Adult male Sprague-Dawley rats weighing 250–300 g were provided by the Experimental Animal Center of Chongqing Medical University (Chongqing, China) and were provided with food and water ad libitum. The animal experiments were conducted in accordance with the guidelines of the Institutional Animal Care and Use Committee of Chongqing Medical University. Food and water were withdrawn 24 h prior to surgery. Resveratrol was purchased from Hongda Biotechnology Co., Ltd. (Linxia, China) and dissolved in water to prepare a 20% (w/v) solution. A superoxide dismutase (SOD) detection kit and an iNOS ELISA kit were purchased from Uscn Life Sciences, Inc. (Wuhan, China). Polyclonal anti-AQP4 antibodies (cat. no., LS-C3805-50; dilution, 1:50) were obtained from LifeSpan BioSciences (Seattle, WA, USA) 1:50). Additional general reagents were obtained from Nanjing Jiancheng Bioengineering Institute (Nanjing, China).

Transient focal cerebral I/R was produced by right middle cerebral artery occlusion (MCAO) as described previously (17). In brief, the rats were anesthetized with 300 mg/kg chloral hydrate (ChengDu Kelong Chemical Co., Ltd., Chengdu, China). Under an operating microscope, the right common carotid artery (CCA), the right external carotid artery (ECA) and the right internal carotid artery (ICA) were isolated and a 6–0 silk suture (Johnson & Johnson (Shanghai) Medical Equipment Co., Ltd., Shanghai, China). was tied at the proximal and distal ends of the right ECA. The right ECA was then occluded. A 4-0 nylon monofilament (Johnson & Johnson (Shanghai) Medical Equipment Co., Ltd.). was introduced into the right ECA and pushed up the ICA until light resistance was felt. The filament was inserted approximately 18–20 mm from the carotid bifurcation to effectively block the middle cerebral artery (MCA). The diameter of the tip of the suture was ~300 µm. Following 90 min of MCAO, the suture was removed to restore blood flow (reperfusion confirmed by laser-Doppler flowmeter; Periflux System 5010, Perimed AB, Järfälla, Sweden). Sham-operated rats were manipulated in the same way, however the MCA was not occluded. Relative cerebral blood flow (rCBF) was monitored using the laser-Doppler flowmeter with the use of a flexible probe over the skull. rCBF was measured prior to ischemia, during MCAO and during reperfusion. Animals that did not demonstrate a CBF reduction of a minimum of 70% and animals that perished following ischemia induction were excluded from the experimental group. Neurological deficits were assessed in each animal on a numerical scale of 0–4 at the moment of reperfusion and 24 h following reperfusion (18). Rating criteria: 0, no detectable deficits; 1, forelimb flexion and torso turning to the contralateral side when lifted by the tail; 2, same behavior as grade 1 and decreased resistance to a lateral push; 3, same behavior as grade 2 with unilateral circling; 4, no spontaneous walking and a depressed level of consciousness. Rats with a neurological deficit less than 2 or greater than 3 were excluded from the study.

Male rats (n=60) were randomly assigned to four groups (n=10/group): i) Sham-operation group, received a single intraperitoneal injection of 0.9% NaCl; ii) I/R group, received an intraperitoneal injection of 0.9% NaCl 10 min prior to operation in addition to at 0 and 20 h following reperfusion; iii) edaravone group, received an intraperitoneal injection of 10 mg/kg edaravone (Jilin Province Huinan Changlong Biopharmacy Co., Ltd., Jilin, China) 10 min prior to operation in addition to at 0 and 20 h following reperfusion; vi) resveratrol group, received an intraperitoneal injection of 20 mg/kg resveratrol at 10 min prior to operation in addition to at 0 and 20 h following reperfusion. The remaining 20 rats were removed from the experiments.

Animals were sacrificed 48 h following reperfusion and brains were rapidly removed and frozen at −20°C for 20 min. The brains (n=8/group) were sliced into 2 mm-thick coronal sections and immersed in 2% 2,3,5-tripenyltetrazolium chloride (TTC; Beijing Solarbio Science & Technology Co., Ltd., Beijing, China) at 37°C for 50 min for staining of damaged tissue. Infarct size was measured using an image analysis system (Image-Pro Plus software, version 6.0; Media Cybernetics, Inc., Rockville, MD, USA). For the measurement of brain edema, the water content of the brain was assessed by a wet-dry method. Briefly, the brain slices were dried at 110°C for 24 h and the tissue water content was calculated as the percentage of the wet tissue weight as follows: % water = [ ( wet weight − dry weight ) × 100 ] / wet weight

Following transcardiac perfusion with 200 ml 4% formaldehyde in 0.1 M phosphate-buffered saline (PBS; pH 7.4), the rat brains were removed, post-fixed in 4% formaldehyde overnight and processed for paraffin embedding and cutting. Serial coronal sections (4–6 µm thick) were selected and processed for hematoxylin and eosin staining (Shanghai Hualan Chemical Technology Co., Ltd., Shanghai, China).

