CREB-Mediated Bcl-2 Protein Expression after Ischemic Preconditioning

Ischemic Tolerance Model In Vivo

All procedures were performed in a facility accredited by the Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC), in accordance with protocols approved by the Legacy Institutional Animal Care and Use Committee and the principles outlined in the National Institute of Health Guide for the Care and Use of Laboratory animals. Middle cerebral artery occlusion (MCAO) was induced in adult male Sprague–Dawley rats (280–350 g) using the intraluminal filament technique, as described previously (Chen et al, 1996). Rats were anesthetized with 4% isoflurane and ventilated with 1.5% isoflurane (70% N₂O, 30% O₂) to maintain blood gas variables within physiological range. Rectal and temporalis muscle temperature were kept at 37.0°C to 37.5°C. A 3-0 surgical monofilament nylon suture with rounded tip was introduced into the left internal carotid artery and advanced 20 to 21 mm past the carotid artery bifurcation to occlude the middle cerebral artery (MCA). After completion of the MCAO (30 or 100 mins), the suture was withdrawn to allow reperfusion. Second MCA occlusions were performed 72 hours after preconditioning with 30-min MCAO. As a control, sham-operated rats underwent identical surgery but did not have the suture inserted. Animals were euthanized (phenobarbitone overdose) at various time points after MCAO or after 24 hours in nonischemic controls. For Western blotting and electrophoretic mobility shift assay, brains were rapidly removed, placed on ice and ipsilateral cortex was dissected, and stored at −80°C until assay. For immunohistochemistry studies, whole brains were removed and then frozen at −30°C in methylbutane.

Infarct Volume Measurement

Infarct volume was assessed using a triphenyltetrazolim hydrochloride (TTC) method, as described previously (Chen et al, 1996). Animals were subject to (1) 30-min MCAO, (2) 100-min MCAO, or (3) 30-min MCAO followed by 72 hours reperfusion and 100-min MCAO. Animals were killed 24 hours after the last period of MCAO by an overdose of phenobarbitone (i.p.); brains were rapidly removed and sectioned coronally at 2-mm intervals. Sections were immersed in TTC (2%) at 37°C for 20 mins, followed by formaldehyde (4%) for 15 mins. The hemispheric infarct area in each section was calculated by subtracting the area of normal TTC-stained brain in the ipsilateral cortex from the contralateral area, as described previously (Swanson et al, 1990).

Western Blotting

Western blotting was performed as described previously (Meller et al, 2003). Tissue samples were lysed in a nondenaturing buffer containing the protease inhibitors phenylmethylsulfonylfluoride (100 μg/mL), aprotinin (1 μg/mL), leupeptin (1 μg/mL), pepstatin (1 μg/mL), NaF (50 mmol/L), Na₃VO₄ (2 mmol/L) and phosphatase cocktail inhibitor (Sigma, St Louis, MO, USA). Protein concentration was determined by Bradford reagent spectrophotometrically at A⁵⁹⁵. Protein samples (50 μg) were denatured in a gel-loading buffer at 100°C for 5 mins and then loaded onto 12% SDS-polyacrylamide gels to detect Bcl-2, CREB, or phosphorylated CREB proteins or onto 7.5% SDS-polyacrylamide gels to detect CBP protein. Proteins were transferred to polyvinylodene difluoride membranes and incubated with primary antibodies at 4°C overnight: anti-CREB and phospho-CREB (Ser 133) (Cell Signaling Technology, Beverly, MA, USA), anti-Bcl-2 and CBP (Santa Cruz Biotechnology, Santa Cruz, CA, USA). Membranes were incubated with anti-rabbit IgG conjugated to horseradish peroxidase (Cell Signaling Technology, Beverly, MA, USA) followed by chemiluminescense detection (NEN Life Science Products, Boston, MA, USA) and then exposed to film. Images were captured using a Dage 72 camera.

