Introduction

Stroke is the second leading cause of death, the second most common cause of dementia and the most common cause of adult disabilities among elderlies [1–3]. Globally, it is responsible for nearly 5.5 million deaths every year, with 44 million disability-adjusted life-years lost which will increase up to 61 million in 2020 [2, 4]. Of this, ischemic stroke accounts for approximately 73 to 86 %, whereas hemorrhagic stroke is responsible for 8 to 18 % of the cases [5].

Brain ischemia initiates a cascade of pathological events that finally lead to irreversible neuronal damage [6]. It involves a complex series of biological and molecular mechanisms such as excitotoxicity, oxidative stress, inflammation, ionic imbalance, blood–brain barrier disruption and apoptosis [7, 8]. Among these mechanisms, excitotoxicity is notable. It is mainly a glutamate-mediated specific type of neurotoxicity and is the overlooked link between ischemic stroke and neuronal damage and cell death [9]. It stems from excessive accumulation of excitatory amino acids such as glutamate and leads to toxic increases in intracellular calcium [7, 10] which mediates neuronal damage in stroke [11].

miRNAs are small (~19 to 23 nucleotides) noncoding endogenous molecules that bind messenger RNAs (mRNA) and promote their degradation or repression of translation [12]. They have been found to be responsible for neuronal differentiation and development [6] and their role as mediators of silencing of post-transcriptional gene in pathological aspects of ischemic stroke has been proved [13]. Changes in the microRNAs (miRNA) transcriptome occur right after focal cerebral ischemia which indicates miRNAs role in the pathological cascades of ischemic stroke [13–15]. Alteration in miRNAs level after brain ischemic injury was first noticed in focal brain and forebrain ischemia and recent studies have assessed the significance of local changes in miRNA expression level [12, 16]. Furthermore, microRNAs aberrant, both up and down, regulation has been reported during cerebral ischemia and/or reperfusion [6, 17]. Also, it has been shown that in the aging brain, miRNAs play a neuroprotective role and are effective in controlling synaptic function and plasticity [18].

Although accumulating evidence shows that trafficking and phosphorylation of receptors have an important role in glutamate receptor’s regulation during and after stroke, the exact mechanisms behind remain poorly understood. Recent data indicate that specific miRNAs regulate glutamate receptors expression level [11, 19, 20] and excitotoxicity during stroke [21]. Among these miRNAs, miR-223, miR-181, miR-125a, miR-125b, miR-1000, miR-132 and miR-124a seem to play a role in glutamate receptors and their associated proteins regulation in neurons and astrocytes after an episode of stroke [11, 18, 19, 22, 23] and thus can influence post-stroke excitotoxicity [21]. Apart from excitotoxicity, changes in the miRNA transcriptome right after cerebral ischemia would affect acute phase molecular mechanisms linked to oxidative stress, inflammation, blood–brain barrier breakdown, cytoskeletal remodeling, metabolic changes, mitochondrial dysfunction, and apoptosis all of which are associated with post-stroke damage [13, 17].

Among these changes, excitotoxicity is an important trigger of tissue damage in both focal experimental, and clinical ischemia. Furthermore, glutamate receptors are the key mediators of death during excitotoxic injury. Besides as noted above, microRNAs play an important role in the etiology and pathophysiology of post-stroke excitotoxicity and [23, 24] these small regulatory RNAs will allow the assessment of their application as promising therapeutic targets to reduce the ischemic penumbra volume, post-ischemic brain damage and neurological dysfunction [25, 26]. As shown in Table 1, this review focuses on miRNAs function and their roles in post-ischemic stroke excitotoxicity.

Table 1 Summary of miRNAs and their relevant targets in post-ischemia excitotoxicity
Full size table

miRNAs: a novel therapeutic target in stroke medicine and the problems ahead

miRNAs could be considered a major therapeutic breakthrough in stroke medicine [6, 27] for two main reasons: (1) a single miRNA can control and influence many target genes, and (2) they can be inhibited both in vitro and in vivo [15].