Brain tissues of the right hemisphere were homogenized in cold 0.1 mol/l sodium phosphate buffer (pH 7.4; Wuhan Boster Biological Technology, Ltd., Wuhan, China) to make a 10% homogenate. The homogenate was centrifuged for 15 min at 3,000 × g and 4°C, and the supernatant was used for bioassays. The absorbance of the supernatant was measured spectrophotometrically (Shanghai Flash Spectrum Biological Technology Co., Ltd., Shanghai China) at 532 nm at room temperature and the results are expressed as nanomoles of MDA/mg protein.

The production of superoxide was assessed by measuring the SOD-inhibited reduction of a water soluble tetrazolium salt (WST-1; Wuhan USCN Business Co., Ltd., Wuhan, China) as described previously (19,20). Briefly, xanthine-xanthine oxidase (Wuhan USCN Business Co., Ltd.) was used to generate a superoxide flux. The absorbance obtained from a nitroblue tetrazolium (NBT; Wuhan USCN Business Co., Ltd.) reduction to blue formazan by superoxide was determined at 560 nm spectrophotometrically at room temperature. SOD in the sample competes for superoxide, inhibiting the rate of the reaction of superoxide with NBT. The rate of NBT reduction in the absence of tissue was used as the reference rate (0.020±0.005 absorbance units/min). The standard assay substrate mixture contained in 0.2 ml 1 mol/l xanthine (Wuhan USCN Business Co., Ltd.), 0.1 mol/l EDTA (Nanjing Jiancheng Bioengineering Institute), 5.6×10⁻² mol/l NBT and 1 mg/ml bovine serum albumin in 0.1 mol/l sodium phosphate (pH 7.8) (both from Nanjing Jiancheng Bioengineering Institute).

iNOS levels were measured with an iNOS ELISA kit according to the manufacturer's instructions. Briefly, the standards and test samples were added to pre-prepared rat iNOS monoclonal antibody ELISA plates, incubated with the plates for 2 h at 37°C and washed with scrubbing solution to remove unbound components. Enzyme working solution was then added, incubated for a sufficient amount of time, and washed with the wash buffer to remove unbound components. Solutions A, B and the 3,3′,5,5′-tetramethylbenzidine (TMB) substrate were then sequentially added, followed by the stop solution which catalyzed the conversion of the blue substrate into a yellow product. iNOS levels were calculated from optical density measurements taken at a wavelength of 450 nm, according to the values for the standards using Image-Pro Plus 6.0 (Media Cybernetics, Inc., Rockville, MD, USA)

Brain tissues were removed following perfusion with 4% formaldehyde for tissue fixation, were embedded in paraffin and then serially sectioned (4 µm). The rat brain sections were washed in 0.1 mol/l PBS 10 times and placed in 3% H₂O₂ to quench endogenous peroxidase activity for 15 min. The sections were pre-incubated with 10% normal goat serum (Nanjing Jiancheng Bioengineering Institute) for 30 min at 37°C to block non-specific binding and incubated overnight at 4°C with the rabbit AQP4 antibody. Subsequently, the sections were washed in PBS and incubated with biotinylated anti-rabbit IgG antibodies (cat. no SA1028; Wuhan Boster Biological Technology, Ltd.) for 1 h at room temperature. Following washing, the sections were incubated for 30 min with streptavidin-biotin and horseradish peroxidase (Wuhan Boster Biological Technology, Ltd.) and developed with 3,3′-diaminobenzidine (Beijing Solarbio Science & Technology Co., Ltd.). For negative controls, the primary antibodies were replaced with PBS. AQP4 expression was measured by the mean optical density (MOD) using Image-Pro Plus software, version 6.0.

