The Detrimental Role of Glial Acidification during Ischemia

Abstract

In this issue of Neuron, Beppu et al. (2014) demonstrate that glial acidification during ischemia contributes to neurotoxicity. Using a suite of in vivo optogenetic tools, the authors are able to exacerbate or attenuate neuronal damage during ischemia with glial acidification or alkalization, respectively.

Ischemic stroke is the second leading cause of death and disability in humans worldwide (Di Carlo, 2009), yet we still understand remarkably little about the mechanisms that perpetuate neuronal toxicity during these events. Ischemic episodes occur due to an interruption of blood supply either focally (artery blockage) or globally (cardiac arrest) and result in oxygen and glucose deprivation to the brain. The downstream consequences of the absence of these essential metabolic components are severe, with neuronal death occurring in as little as 5 min. Despite the prevalence and severe consequences of brain ischemia, there are currently few pharmacological interventions capable of providing significant neuroprotection during a stroke. This problem has motivated significant research into the cellular mechanisms that underlie ischemic brain damage in hopes of revealing new therapeutic avenues to address ischemic injury.

The best-understood mechanism of neuronal death after ischemia is known as “glutamate excitoxicity,” a term initially coined based on the observation that subcutaneous glutamate injections in mice produce intracranial brain lesions (Olney, 1969). Glutamate is the most prevalent neurotransmitter in the brain and acts on a variety of synaptic receptors in order to induce excitatory neuronal depolarization. Despite its relative abundance in the extracellular milieu of the brain, glutamate is toxic at high concentrations—a perilously thin margin between physiology and toxicity. Glutamate toxicity stems from the fact that neuronal glutamate receptors allow calcium ions to flow into the cell when activated by extracellular glutamate. Calcium influx is required for normal synaptic transmission and cellular signaling, but the excess glutamate released during ischemic injury results in intracellular Ca²⁺ concentrations that far exceed physiologic levels. At elevated concentrations, excess Ca²⁺ results in the overactivation of deleterious enzymes and signaling processes that impair neuronal function or initiate cell-death pathways (Szydlowska and Tymianski, 2010).

While a large body of literature continues to investigate the specific mechanisms implicated in the propagation of excitotoxic signaling, a fundamental question remains debated regarding the cellular source of glutamate and its mechanism of release during ischemia. In this issue of Neuron, elegant work by Beppu et al. (2014) suggests that glial acidosis after ischemia may act as a trigger for ensuing neurotoxicity. Furthermore, the authors provide evidence that interventions to alkalize glia during ischemic episodes are capable of attenuating neuronal injury.

Astrocytes, the most abundant population of nonneuronal cells in the brain, are first responders to ischemic stress. Furthermore, astrocytes have diverse and significant roles in glutamatergic signaling. For example, astrocytic glutamate transporters are the primary controllers of ambient extracellular glutamate. These qualities suggest that they may play an integral role in the glutamate-dependent excitotoxicity that accompanies ischemia. After ischemic injury, oxygen and glucose deprivation cause astrocytes to anaerobically metabolize stored glycogen, thus producing significant intracellular acidosis as a consequence of accumulating lactate concentrations. Beppu et al. (2014) asked whether this glial acidosis might contribute to ischemia-induced neuronal toxicity in the brain. In order to address this question they first loaded acute cerebellar slices with a pH-sensitive fluorescent dye and exposed the cells to oxygen and glucose deprivation (OGD). After OGD,Beppu et al. (2014) observed quick acidification of the Bergmann glia (a specialized subtype of astrocytes in the cerebellum). When paired with the observation that astrocytes play pivotal roles in neuronal survival and glutamate homeostasis, these initial results suggested that a dip in glial pH after OGD may serve as a trigger for subsequent neurotoxicity.

