupling has not been demonstrated, ASIC1a-TRPM relationships are treated here as functional convergence.

Cellular context of proton-sensitive signaling in the neurovascular unit

Although ASIC1a-mediated ischemic injury has been primarily characterized in neurons, ischemic acidosis occurs within a multicellular neurovascular environment where multiple cell types express proton-sensitive channels and receptors. For example, neurons contain endogenous enzymes for synthesizing bilirubin which can be released during oxygen glucose deprivation, independent of classical heme metabolism pathway in circulation [40]; bilirubin potentiates ASIC1a currents but directly activates TRPM2 channels [40]. Therefore, the functional consequences of pH changes are likely determined not only by channel properties but also by cell-type-specific expression patterns, subcellular localization and temporal responses after ischemic onset.

Neurons

Neurons represent the best-characterized cellular context for ASIC1a-mediated acidotoxicity. Neuronal ASIC1a activation contributes to proton-induced cation influx, including Na⁺ and functionally relevant Ca²⁺ entry, and is functionally coupled to excitotoxic NMDAR signaling during ischemia [19, 94, 98]. TRPM channels also influence neuronal vulnerability: TRPM2 contributes to oxidative-stress-associated ischemic injury [40], TRPM4 participates in NMDAR-dependent pro-death signaling [95], and TRPM7 promotes divalent-cation overload and anoxic neuronal death [1].

Astrocytes

Astrocytes regulate extracellular glutamate, K⁺ and ion homeostasis, provide metabolic support, and participate in inflammatory responses during ischemia [86]. Functional GluN2B-containing NMDARs have been reported in astrocytes after ischemia in vivo and anoxia in vitro, although their pathological roles remain less well defined than those of neuronal NMDARs [33]. Proton-sensitive channels and pH-regulatory transport systems are also present in astrocytes and may influence intracellular signaling and ion/volume homeostasis; however, their specific contribution to ASIC1a-associated ischemic injury networks remains incompletely defined [9, 86].

Microglia

Microglia are major regulators of post-ischemic inflammation, clearance of damaged tissue and subsequent remodeling [66]. TRPM2 is functionally expressed in microglia and is activated by oxidative signals such as H₂O₂ and ADP-ribose; its expression increases after experimental focal ischemia, supporting a role in oxidative-stress-associated microglial responses [18]. Although ASICs are expressed in microglia [9], direct evidence defining a cell-autonomous role for microglial ASIC1a in ischemic stroke remains limited.

Endothelial cells

Cerebral endothelial cells are central to blood-brain barrier (BBB) integrity, vascular homeostasis and inflammatory-cell trafficking. Endothelial NMDAR signaling can regulate BBB permeability, while endothelial TRPM2 promotes oxidative stress, Ca²⁺ overload, tight-junction disruption and BBB hyperpermeability after experimental ischemia [48, 101]. TRPM4 is also upregulated in ischemic vascular endothelium and has been implicated in capillary injury and impaired vascular integrity after stroke [41]. These observations establish that ischemia-associated ion-channel signaling extends beyond neurons, although the specific contribution of endothelial ASIC1a remains unresolved.

Cell-type-specific genetic and spatially resolved studies are needed to define how ASIC1a signaling converges with other proton-sensitive pathways across the neurovascular unit and whether therapeutic effects vary by cellular compartment and disease stage.

ASIC1a as a context-dependent integrating node

The available evidence supports a model in which ASIC1a acts as a context-dependent integrating node and may participate in bidirectional amplification within an ischemic injury network. The evidence supporting this model is asymmetric. Direct proton gating of ASIC1a [87, 94] and NMDAR/CaMKII-dependent regulation of ASIC1a [19] are comparatively well established; glutamate-dependent allosteric modulation [36] and bilirubin-associated signaling remain emerging mechanisms [35, 40]; and direct ASIC1a-TRPM coupling remains unproven, with current evidence supporting pathway-level convergence without a physical interaction.

