The Role of ASIC1a in Epilepsy: A Potential Therapeutic Target.

BACKGROUND: Epilepsy represents one of the most common brain diseases among humans. Tissue acidosis is a common phenomenon in epileptogenic foci. Moreover, its role in epileptogenesis remains unclear. Acid-sensing ion channel-1a (ASIC1a) represents a potential way to assess new therapies. ASIC1a, mainly expressed in the mammalian brain, is a type of protein-gated cation channel. It has been shown to play an important role in the pathological mechanism of various diseases, including stroke, epilepsy, and multiple sclerosis.
METHODS: Data were collected from Web of Science, Medline, PubMed, through searching for these keywords: "Acid-sensing ion channels 1a" or "ASIC1a" and "epilepsy" or "seizure".
RESULTS: The role of ASIC1a in epilepsy remains controversial; it may represent a promising therapeutic target of epilepsy.
CONCLUSION: This review is intended to provide an overview of the structure, trafficking, and molecular mechanisms of ASIC1a in order to elucidate the role of ASIC1a in epilepsy further. Copyright© Bentham Science Publishers; For any queries, please email at epub@benthamscience.net.

INTRODUCTION

Epilepsy is one of the most common brain diseases. Currently, there are more than 70 million patients with epilepsy in the world, of which about 30% of patients have developed drug-refractory epilepsy, representing a large burden on patients, families, and society [1]. Characterized by a persistent tendency to induce spontaneous seizures, epilepsy has many neurobiological, cognitive, and psychosocial effects [2].

In all epilepsy patients, antiepileptic drugs can suppress up to two-thirds of seizures but will not change a long-term prognosis. Epilepsy, especially among those who continue to have seizures, is the main stressor that affects the quality of life, morbidity, and risk of premature death [3].

The pathogenesis of epilepsy is highly complex and has not yet been fully elucidated. Several neurobiological processes are considered potential targets in epilepsy.

Studies in molecular biology and genetics have shown that some types of epilepsy syndrome may be closely related to mutations in genes encoding ion channel proteins [4]. A recent genome-wide mega-analysis identifies the 21 most likely epilepsy genes, including 7 ion-channel genes (SCN1A, SCN2A, SCN3A, GABRA2, KCNN2, KCNAB1, and GRIK1) [5].

Recently, antiepileptic drugs (AEDs) remain the mainstay of epilepsy treatment [6]. Blocking voltage-gated sodium or calcium channels is a crucial strategy in the context of the mechanisms of action of AEDs such as carbamazepine, lamotrigine, gabapentin, and oxcarbazepine [7]. However, there still lacks a comprehensive understanding of the numerous ion channels involved in this pathology.

Acid-sensing ion channels (ASICs) represent an important class of proton receptors in the brain and belong to the voltage-independent,proton-gated cation channel family [8]. Among them, ASIC1a has become a particularly interesting candidate for a variety of physiological and pathological processes, including epilepsy, pain, fear, stroke, neurodegenerative diseases, and other physiological and pathological activities [9-12] due to its wide distribution patterns in the central nervous system [13] and the characteristics of calcium ion permeation [14].

Acidity involvement in the process of seizures and clinical application of acidity modulation has been demonstrated several decades ago [15]; for instance, acetazolamide, a carbonic anhydrase inhibitor, has been approved for the clinical treatment of epilepsy since 1953 [16]. However, the extent of this involvement is not yet fully understood [17].

ASIC1a can be activated in an acidic extracellular environment [13], which in turn leads to a large amount of Na+ and Ca2+ influx, promotes the release of related neurotransmitters, and even causes neuronal damage [18, 19].

Researchers believe that ASIC1a is closely related to seizures. Some scholars have found a significant association between ASIC1a variant alleles and temporal lobe epilepsy (TLE) in the Han population [20]. Furthermore, the expression level of ASIC1a in reactive astrocytes in the hippocampus of TLE patients and epileptic mice was significantly increased. In addition, activated ASIC1a significantly increased the concentration of calcium ions in reactive astrocytes, while knocking down or knocking out ASIC1a expression reduced spontaneous seizures [14]. In addition, the application of ASIC1a inhibitor amiloride in the rat model of pilocarpine-induced epilepsy can effectively suppress the occurrence of epilepsy [21].

