Activation of NR2A Receptors Induces Ischemic Tolerance through CREB Signaling

Abstract

Previous exposure to a nonlethal ischemic insult protects the brain against subsequent harmful ischemia. N-methyl-D-aspartate (NMDA) receptors are a highly studied target of neuroprotection after ischemia. Recently, NMDA receptor subtypes were implicated in neuronal survival and death. We focused on the contribution of NR2A and cyclic-AMP response element (CRE)-binding protein (CREB) signaling to ischemic tolerance using primary cortical neurons. Ischemia in vitro was modeled by oxygen–glucose deprivation (OGD). Ischemic tolerance was induced by applying 45-mins OGD 24 h before 180-mins OGD. Sublethal OGD also induced cross-tolerance against lethal glutamate and hydrogen peroxide. After sublethal OGD, expression of phosphorylated CREB and CRE transcriptional activity were significantly increased. When CRE activity was inhibited by CREB-S133A, a mutant CREB, ischemic tolerance was abolished. Inhibiting NR2A using NVP-AAM077 attenuated preconditioning-induced neuroprotection and correlated with decreased CRE activity levels. Activating NR2A using bicuculline and 4-aminopiridine induced resistance to lethal ischemia accompanied by elevated CRE activity levels, and this effect was abolished by NVP-AAM077. Elevated brain-derived neurotrophic factor (BDNF) transcriptional activities were observed after sublethal OGD and administration of bicuculline and 4-aminopiridine. NR2A-containing NMDA receptors and CREB signaling have important functions in the induction of ischemic tolerance. This may provide potential novel therapeutic strategies to treat ischemic stroke.

Keywords

BDNF CRE ischemic tolerance neuroprotection NMDA receptor 2A

Introduction

Stroke is the third leading cause of death worldwide after ischemic heart attack and cancer, and poses a global health threat (Warlow et al, 2003). It is socially important to conquer this disease; however, multiple clinical trials of neuroprotectants to treat stroke patients have failed. In this dead-end situation, ischemic tolerance represents a strategy that may help attenuate ischemic damage. Ischemic tolerance was first demonstrated in 1990 to protect gerbil hippocampal neurons (Kitagawa et al, 1990), and is now a well-established phenomenon in vitro, in vivo, and in clinical settings (Dirnagl et al, 2009; Kirino, 2002). Although it is necessary to better understand ischemic tolerance to overcome this crucial neurologic problem, the molecular mechanisms underlying the phenomenon remain unclear.

The transcription factor cyclic-AMP response element (CRE)-binding protein (CREB) mediates diverse responses in the nervous system, including learning, memory, neuronal plasticity, and cell survival (Barco et al, 2002; Kandel, 2001; Lonze and Ginty, 2002). Phosphorylation of serine-133 in CREB allows it to contact its co-activator, CREB-binding protein/p300, and is necessary for activation. We previously reported that CRE-mediated gene expression increased in the penumbra using CRE-LacZ transgenic mice. The CREB activation was a critical event in neuroprotection against ischemic injury and in ischemic tolerance (Meller et al, 2005; Sugiura et al, 2004).

Meanwhile, the calcium–glutamate hypothesis suggests that excessive release of glutamate during and after ischemic injury leads to neuronal toxicity and death. The N-methyl-D-aspartate (NMDA) type glutamate receptors are associated with ischemic neuronal injury and ischemic tolerance (Grabb and Choi, 1999). On theoretical grounds, central nervous system disorders, including ischemic stroke, caused by glutamate-induced excitotoxicity might be treatable by blocking NMDA receptors. However, clinical application of NMDA receptor antagonists has not succeeded (Ikonomidou and Turski, 2002). Among a number of possible reasons, this is partly because there are multiple NMDA receptor subtypes, which may have different functions. The NMDA receptors are heteromeric complexes incorporating different subunits of the NR1, NR2, and NR3 subtypes. These tetramers most often incorporate two NR1 and two NR2 subunits, of which there are four different isoforms (NR2A to NR2D) (Paoletti and Neyton, 2007). Recent studies suggest that NR2A- and NR2B-containing NMDA receptors may have different functions in epileptogenesis and ischemic stroke (Chen et al, 2008; Chen et al, 2007; Liu et al, 2007). Synaptic NMDA receptors mainly dominated by NR2A were also shown to promote neuronal survival via CREB signaling (Hardingham et al, 2002).

In light of recent reports, we hypothesized that NR2A-containing NMDA receptors and CREB signaling might have a critical function in the induction of ischemic tolerance. In this study, we established ischemic tolerance in cultured cortical neurons and found that activation of NR2A induced ischemic tolerance via CREB signaling.

Materials and methods

The experimental procedures were approved by the Institutional Animal Care and Use Committee of the Osaka University Graduate School of Medicine.

