New Neuroprotective Approaches for Stroke: Use of Decoy Peptides Targeting Specific Molecular Events

Stroke is characterized by the relatively rapid onset of neurological impairment, most often caused by the reduction or abrupt cessation of blood flow to the brain (cerebral ischemia). The interruption of cerebral blood flow is generally caused by a thrombus formed in an artery or the lodging of an embolus from a distant site. Behind heart disease and cancer, stroke is the third leading cause of death in the USA and is considered to be the most common cause of disability in adults [1–3]. Despite years of research and millions of dollars invested in finding an appropriate therapy, pharmacological intervention has made little, if any, impact in preventing neurological damage in patients suffering an acute stroke [4]. Thus, stroke clearly qualifies as a severe and prevalent national health problem with inadequate existing therapies.

It is now generally believed that irreversible neurological damage occurs over a period of hours following stroke [1]. During this time, a cascade of cellular and biochemical events, triggered by the initial vascular insult, leads to the production of neurotoxic mediators and, ultimately, a continuing wave of destruction of neural tissue. This cascade is believed to cause neurological damage that spreads beyond the initial necrotic core (i.e., the tissue directly supplied by the occluded vasculature) into the surrounding penumbral tissue [1]. Some therapeutic strategies for stroke, such as the use of the thrombolytic enzyme tissue plasminogen activator, have focused on removing the initial vascular obstruction. Another class of strategies has been directed at more ‘downstream‘ events, and aimed at blocking or reversing steps hypothesized to be crucial in the cascade of tissue destruction that occurs in the hours following the occlusion. The hope is that the latter type of intervention will halt or reduce neurological damage, even if administration is delayed until hours after stroke has occurred. This is important because clinical studies have shown that, most often, hours have elapsed by the time a stroke is diagnosed [2]. This implies that, in many clinical circumstances, removal of the vascular obstruction might not be accomplished early enough to prevent significant neurological impairment. Thus, pharmacological strategies for blocking later neurotoxic events might serve to increase the crucial ‘window of opportunity‘ for the treatment of stroke.

To facilitate design of such downstream therapeutic strategies, much research has been devoted to understanding the cascade of events that are responsible for brain-tissue destruction following the initial stroke, and on identifying key steps for potential intervention. Not surprisingly, this research has shown that the cascade is extremely complicated [5]. Nonetheless, several potential targets of intervention have been identified.

One leading hypothesis is that neurological damage in stroke results from a process known as excitotoxicity, whereby neurons die as a consequence of overstimulation of the receptors for the excitatory amino acid glutamate, in particular the NMDA receptors (NMDARs) by excessive release of glutamate following ischemia [6–8]. This has led to the development of glutamate-receptor antagonists that have been shown to be neuroprotective in experimental excitotoxicity models [9,10]. Some of these compounds have been investigated but failed in the clinic [11–14]. One molecular consequence of the activation of some types of glutamate receptors is a widespread increase in cytosolic calcium, which has itself been implicated in ischemic neurotoxicity [15,16].

One of the consequences of calcium influx is the activation of calpain, a neutral calcium-dependent protease that has been shown to participate in ischemia-induced cell death [17]. Moreover, calpain inhibitors have been found to be neuroprotective in animal models of stroke [18–22]. However, no calpain inhibitor has reached the clinic, and this is likely due to the relatively low specificity of existing calpain inhibitors, as well as the broad spectrum of calpain substrates and functions in which this family of enzymes is implicated. The nature of the calpain substrate(s) that are critically involved in ischemia-mediated neuronal death has also not yet been characterized, although it has recently been suggested that calpain-mediated cleavage of the Na⁺–Ca²⁺ exchanger participates in excitotoxic cell death [23–25]. Calpain also cleaves a number of antiapoptotic proteins, such as Bcl-2, Bcl-XL and Bid, and such cleavage could account for the role of calpain in apoptosis [26].

Metabotropic glutamate receptors (mGluRs) have also recently emerged as new players in ischemic neuronal death [27]. In particular, activation of group I mGluRs appears to be neuroprotective under various conditions [28]. The agonist of group I mGluRs, dihydroxyphenylglycine, prevented and reversed NO-induced neurotoxicity in primary cultures of hippocampal neurons [29], as well as the neurotoxic effects of hydrogen peroxide or platelet-activating factor in cortical neuronal cultures [30]. Activation of mGluRs also protected neurons from oxidative stress [31]. In organotypic hippocampal slice cultures, activation of mGluR1 before exposure to NMDA protected against NMDA-induced excitotoxicity [32].

