The experimental protocol of the present study was approved by the ethical committee of the Animal Care and Experimental Committee of the School of Medicine of Shanghai Jiao Tong University (Shanghai, China). A total of 136 age-matched adult male Sprague-Dawley rats (weight, 270–300 g; age, 7–8 weeks), were housed in a temperature- and light-controlled environment with a 14/10-h light/dark cycle. The rats were purchased from the Animal Center of the School of Medicine of Shanghai Jiao Tong University (Shanghai, China). The rats were randomly divided into five groups: Sham-operated (sham; n=24), middle cerebral artery occlusion (MCAO; n=30), early ischemic postconditioning (IPe; n=28), delayed ischemic postconditioning (IPd; n=28) and combinative ischemic postconditioning treatment (IPc; n=26) groups. Transient cerebral ischemia was induced using MCAO, as previously described (13). Briefly, the rats were anesthetized by inhalation of 5% isoflurane (Ohio Medical Corporation, Gurnee, IL, USA), and were treated with 2% isoflurane during surgery and early reperfusion. A heating blanket was applied to the rats, in order to maintain a rectal temperature of 37.0±0.5°C throughout the experiments. A middle neck incision was made to expose the left common carotid artery (CCA). The external carotid artery was ligated using a bipolar coagulator (Hkpromed, Beijing, China). A nylon filament (diameter 0.24–0.28 mm) was gently inserted into the left internal carotid artery, until resistance was felt. Subsequently, the CBF dropped sharply, indicating that the ischemic model was successfully established. Reperfusion was induced by removing the filament after 90 min of ischemia. Sham-operated animals were subjected to the same anesthetic, incision and exposure of the vessels, without arterial occlusion. Ischemic postconditioning was established by repeatedly occluding and releasing the left CCA using aneurysm clips (cat. no. FE 681 K; Aesculap, Inc., Center Valley, PA, USA). In order to explore the combinative effects of early and delayed postconditioning, the two types of postconditioning were combined. The protocol for early plus delayed ischemic postconditioning is shown in Fig. 1A. Both types of postconditioning consisted of 10 cycles of 10-sec reperfusion/10-sec occlusion; early postconditioning was conducted immediately following reperfusion, and late postconditioning took place 3 h after reperfusion.

A laser Doppler flow meter (PeriFlux System 5000; Perimed AG, Järfälla, Sweden) was used to measure CBF throughout the experiments, and was connected to a standard laser Doppler monitor (PeriFlux 5000 Laser Doppler Perfusion Monitor and PeriFlux 5000 main unit; Perimed AG). The holder for the Doppler probe was fixed to the left side of the skull, in the region of the ischemic penumbra (2 mm lateral and 2 mm posterior to the bregma). The baseline value was regarded as the recorded CBF prior to occlusion. All CBF values were expressed as a percentage relative to the baseline.

A fully computerized electronically controlled rotarod treadmill for rats (Accuscan Instruments, Inc., Columbus, OH, USA) was used to investigate motor function. Each rat was placed in a neutral position on a cylinder with a diameter of 3.75 inches. The speed of the rod accelerated linearly between 0 and 24 rpm within 60 sec. The time that rats maintained on the rotarod was recorded automatically. The maximum score given to an animal was fixed at 60 sec. All rats were given three trials and the average score from these three trials was calculated.

The rats were euthanized 3 days after surgery using CO₂ and the brain tissue was harvested after transcardial perfusion with cold phosphate-buffered saline (PBS). Following 3 min of freezing, the brains were cut into six 2-mm coronal sections. The tissue sections were then immediately stained with 2% 2,3,5-triphenyl tetrazolium chloride (TTC) in PBS at 37°C for 20 min. Images of the stained sections were captured using a digital camera (Nikon E5100; Nikon Corporation, Tokyo, Japan), and were then quantified using ImageJ version 1.37c software (National Institutes of Health, Bethesda, MA, USA). The infarct volume was the sum of the measured infarct areas of evenly sliced (2-mm) brain sections, according to Simpson's rule (14).

The cerebral hemispheres, cerebellum and brain stem were separated. The wet weight was regarded as the immediate weight, and the dry weight was regarded as the weight after drying at 104°C for 48 h. The water content percentage was determined using the following formula: [(wet weight-dry weight)/wet weight] × 100.

