The role of ASICS in cerebral ischemia

Abstract

Cerebral ischemia is a leading cause of death and long-term disabilities worldwide. Excessive intracellular Ca²⁺ accumulation in neurons has been considered essential for neuronal injury associated with cerebral ischemia. Although the involvement of glutamate receptors in neuronal Ca²⁺ accumulation and toxicity has been the subject of intensive investigation, inhibitors for these receptors showed little effect in clinical trials. Thus, additional Ca²⁺ toxicity pathway(s) must be involved. Acidosis is a common feature in cerebral ischemia and was known to cause brain injury. The mechanisms were, however, unclear. The finding that ASIC1a channels are highly enriched in brain neurons, their activation by ischemic acidosis, and their demonstrated Ca²⁺ permeability suggested a role for these channels in Ca²⁺ accumulation and neuronal injury associated with cerebral ischemia. Indeed, a number of studies have now provided solid evidence supporting the involvement of ASIC1a channel activation in ischemic brain injury.

Introduction

Ischemic stroke, or cerebral ischemia, is the third most common cause of death in most industrialized countries. Although major advances have occurred in the prevention of stroke during the past several decades, no effective treatment is now available. Current clinical practices for stroke patients utilize thrombolytic agent tissue plasminogen activator (tPA) to reopen the clotted vessels ¹. This approach, however, has very limited success due to a short therapeutic time window of 3h and side effect of intracranial hemorrhage. On the other hand, cell death is prominent following stroke. Therefore, the need for a continuous search of neuronal damage mechanisms and effective therapeutic strategies for neuroprotection remains high.

Although multiple pathways and biochemical changes contribute to ischemic brain injury, excessive intracellular Ca²⁺ accumulation and resultant toxicity has been considered essential in the pathology of cerebral ischemia ². In the resting conditions, free intracellular Ca²⁺ concentration ([Ca²⁺]_i) in neurons is maintained at nanomolar range. Following cerebral ischemia, however, [Ca²⁺]_i can rise to as high as several micromoles. Excessive accumulation of Ca²⁺ in neurons leads to uncontrolled activation of various enzymes causing breakdown of proteins, lipids and nucleic acids, and the destruction of neurons ^3-5. In addition, overloading Ca²⁺ in mitochondria can cause opening of mitochondria permeability transition pore (PTP), promoting apoptosis through release of cytochrome c and activation of caspases ⁶.

Ca²⁺ can enter neurons through various pathways, among which glutamate receptor-gated channels have received the most attention. Unfortunately, clinical trails targeting these channels have shown little effect in improving the outcome of cerebral ischemia ⁷. Multiple factors may have contributed to the failure of the trials. In particular, additional glutamate-independent Ca²⁺ entry and toxicity pathways must be considered.

Brain acidosis in cerebral ischemia

Acidosis, a condition characterized by too much acid in the tissue or body fluid, is one of the most common pathophysiological changes in the brain associated with acute neurological conditions such as cerebral ischemia ^8,9. In the ischemic core, for example, a rapid drop of brain pH to 6.5 or lower is frequently observed ^10,11. The lack of oxygen supply promotes anaerobic glycolysis which leads to increased production of lactic acid ¹¹. Accumulation of lactic acid, along with increased production of H⁺ from ATP hydrolysis, and release of H⁺ from presynaptic terminals ¹², contributes to the acid buildup in the brain. Acidosis has long been recognized to aggravate brain injury associated with cerebral ischemia ^8,9. However, the detailed mechanism(s) remained elusive, although a number of possibilities have been suggested, well before the role of ASICs was recognized ^8,13,14.