In this review, we sought to provide an overview of the structure, trafficking, and molecular mechanisms of ASIC1a in order to further elucidate the role of ASIC1a in epilepsy.

ASIC1 STRUCTURE AND FUNCTION

In 1980, researchers discovered the presence of acid-induced currents in mammalian neurons [22]. A few months later, OA Krishtal reported a receptor for protons in the membrane of nerve cells [23]. In 1997, when R Waldmann cloned the first H+-gated ion channel (ASIC) [24], more and more scholars paid attention to the role of ASIC in neurological diseases.

ASICs belong to the Na+ channel/degenerin family of ion channels [24]. There are 6 subunits of ASICs which have been identified so far, 1a, 1b, 2a, 2b, 3, and 4, encoded by 4 genes (ACCN1-4) [25-28]. All subunits have two transmembrane domains, a large extracellular loop, and short intracellular N and C-terminals.

Different ASIC subtypes have different sensitivities to extracellular acidosis, in which ASIC1a and ASIC3 are the most sensitive subtypes [29], and ASIC1a desensitization occurs the fastest [24]. In the central nervous system, ASIC1a, ASIC2a, and ASIC2b are mainly expressed, and these three subunits are the main contributors to acid-sensitive currents, of which ASIC1a plays a key role [30].

ASICs are generally impermeable to Ca2+; however, the homomeric ASIC1a is the only family member reported to be permeable to Ca2+ [19]. The increase in intracellular Ca2+ concentration is crucial for many physiological and pathological processes of neurons, such as synaptic signaling and energy metabolism [31]. The Ca2+ conductance of homomeric ASIC1a reveals the specific functions that this complex performs in Ca2+-related physiological and pathological processes.

ASIC1a mainly exists in the hippocampus, amygdala, cingulate cortex, somatosensory cortex, and striatum [32-34]. In neurons, ASIC1a is located in the soma region and the dendritic spines [35]. High-density Ca2+-permeable ASIC1a is also found on the surface of the NG2 hippocampal glial cells [36].

MOLECULAR MECHANISMS OF ASIC1A

Trafficking of ASIC1a

In neurons, ASIC1a is mainly expressed in the cell membrane and cytoplasm; however, in astrocytes, ASIC1a is mainly distributed in the nucleus [37]. Interestingly, a study found ASIC1a highly expressed in the membrane and cytoplasm of reactive astrocytes in epileptic mice [14], pointing to the possible trafficking of ASIC1a in astrocytes during chronic epilepsy pathology. Zeng et al. revealed the molecular mechanisms of ASIC1a internalization and its protective function in acidosis-induced neuron death [38]. In the chronic pain model, blocking ASIC1a transport to the membrane reduces the sensitivity of pain [34], indicating that the function of ASIC1a is closely related to its dynamic transport.

Jing et al. found that the first transmembrane domain (TM1) of ASIC1a plays a critical role in the synaptic targeting and surface expression of ASIC1a [39]. N-glycosylation on Asn393 has also been shown to participate in the regulation of ASIC1a dendritic trafficking [40]. In the brain, ASIC2a facilitates ASIC1a surface trafficking [41]. In pulmonary arterial smooth muscle cells, the activation of RhoA promotes ASIC1a-mediated Ca2+ influx by promoting ASIC1a plasma membrane localization [42].

Recently, Song et al. revealed the movement of hASIC1a on the surface of excitatory synapses by using a combination of single-particle tracking and anti-hASIC1a extracellular domain antibodies [43]. This toolbox provides a new method for studying the role of ASIC1a in pathological processes and exploring new clinical treatments.

ASIC1a in Synaptic Plasticity

Synaptic plasticity refers to the modification of the strength or efficacy of synaptic transmission in response to activity, which includes two main forms, long-term potentiation (LTP) and long-term depression (LTD), that are closely related to ASIC1a [44].

Some observations indicate that ASIC1a contributes to synaptic plasticity in different regions of the brain. In the striatum, ASIC1a is involved in regulating the synaptic plasticity of striatal medium spiny neurons, which is critical for striatum-dependent memory and procedural learning [45]. In the hippocampal-prefrontal circuitry, ASIC1a contributes to pain-related cortical plasticity [46]. In the anterior cingulate cortex, ASIC1a participates in modulating pain hypersensitivity by mediating synaptic potentiation critically [47].