In order to determine the level of apoptosis in the penumbral area, TUNEL staining was performed on each tissue section, according to the manufacturer's instructions (Roche Diagnostics GmbH, Mannheim, Germany). Briefly, the slides were washed with PBS at room temperature for 30 min and incubated with 0.1% Triton X-100 in 0.1% sodium citrate solution at 4°C for 2 min. Approximately 50 µl TUNEL mixture solution was applied to each slide at 37°C for 1 h. All of the sections were overlaid with prolonged mounting reagent and DAPI (Invitrogen Life Technologies, Carlsbad, CA, USA), and were covered with coverslips. The Nikon ECLIPSE Ti fluorescence microscope (Nikon Corporation) and CoolSNAP photometrics camera (Nikon Corporation) were used to visualize and capture the images. TUNEL-positive cells, indicated by bright green fluorescence, were quantified using the NIS-Elements BR diagnostic software version 3.2 (Nikon Corporation). The number of TUNEL-positive cells was counted from three random 1×1 mm² areas.

Brain sections (10 µm) were incubated with 10% normal goat serum/0.1% Triton-X 100 in PBS blocking solution at room temperature for 1 h. Slides were then incubated with the following corresponding primary antibodies: Polyclonal rabbit anti-rat BDNF (1:200; cat. no. ab6201; Epitomics, Inc., Burlingame, CA, USA), monoclonal anti-neuron-specific nuclear protein (NeuN; 1:500; cat. no. MAB377B; EMD Millipore, Bedford, MA, USA) and monoclonal anti-glial fibrillary acidic protein (GFAP; 1:200; cat. no. G3893; Sigma-Aldrich, St. Louis, MO, USA), overnight at 4°C. Following incubation, slides were then washed and incubated for 1 h with the following secondary antibodies: Alexa Fluor 488 goat anti-rabbit Immunoglobulin G (1:300; cat. no. A31267; Invitrogen Life Technologies, Camarillo, CA, USA) and Alexa Fluor 568 goat anti-rat (1:500; cat. no. A10042; Invitrogen Life Technologies). DAPI was used to stain the nuclei (Sigma-Aldrich). All of the slides were visualized under a Nikon ECLIPSE Ti fluorescence microscope, loaded with a CoolSNAP photometrics camera at 400× magnification.

The expression levels of CREB-ERK pathway-associated proteins were detected by western blot analysis. Protein extraction reagents (Sigma-Aldrich) were applied for total protein isolation and purification. Protein samples (~50 µg) were separated by 11.5% SDS-PAGE (Sigma-Aldrich). Polypeptides were electrophoretically transferred onto an Immobilon polyvinylidene fluoride membrane (EMD Millipore) at 300 mA for 1 h on ice. The membranes were then blocked with 5% non-fat milk in Tris-buffered saline with Tween (TBST). After blocking, the membranes were rinsed with PBS and incubated with the following antibodies: Anti-BDNF (1:500; Epitomics, Inc.), anti-phosphorylated (p) ERK1/2 (1:200; cat. no. 14227s; Cell Signaling Technology, Inc., Danvers, MA, USA), anti-p-CREB (1:600; cat. no. 8212s; Cell Signaling Technology, Inc.) and anti-β-actin (1:1,000; cat. no. A2022; Sigma-Aldrich) overnight at 4°C. The membranes were then washed with TBS and were treated with horseradish peroxidase-conjugated secondary antibodies (1:800; cat. no. sc-2448; Santa Cruz Biotechnology, Inc., Dallas, TX, USA) using 1% non-fat milk in TBST for 2 h at room temperature. The blots were visual-ized using the ChemiDoc™ XRS Imaging system (Bio-Rad Laboratories, Inc., Hercules, CA, USA), and the band densities were quantified using the Multi Gauge Software of Science Lab 2006 (Fujifilm Corporation, Tokyo, Japan). Six animals were analyzed from each group for the western blot analyses.

Data are expressed as the mean ± standard error of the mean. The neurological scores were analyzed using the Mann-Whitney non-parametric test. Statistical analysis was performed using analysis of variance, followed by Dunnett's post hoc test. P<0.05 (95% confidence interval) was considered to indicate a statistically significant difference. SPSS version 15.0 software (SPSS, Inc., Chicago, IL, USA) was used to conduct statistical analyses.

In order to determine the effects of the combination of early and delayed ischemic postconditioning on brain ischemic injury, the infarct size due to ischemic insult was measured. Representative images of TTC-stained brain sections from rats subjected to MCAO are shown in Fig. 1B. The rats in the sham group exhibited no cerebral infarcts. In the MCAO group, large infarcts developed, which were predominantly located in the cerebral cortex and striatum. The normalized infarct volume in the MCAO group was 28.16±1.79%. Early and delayed ischemic postconditioning reduced the infarct volume, and the normalized infarct volumes in these groups were 9.38±0.96 and 10.33±1.31%, respectively. The combination of early and delayed ischemic postconditioning further reduced the infarct volume to 2.22±0.65%, which was approximately 22% lower than that of the MCAO group (P<0.01), 7% lower than that of the IPe group (P<0.05) and 8% lower than that of the IPd group (P<0.05, Fig. 1C).