Electrophoretic Mobility Shift Assay (EMSA)

Nuclear protein for EMSA assay was prepared by lysing brain or cell samples in a high-salt buffer containing Hepes (20 mmol/L; pH 7.9), NaCl (420 mmol/L), NaF (20 mmol/L), Na₃VO₄ (1 mmol/L), Na₄P₂O₇ (1 mmol/L), EDTA (1 mmol/L), EGTA (1 mmol/L), DTT (1 mmol/L), glycerol (20%), PMSF (0.5 mmol/L), leupeptin (1 μg/mL). The samples were freeze-thawed on dry ice and the cell lysate was centrifuged at 14,000g for 20 mins. The bcl-2 CRE probe sense (5′-GAACCGTGTGACGTTACGC-3′) and antisense (5′-TGCGTAACGTCACACGGTTC-3′) oligonucleotides (200 pmol) were annealed in 100 μL of an annealing buffer containing Tris (10 mmol/L; pH 8.0), NaCl (100 mmol/L) and EDTA (1 mmol/L) by heating at 90°C for 5 mins and then reducing the temperature to 25°C over 30 mins. The annealed oligonucleotides (30 pmol) were labeled by TdT kinase reaction and purified using a G-25 column.

Nucleic proteins (10 μg) were incubated with oligonucleotides and poly(dI-dC)-poly(dI-dC) in a buffer containing Hepes (12 mmol/L; pH 7.9), NaCl (60 mmol/L), DTT (1 mmol/L), EDTA (0.2 mmol/L), and glycerol (8%) at room temperature for 20 mins. The DNA–protein complexes were resolved by a 5% polyacrylamide gel containing 0.5 × TBE and 2.5% glycerol, at 200 V for 2 hours. The gel was dried and exposed to an X-ray film at −80°C with an intensifying screen. In some experiments, an excess of cold (unlabeled) oligonucleotide or CREB antibody was added to the reaction mixture.

Neuronal Cell Culture

Cortical neuronal cultures were prepared from 1- to 3-day-old Sprague–Dawley rat pups using the method of Goslin (1998) as described previously (Stenzel-Poore et al, 2003). Briefly, cortices from 10 to 12 rat pups were dissected, dissociated with papain (Worthington Biochemicals, Lakewood, NJ, USA) and plated out at a density of 400,000 cells per coverslip in Neurobasal-A/B27 media (Invitrogen, Carlsbad, CA, USA). The cells were used after 7 to 10 days in culture, when cultures consist of approx. 80% to 90% neurons. For Western blotting, EMSA and chromatin immunoprecipitation assays, cells were plated out onto 10-cm polylysine-coated culture dishes (Primara; Becton Dickinson, San Jose, CA, USA) and samples were collected at various time points after oxygen–glucose deprivation (OGD).

In Vitro Ischemic Tolerance Model

Oxygen–glucose deprivation was performed by washing the cells with phosphate-buffered saline (PBS) (0.5 mmol/L CaCl₂, 1 mmol/L MgCl₂; pH 7.4) and placing culture dishes in an anaerobic chamber (Forma Scientific, Marjetta, OH, USA) (85% N₂, 5% H₂, 10% CO₂; 35°C). Anaerobic conditions in the chamber were monitored using Gaspack anaerobic indicator strips (Becton Dickinson, San Jose, CA, USA). Oxygen–glucose deprivation was terminated by removing cells from the chamber, replenishing with Neurobasal A media and placing them back into the normoxic incubator. Cultured cells were subjected to the following treatments: (1) control cells were washed with PBS and maintained in Neurobasal A media, (2) 30-min OGD (preconditioning) and 48-hour recovery in Neurobasal A media, (3) 120-min OGD (toxic ischemia) followed by 24-hour recovery in Neurobasal A media, (4) 30-min OGD followed by 24-hour recovery, then 120-min OGD followed by 24-hour recovery in Neurobasal A media.