Primary Neuronal Cell Culture

Cortical neuronal cultures were prepared from 16-day-old Wistar rat embryos (Charles River, Yokohama, Japan) using methods described earlier (Mabuchi et al, 2001). Briefly, cortices were dissected and dissociated using papain dissociation system (Worthington Biochemical Corporation, Lakewood, NJ, USA). Cells were plated on dishes coated with polyethylenimine in high-glucose DMEM (Sigma, St Louis, MO, USA) containing 5% fetal bovine serum (Invitrogen, Carlsbad, CA, USA) and 1% Antibiotic-Antimycotic (Invitrogen) at a density of 7 × 10⁵ per cells/mL. At 24 h after seeding, the medium was changed to Neurobasal medium (Invitrogen) supplemented with B-27 (Invitrogen) and 1% Antibiotic-Antimycotic. Cells were cultured at 37°C in a humidified chamber of 95% air and 5% CO₂. Cultures were used for experiments 10 to 12 days after seeding.

Oxygen–Glucose Deprivation

For oxygen–glucose deprivation (OGD)/reoxygenation, cells were washed with phosphate-buffered saline and incubated with glucose-free Earle's balanced salts solution medium (Biological Industries Israel Beit Haemek Ltd, Beit Haemek, Israel) in an anaerobic chamber containing 5% CO₂ and 95% N₂ (< 1% O₂). The OGD was terminated by bringing the cultures back to the original medium and placing them in a normoxic chamber. Sublethal ischemia was applied by OGD for 45 mins, and lethal ischemia was applied for 180 mins. For cross-tolerance experiments, cells were treated with 100 μmol/L glutamate or 50 μmol/L hydrogen peroxide for 15 mins. After treatment, neurons were incubated in the original medium.

To investigate the involvement of NR2A in ischemic tolerance, cultures were pretreated for 2 h and treated between sublethal and lethal OGD with one of the following chemicals: NMDA receptor antagonist D(−)-2-Amino-5-phosphopentanoic acid (APV) (100 μmol/L), non-NMDA receptor antagonist CNQX (10 μmol/L), L-type Ca²⁺ channel blocker nifedipine (5 μmol/L), NR2A-specific antagonist NVP-AAM077 (200 nmol/L), and NR2B-specific antagonist Ro 25-6981 (100 nmol/L). Neurons were not coincubated with these drugs during sublethal OGD, during and after lethal OGD. The potency and specificity of NVP-AAM077 were previously reported that it indicated a 100-fold preference for NR2A-containing receptors over NR2B-containing receptors (Auberson et al, 2002; Mallon et al, 2005). To stimulate neuronal synapses, neurons were pretreated for 24 h before lethal OGD with (−)-bicuculline methiodide (30 μmol/L), a GABA_A antagonist, and 4-aminopyridine (4-AP, 2500 μmol/L), a weak potassium channel blocker. NVP-AAM077 was a kind gift from Dr Yves Auberson, Novartis. Ro25-6981 and (−)-bicuculline methiodide were purchased from Tocris (Ellisville, MO, USA). Other agents were purchased from Sigma.

Cell Viability Assays

Neuronal injury was measured 24 h after treatment by lactate dehydrogenase (LDH) using the Cytotoxicity Detection Kit (Roche Applied Science, Mannheim, Germany). In sister culture, 100% cell death was induced with 2 mmol/L NMDA. The relative assessments of neuronal injury were normalized by comparison with 100% cell death.

Adenovirus Transfection and Measurement of Cyclic-AMP Response Element Transcriptional Activity

To investigate an association of CREB phosphorylation with ischemic tolerance, a dominant negative CREB mutant in which serine-133 was mutated to alanine (CREB-S133A, M-CREB) was used. Neurons were transfected with recombinant adenovirus vectors expressing CREB-S133A (adeno-CREB-S133A) 72h before experiments.

The CRE transcriptional activity was quantified by luciferase assay. As plasmid-based reporters did not provide an adequate level of activity in the neurons, we prepared an adenovirus-mediated reporter system. Cortical neurons were transfected with purified adenovirus, adeno-CRE-Luc (firefly) and adeno-TK-Luc (renilla). As an internal control, an adenovirus driving renilla luciferase was used (Promega, Madison, WI, USA). Luciferase activities were measured using the Dual-Luciferase Reporter Assay System (Promega). For luciferase assay, cell lysates were collected with 1 × passive lysis buffer (Promega). The relative firefly luciferase-based CRE-reporter activities were standardized to the corresponding Renilla luciferase in all cases.

Construction of a Mouse Brain-Derived Neurotrophic Factor Exon IV Promoter-Luciferase Reporter

Mouse genomic DNA was used as a template for polymerase chain reaction to generate an ∼400 bp brain-derived neurotrophic factor (BDNF) exon IV fragment with the following primers: 5′-GAGCTCGCCCTCCAGAACCTAGT CA-3′ and 5;-AAGCTTAGAGCAGTCCTCTCCTCGG-3′. The fragment was cloned into the Sac I/Hind III sites of the pGL4 luciferase reporter vector (Promega). A BDNF exon IV promoter-luciferase mutant lacking a CRE motif was generated by exchanging nucleotides using Quick-change Mutagenesis Kit (Stratagene, Cedar Creek, TX, USA) as follows: consensus CRE, TCACGTCA, and mutant CRE, TAAGCTTA. All constructs were verified by sequencing. Cortical neurons were transfected with either adeno-wild type or adeno-mutant-BDNF-luc and adeno-TK-Luc as an internal control.