Furthermore, selective blockade of mGluR1 was shown to exacerbate amyloid-β toxicity [33]. Recent studies indicated that the neuroprotective effects of mGluRI were mediated through the formation and activation of the mGluRI-Homer–PIKE-L signaling complex (PIKE-L is the longer isoform of PI3K-enhancer) [34]. Activation of PI3K and Akt by mGluRI was also reported in another independent study [35]. However, in cerebral ischemia, the exact role of mGluRI remains confusing; some studies indicate a protective role of antagonists [36], while others a protective role of agonists [37]. The neurotoxic effects of mGluR1 agonists might be due to their effects on cytosolic free Ca²⁺ as well as their stimulation of glutamate release [27]. Thus, it is not clear whether agonists or antagonists of mGluR1 should be tested in animal models of stroke.

We recently discovered a new mechanism that links all of the elements discussed above, which suggests the existence of a positive feedback loop that could play a critical role in ischemia-induced neuronal death and, potentially, in other forms of neurodegenerative diseases. In brief, we have found that NMDAR activation results in calpain-mediated truncation of the C-terminal domain of mGluR1α, one of the isoforms of mGluR1 [38]. As a result of this truncation, mGluR1α loses its neuroprotective signaling through the PI3K–Akt pathway, but it maintains normal calcium-signaling function through phospholipase C activation and inositol triphosphate formation.

In other words, truncated mGluR1α becomes an exclusively neurodegenerative receptor. In support of this idea we demonstrated that transfection of neurons in cultures with the calpain-mediated truncated mGluR1α increased glutamate toxicity. These results, therefore, account for the previously discussed discrepancy in the literature regarding the potential role of mGluR1 in ischemia-induced damage. They also indicate that the development of classic agonists or antagonists of mGluR1 into therapeutic agents for the treatment of stroke or other neurodegenerative diseases could be extremely challenging. We identified the site of calpain-mediated truncation, and we generated a small peptide consisting of a sequence of amino acids surrounding the cutting site and linked it to the HIV tat peptide, which has recently been demonstrated to represent a good strategy by which to carry peptides across cell membranes [39]. Remarkably, this tat–mGluR1 peptide prevented mGluR1α truncation and was neuroprotective against glutamate toxicity in neuronal cultures. It is important to stress that this peptide is not a general calpain inhibitor; in fact, in synaptic membrane fractions, the peptide had no effect on calpain-mediated degradation of spectrin, one of the most prevalent calpain substrates. Finally, systemic injection of this peptide in mice before kainic acid injection also prevented mGluR1α truncation and neurotoxicity. These results clearly suggest that prevention of mGluR1 truncation represents a major step in reducing excitotoxic damage. They also indicate that novel approaches that target the functions of mGluR1α, such as the one using a tat–mGluR1 peptide, might be extremely promising to reduce neuronal damage resulting from overstimulation of glutamate receptors. The target is downstream of NMDARs and, therefore, eliminates all of the problems associated with widespread inhibition of NMDARs. Such an approach also avoids the nonspecific side effects of calpain inhibitors, as it potentially allows the selective blockade of calpain-mediated truncation of mGluR1 without inhibiting overall calpain activity. As the peptide itself should be devoid of physiological activity, it might even be possible to envisage a chronic treatment for slowly developing neurodegenerative diseases. Although there are still some questions regarding the mechanisms of entry of the cargoes attached to the tat peptide, this peptide has proven to be a remarkable tool to facilitate the cell entry of a variety of molecules, from peptides to protein to oligonucleotides [39]. Moreover, several approaches are using a similar strategy (coupling of the tat peptide with various peptides or oligonucleotides) to provide protection against ischemia (peptide inhibitor of the Jun-C-terminal kinase [40]), increased neurotrophin signaling (peptide inhibitor of the protein tyrosine phosphatase [41]), and protection against radiation-induced apoptosis (BH4 peptide domain of the antiapoptotic protein Bcl-xL) [42]. We believe that the current tat–mGluR1 peptide provides a useful tool to obtain the proof-of-concept that the mechanism we identified might be involved in a variety of disorders in which excitotoxicity has been implicated. Future studies will be directed at testing the effects of this peptide in animal models of stroke and at identifying modifications of the peptide sequence that could improve its potency and selectivity.