Some cells were incubated with a CRE decoy oligonucleotide based on the bcl-2 CRE sequence or control oligonucleotides, similar to those previously described by Park et al (1999) (all 0.5 μmol/L) (CRE decoy oligonucleotide 5′-TGACGTCAGAGAGCGCTCTCTGACGTCA-3′; Control CRE mis-match sequence 5′-TAGCTGCAGAGAGCGCTCTCTGCAGCTA-3′; (all linkages phosphorothioate protected) (Invitrogen Carlsbad, CA, USA or Sigma-Genosys, The Woodlands, TX, USA). Oligonucleotides were heated to 95°C and then slowly cooled to room temperature before incubation with the cells. Some cells were incubated with the following protein kinase inhibitors for 24 hours after 30-min OGD preconditioning and before the 120-min OGD challenge; PKA inhibitor H89 (N-[2-(p-bromocinnamylamino)ethyl]-5-isoquinolinesulfonamide dihydrochloride) 0.5 μmol/L, CAMK inhibitor KN 62 (1-[N,O-bis(5-isoquinolinesulfonyl)-N-methyl-l-tyrosyl]-4-phenylpiperazine) 1.0 μmol/L (Sigma, St Louis, MO, USA), MEK (MAP kinase kinase) inhibitor U0126 (1,4-diamino-2,3-dicyano-1,4-bis(o-aminophenylmercapto)butadiene) 10 μmol/L (Sigma, St Louis, MO, USA), and phosphatidylinositol 3-kinase inhibitor LY294002 (2-(4-morpholinyl)-8-phenyl-1(4H)-benzopyran-4-one hydrochloride) 10 μmol/L (Cell Signaling Technology, Beverly, MA, USA). Concentrations of compounds were chosen on the basis of their selectivity in previous studies (Chijiwa et al, 1990; Davies et al, 2000; Meller et al, 2003).

Cell Viability Assay

For cell viability assays, coverslips containing cortical cells were incubated with propidium iodide (1.5 μg/mL) for 2 mins, washed with PBS and fixed with 4% formaldehyde. Cells were permeabilized with 0.1% Triton X-100 and then mounted onto glass slides using Vectashield mounting medium containing 4′,6-diamidino-2-phenylindole (DAPI) (Vector Labs, Burlingame, CA, USA). Cell viability was determined as the ratio of propidium iodide-stained cells, to the total number of DAPI-stained cells visualized using a fluorescence microscope.

Immunohistochemistry and Immunocytochemistry

Fresh frozen brain sections (12 μm) or cultured cells on coverslips were permeabilized with 3% Triton X-100. The samples were preblocked with 2% goat serum and 1% BSA, followed by incubation with primary antibody at 4°C overnight. The samples were washed three times with PBS and incubated with secondary antibody for 1 hour at 37°C (Jackson Labs). Slides were washed and mounted in a medium containing DAPI (Vector Laboratories, Burlington, CA, USA). Terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling was performed according to manufacturer's instructions (Roche, Indianapolis, IN, USA). Immunolabeling was studied using a Leica microscope under Ex/Em wavelengths of 340/425 nm (blue), 500/550 nm (green) and 580/630 nm (red), respectively. Images were collected using an Optronics DEI-750 3-chip camera equipped with a BQ 8000 sVGA frame grabber and analyzed using Bioquant (Nashville, TN, USA).

Chromatin Immunoprecipitation Assay (ChIP)

ChIP assays were performed according to the method described by Impey et al (2002). After preconditioning OGD, cells were fixed with paraformaldehyde (2.0%, 2 mins) and lysed using the following buffer; Tris (10 mmol/L, pH 8.0), NaCl (140 mmol/L), glycerol (5%), sodium deoxycholate (0.1%), SDS (0.1%), Triton X-100 (1%), and Protease inhibitors with EDTA (Roche Complete™, Roche, Indianapolis, IN, USA). Samples were cleared (12,000g; 15 minutes) and 800 μg of tissue was subjected to immunoprecipitation with anti-CREB (Abcam, Cambridge, MA, USA) or anti-CBP (Santa Cruz Biotechnology, Santa Cruz, CA, USA) in a total volume of 1.0 ml. After precipitation with protein A/G agarose beads (Santa Cruz Biotechnology, Santa Cruz, CA, USA), the DNA crosslinked to the CREB was released by heating in 0.1 mol/L NaHCO₃ for 16 hours at 65°C, and purified using a PCR clean-up kit (Qiagen, Valencia, CA, USA). The DNA was subject to PCR with primers specific to the bcl-2 CRE promoter region (antisense 5′-TCGCACACTCACTGGGTTACG-3′ and sense 5′-GCGCCTGATTTCCTGTACG-3′) for 38 cycles using Fast start Taq (Roche). The bcl-2 promoter region was identified on the rat genbank (http://www.hgsc.bcm.tmc.edu) as a part of a BAC Sequence (Rat CH236 194D24 contig 15-18). The start site and promoter region were predicted using the Mat-Inspector software (Genomatix, Germany). BlastN searches show low homology with rat DNA sequences, except for the rattus norvegicus bcl-2 gene and promoter (accession number AF531426). (Note there is a single base mismatch between the rat GenBank sequence and AF531426 for the sense oligonucleotide.) Amplicons were run on 2% agarose or 10% acrylamide gels and stained with ethidium bromide. Amplicons for bcl-2 CRE ran as a single band with a weight of 250 bp. The coding bcl-2 region (accession number NM_016993) was amplified with the following primers (sense 5′-TTCCAGCCTGAGAGCAACC-3′; antisense 5′-CCACAGAGCGATGTTGTCC-3′); the amplicon ran as a single band at a weight of approximately 370 bp.