Western Blot Analysis and Immunohistochemistry

Whole-cell protein extracts of cortical neurons were harvested and prepared for Western blot analysis as described earlier (Mabuchi et al, 2001). Proteins were extracted in lysis buffer composed of 20 mmol/L Tris-HCl, 150 mmol/L NaCl, 2 mmol/L EDTA, 1% Nonidet P-40, protease inhibitor cocktail (Complete Mini, Roche Applied Science) and phosphatase inhibitor cocktail (PhosSTOP, Roche Applied Science). After mixing and boiling for 5 mins with 2x sample buffer, an equal amount of protein for each sample was separated by 7.5% SDS-PAGE, and transferred onto polyvinylidine difluoride membrane (Millipore, Billerica, MA, USA). Blots were blocked in phosphate-buffered saline and 0.1% Tween-20 containing 5% skim milk for 1 h and probed with anti-phospho-CREB (1:1000; Upstate, Temecula, CA, USA), anti-CREB (1:1000; Cell Signaling Technology, Beverly, MA, USA), anti-NMDAR2A (1:1000; Calbiochem, La Jolla, CA, USA), anti-NMDAR2B (1:1000; Chemicon, Temecula, CA, USA) anti-GluR2 (1:1000; BD Biosciences Pharmingen, San Diego, CA, USA), and anti-β-actin (1:100000; Sigma) in Can Get Signal Solution 1 (Toyobo Co. Ltd, Osaka, Japan) at 4°C overnight. The membranes were washed three times in phosphate-buffered saline and 0.1% Tween 20 and incubated for 1 h with anti-rabbit IgG conjugated to HRP (1:5000; Amersham Pharmacia Biotech, Uppsala, Sweden) for anti-phospho-CREB, anti-CREB, anti-NMDAR2A, and anti-NMDAR2B, or with anti-mouse IgG conjugated to HRP (1:2000; Amersham Pharmacia Biotech) for anti-GluR2 and anti-β-actin in Can Get Signal Solution 2 (Toyobo). Blots were developed by enhanced chemiluminescence (GE Healthcare UK limited, Buckinghamshire, UK) and exposed to x-ray film. Digital images were produced by densitometric scans of autoradiographs and quantified by Image J 1.40 software (National Institutes of Health, Bethesda, MD, USA).

For immunohistochemical studies, neurons were fixed in 4% paraformaldehyde for 30 mins, washed and permeabilized in 0.01% Triton X-100 for 30 mins. After blocking with 10% donkey serum, cells were incubated with anti-microtubule-associated protein 2 (MAP2) (1:500; Sigma) at 4°C overnight. Neurons were washed, probed with Alexa Fluor 488 conjugated donkey anti-mouse secondary anti-body (1:300; Invitrogen) at room temperature for 30 mins, and viewed using a confocal microscope.

Statistical Analysis

Data in the text and figures are described as mean ± s.d. Statistical analysis was performed by one-factor analysis of variance followed by Scheffe's post hoc test. Statistical significance was defined as P < 0.05.

Results

Ischemic Tolerance and Cross-Tolerance In Vitro

To elucidate the mechanisms of induction of ischemic tolerance, we established an ischemic tolerance model using in vitro OGD. With regard to ischemic preconditioning (+) group, the cultures were subjected to 45-mins OGD before 180-mins OGD. The medium of ischemic preconditioning (−) group was changed as sham wash instead of 45-mins OGD (Figure 1A). When cells were subjected to 180-mins OGD, we observed ∼70% cell death. When the cultures were preconditioned with a 45-mins OGD before the 180-mins OGD, cell death was reduced (Figure 1B). The most effective interval between sublethal and lethal OGD to obtain ischemic tolerance was 24 h (Figure 1C), whereas, at intervals of 1 and 4 h, the protective effect was not observed. Importantly, the preconditioning ischemia alone did not induce cell death (data not shown). Interestingly, we observed cross-tolerance in these cultures: sublethal OGD reduced neuronal injury by lethal insults such as 15 mins treatment with 100 μmol/L glutamate or 50 μmol/L hydrogen peroxide (Figures 1G–1H).

Figure 1

Model of ischemic tolerance and cross-tolerance in primary neuronal cell culture. (A) Experimental procedures. Ischemic preconditioning (IP) (+), cells were subjected to 45-mins oxygen–glucose deprivation (OGD) before lethal OGD. Ischemic preconditioning (−), cells were provided with a control wash instead of sublethal OGD. (B) Ischemic tolerance was induced by subjecting cells to sublethal 45-mins OGD, followed by a 24 h recovery period before lethal 180-mins OGD. (C) Ischemic tolerance was observed when the interval between sublethal and lethal injury was 12 to 48 h, and the most effective interval was 24 h. (D–F) Immunohistochemistry images of MAP2 were acquired under the following conditions: D, no stimulation; E, IP (+); F, IP (−). (G, H) Cross-tolerance was induced in the culture. Following sublethal OGD, neuronal injury induced by an otherwise lethal 15-mins 100 μmol/L glutamate (G) or 15-mins 50 μmol/L hydrogen peroxide (H), treatment was reduced. *P < 0.01, ^†P < 0.05 compared with IP (−). n = 8 to 12.