ASIC1a activation is involved in acidosis-mediated ischemic brain injury

Based on the evidence that ASIC1a subunits are highly expressed in brain neurons, their activation by pH drops to the level commonly seen in cerebral ischemia, and their permeability to Ca²⁺ and Na⁺, Xiong and colleagues tested the hypothesis that activation of ASIC1a channels is involved in neuronal Ca²⁺ accumulation and injury associated with cerebral ischemia ¹⁵. Using patch-clamp recording and fast-perfusion technique, large inward currents were recorded in cultured mouse cortical neurons in response to rapid perfusion of acidic solutions at pH levels relevant to cerebral ischemia. The acid-activated currents in cortical neurons were sensitive to non-specific ASIC blocker amiloride and partially inhibited by ASIC1a-specific inhibitor PcTX1, suggesting that the currents were mediated by ASIC1a-containing channels. Consistent with the presence of functional homomeric ASIC1a channels which are Ca²⁺-permeable ¹⁶, perfusion of acidic solution in these neurons increased intracellular Ca²⁺ concentration, even in the presence of blockers of voltage-gated Ca²⁺ channels and glutamate receptors. As expected, the acid-induced increase of intracellular Ca²⁺ was inhibited by PcTX1 and completely eliminated in neurons from ASIC1 knockout mice. Thus, acidosis can cause intracellular Ca²⁺ accumulation through activation of homomeric ASIC1a channels, although a secondary activation of other channels cannot be excluded ¹⁷.

To provide a link between ASIC1a activation and ischemic brain injury, both in vitro neuronal injury and in vivo cerebral ischemia models were employed. A brief (1 h) acid incubation, in the presence of blockers of glutamate receptors and voltage-gated Ca²⁺ channels, was able to induce substantial neuronal injury measured at 6h or 24h after acid treatment. This acid-induced, glutamate-independent neuronal injury was inhibited by amiloride or PcTX1, supporting the involvement of homomeric ASIC1a channels. Consistent with an essential role for ASIC1a subunit in acid injury, neurons cultured from the ASIC1 knockout mice were resistant to acid incubation ^15,18. Reducing the concentration of extracellular Ca²⁺, which decreases the driving force for Ca²⁺ entry, also ameliorated acid injury. To know whether ASIC1a-mediated injury can also take place in ischemic condition, acid-mediated neuronal injury was studied in the condition of oxygen glucose deprivation (OGD). OGD, in the presence of blockers of glutamate receptors and voltage-gated Ca²⁺ channels, further enhanced the acid-induced neuronal injury which was inhibited by amiloride and PcTX1. Thus, in vitro studies support a role for ASIC1a activation in acidosis-mediated, Ca²⁺-dependent, ischemic neuronal injury. Since a recent study showed that PcTX1 also inhibits heteromeric ASIC1a/ASIC2b channels in addition to homomeric ASIC1a channels ¹⁹, the potential contribution of heteromeric ASIC1a/ASIC2b channels to acidosis-mediated neuronal injury cannot be excluded.

Does activation of ASIC1a channels also play a role in ischemic brain injury in vivo? To answer this question, two sets of experiments were performed. The first experiment examined whether application of ASIC inhibitors reduces ischemic brain injury, and the second tested whether knockout of the ASIC1 gene renders the animal resistant to cerebral ischemia. Rodent model of focal ischemia, by middle cerebral artery occlusion (MCAO), was employed for both experiments. In rats and mice, intracerebroventricular injection of ASIC1a inhibitor PcTX1 reduced the infarct volume by up to 60%, measured at 24h after MCAO ^15,20. Similar to the pharmacological blockade, ASIC1 gene knockout provided a comparable degree of protection against ischemic brain injury ¹⁵. Remarkably, in contrast to most glutamate antagonists which have only a short time window of less than 1 hour ²¹, the protection by ASIC1a blockade showed a prolonged effective time window of ~5 hours ²⁰.

In addition to rodent cells, functional Ca²⁺-permeable ASICs have been recently described in human brain neurons ²². Activation of these channels, as expected, also contributes to acidosis-mediated neuronal injury. Thus, Ca²⁺-permeable ASIC1a channels represent a promising therapeutic target for human cerebral ischemia.

Ischemia-related signals enhance the activation of ASICs

Even though the results from both in vitro and in vivo pharmacological and molecular biological interventions clearly supported an important role for ASIC1a activation in neuronal injury associated with cerebral ischemia, several questions remain to be answered.

