Cur was purchased from Sigma-Aldrich; Merck KGaA (Darmstadt, Germany). Antibodies against phosphorylated anti-(p)-MEK), anti-MEK, anti-p-ERK, anti-ERK, anti-CREB, anti-p-CREB, B-cell lymphoma-2 (Bcl-2), Bcl-2 associated X protein (Bax), β-actin antibodies and horseradish peroxidase (HRP)-conjugated secondary antibodies were all provided by Sigma-Aldrich (Merck KGaA). Terminal dexynucleotidyl transferase (TdT)-mediate d dUTP nick end labeling (TUNEL) assay kits were obtained from Roche Diagnostics (Basel, Switzerland). Lactate dehydrogenase release assay (LDH) kits were purchased from Nanjing Jiancheng Bioengineering Institute (Nanjing, China). Dulbecco’s modified Eagle media: Nutrient Mixture F-12 (DMEM/F12) medium, fetal calf serum and trypsin were purchased from Hyclone (Logan, UT, USA). Unless otherwise stated, triphenyl tetrazolium chloride (TTC) powder, PBS buffer, paraformaldehyde, western blot reagents, tetrazolium blue tetrazolium bromide (MTT) and all other chemicals used were provided by Sigma Aldrich; Merck KGaA.

A total of 60 male Sprague-Dawley (SD) rats (280-320 g) for in vivo experiment and a total of 5 SD rats (1-2 days old) for primary astrocyte cell culture were provided by the Experimental Animal Center of Chongqing Medical University (Chongqing, China) Specific Pathogen Free [animal production license no. SCXK (Chongqing) 2011-0001]. Animals were kept in a controlled room with adequate food and water, constant temperature and humidity (22°C and 55%, respectively) as well as a 12/12 h light/dark cycle.

Experimental protocols were all approved by the Chongqing Medical University’s Institutional Animal Care and Use Committee. Using Longa’s method, focal cerebral ischemia and reperfusion of middle cerebral artery occlusion models of rats were performed. Rats were anesthetized with 4% chloral hydrate through intraperitoneal injection, then an incision in the skin was made to expose the right common carotid artery, internal carotid artery (ICA) and external carotid artery. The right middle cerebral artery occlusion model was established by introducing an embolus through the ICA. The embolus was inserted into the internal carotid artery until the tip reached the origin of the middle cerebral artery (18-22 mm). After 120 min of ischemia, the embolus was removed. After the procedure, the rats recovered in pre-warmed cages. In total, 60 rats were divided into 4 groups randomly as follows: i) Sham operation control group (SC) (n=15); ii) ischemia-reperfusion group (IR) (n=15); iii) Cur-treatment group (IC) receiving cur at 100 mg/kg (n=15); iv) IC receiving Cur at 300 mg/kg (n=15). Pre-treatment with Cur took place 30 min prior to surgery and daily via intraperitoneal injection. The sham group and ischemia-reperfusion group received similar pre-treatment with normal saline.

A total of 24 h following ischemia/reperfusion, the neurological deficit score was examined by referencing the standard 5 point system established by Longa et al (20): i) No obvious signs of deficit (score 0); ii) failure to extend the right forepaw fully (score 1); iii) circling to the right (score 2); iv) falling to the right (score 3); and v) loss of walking abilities (score 4).

After evaluation of neurological deficit scores, 3 rats were anesthetized with 4% chloral hydrate via intraperitoneal injection and rapidly decapitated. Brain tissue was removed quickly, followed by the pia mater and brain tissue blood. Tissue wet weight (A) and dry weight (B, tissue was dried in an oven for 24 h) were weighed (accurate to 0.1 mg). Finally, the brain water content was calculated in accordance with the Blliot formula: Brain water content = (A - B)/A ×100%.