Considering these results, we found that preconditioning OGD could induce both ischemic and cross-tolerance.

Phosphorylation of Cyclic-AMP Response Element-Binding Protein and Cyclic-AMP Response Element Activation are Necessary for Ischemic Tolerance

It has been shown earlier that CREB was activated after ischemic preconditioning (Grabb and Choi, 1999); accordingly, we investigated the effects of preconditioning ischemia on phosphorylation of CREB and CRE transcriptional activity. The CREB was expressed in control cells, and total CREB expression levels did not change after 45-mins OGD. In contrast, phosphorylated CREB levels significantly increased 1 to 6 h after 45-mins OGD (Figure 2A). In the same time course, transcriptional activity of CRE was markedly enhanced 1 h after sublethal OGD, and maintained at 24 h (Figure 2B).

Figure 2

Temporal profiles of expression of phosphorylated CREB and CRE transcriptional activity after sublethal ischemia. (A) Phosphorylation levels of CREB significantly increased 1 to 6 h after sublethal ischemia and subsequently decreased. The same amount of protein was loaded per lane (n = 4). (B) The CRE transcriptional activity markedly increased 1 h after sublethal OGD and was maintained for 24 h (n = 8). *P < 0.01, ^†P < 0.05 compared with control.

To further investigate the importance of the CREB cascade, we examined whether a CREB-S133A mutant would affect CRE activity under these conditions. When cells were transfected with adeno-CREB-S133A, CRE transcriptional activity was lower compared with the EGFP-transfected control group after sublethal OGD (Figure 3A). Under these conditions, the adeno-CREB-S133A abolished ischemic tolerance (Figure 3B). In contrast, the EGFP controls showed ischemic tolerance.

Figure 3

Effect of CREB-S133A on CRE reporter activity and ischemic tolerance. (A) When cells were transfected with CREB-S133A-containing adenovirus, CRE transcriptional activity was significantly reduced, compared with the EGFP-transfected control group. (B) EGFP-transfected cells showed ischemic tolerance, whereas tolerance was abolished in CREB-S133A-transfected neurons. *P < 0.01. n = 8.

Consistent with earlier reports (Chen et al, 2008; Meller et al, 2005), these findings suggest that phosphorylation of CREB and the subsequent enhancement of CRE activity are necessary for induction of ischemic tolerance.

NR2A Activation Protects Neurons Against Lethal Ischemia via Cyclic-AMP Response Element-Binding Protein Signaling

Among the routes of Ca²⁺ entry, NMDA receptors are of particular interest because of their ability to gate high levels of Ca²⁺ influx, and they have been extensively studied for their critical function in ischemia. Recently, several studies have suggested that NMDA receptor subtypes (NR2A/synaptic or NR2B/extrasynaptic) are key regulators of neuronal viability (Hardingham et al, 2002). To determine whether different types of glutamate receptor are associated with ischemic tolerance and CREB signaling, we administered several types of glutamate receptor antagonist (Figure 4A). First, we examined the effect of sublethal OGD on the expression of NMDA and AMPA receptors with Western blot analysis. The expression levels of NR2A, NR2B, and GluR2 did not change after sublethal OGD (Figure 4B). Under such a condition, inhibiting NR2A-containing NMDA receptors using NVP-AAM077 attenuated preconditioning-induced neuroprotection (Figure 4C). In contrast, other types of glutamate receptor antagonist such as APV, CNQX, and Ro25-6981, or an L-type calcium channel blocker, nifedipine, did not have much effect on ischemic tolerance. Treatment with these drugs alone exerted little effect on the basal LDH levels. Among the neuronal signaling pathways, NR2A-containing NMDA receptors have been shown to activate the CREB cascade (Hardingham et al, 2002). To further elucidate the pathway downstream of NR2A receptors, we investigated the effect of inhibiting NR2A receptors on phosophorylation of CREB. When cells were treated with NVP-AAM077, phosphorylated CREB levels after sublethal OGD were significantly attenuated compared with those treated with vehicle (Figure 4D). Furthermore, we examined the association between antagonizing NR2A receptors and CRE transcriptional activity. Compared with vehicle treatment, inhibition of NR2A receptors by NVP-AAM077 partially but significantly decreased CRE activity 3 and 6 h after 45-mins OGD (Figure 4E).