However, a central unresolved question in ischemic neurobiology is how multiple proton-sensitive channels collectively interpret the dynamic changes in extracellular pH. Because acidosis activates ASIC1a while suppressing NMDAR and TRPM2 activity and enhancing TRPM7 currents, its network effects are context dependent. Crosstalk may be synergistic, as in reciprocal ASIC1a-NMDAR amplification, or functionally opposing/compensatory through proton-mediated suppression of NMDAR and TRPM2 activity. The balance likely varies across the ischemic core, penumbra and reperfusion phase.

From this perspective, ASIC1a may function as an integrating node within a broader acid-sensing network by linking extracellular proton accumulation to intracellular injury amplification. Key unresolved questions are whether these interactions occur within shared signaling microdomains, whether ASIC1a directly interfaces with TRPM channels, and how they vary across neuronal and non-neuronal populations during ischemia and reperfusion (Fig. 1).

Fig. 1.

Hypothetical model of ASIC1a as a node for ischemic neurotoxicity. During cerebral ischemia, energy failure and anaerobic metabolism produce extracellular acidosis, glutamate accumulation, while bilirubin-associated signaling may arise under specific pathological conditions. The resulting proton accumulation activates ASIC1a-containing channels, leading to Ca²⁺ influx and membrane depolarization, while intracellular Ca²⁺ and ROS accumulation and ADP-ribose-dependent TRPM2 activation contribute to downstream injury signaling. In parallel, ASIC1a functionally interacts with NMDAR-dependent signaling, including CaMKII-mediated ASIC1a phosphorylation, thereby linking acidotoxicity with glutamate-dependent excitotoxicity. ASIC1a-associated signaling may converge with TRPM2/4/7 and NMDAR-associated pathways within a broader injury network that contributes to neuronal death. Question marks denote relationships that remain incompletely defined

Therapeutic potential of targeting the ASIC1a node in ischemic stroke

ASIC1a-targeted blockers show neuroprotective effects across several preclinical stroke settings, including delayed-administration paradigms. These findings support therapeutic exploration of ASIC1a but remain preclinical; clinical benefit has not been established.

Current pharmacological toolbox

Current ASIC1a-targeted tools include peptide toxins, small molecules and biologic inhibitors.

Peptide toxins

PcTx1 is a potent and relatively selective ASIC1a blocker that has been evaluated in several preclinical stroke models [3, 46, 63]. In conscious hypertensive rats, central administration of PcTx1 2 h after stroke onset significantly reduced infarct volume and improved motor and neurological function [46]. More recent studies using intranasal delivery of PcTx1 in mice after 30-min middle cerebral artery occlusion (MCAO) showed that acute inhibition of ASIC1a not only reduced acute infarct volume but also produced sustained improvements in long-term behavioral recovery and functional outcomes assessed up to 28 days post-stroke [3].

The spider-venom peptide Hi1a has emerged as another potent preclinical ASIC1a inhibitor. Hi1a inhibits ASIC1a with sub-nanomolar potency and a slow off-rate [10]. A single low dose administered up to 8 h after stroke onset markedly reduced infarct volume in both cortical and striatal regions and significantly improved neurological and motor function in rodent MCAO models [10]. In an endothelin-1-induced focal stroke model in rats, Hi1a improved sensorimotor outcomes and reduced lesion-related deficits. However, in a filament-induced MCAO mouse model, Hi1a did not significantly improve functional outcomes, although a trend toward reduced infarct volume was observed [32]. These findings highlight that the therapeutic efficacy of ASIC1a inhibition may depend on experimental model, species, injury characteristics, and outcome assessment, emphasizing the need for further validation across clinically relevant stroke paradigms.

Mambalgins, a family of three-finger peptides isolated from black mamba venom, constitute another important class of ASIC1a inhibitors [49, 80]. Mambalgin-1 and mambalgin-2 potently inhibit ASIC1a homomers and ASIC1a-containing heteromers [13, 49]. Although direct stroke evidence remains limited, these pharmacological properties support their utility as mechanistic tools and their potential as lead peptides. Recent protein-engineering efforts have generated a mambalgin-1 analog with more efficient recombinant production and enhanced inhibitory potency at ASIC1-containing channels [58].