Under in vitro experimental conditions, the currents of most ASIC subtypes, particularly the homomeric ASIC1a channels, decay rapidly in the continuous presence of acidic pH ²³, a phenomena of channel desensitization. In addition, pre-exposure of these channels to small pH drops (e.g. from 7.4 to 7.2) that do not activate the channel also suppresses the channel activity in response to subsequent, large pH drops, a process termed “steady-state desensitization” ^24,25. Besides, the activities and/or expression of some ion channels could be dramatically down-regulated following the ischemia/hypoxia, as exemplified by n-methyl d-aspartate (NMDA) channels in hypoxic turtle brain ²⁶. Thus, whether a significant amount of ASIC1a current can be activated in ischemic conditions and whether the effects of ASIC1a activation (e.g. membrane depolarization and intracellular Ca²⁺ accumulation) could be long-lasting and detrimental are crucial in determining the pathological functions of these channels.

To this end, Xiong and colleagues first showed that OGD treatment can dramatically potentiate the activity of neuronal ASICs - the amplitude of the acid-activated current was enhanced while the decay of the current was reduced following 1h OGD ¹⁵. The overall outcome of these two effects would be a dramatically increased ASIC-mediated response, which might be translated into enlarged and longer-lasting intracellular Ca²⁺ elevation. What could be the underlying mechanisms responsible for the changes of electrophysiological property of ASICs observed after OGD or ischemia? This question was not directly answered by the studies of Xiong et al. However, later studies by different laboratories have provided important missing links. It is now known that various ischemia-related signals, e.g. arrachidonic acid, dynorphine, lactate, and spermine, can dramatically enhance the amplitude and/or reduce the desensitization of ASICs ²⁷ (Figure1).

Activation of ASIC1a channels and outcomes in non-ischemic and ischemic conditions. In non-ischemic condition, activation of ASIC1a channels and slight increase of cellular Ca²⁺ is involved in synaptic plasticity. In ischemic condition, ASIC1a activation is dramatically potentiated by various factors. Increased concentration of H⁺ per se activates more current. In addition, various ischemia-related signals not only increased the amplitude but also reduce the desensitization of the ASIC responses. Overload of neurons with Ca²⁺ leads to cell death.

(1) Arachidonic acid

Arachidonic acid (AA) is a polyunsaturated fatty acid present in the phospholipids of all cell membranes and one of the most abundant fatty acids in the brain. In addition to being involved in cellular signaling as a lipid second messenger ²⁸, AA plays important roles in pathological conditions including brain ischemia ^28,29. Following brain ischemia, the rise of [Ca²⁺]_i leads to the activation of phospholipase A2, resulting in increased production of lipid mediators including AA. Although the exact mechanisms are unclear, high concentrations of lipid mediators cause neurotoxicity ²⁸.

Based on demonstrated effects of AA on a variety of voltage-gated and ligand-gated ion channels, e.g. potentiation of NMDA channel currents ³⁰, Allen and Attwell tested the effect of AA on ASICs. In rat cerebellar Purkinje cells, bath perfusion of 5 or 10 μM arachidonic acid produced a large increase in the amplitude of the ASIC current. In addition to potentiating the peak amplitude, AA enhanced or induced an additional sustained component of the ASIC current ³¹. In heterologous expression systems, AA potentiates both homomeric ASIC1a and ASIC2a channels ³². Thus, promoting the activation of ASIC1a channels could be one of the mechanisms mediating the neurotoxicity of AA.

(2) CaMKII

Ca²⁺/calmodulin (CaM)-dependent protein kinase II (CaMKII) is the most abundant kinase isoform in the brain and a major mediator of the function of excitatory glutamate receptors. Influx of intracellular Ca²⁺ through glutamate receptors triggers an autophosphorylation of CaMKII and activation of the enzyme. Although activation of CaMKII plays a prominent role in synaptic plasticity, increased CaMKII activity has been implicated in the regulation of neuronal cell death after cerebral ischemia ³³.

Studies by Gao et al demonstrated a link between CaMKII activation and acidotoxic neuronal death ³⁴. They showed that global ischemia in rats results in an increased phosphorylation of ASIC1a at Ser478 and Ser479 by CaMKII. This phosphorylation sensitizes the channel to low pH, exacerbating cell death by allowing prolonged and increased Ca²⁺ entry.