Figure 4

Effects of NVP-AAM077, a NR2A-specific antagonist, on ischemic tolerance and CRE transcriptional activity after sublethal OGD. (A) Experimental procedure. Cultures were pretreated for 2 h and treated between sublethal and lethal OGD with inhibitors. Cells were not coincubated during sublethal OGD, during and after lethal OGD. (B) The expression levels of NR2A, NR2B, and GluR2 did not change after sublethal OGD. (C) NVP-AAM077 alone abolished ischemic tolerance. Other types of glutamate receptor antagonists or an L-type calcium channel blocker did not affect ischemic tolerance. (D) When cells were treated with NVP-AAM077, phosphorylated CREB levels after sublethal OGD were significantly attenuated compared with those treated with vehicle. (E) When NR2A was antagonized by NVP-AAM077, CRE transcriptional activity was partially but significantly reduced compared with vehicle-treated controls 3 and 6 h after sublethal OGD. NVP, NVP-AAM077; Ro, Ro25-6981. *P < 0.01, ^†P < 0.05. n = 6 (C) and 8 (E).

These results show that inhibiting NR2A-containing NMDA receptors extinguishes ischemic tolerance with a correlated reduction of CRE activity. To determine whether activation of NR2A receptors would induce ischemic tolerance, we stimulated neurons by treatment with bicuculline and 4-AP, which has been shown earlier to stimulate neuronal synapses and activate NR2A receptors (Figure 5A) (Liu et al, 2007). As shown in Figure 5B, CRE activity levels significantly increased 3 and 6 h after administration of bicuculline and 4-AP, compared with vehicle-treated controls. Further, without preconditioning ischemia, ischemic tolerance was induced by pretreatment with bicuculline and 4-AP for 24 h before lethal ischemia (Figure 5C). To evaluate whether NVP-AAM077 blocked neuroprotective effect of bicuculline and 4-AP, we designed additional experiments; cells were treated with NVP-AAM077 or vehicle together with bicuculline and 4-AP for 24 h before lethal OGD (Figure 5D). The CRE activity levels significantly decreased 3 and 6 h after administration of NVP-AAM077 group compared with vehicle group (Figure 5E). Further, the neuroprotective effect of bicuculline and 4-AP against lethal OGD was attenuated by treatment with NVP-AAM077 (Figure 5F). These results suggest that activation of NR2A-containing NMDA receptors is sufficient to induce ischemic tolerance mediated by enhancement of CRE activity.

Figure 5

Induction of ischemic tolerance by stimulating neuronal synapses. (A) Experimental procedure for (B) and (C). Bicuculline and 4-aminopridine (4-AP) were administered 24 h before lethal OGD. (B) When synapses were stimulated by bicuculline and 4-AP, CRE transcriptional activity levels were significantly enhanced compared with vehicle treatment. (C) Stimulating neuronal synapses by pretreatment with bicuculline and 4-AP reduced neuronal injury after lethal ischemia. (D) Experimental procedure for (E) and (F). Cells were treated with NVP-AAM077 or vehicle together with bicuculline and 4-AP for 24 h before lethal OGD. (E) The CRE activity levels significantly decreased in NVP-AAM077 group compared with vehicle group. (F) The neuroprotective effect of bicuculline and 4-AP against lethal OGD was attenuated by treatment with NVP-AAM077. Bic, bicuculline; 4-AP, 4-aminopyridine; Veh, Vehicle. *P < 0.01. n = 8 (B), 12 (C), 8 (E), and 12 (F).

NR2A Activation Enhances Brain-Derived Neurotrophic Factor Exon IV Promoter Activity Through Cyclic-AMP Response Element-Binding Protein Signaling

The CREB leads to neuroprotection mediated by transcription of downstream gene targets containing CRE motifs in their promoters; targets include BDNF, TrkB, and Bcl-2 (Deogracias et al, 2004; Riccio et al, 1999; Shieh et al, 1998). In particular, BDNF is a strong neurotrophic factor and TrkB, a receptor of BDNF also is a target for CREB regulation. Therefore, to elucidate the involvement of factors downstream of CREB, we investigated bdnf exon IV and mutant-bdnf (ΔCRE) exon IV transcriptional activity (Figure 6A). Promoter IV obviously contributes to activity-dependent BDNF transcription (Timmusk et al, 1993). More recently, promoter IV-driven BDNF transcription has been shown to have a critical function on synaptic plasticity with promoter IV mutant mice (Sakata et al, 2009). To dissect the function of the CRE cis-element within the bdnf promoter, we prepared a bdnf reporter gene construct with a mutation in the CRE-binding site. After sublethal OGD, BDNF levels significantly increased compared with basal levels, whereas mutant-BDNF levels were unaffected (Figure 6B). When NR2A was inhibited by NVP-AAM077, BDNF activity levels was significantly reduced compared with vehicle treatment (Figure 6B). In contrast, when the CRE was mutated, the inhibitory effect of NVP-AAM077 was not observed. The BDNF activity levels significantly increased 6 and 12 h after administration of bicuculline and 4-AP compared with mutant-BDNF activity (Figure 6C). These results suggest that BDNF transcriptional activity depends on CRE motifs and that a CREB cascade originated from NR2A enhances BDNF expression.