Following evaluation of neurological deficit scores, 3 other rats were anesthetized with 4% chloral hydrate via intraperitoneal injection and rapidly decapitated. The brain was removed and coronally sectioned into 2 mm slices, which were placed into 2% TTC solution at 37°C. After 20 min, slices were stained and images were captured with a camera. Noninfarcted areas of the brain tissue were stained red; infarcted areas appeared white. CMIAS-008 image analysis software (Institute of Beijing University of Aeronautics and Astronautics, Beijing, China) was used to calculate the ratio of the infarct size/the total area.

After evaluation of neurological deficit scores, 3 rats were perfused transcardially with 0.9% NaCl and 2.5% glutaraldehyde and rapidly decapitated. Brain tissue sized 1 mm³ were quickly removed and fixed in 2.5% glutaraldehyde at 37°C. After 1 h, specimens were dehydrated with acetone and embedded by Epon812 at 60°C for 24 h. The 60 nm ultrathin sections of the cubes were stained with uranyl acetate and lead citrate at 25°C for 30 min, and then observed under TEM (7100; Hitachi, Tokyo, Japan).

A total of 3 anesthetized rats were perfused transcardially with 0.9% NaCl and 4% paraformaldehyde through the left ventricle and rapidly decapitated. After brains were removed and embedded in paraffin, they were processed into 5-µm-thick slices. Finally, H&E was applied at 25°C for 15 min to the tissue and the pathological changes of cells in hippocampal CA1 region were observed under an optical microscope (Olympus Corporation, Tokyo, Japan).

TUNEL was used to detect neuron apoptosis in hippocampal CA1 in according to the TUNEL assay kit protocol. Stained slices with 5-µm-thick were imaged under the optical microscope. The extent of brain damage was then evaluated by apoptotic index, which was the arithmetic mean of positive cells counted in 5 microscopic fields in each CA1 region of the hippocampus section.

A total of 24 h following reperfusion, the hippocampal CA1 of remaining three rats in each group was removed and put on ice. Total protein extraction and protein determination was performed with a protein extraction kit (Beyotime Institute of Biotechnology, Shanghai, China). Subsequently, using western blot analysis, the expression of MEK, p-MEK, p-ERK, ERK, CREB, p-CREB, Bcl-2 and Bax was measured. Protein (30 µl) were separated by 12% SDS-PAGE and transferred onto a nitrocellulose membrane. The membrane was blocked with 10% goat serum (Beyotime Institute of Biotechnology) at 25°C for 2 h and the anti-rabbit antibodies against MEK (cat. no. SAB4502407), p-MEK (cat. no. HPA026430), ERK (cat. no. M7927), p-ERK (cat. no. M7927), CREB (cat. no. SAB4500444), p-CREB (cat. no. AV01026), Bcl-2 (cat. no. SAB4500003), and anti-mouse antibodies against Bax (cat. no. B8554) and β-actin (cat. no. A5441) (1:1,000) were added to the membrane overnight at 4°C. After washing, the membranes were incubated with anti-rabbit HRP-conjugated secondary antibodies (cat. no. AP182P; 1:2,000) for 40 min at 37°C. Subsequently, the membranes were processed with an ECL kit (Beyotime Institute of Biotechnology) to detect immune reactivity. Finally, images were analyzed by a Versa Doc Model 3000 (Bio-Rad Laboratories, Inc., Hercules, CA, USA).

Cerebral cortexes from 1-2 days old SD rats were separated under sterile conditions. The cell suspensions were seeded (2×10⁶ cells/cm²) into culture plates and complete medium was added. Cultured cells were grown in an incubator at 37°C with 5% CO₂ and 100% humidity. After 24 h, astrocytes were replaced with new medium and the medium was replaced every three days. After 11 days astrocytes were used for future experiments. The purity of astrocytes was identified as above 95% by fluorescent antibody technical analysis with anti-glial fibrillary acidic protein (anti-GFAP).

