The drugs and reagents used in the present study included propofol (Diprivan; AstraZeneca, London, UK), Evans blue (EB) and dimethylformamide (Sigma-Aldrich, St. Louis, MO, USA). Antibodies included rabbit anti-mouse MMP-9 polyclonal antibody and JNK/pJNK polyclonal antibody (Santa Cruz Biotechnology, Inc., Dallas, TX, USA) rabbit anti-mouse AQP-4 antibody (Chemicon^®, EMD Millipore, Billerica, MA, USA) and rabbit anti-mouse β-actin antibody (Sigma-Aldrich). A western blotting kit was purchased from KPL (Gaithersburg, MD, USA).

Ninety-six healthy adult Wistar rats, weighing 260–300 g and aged 4–6 weeks (Experimental Animal Centre of Zhongshan University, Guangzhou, China) were kept in a 12 h/12 h light/dark cycle at 28–32°C with free access to food and water. Rats were randomly divided into four groups: The sham operation group (group S, n=24), the ischemia reperfusion group (group I, n=24) and two propofol-treated groups, which included group P1 (20 mg/kg/h profopol, n=24) and group P2 (40 mg/kg/h profopol, n=24). The study was approved by the Animal Care Committee of Sun Yat-sen University (Guangzhou, China) and performed in strict accordance with the National Institutes of Health Guide for the Use of Laboratory Animals.

The rats were subjected to fasting without drinking for 12 h prior to surgery. Anesthesia was induced in a Plexiglass chamber with 4% halothane (Sigma-Aldrich), followed by tracheal intubation and mechanical ventilation with 1.3% halothane in 30% O₂/70% N₂O. The left femoral artery was cannulated to monitor blood pressure and spontaneous breath was conserved. The rat was prepared for surgery by the removal of hair in the middle of the neck in the supine position. Subsequently the skin was prepared with 75% alcohol and a surgical incision was made at the flank of the neck. The right common carotid artery, right external carotid artery, right internal carotid artery and wing palatal artery were isolated. The external carotid and wing palatal arteries were ligatured and the right common carotid artery was blocked by a bulldog clamp, while the right external carotid artery was punctured 0.5 cm from the ligation at the proximal end to heart. A nylon thread was inserted into the wound and pushed through the right internal carotid artery in the direction of the brain and the opposite side of the artery was pierced 18–20 mm further down. The middle cerebral artery was then blocked. Following surgery, cerebral blood flow was required to be reduced to ~30%; otherwise, the rat model was considered unsuccessful and removed (10). Two hours following the arterial block, the thread was removed, the residual end of the artery was fastened and the skin and subcutaneous tissue were sutured. Once the rats regained consciousness, behavioral disorders were assessed according to the Longa method (11) using the following scale: No visible neural function loss, 0; left forepaw unable to straighten, 1; walking impaired with an inclination towards ischemic hemisphere, 2; repeated left turning, 3; falling left, 4; unable to walk, 5.

The P1 and P2 groups were treated with 20 mg/kg/h or 40 mg/kg/h propofol, respectively, for 30 min via the tail vein, 30 min following reperfusion. Group I was treated with an equal amount of physiological saline (Sigma-Aldrich) and in group S, the incision was made but no arterial occlusion was performed.

The water content of the ischemic hemisphere of the brain was detected using the wet/dry weight ratio method with the following formula: Water content (%)=100×[(wet weight)−(dry weight)]/wet weight.

The permeability of the BBB was determined according to the protocol described by Uyama et al (12), by the detection of EB fluorescence. Twenty-three hours following reperfusion, rats were anesthetized with 10% chloral hydrate and 2% EB (3 ml/kg) was injected into the femoral vein. One hour later, the ventriculus sinister was washed with physiological saline for 0.5 h. The brain was removed by decollation and surface moisture was blotted with filter paper. The cerebral cortex of the ischemic side was excised and the wet weight was measured with an electronic balance prior to incubation with 4 ml dimethylformamide (Sigma-Aldrich) at 50°C for 48 h followed by centrifugation at 1509 × g for 15 min. The absorbency of the supernatant at 620 nm was determined using a Hitachi U2001 (Hitachi, Ltd, Tokyo, Japan). The content of EB (µg/g) was subsequently calculated according to the standard curve method (12).