Figure 6

Increased BDNF transcriptional activity levels downstream of CREB after sublethal ischemia and synaptic stimulation. (A) Construction of the bdnf exon IV promoter reporter. (B) The BDNF activity levels markedly increased after 45-mins OGD, but not mutant BDNF. When NR2A was inhibited by NVP-AAM077, BDNF activity level was significantly reduced compared with vehicle treatment. (C) The BDNF activity significantly increased 6 and 12 h after administration of bicuculline and 4-AP (Bic/4-AP) compared with mutant-BDNF activity. *P < 0.01, ^†P < 0.05. n = 4 to 8.

Discussion

In this study, we present two major evidences. First, CREB signaling originated from NR2A is an essential step in the induction of ischemic tolerance. In addition, sublethal ischemia induced cross-tolerance against otherwise lethal treatment with glutamate or hydrogen peroxide. Thus, this preconditioning stimulus may make neurons resistant to various insults related to excitotoxicity and oxidative stress. Second, preconditioning-induced neuroprotection was only attenuated by treatment with NVP-AAM077, but not other glutamate receptor antagonists. Activation of NR2A-containing NMDA receptors by administration of bicuculline and 4-AP increased CRE activity levels and led to neuroprotection. Taken together, these results suggest that CREB signaling originated from NR2A-containing NMDA receptors has a pivotal function in ischemic tolerance.

Considerations that should be taken into account when discussing stimulation of synapses with bicuculline and 4-AP include developmental and regional expression of NMDA subtypes and proportion of synaptic NMDA receptors containing the NR2B subunit (Monyer et al, 1994). NR2A-containing NMDA receptors are predominantly expressed at synapses; however, NR2B-containing receptors are thought to exist predominantly at extrasynaptic sites, but are also expressed at synapses (Massey et al, 2004). In addition, cortical neurons are thought to express NR1/NR2A/NR2B heterotrimeric NMDA receptors (Luo et al, 1997). As the function of NR1/NR2A/NR2B heterotrimeric receptors remains unclear, further studies to clarify their effects on neuronal survival or death and ischemic tolerance are necessary.

Tissue damage in ischemia is the result of a complex pathophysiological cascade, which comprises many distinct pathological stresses including hypoxic stress, oxidative stress, inflammatory stress, and glutamate excitotoxic stress. Ischemic tolerance protects the brain itself mediated by endogenous multiple pathways against these noxious stresses after ischemia (Dirnagl et al, 2009). One of the key regulators involved with ischemic tolerance is hypoxia-inducible factor. Ischemic tolerance activates hypoxia-inducible factor-1 and leads to neuroprotection via its target gene (Bernaudin et al, 2002). Also, ischemic tolerance attenuates oxidative stress by modification of nuclear factor κ B-driven gene and regulation of heat shock protein (Blondeau et al, 2001; Kirino et al, 1991). The preconditioning with lipopolysaccharide, a ligand of Toll-like receptors, stimulates antiinflammatory processes, suppresses proinflammatory pathways, and induces ischemic tolerance (Stenzel-Poore et al, 2007; Stevens and Stenzel-Poore, 2006). In addition, glutamate excitotoxicity is important in the pathophysiology of ischemic brain injury, which induces massive release of glutamate into extracellular space. Ischemic preconditioning might induce tolerance either by lowering excessive glutamate release or by increasing glutamate uptake (Romera et al, 2007). Activation of glutamate receptors causes calcium influx, activates calcium/CaM kinase, PI3K-Akt pathway, and Ras-MAPK cascade (Soriano et al, 2006; Wang et al, 2007), and in turn converge on CREB activation and upregulation of its target genes. Also, NR2A-containing synaptic NMDA receptors activate the extracellular signal-regulated kinase (ERK) pathway (Leveille et al, 2008). In fact, both Akt and ERK pathways not only activate CREB signaling but also mediate ischemic tolerance, and then, the target genes of CREB, such as Bcl-2, have a crucial function on ischemic tolerance via its antiapoptotic mechanism. Therefore, CREB signaling through NR2A could be one of the most important pathways involved in ischemic tolerance.

In excitotoxic conditions including ischemia, abnormal intracellular calcium entry triggers cell death through calcium overload (Bano and Nicotera, 2007). As shown recently, after an ischemic insult, other plasma membrane channels such as transient receptor potential (TRP) channel and acid sensing ion channel can contribute to calcium accumulation that induces neuronal death (Sun et al, 2009; Xiong et al, 2004). Also, earlier study reported that TRPC6, one of the TRP channel subfamilies, was associated with CaMKIV-CREB-dependent gene transcription including BDNF (Zhou et al, 2008). Thus, under ischemic condition, there is a possibility that calcium inflow routes besides NMDA receptors contribute to CREB pathway. Therefore, as shown in Figure 4E, transcriptional activity of CRE was not completely suppressed with NVP-AAM077.