Data Analysis

Data are presented as mean±standard deviation of n determinations. Data were analyzed using repeated-measures one-way analysis of variance (ANOVA) with Bonfferoni's post hoc test (Graphpad Prism, Graphpad Software, San Diego, CA, USA). Significance was accepted at P<0.05.

Ischemic Preconditioning Reduces Neuronal Damage after Harmful Ischemia and Increases Bcl-2 Expression In Vivo

We investigated the role of CREB in mediating the increase of Bcl-2 expression after preconditioning in a focal model of ischemic tolerance (Figure 1A). Ischemic infarct size in rats after MCAO was determined using the vital dye TTC. Cortical infarction after 100 mins MCAO was reduced by preconditioning the animals with 30-min MCAO, 72 hours before the injurious period of ischemia (Figure 1B). In contrast, 30-min MCAO did not cause infarction. We also assessed DNA damage using TUNEL. There was no TUNEL labeling in sham control or animals subject to 30-min MCAO. In contrast, there was a robust increase in TUNEL-positive cells in the brains of animals subjected to 100-min MCAO. The increase in TUNEL-positive cells was reduced in animals subjected to preconditioning 30-min MCAO, 72 hours before 100-min MCAO (Figure 1C).

OPEN IN VIEWER

Figure 1

Preconditioning the brain with mild ischemia results in the brain being tolerant to harmful ischemia. (A) Schematic diagram of in vivo ischemic tolerance paradigm. Tolerance is induced by subjecting animals to 30-minute MCAO 72 hours before a harmful 100-minute MCAO. Assessment of infarct is performed 24 hours after the harmful ischemic insult. The effects of the 30-minute MCAO preconditioning stimulus were investigated in further experiments. (B) Preconditioning reduces cerebral infarct after harmful focal ischemia. Animals were subjected to (1) sham, (2) 100-minute MCAO, or (3) 30-minute MCAO followed by 72-hour reperfusion and then a further 100-minute MCAO. Animals were killed 24 hours after the last MCAO and infarct determined by TTC staining. Note the decrease in infarct in the animals preconditioned with 30-minute MCAO. (C) Preconditioning reduces DNA damage after harmful focal ischemia. Sections of brain from animals subject to (1) sham, (2) 30-minute MCAO, (3) 100-minute MCAO, or (4) 30-minute MCAO followed by 72-hour reperfusion and then a further 100-minute MCAO were processed for DNA damage using TUNEL staining. As can be seen in animals subject to preconditioning, the number of TUNEL-positive cells was reduced.

Protein expression in the protected region of the cortex was measured using Western blotting. Consistent with previous reports (Bossenmeyer-Pourie and Daval, 1998; Shimizu et al, 2001), we detected an increase in Bcl-2 expression after 30-min MCAO and levels remained elevated for up to 72 hours (Figure 2A). In comparison, after harmful ischemia (100-min, MCAO), there was no increase in Bcl-2 expression (not shown).