Thus, PcTx1 and Hi1a have model-dependent stroke evidence, whereas mambalgins remain primarily mechanistic leads.

Small-molecule inhibitors

Small-molecule ASIC1a inhibitors may offer translational advantages over peptide toxins, including the potential for systemic bioavailability, BBB penetration, oral or intravenous administration, and lower immunogenicity risk.

Earlier non-selective agents such as amiloride and its derivatives have served as important tool compounds [39, 98]. While amiloride inhibits multiple ASIC isoforms and shows neuroprotective effects in stroke models, its lack of selectivity and off-target effects on other ion channels have limited its therapeutic development.

The toxin-inspired compound 5b (C5b) is a highly selective ASIC1a inhibitor that blocks both homomeric ASIC1a and ASIC1a-containing heteromers in a pH-dependent manner [7]. C5b reduced acid-induced neuronal death in vitro, and intravenous administration reduced infarct volume and improved neurological function in transient MCAO models in vivo [67].

A recent study identified LK-2 through structure-based virtual screening of NMDAR-antagonist analogues [36]. LK-2 interacts with a putative glutamate-binding cavity within the extracellular domain of ASIC1a and selectively suppresses glutamate-dependent potentiation of ASIC currents (i.e. the PAM effect) while sparing basal activity of ASICs and NMDARs. In mouse MCAO models, systemic administration of LK-2 significantly reduces infarct volume and improves sensorimotor recovery, producing effects comparable to those observed in ASIC1a-knockout animals and in animals treated with the competitive glutamate-site antagonist CGS19755. Because CGS19755 targets glutamate-binding sites on both ASIC1a and NMDARs [36], this speaks of the pivotal importance of PAM effect by glutamate on neurotoxicity. Because LK-2 targets the allosteric glutamate site near the acid pocket, it may preferentially suppress pathological glutamate-dependent ASIC1a potentiation while sparing basal proton-gated activity. However, LK-2 was characterized in the same study that reported glutamate-dependent ASIC potentiation, it should be regarded as a proof-of-concept compound. Whether allosteric inhibition can achieve sufficient selectivity and therapeutic benefit across different ischemic contexts remains to be determined.

Further medicinal chemistry will be required to optimize the potency, selectivity, brain penetration and pharmacokinetic properties of these compounds.

Biologic and antibody-based approaches

One preclinical biologic approach is the human combinatorial antibody ASC06 (ASC06-IgG1), which was selected from a large phage-display library for its potent and selective blockade of ASIC1a [68]. This antibody binds with high affinity to the extracellular domain of ASIC1a, inhibits proton-gated currents with nanomolar potency, and protects cultured neurons from acid-induced death in cell-based assays. In a mouse MCAO model, intracerebroventricular infusion of ASC06-IgG1 3 h after ischemia onset significantly reduced infarct volume by approximately 23% compared with vehicle-treated controls [68]. Nanobody-based formats have also been explored as compact ASIC1a-binding reagents. An alpaca-derived nanobody, Nb.C1, specifically binds human ASIC1a (hASIC1a) and has been structurally characterized by cryo-electron microscopy at near-atomic resolution, defining its extracellular binding site on a closed/resting-state hASIC1a structure [93]. Fusion of Nb.C1 to PcTx1 increased the potency of PcTx1 in vitro [93]. In vivo stroke efficacy and brain delivery of Nb.C1-based constructs have not yet been established; their therapeutic relevance therefore remains prospective.

Therapeutic implications and challenges of targeting the ASIC1a node

Previous attempts to translate neuroprotective strategies, including NMDAR-directed approaches, have highlighted the difficulty of converting robust preclinical efficacy into clinical benefit across heterogeneous human stroke populations [29, 50, 51, 81]. In ESCAPE-NA1, nerinetide did not improve the primary functional outcome in the overall population undergoing endovascular thrombectomy, and ESCAPE-NEXT subsequently failed to demonstrate benefit in thrombectomy-treated patients who had not received thrombolysis [25, 26].