The rotarod test was performed in order to test the neurological function of rats following ischemic injury. The results indicated that early, delayed and combinative ischemic postconditioning significantly attenuated the neurological deficits, compared with the MCAO group (P<0.05, Fig. 1D). However, no significant differences were detected among the three ischemic postconditioning groups.

During ischemia, CBF was decreased to equivalent extents (51.64±3.08% of baseline) in all groups, apart from the sham group. Following reperfusion, a short period of hyperperfusion (168.67±6.11%), which lasted ~50 min, followed by a long period of hypoperfusion, was observed in the MCAO group (Fig. 2A). Although early postconditioning did not shorten the time of hyperperfusion, the level of hyperperfusion was reduce to 138.67±3.28% in the IPe group (P<0.05 compared with the MCAO and IPd groups) and 132.87±3.31% in the IPc group (P<0.05 compared with the MCAO and IPd groups). Following ~3 h of hypoperfusion, delayed postconditioning increased the CBF value to 119.67±1.17% in the IPd group (P<0.05 compared with the MCAO and IPe groups), and 122.17±2.74% in the IPc group (P<0.05 compared with the MCAO and IPe groups). These results suggest that combinative ischemic postconditioning significantly attenuated early hyperperfusion following ischemic reperfusion, as well as the subsequent hypoperfusion.

The wet-dry method was used to assess the level of brain edema 3 days after MCAO. The left brain hemisphere (84.36±0.72%) contained more water than the right brain hemisphere (80.81±0.88%) in the MCAO group. Early, delayed and combinative ischemic postconditioning all significantly reduced the water content of the left brain hemisphere (P<0.05), compared with the MCAO group. However, no significant differences were detected among the three ischemic postconditioning groups (Fig. 2B).

In pathological and anatomical terms, the penumbra is an area surrounding a severely ischemic core. This area is where pharmacological interventions are most likely to attenuate the neurological injury and improve clinical outcome (Fig. 3A). TUNEL-positive staining was absent in the sham-treated group (Fig. 3B). Numerous bright green dots, which indicate the presence of TUNEL-positive cells, were observed in the MCAO group (Fig. 3C). There were few TUNEL-positive cells observed among rats that were administered with early, delayed and combinative ischemic postconditioning treatment (Fig. 3D–F). Quantitative analysis indicated that the number of TUNEL-positive cells in the MCAO group was 409.78±45.23. Early and delayed ischemic postconditioning significantly reduced the level of apoptosis, compared with the MCAO group (P<0.05, Fig. 3G). Combinative ischemic postconditioning further decreased the rate of apoptosis following ischemia in the penumbral area (P<0.01, compared with the MCAO group; P<0.05, compared with the IPe and IPd groups; Fig. 3G).

The present study sought to determine whether ischemic postconditioning induces BDNF in the penumbral area, using immunofluorescence staining. Neurons in the sham control group exhibited BDNF-positive immunoreactivity, as assessed by BDNF+NeuN double staining (Fig. 4A). Astrocytes in this groups were also positive for BDNF, as assessed by BDNF+GFAP double staining (Fig. 4F). BDNF immunoreactivity was significantly lower in the MCAO group, compared with the sham group (Fig. 4B and G). Neither early nor delayed ischemic postconditioning alone had a significant effect on expression of BDNF (Fig. 4C, D, H and I). However, BDNF immunoreactivity was higher in the combinative ischemic postconditioning group, compared with that in the MCAO, IPe and IPd groups (Fig. 4E and G). Quantitative analysis also indicated that BDNF-positive signals were higher in the IPc group, compared with those in the MCAO (P<0.01), IPe (P<0.05) and IPd groups (P<0.05; Fig. 4K).

Western blot analyses indicated that combinative ischemic postconditioning increased the protein expression levels of BDNF in the penumbral area (Fig. 5A). Western blot analysis was also performed to investigate whether combinative ischemic postconditioning results in the phosphorylation of CREB and ERK1/2 within the cortex. No significant differences in the protein expression of total CREB and ERK1/2 were detected among the groups. However, administration of combinative ischemic postconditioning significantly increased the expression of p-CREB, compared with that of the MCAO (P<0.01), IPe (P<0.05) and IPd groups (P<0.05; Fig. 5A and B). Similar results were obtained for the expression of p-ERK1/2 (Fig. 5A and B), which is the upstream phosphorylating enzyme of CREB.