OPEN IN VIEWER

Figure 2

Preconditioning ischemia activates CREB in the cortex. (A) Western blotting showing expression of Bcl-2, CREB, phosphorylated CREB (Ser 133) and CBP after preconditioning 30-minute MCAO. Images shown are representative blots from three independent experiments using pooled tissue from two animals. (B) Immunohistochemistry showing CREB, phosphorylated CREB and CBP expression after 30-minute MCAO in the region of the cortex made tolerant by preconditioning ischemia, compared with control (sham-operated animals). The schematic brain to the right shows the region of brain made tolerant by preconditioning in gray, and the region studied in red. Images shown are × 40 and representative of two independent experiments, scale bar represents 50 μm. (C) Increased interaction of CREB with bcl-2 CRE after 30-minute preconditioning. After preconditioning, nuclear extracts are incubated with a labeled bcl-2 CRE oligonucleotide and separated by electrophoresis. Note the increased retention of oligonucleotide probe by nuclear extracts. The interaction was reduced when nuclear extracts were incubated with excess cold oligonucleotide (right lane). Data shown are a representative blot of two experiments.

Preconditioning Ischemia Increases CREB Activation In Vivo

It has been shown that Bcl-2 expression might be regulated by the transcription factor CREB (Wilson et al, 1996); hence, we investigated the effect of preconditioning ischemia on CREB expression and phosphorylation, using Western blotting. CREB was expressed as a 43-kDa protein in control (sham) animals. CREB expression did not change significantly after either 30-min MCAO (Figure 2A) either. CREB phosphorylated on Ser 133 has been shown to increase its activity (Walton and Dragunow, 2000), and phosphorylation of CREB (Ser 133) levels increased 1 hour after 30-min MCAO and remained elevated for up to 72 hours (Figure 2A). CREB-binding protein (CBP) is expressed at low levels as a 300-kDa protein in control cortex. CREB-binding protein expression was slightly increased at 24 and 72 hours after 30-min MCAO (Figure 2A). These results suggest that, after preconditioning ischemia, CREB is phosphorylated at Ser 133, which may increase its activity.

We used immunohistochemistry to investigate the expression of CBP, CREB, and phosphorylation of CREB after preconditioning. Sections taken from the cortical area protected by 30-min MCAO (as shown in Figures 1B and 2B) show an increase in phospho-CREB (Ser 133) immunoreactivity after 30-min MCAO and 24 hours reperfusion (Figure 2B). CREB-binding protein and CREB immunoreactivity show no difference between sham control and animals treated with 30-min MCAO with reperfusion for 24 hours (Figure 2B).

To investigate the regulation of bcl-2 by CREB, we used an EMSA to assess the binding of transcription factors to the bcl-2 CRE site after preconditioning. Nuclear extracts were prepared from brain samples of rats subjected to 30 minute MCAO and various reperfusion times (1 to 72 hours). We observed low basal interaction of nuclear proteins with the bcl-2 CRE oligonucleotide (Figure 2C). The interaction was increased after 30-min preconditioning ischemia for up to 72 hours (Figure 2C). Incubating the extracts with excess cold oligonucleotide (Figure 2C) abolished the interaction of the nuclear extracts with the bcl-2 CRE. These results suggest that increased binding of transcription factors to the bcl-2 CRE region may regulate the increase in Bcl-2 expression after preconditioning.

Preconditioning Ischemia Increases CREB Activation In Vitro

The role of CREB in regulating ischemic tolerance was further investigated in an in vitro model of ischemia (Figure 3A). Preconditioning cells with 30-min OGD reduces 120-min OGD-induced cell death when the interval between preconditioning and harmful OGD is 24 hours (Figure 3B). The total number of cells in the cultures did not vary after either 30- or 120-min OGD (Figure 3C). The protective effect of the preconditioning was eliminated in the presence of the protein synthesis inhibitor cycloheximide (0.5 μmol/L) but not vehicle control (0.1% DMSO) (not shown).

OPEN IN VIEWER

Figure 3

Model of ischemic tolerance in cultured cortical cells. (A) Schematic of in vitro ischemic tolerance paradigm. Tolerance is induced by subjecting the cells to 30-minute oxygen and glucose deprivation, followed by 24-hour recovery in normoxic conditions before a harmful 120-minute OGD challenge. Cell death is assessed 24 hours after 120-minute OGD challenge by propidium iodide (PI) staining. (B) Preconditioning cells with 30-minute OGD reduces cell death after 120-minute OGD. Note the lack of effect on cell viability of 30-minute OGD alone. Data shown are mean±s.d. (n=6); ^∗ denotes significance of P<0.01 versus control and ^∗∗ denotes P< 0.01 versus 120-minute OGD. (C) Lack of effect of 30- or 120-minute OGD on cell number in cortical cultures. Cells were subject to the same conditions in (B) and the total number of cells assessed in three views. As can be seen, there was no significant effect of OGD treatments on cell number. Data shown are mean±s.d (n=6).