Within minutes after OGD onset, Purkinje cell neurons in the cerebellum exhibit an inward excitatory current that results from accumulating extracellular glutamate. This deleterious depolarization is almost entirely blocked with glutamate receptor blockers and is unaffected by the blockade of Na⁺ channels with tetrodotoxin (TTX), demonstrating that the increasing inward current is indeed a result of elevated extracellular glutamate levels and not merely a consequence of excess neuronal activity. Beppu et al. (2014) used whole-cell patch-clamp recordings in Purkinje cells to monitor this OGD-evoked excitotoxic drive during various experimental conditions to quantify the significance of the excitotoxic phenomenon. To bypass the OGD process and directly acidify glial cells during physiologically normal conditions,Beppu et al. (2014) utilized the optogenetic tool channelrhodopsin 2 (ChR2), a light-sensitive cation channel. Although ChR2 has been extensively used in the neuroscience field as a tool to temporally control neuronal membrane depolarization, this channel has increased permeability to H⁺ ions than it does to Na⁺ (Nagel et al., 2003) and can be essentially considered an optogenetic pH manipulator. Thus, the authors attempted to utilize ChR2 as a mechanism of inducing glial acidification in the absence of OGD. Beppu et al. (2014) transgenically expressed ChR2 exclusively in glia and observed intracellular glial acidification quickly after light activation of the channels. In the absence of OGD, this ChR2-mediated glial acidification produced the characteristic inward excitatory current in surrounding Purkinje cells, suggesting that glial acidification plays a causal role in the induction of the Purkinje cell excitotoxic drive.

Since glial acidosis contributes to glutamatergic excitotoxicity during OGD,Beppu et al. (2014) next wondered whether glial alkalization could suppress OGD neurotoxicity. In order to address this question, the authors utilized a transgenic mouse that expresses the light-activated outward proton pump archaerhodopsin (ArchT) under an astrocyte- specific promoter. Experiments with the pH-sensitive indicator confirmed that ArchT stimulation potently and specifically alkalized glial cells. Remarkably, when Beppu et al. (2014) stimulated ArchT during OGD conditions, the glutamate- mediated Purkinje cell inward current was significantly attenuated. In fact, the inward current dissipated to nearly the same level as observed during pharmacologic antagonization of the glutamate receptors.

In a final experiment,Beppu et al. (2014) addressed whether in vivo ArchT activation is neuroprotective during an ischemic event as predicted from their previous slice culture experiments. The authors induced an ischemic stroke with the commonly used photosensitive dye rose bengal, which causes thrombosis when focally excited with an optic fiber. Those animals lacking ArchT activation exhibited severe neuronal degeneration at the ischemic injury site, but the cohort receiving ArchT-mediated glial alkalization demonstrated significantly reduced infarct sizes. This is remarkable in vivo proof-of-principal evidence that strongly implicates the glial contribution to neurotoxicity following the onset of ischemia. Figure 1 illustrates a summary model of the cellular and molecular events that involve neuroglial interactions during ischemia.

Figure 1. A Model for Ischemia-Induced Excitotoxicity.

Without a steady blood supply, oxygen, glucose, and ATP levels decline quickly in the ischemic region. As a result, lactate begins to accumulate in astrocytes, acidifying the intracellular space. Glial acidification is correlated with neuronal cell death, and alkalization is capable of attenuating the full neurotoxic effects. Whether a mechanism for glial glutamate release exists in response to declining cytoplasmic pH remains to be elucidated. The absence of ATP prevents the energy-dependent maintenance of normal ionic gradients. Consequently, glutamate transporters that are typically responsible for clearing extracellular glutamate begin to run in reverse and expel glutamate from astrocytes and neurons. Increased extracellular glutamate levels activate glutamate receptors on neurons, which then leads to the toxic influx of calcium.

Though the elegant experiments performed by Beppu et al. (2014) demonstrate an important role of glial acidification during ischemia, several questions remain regarding the cellular source of glutamate release during OGD. The authors reach the provocative conclusion that glial acidification is directly upstream of glial glutamate release during ischemia. The mechanism of the proposed glial glutamate release, however, remains a mystery. To date, there is dearth of evidence that astrocytes contain the machinery required to release glutamate under physiologic conditions. Astrocytes lack the transport proteins required for vesicular glutamate release, and export of cytoplasmic glutamate seems unlikely considering the high quantity of astrocytic glutaminase that constantly degrades cytoplasmic glutamate. Overall, there is sparse data that under physiologic conditions glutamate concentrations accumulate sufficiently in astrocytes for glial glutamate release to be a significant contributor to rising extracellular concentrations of glutamate. Although Beppu et al. (2014) correlated glial acidification with neural excitatory current, this is an indirect observation that is not conclusive of astrocytic release of glutamate.