It is worth noting that study results on the role of ASIC1a in synaptic plasticity vary according to experimental conditions, including the recording method, ASIC1a inhibitor concentration, and observation area [48, 49]. An objective method is needed to assay the synaptic plasticity mechanisms in the brain.

NMDAR-CaMKII/ASIC1a Coupling

A study on hippocampal slices shows that ASIC1a is localized on dendritic spines, and affects the density of dendritic spines by adjusting intracellular Ca2+ concentrations and calmodulin-dependent protein II (CaMKII) phosphorylation [35]. Spines are the postsynaptic site of most excitatory synapses. Further research indicates that ASIC1a knockout increases the density of dendritic spines, but is detrimental to the spine morphology and post-synaptic structure, which largely rely on inhibiting the activation of CaMKII and extracellular signal-regulated protein kinases [45].

ASIC1a currents form part of the excitatory post-synaptic currents [50]. During synaptic transmission, protons are released from synaptic vesicles, resulting in the acidification of synaptic clefts, thereby activating ASIC1a-dependent currents. These currents can cause action potentials and induce calcium influx, independent of glutamatergic currents [51].

Several findings indicate that ASIC1a and N-methyl-D-aspartate receptor (NMDAR) has functional interactions in regulating hippocampal synaptic plasticity [48, 49]. ASIC1a facilitates the activation of the NMDAR, and the knockout of ASIC1a can restrain the activation of NMDAR at excitatory synapses in mice, which largely depends on the phosphorylation of CaMKII [45]. Specifically, ASIC1a participates in synaptic plasticity through NMDAR-CaMKII/ASIC1a coupling. Furthermore, the NMDAR-CaMKII cascade is functionally coupled to ASIC1a and exacerbates neuronal death caused by acidosis [52, 53].

It has been proven that the activation of ASIC1a facilitates NMDAR currents [54]. Interestingly, extracellular acidosis is reported to decrease the opening frequency of NMDAR [55], thereby suppressing seizures [56], and NMDAR antagonists attenuated the acid’s effect on epileptiform activity in brain slices. The role of ASIC1a/NMDAR-CaMKII coupling in epilepsy still warrants further experimental study.

Additionally, several studies showed that some other molecular partners are involved in the role of ASIC1a in synaptic plasticity. For instance, Mango et al. proved that ASIC1a is involved in metabotropic glutamate receptor-induced LTD [57].

ASIC1a in Mitochondria

In the past, it was thought that ASIC1a only existed in the plasma membrane. However, a study in 2013 indicated that ASIC1a protein is also present in the mitochondria of mouse cortical neurons and may serve as an important regulator of mitochondrial permeability transition pores [58]. Since several studies have demonstrated that mtASIC1a plays an important role in acid-induced neuronal injury.

In 2014, Li et al. demonstrated that extracellular acidosis induces cytochrome C transport from mitochondria to cytoplasm, contributing to the apoptosis of endplate chondrocytes; this process can be inhibited by ASIC1A-siRNA or psalmotoxin 1(PcTX1) [59].

Liu et al. designed a DNA nanoprobe loaded with a Ca2+ fluorescent probe, pH response, internal reference, and mitochondria-targeting molecules for real-time imaging and simultaneous quantification of Ca2+ and pH. Using this probe, they found that cytoplasmic acidosis activates ASIC1a in the mitochondrial membrane, leading to mitochondrial Ca2+ overload and pH abnormalities. Moreover, PcTx1 provides effective protection against neuronal damage in this process [60].

In 2020, Ivana et al. found that in primary mouse cortical neurons, a mild extracellular pH decrease can activate ASIC1a, which triggers cytoplasmic Na+ and Ca2+ waves that propagate to the mitochondria, enhance mitochondrial respiration, and cause mitochondrial Na + signaling. This process can be blocked by PcTx1 or ASIC1a gene knockout [61].

Conductance-independent Functions of ASIC1a

Traditionally, the ion-conducting function of ASIC1a, especially Ca2+ influx leading to Ca2+ overload, was considered be the main reason for neuronal acidosis [18, 19, 62]. However, this hypothesis is not sufficient to explain the mismatch in the degree of ASIC1a-mediated Ca2+ currents and neuronal damage [63]. Recently, evidence has demonstrated that ASIC1a mediates acidic neuronal necroptosis via the disruption of auto-inhibition, independently of the channel’s ion-conducting function [64].