In the CNS miRNAs can be manipulated by several methods [28]. Systemic delivery of modified oligonucleotides through either intravenous injection or cerebrospinal fluid infusion is one of the methods of miRNA manipulation. Sense, antisense miRNAs and artificial miRNAs have been used in mammalians with different effects on targets [29]. Lately, miRNA sponge, aimed to a specific miRNA, has been manipulated as an alternative to antisense oligonucleotides. These sponges could be delivered using viral vectors to allow for stable inhibition of specific, disease-related miRNAs [30]. Blood brain barrier (BBB) is a major problem to any therapy which involves systemic delivery of oligonucleotides. However, rabies-virus peptide solves this problem and helps oligonucleotides to cross the BBB and target neurons [29]. Viral methods also exist and are more target specific and avoid the problems of systemic delivery of oligonucleotides, however, toxicity-related problems and immunogenicity limit their clinical applications [28].

The use of miRNA technology to treat excitotoxic neuronal injury is also being pursued. Because a large number of miRNAs change between 3 h and 3 days after ischemic insult to the brain, expression of miRNAs has been also observed to change with reperfusion the early use of miRNA-based therapies could change the course of the disease [29, 31]. Thence, limitations such as poor in vivo stability, side effects including toxicity and immunogenicity and inappropriate biodistribution in miRNA-based therapies still exist and should be overcome before clinical application [28, 32].

Also, approved and appropriate animal models are important to validate miRNA drugs’ efficacy and unfortunately most animal models cannot completely recapitulate some aspects of human diseases such as stroke which limits the clinical usage of these models [33].

Role of miRNAs in glutamate-induced excitotoxicity in ischemic stroke

Excitotoxicity is the central missing link between ischemic insult in stroke and neuronal death [34–37]. Glutamate receptors have been hypothesized to play a major role in excitotoxicity-induced neuronal death in neurodegeneration and stroke [37, 38]. Following the activation of glutamate receptors during an ischemic insult, intracellular Ca2+ load increases, which in turn, activates Ca2+-dependent enzymes such as proteases, lipases, phosphatases and endonucleases [34], increases reactive nitrogen and oxygen species, disrupts cell membrane and mitochondria, induces fragmentation of DNA and initiates cell death cascades [34, 39].

Following cerebral ischemic episode, changes in the miRNA transcriptome indicate miRNAs role in the pathological events such as excitotoxicity which ensue the episode and might affect the disease outcome [12, 13, 17, 21, 31]. miRNAs affect post-transcriptional gene expression by interaction, repression or degradation, with their target mRNAs [40]. Although many studies have reported the suppressive role of miRNAs, some have also suggested their inductive role in post-translational level of gene expression [40–42].

miRNAs are also thought to be neuroprotective in stroke [43–45] and among them, glutamate receptors-specific miRNAs such as miR-223, miR-1000 and miR-125b are interesting targets for this purpose [11, 18, 22, 46].

Neuroprotective role of miR-223 in ischemic stroke by targeting glutamate receptors

Due to its various functions in the cell, miR-223 is important [47]. It is also expressed in central nervous system (CNS) and has been shown to be neuroprotective by targeting α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor subunit GluR2 and N-methyl-d-aspartate (NMDA) receptor subunit NR2B. Indeed, miR-223-mediated change in NMDA and AMPA function and composition controls neuronal excitatory response to glutamate in hippocampus [11]. Evidence shows that miR-223 is upregulated following the middle cerebral artery occlusion in a rat model of stroke [31, 48, 49]. Moreover, miR-223 expression is increased in circulating blood samples of patients with acute ischemic stroke and the severity and volume of infarct were lesser in patients who had more expression of miR-223 [21, 31]. Evidence shows that overexpression of miR-223 attenuates neuronal loss after excitotoxic insult [11] and its silencing is paralleled with increasing in GluR2 and NR2B subunits which subsequently increases miniature excitatory post-synaptic currents (mEPSCs) amplitude and decay time and enhances NMDA-mediated Ca2+ influx and results in nitric oxide production and excitotoxicity. This finally increases neuronal cell death following middle cerebral artery stroke in hippocampus or global brain ischemia (Table 1) [11, 22, 50]. Therefore, it appears that miR-223 direct delivery to the neuronal cell could be a novel therapeutic strategy to reduce excitotoxicity and neuronal cell death after ischemic stroke [51].