Consistent with the report by Xiong and colleagues ¹⁵, Gao and colleagues observed an enhancement of ASIC currents by OGD. Inhibition of CaMKII with KN-93 or CaMKIINtide abolished the enhancement of ASIC currents, indicating an involvement of CaMKII. They further demonstrated an increased phosphorylation of ASIC1a protein after transient global ischemia, which was blocked by intracerebroventricular administration of KN-93 or CaMKIINtide. Pharmacological inhibition of CaMKII phosphorylation of ASIC1a with KN-93, or mutation of ASIC1a at Ser478 and Ser479, produced neuroprotection. Thus, phosphorylation of ASIC1a by CaMKII deteriorates ASIC-mediated neuronal injury in ischemia.

(3) Dynorphins

Dynorphins are endogenous neuropeptides abundantly expressed in the central nervous system (CNS). They are involved in a variety of physiologic functions. Under pathophysiological conditions where their levels are substantially elevated, these peptides can be neurotoxic, partially through glutamate receptors ³⁵. Recently, Sherwood and Askwith reported that, at high concentrations dynorphins such as big dynorphin potentiate acid-activated currents in mouse cortical neurons and in CHO cells expressing homomeric ASIC1a channels ³⁶. The potentiation of the ASIC1a activity was mediated through a reduction of the steady-state desensitization of these channels. In the absence of big dynorphine, pre-exposing neurons to conditioning pH of 7.0 completely desensitizes the channels, resulting in no responses to subsequent larger decrease in pH (e.g. to 5.0). In the presence of big dynorphine, however, ASIC1a currents were readily activated. As expected, big dynorphine enhanced ASIC1a-mediated neuronal injury during prolonged acidosis.

(4) Lactate

Back in 2001, Immke and McCleskey demonstrated that, in sensory neurons that innervate the heart and COS-7 cells transfected with AIC1a channels, addition of lactate, at the level seen in ischemia, dramatically increased the amplitude of the ASIC current activated by a moderate pH drop to ~7.0 ³⁷. The increase of the current amplitude was accompanied by a reduced current desensitization. Applications of lactate at pH values that do not activate ASICs caused no response. Thus, lactate acted by potentiating but not activating the ASICs. The effect of lactate persisted in excised cell-free membrane patches indicating the lack of second messenger involvement. Since lactate has the ability to chelate the divalent cations which have a modulatory role of various membrane receptors and ion channels ^38-40, it was logical to hypothesize that potentiation of the ASIC currents could be due to a chelation of Ca²⁺ and Mg²⁺ in the solution. Indeed, adjusting the concentrations of Ca²⁺ and Mg²⁺ eliminated the potentiating effect of lactate ³⁷. Similar to the cardiac sensory neurons, potentiation of the ASIC current by lactate has been reported in other neurons ³¹.

(5) Nitric oxide

Nitric oxide (NO) is an important reactive oxygen/nitrogen species which has a variety of physiological and pathological functions ⁴¹. During ischemia, intracellular Ca²⁺ overload leads to activation of the Ca²⁺-dependent neuronal form of nitric oxide synthase (nNOS), resulting in an increased production of NO ^42,43. NO can also be released by activated microglia ⁴⁴. Excessive NO production is known to increase neuronal injury ⁴⁴. Although the formation of a strong oxidant of peroxynitrite is likely involved in cell injury, other mechanisms cannot be excluded. Cadiou and colleagues reported that NO donor S-nitroso-N-acetylpenicillamine (SNAP) potentiates ASIC currents in DRG neurons and in CHO cells expressing ASIC subunits. Modulators of the cGMP/PKG pathway had no effect on the potentiation, but in excised patches from CHO cells expressing ASIC2a, the potentiation could be reversed by externally applying reducing agents. NO therefore has a direct external effect on ASICs, probably through oxidization of cysteine residues ⁴⁵.