Wang et al. reported a proposed mechanism of conformational signaling of ASIC1a in acidotoxicity [65]. Under normal conditions, the death motif of ASIC1a C terminal (CT) is masked by a cytoplasmic N terminal (NT) and CT interaction. Acidosis disrupts the interaction and unmasks the CT death motif, which in turn recruits receptor-interacting serine/threonine-protein kinase 1 (RIPK1) and activates it by phosphorylation, resulting in neuronal necroptosis. This transition is facilitated and stabilized by N-ethylmaleimide-sensitive fusion ATPase binding to ASIC1a-NT.

Studying the auto-inhibition mechanism of ASIC1a provides a new perspective to inform a therapeutic target for ASIC1a-related illnesses. Inhibitory peptides which specifically target the death motif without affecting its ionic conduction, have a huge therapeutic potential. For instance, peptide NT1–20 can significantly reduce infarct volumes in vivo ischemia experiments [65].

ASIC1a AND EPILEPSY

The exact role of ASICs in seizure generation, propagation, and termination has remained controversial [66]. ASIC1a is believed to be involved in the termination of seizures induced by acidosis [67]. However, research confirms that amiloride, a nonspecific blocker of ASIC1a, may have anticonvulsant effects in animals through mechanisms related to acid-sensing ion channels [68], indicating that ASIC1a contributes to seizures.

Table represents the data on ASIC1a in epilepsy model animals experiments in order to more intuitively understand the controversial role of ASIC1a in epilepsy.

ASIC1a in Acidosis-mediated Seizure Self-limitation

Seizures reduce brain pH [69], and extracellular low pH suppresses seizures induced by low Mg2+ in brain slices [56]. As a major contributor to acid-induced currents, what role does ASIC1a play in stopping seizures?.

ASIC1a mRNA was reported to be decreased in the hippocampal CA1-2 region of rats with status epilepticus [70]. A lower expression and altered cellular distribution of ASIC1a were also detected in cortical lesions of patients with focal cortical dysplasia (FCD), a well-recognized cause of medically intractable epilepsy [71]. Previous studies confirm that disrupting ASIC1a can eliminate acidosis-induced currents [33, 72]; in other words, in neurons, acidosis depends on ASIC1a to generate currents. Knockout of the gene encoding ASIC1a in mice increased the duration of seizures, without affecting the seizure threshold. Consistently, the overexpression of ASIC1A in mice has the opposite effect and suppresses seizures [67]. These results suggest that ASIC1a plays an important role in the self-limiting nature of seizures.

ASIC1a Contributes to Seizure Activity

Interestingly, although acidosis is involved in stopping seizures, some studies have found that acidosis injures neurons through a mechanism that depends on the membrane receptor ASIC1a [18].

Amiloride can play a protective role by increasing seizure threshold on seizures induced by increasing current electroshock or pentylenetetrazol (PTZ) in mice [7, 73]. Additionally, an experimental demonstration that amiloride has similar protective effects on PTZ-induced status epilepticus in mice [74]. In rats, the first episode of limbic seizures and the occurrence of status epilepticus after pilocarpine was significantly delayed by amiloride [75].

Since amiloride can also block other ion channels, such as the sodium–proton exchanger and epithelial sodium channel [76], more work needs to be done to determine whether the function of amiloride is directly attributable to ASICs.

The use of the ASIC1a selective blocker PcTX1 and ASIC1a gene knockout made it possible to assess the exact role of ASIC1a in epilepsy.

Yang Feng found that ASIC1a is highly expressed in hippocampal reactive astrocytes in TLE patients and epileptic mice [20]. Inhibiting the expression of astrocytic ASIC1a can reduce spontaneous seizures, and restoring astrocytic ASIC1a expression in the ASIC1a knockout mouse can increase the frequency of spontaneous seizures, suggesting that ASIC1a in astrocytes contributes to the development of chronic epileptogenesis [14].