Western blot analysis was used to measure AQP4 protein expression 48 h following I/R injury. Following transfer, the membrane was blocked in 5% non-fat milk (Beijing Solarbio Science & Technology Co., Ltd.) in TBST [1X Tris-HCL (1,000 ml), Tween 20 (1 ml)] for 1 h at room temperature. The blots were incubated with rabbit polyclonal anti-AQP4 (1:2,000), or rabbit monoclonal anti-β-actin (1:2,000; sc-47778; Santa Cruz Biotechnology, Inc., Dallas, TX, USA) in 5% non-fat milk and TBST overnight at 4°C. Following three washes with 1X TBST for 5 min each, the blot was incubated with the horseradish peroxidase-conjugated goat anti-rabbit IgG antibody (1:4,000; cat. no. GAR007) MultiSciences, Lianke, Biotech Co., Ltd., Hanzhou, China) in TBST for 1 h at room temperature. Following washing with TBST, immunoreactive proteins were revealed with the enhanced chemiluminescence SuperSignal West Pico kit (Bio-Rad Laboratories, Inc., Hercules, CA, USA) on X-ray film (Bio-Rad Laboratories, Inc.), and the expression level was measured by densitometry. Band density values were normalized to that of β-actin.

Data are expressed as the mean ± standard deviation. Statistical analyses of data were performed using one-way analysis of variance followed by the Student-Newman-Keuls test. P<0.05 was considered to indicate a statistically significant difference. All data analyses were conducted using SPSS software, version 13.0 (SPSS, Inc., Chicago, IL, USA).

At 24 h following reperfusion, the total infarct size was assessed by TTC staining with volumetric analysis. As presented in Fig. 1, there were no observed lesions or non-viable tissue in the sham-operated brains. By contrast, brain sections from the I/R group exhibited lesions visible as white patches in the areas supplied by the MCA, which were observed overlaying the cortex and penetrating into the subcortical tissue. Resveratrol treatment significantly reduced (P<0.05) the lesions to 14.40±2.94% in the cerebral cortex and subcortical brain tissue compared with the I/R group. In the edaravone group, lesion size was significantly reduced (P<0.01) to 13.49±1.26% (Fig. 1F). Furthermore, cerebral I/R-induced injury significantly increased brain edema, which was inhibited by edaravone and resveratrol (P<0.01; Fig. 1G). Neurological deficits were measured 22 h following reperfusion. An increase in the severity of the neurological deficits was observed in the I/R group, however these were improved by 20 mg/kg resveratrol (P<0.01 vs. the I/R group; Fig. 1E).

As presented in Fig. 2, neurons of the hippocampal CA1 region in the sham-operated group were tightly packed with clear boundaries and no apoptosis or vacuolar degeneration was observed (Fig. 2A). Following cerebral I/R, the number of neurons was reduced and cell morphology became disrupted, with evidence of necrosis (Fig. 2B). In the I/R + R and I/R + E groups, the number of neurons was not observed to reduce, intact cell morphology was clearly visible and only a small fraction of cells exhibited cell death or vacuolar degeneration (Fig. 2C and D).

Chemocolorimetry was used to measure the brain MDA levels. As presented in Fig. 3A, cerebral I/R increased the brain MDA levels to 11.49±4.21 mg/ml, whereas edaravone and resveratrol significantly reduced this level to 7.15±2.95 and 6.53±1.94 mg protein/ml, respectively (P<0.01; Fig. 3A). Furthermore, cerebral I/R increased the iNOS levels in the focal ischemic cerebral cortex and hippocampus to 28.48±7.05 U/mg protein, compared with the sham control group (19.01±5.56 U/mg protein). Resveratrol and edaravone significantly reduced iNOS levels to 20.17±6.49 and 19.05±7.36 ng/ml respectively (Fig. 3B). Total SOD activity was measured in the cortex and hippocampus. As presented in Fig. 3C, SOD activity was increased with 24 h of reperfusion following MCAO. Furthermore, at 48 h of reperfusion following MCAO, the SOD activity level in the I/R group reduced to a lower level compared with the sham-operated group. Consistently, edaravone and resveratrol were highly effective at increasing SOD activity at 24 and 48 h of reperfusion following MCAO (Fig. 3C and D).

At 48 h of reperfusion following MCAO, immunostaining for AQP4 expression was conducted in the cerebral hemispheres and the hippocampus in the four groups. As presented in Fig. 4, in the sham-operated group few AQP4-positive regions were observed (Fig. 4A). In the I/R group, the distribution of AQP4 immunostaining indicated diffuse, positive expression (Fig. 4B). In the edaravone and resveratrol treatment groups, reduced AQP4 expression was observed in astrocytes compared with the I/R group (Fig. 4C and D). To semi-quantify AQP4 expression, the MOD and the integrated optical density over the measurement area was calculated using Image-Pro Plus software, version 6.0 (Fig. 4E). Western blotting indicated that AQP4 expression was significantly increased following cerebral I/R compared with the control group, however this was inhibited by pre-treatment with edaravone and resveratrol (Fig. 4F and G). Both edaravone and resveratrol significantly reduced the infarct size; however, in the aspect of reducing neurological deficits, no statistically significant differences were detected between the edaravone and control treatment groups.