An enormous number of studies in vitro and in vivo have suggested that NMDA receptor antagonists are effective in ischemic neuronal death. However, clinical stroke trials using NMDA antagonists have all failed. At least in part, narrow time windows for treatment may be important. Indeed, in the early phase of stroke, overactivation of NMDA receptors is detrimental, but in the delayed phase, NMDA signals may be required for recovery, including neuroplasticity and protection. Meanwhile, activation of NR2A-containing NMDA receptors may have some benefits compared with NMDA receptor antagonists that have been applied to stroke patients in previous trials. It remains possible that stimulating NR2A in clinical settings may attenuate neuronal injury and lead to a more favorable outcome than not stimulating NR2A. In addition to ischemic stroke, activation of NR2A-containing NMDA receptors may be a potential avenue to treat chronic neurodegenerative diseases (Ikonomidou and Turski, 2002).

Ischemic tolerance, an endogenous adaptive response, has received intense interest both clinically and nonclinically. Thus, we have clarified that stimulation of NR2A-containing NMDA receptors protect the brain against lethal ischemia mediated by neuroprotective signaling via CREB. This understanding of the mechanisms engaged in ischemic tolerance may allow novel therapeutic strategies for neuroprotection.

Footnotes

Acknowledgements

This work was supported by a Grant-in-Aid from the Ministry or Education, Science, and Culture in Japan. We thank Ms C Kurano for secretarial assistance and Ms K Nishiyama for laboratory assistance.

The authors declare no conflict of interest.

References

Auberson

Allgeier

Bischoff

Lingenhoehl

Moretti

Schmutz

(2002) 5-Phosphonomethylquinoxalinediones as competitive NMDA receptor antagonists with a preference for the human 1A/2A, rather than 1A/2B receptor composition. Bioorg Med Chem Lett 12:1099–102

Bano

Nicotera

(2007) Ca2+ signals and neuronal death in brain ischemia. Stroke 38:674–6

Barco

Alarcon

Kandel

(2002) Expression of constitutively active CREB protein facilitates the late phase of long-term potentiation by enhancing synaptic capture. Cell 108:689–703

Bernaudin

Nedelec

Divoux

MacKenzie

Petit

Schumann-Bard

(2002) Normobaric hypoxia induces tolerance to focal permanent cerebral ischemia in association with an increased expression of hypoxia-inducible factor-1 and its target genes, erythropoietin and VEGF, in the adult mouse brain. J Cereb Blood Flow Metab 22:393–403

Blondeau

Widmann

Lazdunski

Heurteaux

(2001) Activation of the nuclear factor-kappaB is a key event in brain tolerance. J Neurosci 21:4668–77

Chen

Zhou

Chen

Feng

Duan

Xiong

(2008) Differential roles of NMDA receptor subtypes in ischemic neuronal cell death and ischemic tolerance. Stroke 39:3042–8

Chen

Zhou

Zheng

Zhang

Duan

Xiong

(2007) Differential roles of NR2A- and NR2B-containing NMDA receptors in activity-dependent brain-derived neurotrophic factor gene regulation and limbic epileptogenesis. J Neurosci 27:542–52

Deogracias

Espliguero

Iglesias

Rodriguez-Pena

(2004) Expression of the neurotrophin receptor trkB is regulated by the cAMP/CREB pathway in neurons. Mol Cell Neurosci 26:470–80

Dirnagl

Becker

Meisel

(2009) Preconditioning and tolerance against cerebral ischaemia: from experimental strategies to clinical use. Lancet Neurol 8:398–412

10.

Grabb

Choi

(1999) Ischemic tolerance in murine cortical cell culture: critical role for NMDA receptors. J Neurosci 19:1657–62

11.

Hardingham

Fukunaga

Bading

(2002) Extrasynaptic NMDARs oppose synaptic NMDARs by triggering CREB shut-off and cell death pathways. Nat Neurosci 5:405–14

12.

Ikonomidou

Turski

(2002) Why did NMDA receptor antagonists fail clinical trials for stroke and traumatic brain injury? Lancet Neurol 1:383–6

13.

Kandel

(2001) The molecular biology of memory storage: a dialogue between genes and synapses. Science 294:1030–8

14.

Kirino

Tsujita

Tamura

(1991) Induced tolerance to ischemia in gerbil hippocampal neurons. J Cereb Blood Flow Metab 11:299–307

15.

Kirino

(2002) Ischemic tolerance. J Cereb Blood Flow Metab 22:1283–96

16.

Kitagawa

Matsumoto

Tagaya

Hata

Ueda

Niinobe

Handa

Fukunaga

Kimura

Mikoshiba

(1990) ‘Ischemic tolerance’ phenomenon found in the brain. Brain Res 528:21–4

17.

Leveille

El Gaamouch

Gouix

Lecocq

Lobner

Nicole

Buisson

(2008) Neuronal viability is controlled by a functional relation between synaptic and extrasynaptic NMDA receptors. FASEB J 22: 4258–71

18.

Liu

Wong

Aarts

Rooyakkers

Liu

Lai

Tymianski

Craig

Wang

(2007) NMDA receptor subunits have differential roles in mediating excitotoxic neuronal death both in vitro and in vivo. J Neurosci 27:2846–57

19.