Consistent with our in vivo results, we saw an increase in Bcl-2 expression and CREB phosphorylation, after 30-min OGD (preconditioning) (Figure 4A), but not after 120-min OGD (injurious) (not shown). Levels of CBP and CREB were not affected by 30 mins (Figure 4A) either. These data support the in vivo data suggesting that there is a concurrent increase in Bcl-2 expression and CREB phosphorylation after ischemic preconditioning.

OPEN IN VIEWER

Figure 4

Increased expression of Bcl-2 and activation of CREB after preconditioning ischemia in vitro. (A) Bcl-2 expression and CREB phosphorylation increases in cells subject to 30-minute OGD. Western blotting was used to determine the expression of Bcl-2, CREB, phosphorylated CREB (Ser 133), CBP and α-tubulin. Data shown are representative blots from at least four experiments. (B) Increased interaction of CREB with bcl-2 CRE oligonucleotide after preconditioning. Nuclear extracts were prepared from cells subjected to 30-minute OGD and incubated with the same bcl-2 CRE oligonucleotide probe as in Figure 2C.

Interaction of CREB and CREB-Binding Protein with the bcl-2 Cyclic AMP-Responsive Element after Ischemia In Vitro

To show an increased interaction of CREB with the bcl-2 CRE, nuclear extracts were prepared from cultured cells after 30- and 120-min OGD and subjected to EMSA with a bcl-2 CRE oligonucleotide, as used in in vivo experiments. A very faint interaction was detected in control cell extracts (Figure 4B). After preconditioning 30-min OGD, the interaction was increased at 6 and 24 hours. Interestingly, we only observed a single band in these EMSA experiments compared with gels from in vivo tissue (Figure 2C). The significance of this is unclear.

We further investigated the interaction of CREB with the bcl-2 CRE site by ChIP. After 30-min preconditioning, cells were subjected to ChIP with anti-CREB (Abcam) antibodies and subjected to PCR using primers specific for the rat bcl-2 CRE or the coding region of the bcl-2 gene (Figure 5A). Initial experiments showed that the bcl-2 CRE was immunoprecipitated when anti-CREB antibody was added (Figure 5B). In contrast, the coding region of the bcl-2 gene was not amplified by coding region-specific primers (Figure 5B). Genomic DNA was used as a positive control for all experiments (Figure 5B). These data suggest that the ChIP conditions used allow the precipitation of specific promoter regions of the bcl-2 gene and not total genomic DNA.

OPEN IN VIEWER

Figure 5

Chromatin immunoprecipitation reveals recruitment of CBP but not CREB to bcl-2 CRE after preconditioning. Cells were subjected to 30-minute OGD, followed by various recovery times. Cells were fixed with 2% formaldehyde, and DNA precipitated by either anti-CREB or anti-CBP antibodies. After clean-up, DNA was subjected to PCR with primers specific for the rat bcl-2 CRE. (A) Schematic showing location of the CRE and start codon of bcl-2 gene. Primers were designed to the CRE and coding region of the bcl-2 gene as indicated. Note numbers of bcl-2 CRE primers are relative to the translation start, whereas location of coding region primers are relative to the start codon of the bcl-2 gene. (B) Preliminary experiments show the specific nature of chromatin immunoprecipitation. Omitting the IP antibody (−) results in no amplification of bcl-2 CRE, compared with incubation with anti-CREB antibody (+). As a positive control, rat genomic DNA was used (DNA). In comparison, PCR using primers for the coding bcl-2 region show that this region of the gene was not amplified from immunoprecipitated DNA. (C) Immunoprecipitation of CREB reveals that bcl-2 CRE binding to CREB did not change after 30-minute preconditioning. Immunoprecipitation with an anti-CBP antibody shows an increased bcl-2 CRE signal after 30-minute OGD preconditioning. Data shown are representative gels from two experiments.