These experiences emphasize the need for appropriate patient selection, clinically relevant models, pharmacokinetic/pharmacodynamic characterization and demonstration of target engagement [43]. Accordingly, future ASIC1a programs should incorporate aged and comorbid models, large-animal validation, and explicit testing of interactions with thrombolysis and thrombectomy. ASIC1a-directed agents should initially be evaluated as adjuncts to, rather than replacements for, established reperfusion therapies. ASIC1a-targeted approaches may offer mechanistically distinct features, including activation under ischemia-associated acidic conditions [63, 94] and the possibility of pathway-selective modulation [36], but these potential advantages remain preclinical hypotheses requiring rigorous translational evaluation.

Because ASIC1a contributes to normal synaptic plasticity [34], fear-related learning [89], and ASIC family members are expressed and functionally implicated in multiple central and peripheral tissues [6, 8, 20, 69, 73], therapeutic strategies must preferentially suppress pathological ASIC1a activation while preserving physiological channel functions. This requirement is particularly important because constitutive or prolonged ASIC1a inhibition could theoretically interfere with learning, memory or other ASIC-dependent physiological processes. In addition, cross-reactivity with other ASIC isoforms expressed in cardiovascular, vascular, renal, or sensory tissues may increase the risk of off-target effects, especially in stroke patients with common comorbidities [6, 8, 20, 69, 73].

A second major challenge is effective central nervous system (CNS) delivery. Peptide toxins and large biologics often show limited blood-brain barrier permeability [60, 62], and many successful preclinical studies have relied on intracerebroventricular or intranasal administration [3, 46, 63], which may be difficult to translate directly into acute clinical stroke care. Future therapeutic development may benefit from brain-penetrant small molecules, optimized peptide or antibody formats, pH-responsive formulations that exploit ischemic tissue acidosis [57], or receptor-mediated transcytosis strategies for macromolecular delivery across the BBB [61, 64]. These approaches should ideally increase drug exposure in ischemic tissue while limiting systemic exposure and preserving normal ASIC1a function outside the pathological microenvironment.

A third challenge is achieving sufficient isoform specificity and optimal therapeutic timing. ASIC1a-targeted agents must discriminate not only between ASIC1a and other ASIC isoforms, but also between pathological pro-death ASIC1a signaling in ischemic tissue and physiological or context-dependent adaptive ASIC1a functions in non-ischemic brain regions [4]. Moreover, ischemic acidosis is spatially and temporally heterogeneous: severe acidosis predominates in the ischemic core, whereas milder acidosis may occur in the penumbra and during early reperfusion [57]. Complete or prolonged ASIC1a blockade in mildly acidic regions could theoretically interfere with adaptive responses or pH-dependent preconditioning mechanisms [64]. Thus, successful clinical translation will likely require inhibitors that are selective, brain-accessible, and context-dependent, with dosing strategies tailored to the severity, location, and timing of ischemic acidosis (Table 1).

Table 1.