miR-1000-mediated neuroprotection by limiting post-stroke glutamate excitotoxicity

There is very limited information about miR-1000 and its role in post-stroke excitotoxicity. The glutamate transporter expression in vesicles (VGlut) is controlled presynaptically by miR-1000 in an activity-dependent fashion. This transporter retains glutamate in synaptic vesicles. It has been shown that genetically knocked down miR-1000 results in glutamate excitotoxicity and early-onset neuronal damage and death (Table 1) [18]. So, misregulation of miR-1000 after ischemic stroke may worsen excitotoxic damage in the damaged brain tissue of the affected patients. Whether the expression of this miRNA changes during ischemic stroke is not yet well understood and more studies are needed to evaluate its role in post-ischemic stroke excitotoxicity.

miR-125b exacerbates post-stroke excitotoxicity

miR-125 family includes 125a, 125b1 and 125b2 members and their overexpression has been found to have a proliferative effect on cells and diminish the apoptosis rate by proapoptotic genes downregulation [52]. Among these miRNAs, miR-125b is of increasing importance in the CNS. Although miR-125b is commonly expressed in brain and is upregulated during neurogenesis and neuronal differentiation [53], evidence shows that miR-125b overexpression in neurons leads to thinner spines and decreases mEPSCs amplitude [54]. Studies also suggest that NMDA receptors (NR2A) are the direct targets of this miRNA in hippocampal neurons and its overexpression or knock down results in NR2A up and down regulation, respectively (Table 1) [11, 22, 54]. On the other hand, the analysis of stroke patients shows that miR-125b manifests its maximum expression level within the acute phase of stroke in humans and it can be of diagnostic value [46]. So, it can be inferred from the studies that overexpression of miR-125b which occurs after ischemic stroke can upregulate NR2ARs, which promote cell death in mature cultures [55], and exacerbate post-stroke excitotoxicity. Further studies are needed to elucidate miR-125b diagnostic and therapeutic values in post-ischemic stroke excitotoxicity.

miR-124 effects on astroglial glutamate receptors control post-ischemic stroke excitotoxicity

miR-124 is selectively expressed in the CNS and its concentration is one hundred times higher in the CNS than other body tissues and is considered the most abundant CNS miRNA [56–58]. miR-124 has a key role in neuronal differentiation and function [6] and its aberrant expression contributes to pathological conditions of the CNS [59].

Evidence shows that neuronal exosomal miR-124a can control astroglial glutamate type I transporter/excitatory amino acid transporters 2 (GLT1/EAAT2) expression in astrocytes. It has been shown that transfer of this miRNA to astrocytes from neuronal exosomes increases GLT1/EAAT2 expression by altering the translation of GLT1 mRNA to its protein. This process is likely to be indirect is mediated through other factors in astrocytes. GLT1/EAAT2 is one of the important mechanisms of astrocyte protection against excitotoxicity which is mediated by rapid removal of excess glutamate from synaptic sites [58, 60].

Although findings about the role of miR-124 in stroke is controversial [61–63], some studies show that the expression level of miR-124 increases in ischemic penumbra in comparison to non-ischemic territory of middle cerebral artery occlusion model of mice and this miRNA is protective in neurons against the brain injury in experimental stroke models (Table 1) [59, 62, 64]. Also, it has been demonstrated that miR-124 could be detected in plasma after stroke and could be of diagnostic value [17, 65]. Therefore, it is highly plausible that miR-124 upregulation in penumbra can increase GLT1/EAAT2 expression in astrocytes and decrease glutamate levels in synapse and excitotoxicity and thus can be considered as a potential novel target for cerebral ischemia neuronal injuries.