Under basal conditions, the CREB antibody precipitated the bcl-2 CRE region (Figure 5C). However, after 30-min OGD, there was no change in CREB immunoprecipitation of the bcl-2 CRE (Figure 5C). Because CREB requires the recruitment of CBP to activate transcription machinery, we investigated the ability of anti-CBP antibodies to immunoprecipitate the bcl-2 CRE. In comparison with the CREB antibody, the levels of DNA immunoprecipitated with the CBP antibody were very low in control samples (Figure 5C). After 30-min OGD, bcl-2 CRE immunoprecipitated by CBP was increased at 6 and 24 hours (Figure 5c). These data suggest that after 30-min OGD, the binding of CREB to the CRE site does not change, but rather the recruitment of CBP to CREB already bound to the bcl-2 CRE site increases after preconditioning.

Blocking CREB Binding to the bcl-2 Cyclic AMP-Responsive Element Site Inhibits Ischemic Tolerance

To investigate the role of CREB in ischemic preconditioning, we used a CRE decoy oligonucleotide to block the binding of CREB to the bcl-2 CRE, as described by Park et al (1999). Cells were incubated with a hairpin loop of DNA containing the bcl-2 CRE, which differs from the consensus CRE site by one base (TGACGTTA versus TGACGTCA). After preconditioning, cells were incubated for 24 hours with the CRE decoy oligonucleotide or a control oligonucleotide containing a mismatched CRE sequence (both 0.5 μmol/L). As can be seen in Figure 6A, the CRE decoy oligonucleotide blocked the induction of ischemic tolerance, whereas the mismatched sequence had no effect on ischemic tolerance (Figure 6A inset). Of note, there was a slight increase in basal cell death in control decoy oligonucleotide-treated cells, but this was not significant.

OPEN IN VIEWER

Figure 6

Blocking CREB function inhibits ischemic tolerance. Cells were preconditioned with 30-minute OGD and then incubated with a CRE decoy oligonucleotide (0.5 μmol/L) for 24 hours before 120-minute OGD. The neuroprotective effect of preconditioning was blocked by the CRE decoy oligonucleotide, but not the control oligonucleotide (inset). Data shown are mean±s.d. (n=4). ^∗P<0.01 versus untreated cells. (B) CREB phosphorylation was assessed by immunocytochemistry in cells preconditioned with 30-minute OGD and left to recover for 6 hours with CRE decoy oligonucleotide. Cyclic AMP-responsive element decoy oligonucleotide did not block CREB phosphorylation, compared with untreated cells. Images shown are × 40, and representative of two experiments. Scale bar represents 20 μm. (C) Effect of CRE decoy oligonucleotide on bcl-2 CRE binding to nuclear proteins. Nuclear extracts were prepared from cells subjected to 30-minute preconditioning and recovered for 6 or 24 hours with CRE decoy oligonucleotide. Data shown are representative blots of two experiments.

To elucidate the mechanism of action of the CRE decoy oligonucleotide, we investigated the activation of CREB. The CRE decoy oligonucleotide did not inhibit phosphorylation of CREB (Ser 133) after 30-min OGD preconditioning (Figure 6B). The CRE oligonucleotide blocked the CREB bcl-2 CRE interaction after 30-min OGD, as demonstrated by EMSA (Figure 6C). Hence, these data confirm previous reports that blocking CREB function with a CRE decoy oligonucleotide blocks ischemic tolerance (Hara et al, 2003).