ASIC1a-targeted therapeutic agents in experimental stroke models

Agent	Target/selectivity	Experimental model	Route/timing	Main reported effect	Key limitation
PcTx1	Peptide toxin; relatively selective ASIC1a inhibitor	Conscious hypertensive rat stroke model; mouse MCAO studies	Central administration; intranasal delivery in later work	Reduces infarct volume and improves neurological or behavioral outcomes	Peptide delivery, stability, BBB penetration, acute-care practicality
Hi1a	Spider-venom peptide; potent ASIC1a inhibitor	Rodent MCAO and endothelin-1-induced stroke models	Single-dose experimental administration; therapeutic window reported up to hours after onset	Reduces infarct volume and improves sensorimotor outcomes in selected models, but did not improve functional outcomes in filament MCAO mice	Model-dependent efficacy; delivery and biologic development challenges
Mambalgins	Snake-venom peptides inhibiting ASIC1-containing channels	Acid-induced injury and preclinical ASIC pharmacology; stroke relevance less mature than PcTx1/Hi1a	Experimental peptide administration	Blocks ASIC-mediated currents; potential neuroprotective rationale	Limited direct stroke-translation evidence; peptide stability and delivery issues
Amiloride	Nonselective ASIC/ENaC blocker	Acid-induced neuronal injury; rodent cerebral ischemia studies	Systemic or experimental delivery depending on model	Reduces acidotoxic injury but lacks ASIC1a specificity	Off-target ENaC/ion-channel effects; limited translational selectivity
C5b	Toxin-inspired small-molecule ASIC1a inhibitor	Acid-induced neuronal death, OGD, mouse MCAO	ICV or intravenous administration in mice	BBB-permeable; reduces infarct volume and improves neurological function in experimental ischemia	Early-stage preclinical lead; pharmacokinetics, safety, and reproducibility need further validation
LK-2	Antagonist of the glutamate-binding cavity on ASIC1a; blocks glutamate-dependent potentiation of ASIC currents while relatively sparing NMDARs	OGD neuronal injury model; mouse MCAO ischemic stroke model	Intraperitoneal administration, tested at reperfusion after transient MCAO	Reduces infarct volume and improves sensorimotor recovery	Preclinical evidence; broader safety, isoform selectivity, pharmacokinetics, and reproducibility require further validation
ASC06 (ASC06-IgG1)	Biologic inhibitor targeting the ASIC1a extracellular domain	Acid-induced neuronal death in vitro; mouse MCAO	Cell-based exposure; intracerebroventricular infusion 3 h after ischemia onset	Inhibits ASIC1a, protects cultured neurons, and reduces infarct volume by approximately 23% in mouse MCAO	Large biologic; BBB delivery remains a major barrier

ASIC1a acid-sensing ion channel 1a, BBB blood-brain barrier, ENaC epithelial Na⁺ channel, ICV intracerebroventricular, MCAO middle cerebral artery occlusion, OGD oxygen-glucose deprivation

Emerging strategies and future directions

Biomarker-guided strategies based on tissue acidosis, metabolic stress or evolving injury burden may help identify patients and therapeutic windows most appropriate for ASIC1a-targeted treatment. Human induced pluripotent stem cell-derived neurons, multicellular co-culture systems and brain organoids may further strengthen preclinical evaluation in human-relevant and genetically diverse ischemic models [21, 31, 74, 92].

Future mechanistic studies should also define ASIC1a within the broader pH-sensing architecture of the ischemic brain. Although ASIC1a-NMDAR functional coupling is comparatively well characterized, ASIC1a-TRPM convergence remains less direct, and PAC/TMEM206, K2P channels and proton-sensing GPCRs may contribute to cell-type-specific responses to acidosis. Spatial and temporal mapping of these pathways may clarify how ASIC1a participates in the broader proton-sensitive network and guide more selective therapeutic strategies.

Conclusion

Overall, ASIC1a is best viewed as a context-dependent integrating node that directly senses protons, functionally couples with NMDAR-dependent signaling, and converges more indirectly with TRPM-associated injury pathways. Its therapeutic potential remains compelling but unproven and warrants further mechanistic and clinical evaluation.

Author contributions

K.L. wrote the original draft and made both Figure 1; Table 1. K.L. and L.Y.W made multiple rounds of edits and revision. K.L. and L.Y.W. have reviewed the final version of the manuscript and agreed to submit it for publication.

Funding

This work is supported by the National Natural Science Foundation of China (82522024 to K.L.), Canadian Institutes of Health Research (202309PJT-505732 & 202403PJT-518084 to L.Y.W.), Natural Sciences and Engineering Research Council of Canada (RGPIN-2022-05425 to L.Y.W.) and the Canada Research Chairs Program (CRC-95-232466 to L.Y.W.)

Data availability

No datasets were generated or analysed during the current study.

Declarations

Competing interests

The authors declare no competing interests.