Lonze

Ginty

(2002) Function and regulation of CREB family transcription factors in the nervous system. Neuron 35:605–23

20.

Luo

Wang

Yasuda

Dunah

Wolfe

(1997) The majority of N-methyl-D-aspartate receptor complexes in adult rat cerebral cortex contain at least three different subunits (NR1/NR2A/NR2B). Mol Pharmacol 51:79–86

21.

Mabuchi

Kitagawa

Kuwabara

Takasawa

Ohtsuki

Xia

Storm

Yanagihara

Hori

Matsumoto

(2001) Phosphorylation of cAMP response element-binding protein in hippocampal neurons as a protective response after exposure to glutamate in vitro and ischemia in vivo. J Neurosci 21:9204–13

22.

Mallon

Auberson

Stone

(2005) Selective subunit antagonists suggest an inhibitory relationship between NR2B and NR2A-subunit containing N-methyl-D: -aspartate receptors in hippocampal slices. Exp Brain Res 162:374–83

23.

Massey

Johnson

Moult

Auberson

Brown

Molnar

Collingridge

Bashir

(2004) Differential roles of NR2A and NR2B-containing NMDA receptors in cortical long-term potentiation and long-term depression. J Neurosci 24:7821–8

24.

Meller

Minami

Cameron

Impey

Chen

Lan

Henshall

Simon

(2005) CREB-mediated Bcl-2 protein expression after ischemic preconditioning. J Cereb Blood Flow Metab 25:234–46

25.

Monyer

Burnashev

Laurie

Sakmann

Seeburg

(1994) Developmental and regional expression in the rat brain and functional properties of four NMDA receptors. Neuron 12:529–40

26.

Paoletti

Neyton

(2007) NMDA receptor subunits: function and pharmacology. Curr Opin Pharmacol 7:39–47

27.

Riccio

Ahn

Davenport

Blendy

Ginty

(1999) Mediation by a CREB family transcription factor of NGF-dependent survival of sympathetic neurons. Science 286:2358–61

28.

Romera

Hurtado

Mallolas

Pereira

Morales

Romera

Serena

Vivancos

Nombela

Lorenzo

Lizasoain

Moro

(2007) Ischemic preconditioning reveals that GLT1/EAAT2 glutamate transporter is a novel PPARgamma target gene involved in neuroprotection. J Cereb Blood Flow Metab 27:1327–38

29.

Sakata

Woo

Martinowich

Greene

Schloesser

Shen

(2009) Critical role of promoter IV-driven BDNF transcription in GABAergic transmission and synaptic plasticity in the prefrontal cortex. Proc Natl Acad Sci USA 106:5942–7

30.

Shieh

Bobb

Timmusk

Ghosh

(1998) Identification of a signaling pathway involved in calcium regulation of BDNF expression. Neuron 20:727–40

31.

Soriano

Papadia

Hofmann

Hardingham

Bading

Hardingham

(2006) Preconditioning doses of NMDA promote neuroprotection by enhancing neuronal excitability. J Neurosci 26:4509–18

32.

Stenzel-Poore

Stevens

King

Simon

(2007) Preconditioning reprograms the response to ischemic injury and primes the emergence of unique endogenous neuroprotective phenotypes: a speculative synthesis. Stroke 38:680–5

33.

Stevens

Stenzel-Poore

(2006) Toll-like receptors and tolerance to ischaemic injury in the brain. Biochem Soc Trans 34:1352–5

34.

Sugiura

Kitagawa

Omura-Matsuoka

Sasaki

Tanaka

Yagita

Matsushita

Storm

Hori

(2004) CRE-mediated gene transcription in the peri-infarct area after focal cerebral ischemia in mice. J Neurosci Res 75:401–7

35.

Sun

Jackson

Martin

Jansen

Teves

Cui

Kiyonaka

Mori

Jones

Forder

Golde

Orser

Macdonald

Tymianski

(2009) Suppression of hippocampal TRPM7 protein prevents delayed neuronal death in brain ischemia. Nat Neurosci 12:1300–7

36.

Timmusk

Palm

Metsis

Reintam

Paalme

Saarma

Persson

(1993) Multiple promoters direct tissue-specific expression of the rat BDNF gene. Neuron 10:475–89

37.

Wang

Fibuch

Mao

(2007) Regulation of mitogen-activated protein kinases by glutamate receptors. J Neurochem 100:1–11

38.

Warlow

Sudlow

Dennis

Wardlaw

Sandercock

(2003) Stroke. Lancet 362:1211–24

39.

Xiong

Zhu

Chu

Minami

Hey

Wei

MacDonald

Wemmie

Price

Welsh

Simon

(2004) Neuroprotection in ischemia: blocking calcium-permeable acid-sensing ion channels. Cell 118:687–98

40.

Zhou

Tai

Yao

Jia

Ding

Wang

(2008) Critical role of TRPC6 channels in the formation of excitatory synapses. Nat Neurosci 11:741–3