During the development of chronic epileptogenesis, reactive astrocyte hyperplasia, and degeneration of GABAergic interneurons dominate [77-79]. ASIC1a can be activated simultaneously by acidosis in both neurons and astrocytes, and the impact of ASIC1a on the neural network depends on the comprehensive effects on different neuronal cells.

A human genetics study in 2011 has demonstrated an association between single nucleotide polymorphisms in ASIC1 and TLE in a Han Chinese population [20].

Controversial Role of ASIC1a in Epilepsy

Why does ASIC1a seem to play a paradoxical role in epilepsy? Here we provide a number of possible explanations.

First of all, differences between various types of epilepsy should be taken into consideration. As mentioned above, the expression of ASIC1a increases in TLE patients [14] and decreases in FCD patients [71]. More specifically, the expression of ASIC1a is significantly lower in FCD type II than in type I [71].

Per Table , most research has focused on chemoconvulsant insult models and kindling models. For chemoconvulsant insult models, the most notable feature is the damage in the hippocampal formation, which is rarely found in kindling models [80]. In addition, precise molecular signaling cascades and neurotransmitter dynamics in epileptogenesis may vary with different epilepsy models. For instance, the low Mg2+ model is sensitive to gamma-aminobutyric acid (GABA) modulators [81], and the pilocarpine-induced model is sensitive to rapamycin [82].

In summary, ASIC1a might be an attractive new target for the treatment of epilepsy, although further studies are needed to more precisely define certain patient subgroups and distinct treatment targets.

SUMMARY AND OUTLOOK

Being the most sensitive channel to pH change in the central nervous system, ASIC1a has been widely researched in various physiological and pathological processes and has shown great potential as a therapeutic target.

Endogenous modulators, animal toxins, herb extracts, microbial extracts, synthetic chemicals, and ASIC1a-Knockout Mice are widely used to investigate ASIC1a [84-87].

In addition to the amiloride and PcTx1 mentioned above, more and more compounds targeting ASIC1 have been discovered and used in related research and clinical trials. In Table , we list ASIC1a modulators used in animal neurological disease models or clinical studies in humans.

Notably, emerging evidence indicates that ASIC1a may offer a therapeutic target in migraine, especially among migraine patients with aura [88-90]. In addition, migraine, one of the common comorbidities of epilepsy, has been proven to have potential interactions with epilepsy in both basic and clinical research studies [91-93]. Furthermore, the injection of PcTx1 and ASIC1a-knockout has been proven to reduce depression [32], a common comorbidity in patients with epilepsy.

Per this Table , research on epilepsy remains relatively rare. Additionally, the effect of amiloride may be due to characteristics unrelated to the ASIC1a blockade. For instance, in a study using amiloride and PcTx1 at the same time, the protective effect of amiloride was significantly better than that of PcTx1 [94]. Elucidating the precise role of ASIC1a in epilepsy and the development of antiepileptic drugs targeting ASIC1a still warrants more and more precise studies.

Determining the resting state structures of homotrimeric ASIC1a at a high pH [95] and the structure of the ASIC1a in a complex with the gating modifier PcTx1 [96] laid forth more possibilities for exploring new drugs. Additionally, the role of the lipid bilayer in ASICs gating is believed to be another interesting field [97], which could be linked to the different functional properties of ASICs in different cells.

Due to deficiencies in the specificity and biological stability of small molecules and toxin peptide inhibitors, monoclonal antibody drugs of ASIC1a show strong advantages and rapid development trends in the field of biomedicine. As mentioned above,the exact role of ASIC1a in seizures remains controversial. To clarify whether ASIC1a is protective or damaging in epilepsy, more knowledge on the magnitude and duration of acidosis induced by seizures, the activation of ASIC1a in different areas of the brain, and the other factors interacting with ASIC1a is needed.

An earlier study has indicated that, in the rat, inhibitory interneurons possessed larger acidosis-induced currents than excitatory neurons [98]. Proton-gated current densities were reported to be larger in interneurons than in pyramidal neurons [67]. It can be inferred from this work that ASIC1a might inhibit seizure activity by activating inhibitory interneuron and increasing inhibitory tone.