Half of the animals in each group were randomly selected to be sacrificed by administration of 5% halothane (Sigma-Aldrich) and fixed in the dorsal position immediately following behavioral testing. The rat hearts were exposed by opening the chest from the xiphisternum and were perfused through the ascending aorta with 150 ml saline followed by 400 ml 4% paraformaldehyde. Following perfusion, the brain was removed, fixed in 4% paraformaldehyde (Sigma-Aldrich) overnight and subsequently transferred to 30% sucrose (Sigma-Aldrich) and incubated for 3–5 days. The damaged brain area was dissected into 1-mm coronal sections. Sequential 10-µm coronal sections were cut from the sections (from bregma −2.3 to −2.45 mm) by cryomicrotomy (CM1900; Leica Microsystems GmbH, Wetzlar, Germany), the sections were treated with 0.3% H₂O₂ (Sigma-Aldrich) in methanol for 30 min, hydrated gradually in distilled water and incubated for 2 h with 5% goat serum (Sigma-Aldrich) to block nonspecific immune reactions. The sections were subsequently incubated at 4°C overnight with rabbit polyclonal anti-MMP-9 or anti-AQP-4 antibodies (1:2,000). Following washing with Tris-buffered saline (Sigma-Aldrich), the sections were incubated with biotinylated goat anti-rabbit immunoglobulin G (IgG) (1:200; Zhongshan Golden Bridge Biotechnology Co., Ltd, Beijing, China) for 2 h followed by incubation with avidin-biotinperoxidase complex (1:200; Zhongshan Golden Bridge Biotechnology Co.) for 2 h. Finally, the sections were exposed to 0.01% 3,3′-diaminobenzidine (Sigma-Aldrich) for 0.5–2 min, followed by examination under a light microscope (Olympus BX51; Olympus Corp., Tokyo, Japan). The optical density of MMP-9 and AQP-4 immunoreactivity was evaluated in an area of 1×1 mm² in the boundary zone.

Twenty-four hours following reperfusion, rats were sacrificed. The brains were removed and immediately cut into two 2-mm sections on dry ice, excluding 4 mm of the rostral tissue. Slices were cut into the right and left hemispheres, and the right striatum and parietal cortex were isolated and homogenized in tissue lysate buffer (10 ml/mg tissue) containing a mixture of proteinase inhibitors (Thermo Fisher Scientific China Co., Ltd, Beijing, China), prior to dispersion by sonication. The homogenate was centrifuged at 2,250 × g for 30 min at 4°C, and the supernatant was saved. A bicinchoninic acid assay (Pierce Biotechnology, Inc.; Thermos Fisher Scientific, Rockford, IL, USA) was used to determine the protein concentration. Proteins (30 µg/well) were separated on using 10% SDS-PAGE and transferred to nitrocellulose membranes (Sigma-Aldrich) at 100 mV. The membranes were blocked with 5% non-fat milk (Sigma-Aldrich) in phosphate-buffered saline with 0.1% Tween-20 (Sigma-Aldrich) for 1 h at room temperature and incubated overnight at 4°C with rabbit anti-mouse MMP-9 polyclonal antibody (1:2,000; Santa Cruz Biotechnology, Inc.), rabbit anti-mouse AQP-4 antibody (1:2,000; Abcam, Cambridge, MA, USA) or rabbit polyclonal anti-pJNK antibody (1:2,000; Santa Crux Biotechnology, Inc.). Following three washes in Tris-buffered saline for 5 min, the blots were incubated for 1 h at room temperature with the appropriate horseradish peroxidase-conjugated secondary antibody (1:3,000; Santa Cruz Biotechnology, Inc.). MMP-9, AQP-4 and pJNK were subsequently visualized by incubation with enhanced chemiluminescence reagent (Thermo Fisher Scientific, Waltham, MA, USA) for 1 min, and exposure onto hyperfilm (Sigma-Aldrich) for 1–10 min. Membranes which were positive for JNK were re-blocked with 5% non-fat milk for 1 h at room temperature and incubated with rabbit polyclonal anti-JNK antibody (Santa Cruz Biotechnology, Inc.) overnight at 4°C. Visualization of JNK was performed as described above. Image J 1.35 software (National Institutes of Health, Bethesda, MD, USA) was used to analyze the relative density values.

Data were analyzed using SPSS 16.0 (SPSS, Inc., Chicago, IL, USA). Values are expressed as the mean ± standard deviation. Differences between groups were analyzed using one-way analysis of variance and multiple comparisons were made using the least significant difference test. Data of skewed distribution are presented as the median. P<0.05 was considered to indicate a statistically significant difference between values.

The assessment of neurological function indicated that all experimental animals exhibited post-ischemic neurological deficits. Rats in groups I, P1, and P2 scored significantly higher than those in group S (P<0.05). Compared with the rats in group I, those in groups P1 and P2 scored significantly lower (P<0.05). The experimental groups demonstrated higher water content and EB content in the ischemic hemisphere in comparison to that of group S (P<0.05). The water and EB content were significantly reduced in groups P1 and P2 compared to those in group I (P<0.05; Table I).

Immunohistochemical analysis indicated that the expression levels of MMP-9 and AQP-4 in ischemic brain tissue in group I were significantly increased compared to those in group S (P<0.05). Following propofol post-conditioning, the expression levels of MMP-9 and AQP-4 were significantly decreased in groups P1 and P2 compared with those in group I (P<0.05), although expression levels remained significantly higher than those in group S (P<0.05; Figs. 1 and 2).

Twenty-four hours following ischemia/reperfusion, the expression levels of MMP-9, AQP-4 and the levels of pJNK were significantly increased. Following treatment with propofol, the expression levels of MMP-9, AQP-4 and pJNK in groups P1 and P2 were significantly decreased (P<0.05). Furthermore, the expression levels of MMP-9, AQP-4 and pJNK were significantly decreased in group P2 compared with those in group P1 (P<0.05; Figs. 3–5).