Multiple Protein Kinases Regulate Ischemic Tolerance and CREB Phosphorylation

CREB is a substrate for many protein kinases (Beitner-Johnson and Millhorn, 1998; Bonni et al, 1999; Pugazhenthi et al, 2000; Sun et al, 1994; Walton and Dragunow, 2000; Zanassi et al, 2001); hence, we investigated the role of multiple protein kinases in the regulation of ischemic tolerance. Specifically, we examined the effect of the PKA inhibitor H89 (0.5 μmol/L), the calcium/calmodulin kinase inhibitor KN62 (1.0 μmol/L), blocking p42/p44 MAP kinase activation with the Mek inhibitor U0126 (10 μmol/L) and blocking the activation of Akt with the PI3K inhibitor LY294002 (10 μmol/L). After preconditioning with 30-min OGD, cells were incubated with H89, KN62, U0126, or LY294002 for 24 hours and then subjected to a further 120-min OGD. Compared with control untreated cells, H89, KN62, U0126, or LY294002 had no effect on basal cell viability (9.3±4%, 17.5±9%, 11.7±8%, 9.3±6% and 7.7±7% cell death, respectively, all P>0.05) or cell death after 120-min OGD (Figure 7A). Cells treated with H89, KN62 or U0126 showed a significant inhibition of the neuroprotective effect of 30-min preconditioning (Figure 7A). LY294002 did not block ischemic tolerance, consistent with previous reports (Namura et al, 2000). LY294002 had no effect on the induction of tolerance, hence we did not study this compound further (Figure 7A).

OPEN IN VIEWER

Figure 7

Blocking protein kinases inhibits ischemic tolerance, CREB function and Bcl-2 expression. (A) Cells were subjected to 30-minute OGD preconditioning and then incubated for 24 hours with H89 (0.5 μmol/L), KN62 (1.0 μmol/L), U0126 (10 μmol/L) and LY294002 (10 μmol/L), before being subjected to a further 120-minute OGD. Cell death was assessed 24 hours later. H89, KN62, and U0126 all reduced the neuroprotective effect of preconditioning, but LY 294002 did not. Data shown are mean±s.d. (n=5/6). ^∗ denotes P<0.05 versus preconditioned cells subjected to 120-minute OGD. (B) Effect of H89, KN62, and U0126 on CREB phosphorylation after 30-minute preconditioning. Cells were subjected to 30-minute OGD, and then recovered in the presence of the protein kinase inhibitors. CREB phosphorylation was assessed by immunocytochemistry. H89 and KN62 blocked CREB phosphorylation, but U0126 did not. Images are × 40 representative samples from two independent experiments; scale bar represents 40 μm. (C) Effect of H89, KN62, and U0126 on CBP immunoprecipitation of the bcl-2 CRE. Cells were subjected to 30-minute OGD, incubated with protein kinases inhibitors for 24 hours and then CBP binding to the bcl-2 CRE assessed by ChIP assay. (D) Effect of H89, KN62, and U0126 on Bcl-2 protein expression after preconditioning. Cells were subjected to 30-minute OGD and then incubated with protein kinase inhibitors for 24 hours. Some cells were incubated with the CRE decoy oligonucleotide. Bcl-2 expression was assessed by Western blotting. Data shown are a representative blot of two independent experiments.

Since H89, KN62, and U0126 blocked ischemic tolerance, we investigated the effects of these compounds on CREB phosphorylation. After 30-min OGD, there was an increase in the nuclear expression of phosphorylated CREB (Figure 7B). In cells treated with KN62 and H89, the increase in CREB phosphorylation (Ser 133) after 30-min OGD was reduced (Figure 7B). In contrast, in cells subjected to 30-min OGD and then recovered in the presence of U0126, CREB phosphorylation was still observed.

We decided to further investigate the effect of the protein kinase inhibitors using a ChIP assay. The immunoprecipitation of the bcl-2 CRE with anti-CBP antibody was increased after 30-min OGD preconditioning (Figure 7C). The increase in bcl-2 CRE immunoprecipitation after preconditioning was reduced by H89, KN62, and U0126 (Figure 7C). These data suggest that protein kinase inhibitors that block either the phosphorylation of CREB or the recruitment of CBP to the bcl-2 CRE block ischemic tolerance.

Inhibition of Bcl-2 Protein Expression after Preconditioning Reduces Ischemic Tolerance

Finally, we examined the effect of H89, KN62, and U0126 on Bcl-2 protein expression after preconditioning. Bcl-2 protein expression was increased 24 hours after 30-min preconditioning (Figure 7D). The increase in Bcl-2 expression after preconditioning was reduced by H89, KN62, and U0126. Furthermore, the CRE decoy oligonucleotide, which blocks CREB activation, also reduced Bcl-2 expression (Figure 7D). Taken together, these results suggest that blockade of CREB function prevents the increase in Bcl-2 expression after preconditioning, inhibiting ischemic tolerance.