A recent study has shown that the activation of ASIC1a will trigger a large influx of Na+ and Ca2+, which will be further enhanced by voltage-gated Na+ and voltage-gated Ca2+ channels [61, 99]. These subsequent Ca2+ influxes were inhibited upon activation of sigma-1 receptor [100]. The interactions between ASIC1a and other ion channels and receptors are far more complex than we have already identified. More accurate recording methods and more pharmacological-related studies are warranted to unlock this complex regulatory network.

Compared with mouse cortical tissues, human cortical tissues had a higher level of ASIC1a in the membrane and medicated stronger acid-induced responses [101]; more studies based on human neurons or resected tissues of patients with epilepsy are warranted to further assess the pathological mechanisms of ASIC1a in human epilepsy.

CONCLUSION

Cell and animal studies have shown that ASIC1a may play different roles in seizures and the termination of epilepsy. The physiological and pathological activities of ASIC1a are closely related to the location and transport of ASIC1a on cell membranes and mitochondria. Various ion channels and receptors, such as voltage-gated calcium channels and NMDARs are also involved. More data has demonstrated that drugs targeting ASIC1a show great promise in the context of various neurological diseases. A deeper understanding of the role of ASIC1a in epilepsy will lay forth the possibility of ASIC1a becoming a new target for epilepsy treatment.

Table 1

Relevant ASIC1a data in different animal studies of epilepsy.

	Model	Function	Refs.
Amiloride	mice with seizures induced by increasing current electroshock (ICES) /pentylenetetrazol (PTZ)	Amiloride produced a dose-dependent increase in seizure threshold	[73, 74]
Amiloride	maximal electroshock (MES)-induced seizures in mice	Amiloride enhances the antiepileptic effects of valproate and topiramate	[83]
Amiloride	Rats with seizures induced by pilocarpine	Amiloride reduced the frequency of discharge in 60~90 min after injection	[21, 75]
rAAV-ASIC1a-shRNA	wide type TLE mouse model induced by pilocarpine	down-regulation of astrocytic ASIC1a expression decreased the frequency of spontaneous seizures	[14]
rAAV-ASIC1a	ASIC1a knock-out TLE mouse model induced by pilocarpine	restored astrocytic ASIC1a expression increased spontaneous seizures frequency	[14]
PCTX1	kainite injection	PcTx1 increased the incidence of continuous generalized tonic–clonic seizures	[67]
ASIC1a-/-	Kainite/PTZ injection	loss of ASIC1a reduced post–ictal depression, increased seizure severity but did not alter seizure threshold	[67]
overexpressing ASIC1a	Kainite/PTZ injection	reduced seizure severity	[67]

Table 2

ASIC1a modulators used in animal neurological disease models or clinical studies in humans.

	Model	Function	Refs.
Amiloride	migraine models in rats	inhibited trigeminal activation and cortical spreading depression	[88]
Amiloride	7 patients with intractable migraine with aura	reduced both frequencies of aura and headache severity in 4 of 7 patients	[88]
ASIC1a-/-	mouse with tail suspension test and forced swim test	reduced depression-related behavior	[32]
AAV-ASIC1a	ASIC1a knock-out mouse with tail suspension test and forced swim test	eliminated the anti-depression effect in ASIC1a-knockout mice	[32]
Amiloride	models of inflammatory pain in rats	reduced nociceptive behaviors	[102, 103]
Amiloride and its analogs	mouse model of middle cerebral artery occlusion-induced focal ischemia	decreased infarct volume	[104]
Amiloride	14 patients with primary progressive multiple sclerosis	reduced major clinically relevant white matter and deep grey matter structures	[105]
PcTx1	mouse model of Parkinson's disease	attenuated the deficits in striatal DAT binding and dopamine	[94]
Amiloride	mouse model of Parkinson's disease	protect substantia nigra neurons	[94]
PcTx1	model of stroke-induced in conscious spontaneously hypertensive rats	reduced cortical and striatal infarct volumes	[106]
APETx2	rat model of inflammatory pain	evoked peripheral antihyperalgesia partially	[107]
Diminazene	rat models of inflammatory pain	evoked peripheral antihyperalgesia effect	[107]
PPC-5060	healthy males with intraluminal esophageal acid perfusions	reduced sensitization to mechanical stimulation of the esophagus	[108]
NS-383	rat models of pain	reversed pathological painlike behaviors	[109]
ASC06-IgG1	mouse stroke model	protect cells from death after an ischemic stroke	[110]