miR-181 affects post-ischemic stroke excitotoxicity through astrocytes

The miR-181 family includes four members (miR-181a, miR-181b, miR-181c, and miR-181d) [60]. These family members, specifically miR-181a and miR-181b, are abundantly found in the brain [66] and their uncontrolled expression has been found to be responsible for some of important brain diseases [60]. miR-181 is one of the miRNAs whose level changes in the ischemic brain [67]. Evidence shows that following both global and focal ischemia miR-181a expression is upregulated and decreases in the penumbra and hippocampal dentate gyrus which are ischemia-resistant areas of the brain (Table 1) [16, 60, 67, 68]. These up and down regulations are associated with neuronal cell death in hippocampal CA1 and neuronal survival in dentate gyrus after ischemic injury, respectively [60]. Besides, anti-miR-181a diminishes infarct size in focal and hippocampal CA1 neuronal loss in global cerebral ischemia [60, 67, 69]. Thus, it seems to be a key mediator in the evolution of stroke-induced injury and its outcome [70].

It has been shown that direct delivery of anti-miR-181a to astrocytes is associated with increased Bcl-2, an important anti-apoptotic protein level which decreases oxidative stress and preserves GLT-1 receptors on astrocytes. This increases astrocytic glutamate uptake and limits excitotoxicity (Table 1) [69]. These changes have been found be cell-specific and do not happen in neurons after ischemia-like injuries in animal models [60].

Interestingly, studies have shown that GluA2/GluR2 subunit of AMPA receptors is also miR-181a target [23, 71, 72]. Upon miR-181a expression in neurons GluA2 surface expression and mEPSC frequency significantly decrease and neurons that overexpress this miRNA have fewer and smaller spines which proves miR-181a functional role in the synapse [20]. There are no data that show the relationship of miR-181a-related changes of AMPA receptors and excitotoxicity. Also, due to the different effects of miR-181 in neurons vs. astrocytes, future studies should assess the possible role of miR-181 in cell type specific responses and whether this miRNA influences post-stroke excitotoxicity by affecting glutamate receptors or not. Furthermore, although pretreatment with anti-miR-181a has been tried in some studies, investigation of the effect of application of this astrocyte-enriched miRNA after the onset of ischemia is a pivotal step for manipulating this idea in clinical use.

miR-132’s controversial role in post-ischemic stroke excitotoxic damage

MiR132 is a neuron-specific miRNA and its expression is enriched in the brain [73, 74]. It is mandatory for the proper development, morphogenesis and function of neurons and its dysregulation results in several neurological disorders. It has been lately classified as a ‘neurimmiR’, which acts within and between the neural and immune systems in which via targeting acetylcholinesterase, it reduces brain inflammation and increases acetylcholine [75–78]. This miRNA is specifically expressed in active synapses and its knock down causes impaired synapse formation. So, it plays a role in the long term activation of synapse and participates in memorization processes [79, 80].

The role of miR-132 in neuronal death has not yet well understood and some studies indicated that brain-derived neurotrophic factor overexpression during cerebral ischemic attacks, upregulates expression of miR-132 through the MAPK/ERK1/2 pathway in cultured cortical neurons, which in turn induces the expression of NR2A and NR2B subunits of NMDA and GluR1 subunit of AMPA receptors. Hence, the use of miR-132 antagomir in this case may have neuroprotective effects by suppressing glutamate receptor expression, thereby reducing excitotoxicity (Table 1) [23, 81].

Conclusion

miRNAs have emerged as mediators of post-transcriptional silencing of genes in ischemic stroke pathogenesis and pathology. These components exert their effects through changing the transcriptional level of glutamate receptors which are important in post-ischemic stroke excitotoxicity. A growing body of evidence shows that specific miRNAs such as miR-223, miR-181, miR-125a, miR-125b, miR-1000, miR-132 and miR-124a regulate glutamate receptors expression level and excitotoxicity during stroke.

Thus, these miRNAs may also be applied to positively affect stroke outcomes through modulating excitotoxicity. This suggests novel strategies in the treatment of ischemic stroke pathological cascade. However, until now, none of the presented miRNAs and their targets has been proved in the context of ischemic stroke treatment. Besides, the lack of the standardized technology used in different experimental platforms of ischemic stroke mandates the need for more studies in the future.