If ischemic glutamate release does not originate from astrocytes, what other mechanisms could explain such a phenomenon? One particularly pivotal event during brain ischemia is the inhibition of the Na⁺-K⁺ ATPase upon loss of cellular ATP, which decreases to 0%–25% of normal levels (Lipton, 1999). Inhibition of the Na⁺-K⁺ ATPase quickly results in a profound loss of intra- and extracellular ionic gradients. These gradients are essential to the function of glial and neuronal glutamate transporters. Under physiological conditions, the extracellular concentration of glutamate is carefully controlled by a family of high-affinity, Na⁺-dependent glutamate transporters, which couple the energetically unfavorable transport of glutamate with the energetically favorable cotransport of Na⁺, K⁺, and H⁺ down their respective electrochemical gradients. Under normal physiologic conditions, glutamate transporters are responsible for clearing synaptically released glutamate and maintaining the ambient glutamate concentration at subthreshold levels. During brain ischemia, the sudden ATP depletion and subsequent disruption of normal ion gradients can cause glutamate transporters to run in reverse, essentially expelling intracellular glutamate into the extracellular space. Previous in vitro and in vivo studies using glutamate transporter antagonists suggested that extracellular glutamate accumulation after ischemia is largely due to the reversal of these transporters in both neurons and astrocytes (Phillis et al., 2000; Rossi et al., 2000).

It is possible to reconcile older observations about glutamate transporter reversal with glial glutamate release during ischemic events. To address this scenario, several issues need to be resolved. First, ischemia-specific mechanisms of glial glutamate release—typically absent under physiologic conditions—need to be identified. For example, it is possible that glial glutamate concentrations rise significantly during ischemia, thus providing a substantial glutamate supply for export to the extracellular milieu. Next, the mechanism of how glutamate escapes from astrocytes in response to decreasing pH must be elucidated. Neuronal expression of the pH-sensitive neutral amino acid exchanger ASCT1 has been suggested to play a role in the extracellular glutamate increase caused by ischemia (Weiss et al., 2005), but astrocytes express ASCT1 at even higher levels (S.A.S., unpublished data). This transporter is therefore a strong candidate to be involved in the increase in extracellular glutamate levels upon astrocyte acidification. Future studies of ischemic injury in ASCT1-deficient mice could help to implicate these receptors if they are truly involved in such a mechanism.

The interest in the events that precipitate neuronal toxicity after ischemia stems from their potential as therapeutic targets. Over the past several decades, over one thousand clinical trials (O’Collins et al., 2006), many of which involved antagonists of glutamate receptor signaling, have failed in humans despite extremely promising histologic results in animals (Muir, 2006). Due to their significant side effects and lack of clinical efficacy, searching for therapeutics that antagonize glutamate receptor activation may have reached an impasse; instead, new mechanisms that precipitate neurotoxicity may provide a more fruitful avenue for therapeutic success. Accordingly, this work by Beppu et al. (2014) opens novel opportunities to consider small molecules that could ameliorate the effects of neurotoxicity during ischemia. Though they would have to be administered immediately after an ischemic event and delivery to a nonperfused brain region proves new obstacles, drugs that could prevent glial acidification may potently attenuate detrimental neuronal death. Especially when one considers the global epidemiology of stroke, any step forward in clarifying the mechanisms that propagate neuronal toxicity offers immense potential to impact worldwide morbidity. The work presented here by Beppu et al. (2014) provides an outstanding example of such a mechanistic leap—one that propels us ever closer to new therapies and interventions for such a prevalent and devastating condition.