(6) Proteases

Brain ischemia is accompanied by increased protease activity ⁴⁶. Following ischemia, blood-derived proteases have access to the interstitial space from a compromised blood-brain barrier ^46,47. Studies by Poirot and colleagues demonstrated an ASIC1a-specific modulation of the ASIC activity by serine proteases ⁴⁸. Exposure of cells to trypsin, for example, leads to a decreased ASIC1a current if the channel is activated by a pH drop from pH 7.4. However, if acidification occurs from a lower basal pH (e.g. 7.0), a condition pertinent to brain ischemia, protease exposure increases, rather than decreases, the ASIC1a activity ⁴⁸. Further studies demonstrate that trypsin modulates the ASIC1a function by cleaving this subunit at Arg-145, which is in the N-terminus of the extracellular loop overlapping with the PcTX1 binding site ⁴⁹.

(7) Spermine

Spermine is a polyvalent cation involved in various physiological processes. High concentration of spermine can induce neuronal depolarization and cytoplasmic Ca²⁺ overload, which may lead to neuronal damage ⁵⁰. Following ischemia, the activity of ornithine decarboxylase (ODC), a rate-limiting enzyme responsible for polyamine synthesis is enhanced, leading to elevated level of spermine ⁵¹. Although the modulation of NMDA receptor function ^52,53 might explain its neurotoxicity, several studies have yielded inconsistent results Babini and colleagues demonstrated that spermine potentiates the activities of ASICs ⁵⁶.

More recently, Duan and colleagues showed that extracellular spermine exacerbated ischemic neuronal injury through sensitization of ASIC1a channels to acidosis ⁵⁷. In addition to increasing channel activation, spermine reduced channel desensitization and accelerated recovery from desensitization in response to repeated acid stimulation. Thus, extracellular spermine contributes to ischemic neuronal injury, at least in part, by enhancing the activity of ASIC1a channels ⁵⁷.

Conclusion

Stroke or cerebral ischemia is a leading health problem worldwide. Unfortunately, there is still no effective treatment for stroke patients. Searching for new brain injury mechanisms and effective therapeutic strategies is a major challenge and priority. Acidosis is a primary feature associated with cerebral ischemia and is known to cause brain injury. The finding that activation of ASICs contributes to acidosis- and ischemia-induced intracellular Ca²⁺ accumulation and neuronal injury has suggested that these channels may represent potential new therapeutic targets for stroke intervention. In additional to developing pharmacological agents directly targeting ASICs, alternative strategies can be considered by targeting ischemia-related signals known to potentiate the activity of these channels (Table 1). Alternative neuroprotective strategies may also consider targeting the mechanisms and pathways that control the expression level of total protein and/or surface ASIC1a ^58,59.

Table 1. Ischemia-related endogenous modulators known to potentiate ASIC-mediated responses.

Modulator	Neurons or ASIC subunit tested effective	Modulation Site(s)	Functional outcome	References
Arachidonic acid	Cerebellar Purkinje cells Sensory neuron ASIC1a ASIC3	Unknown	Increased amplitude Reduced desensitization	32,60
CaMK II	Hippocampal neurons ASIC1a	Intracellular S478 & S479	Increased amplitude Increased injury	34
Dynorphine	Cortical neurons Hippocampal neurons ASIC1a ASIC1b	Extracellular PcTX1 site	Reduced steady-state desensitization Increased acid-injury	36
Lactate	Sensory neuron, Cerebellar Purkinje neurons ASIC1a ASIC3	Extracellular	Increased amplitude Reduced desensitization	37,60
Nitric oxide	DRG neurons Neuro2A cells ASIC1a ASIC1b ASIC2a ASIC3	Extracellular	Increased current amplitude Increased acid-injury	45,61
Protease	Hippocampal neurons ASIC1a	Extracellular PcTX1 site	Enhanced current amplitude activated from a baseline pH of 7.0 Increased recovery from desensitization	48
Spermine	Cortical neurons Hippocampal neurons ASIC1a	Extracellular PcTX1 site E219 & E242	Increased amplitude, Reduced desensitization Reduced steady-state desensitization Increased recovery from desensitization Increased acid-injury	56,57