Astrocytes were seeded at a density of 5×10⁴ into 24-well plates and cultured for 24 h in an incubator (37°C, 5% CO₂). Subsequently, cells were divided into the normoxic group, oxygen-glucose deprivation/reoxygenation group (OGD/Reoxy) and OGD/Reoxy + Cur group (OGD/Reoxy + C group) randomly. There were 6 wells in each group. In the MTT and LDH experiments, the OGD/Reoxy + C group has been divided into three doses [5, 10, 20 micromolar (µmol)/liter (l)], and setting 6 duplicate holes in each group. The solubilization of Cur was achieved with dimethyl sulfoxide (DMSO), but the final concentration of DMSO did not exceed 0.1% in the medium. In the drug-administered group, 1,000 µl of complete medium containing the corresponding drug was added, the normal group and the model group cells had 1,000 µl of complete medium containing no drug but containing the same amount of solvent (DMSO) added as the administration group. Cells were cultivated for 24 h in normal medium, then the model group and drug-administered group were modeled: Cells were subjected to hypoxia and hypoglycemia for 2 h and then reoxygenated for 24 h.

Cell climbing pieces were washed and fixed by 4% paraformaldehyde at 25°C for 20 min. Then cell climbing pieces were blocked with 5% goat serum at 37°C for 1 h, and the primary antibodies against GFAP (cat. no. G3893) (1:50; Sigma-Aldrich; Merck KGaA) were added overnight at 4°C. After washing, the cell climbing samples were incubated with appropriate fluorescein isothiocyanate fluorescence-labeled secondary antibody (1:200) (Sigma-Aldrich; Merck KGaA) for 40 min at 37°C. 4′,6-diamidino-2-phenylindole (DAPI) was used to stain the nucleus of astrocytes at 37°C for 4 min and cell climbing samples were observed under a fluorescent microscope (Nikon Corporation, Tokyo, Japan).

Cell viability was measured by MTT assay. A total of 24 h following reoxygenation, 20 µl MTT solution was added (5 g/l) to the cell culture plate and after 4 h, 150 µl DMSO was added then oscillated for 10 min. Absorbance (A) value was measured by a microplate reader at a wavelength of 490 nm. Finally, according to the following formula, the relative cell viability can be calculated: Relative cell viability = experimental group/control group ×100%.

A total of 24 h following oxygen-glucose deprivation/reoxygenation, cell cytotoxicity was measured by LDH assay. Each group had 20 µl cell culture fluid removed and LDH (U/l) was measured by colorimetry according to the protocol of the LDH assay kit. Finally, according to the following formula, the LDH leakage rate can be calculated: Relative cell cytotoxicity = experimental group (U/l)/control group (U/l) ×100%. Through MTT experiment and LDH experiment the ideal concentration of Cur was identified and the optimum concentration was used in the following experiments.

Radioimmunoprecipiration assay lysis solution (Beyotime Institute of Biotechnology) was used to lyse the cells at 0°C for 20 min, followed by centrifugation at 12,000 × g at 4°C for 10 min to collect protein supernatants. Using western blot analysis, the expression of MEK, p-MEK, ERK, p-ERK, CREB, p-CREB, Bcl-2 and Bax were measured. Samples (30 µl) were separated by 12% SDS-PAGE then transferred to a nitrocellulose membrane. The membrane was blocked with 10% goat serum at 25°C for 2 h and the anti-rabbit primary antibodies against MEK (cat. no. SAB4502407), p-MEK (cat. no. HPA026430), ERK (cat. no. M7927), p-ERK (cat. no. M7927), CREB (cat. no. SAB4500444), p-CREB (cat. no. AV01026), Bcl-2 (cat. no. SAB4500003), and anti-mouse antibodies against Bax (cat. no. B8554) and β-actin (cat. no. A5441) (1:1,000) were added to the membrane overnight at 4°C. After washing, the membranes were incubated with anti-rabbit HRP-conjugated secondary antibodies (cat. no. AP182P; 1:2,000) at 37°C for 40 min. Subsequently, the membranes were processed with an ECL kit to detect immune reactivity. Finally, images were analyzed with a Versa Doc Model 3000.

All data were expressed as the mean ± standard deviation of the mean and were analyzed by SPSS version 24.0 for Windows (IBM Corps., Armonk, NY, USA). A one-way analysis of variance was used to compare the difference of measurement data among multiple groups. The post hoc test was performed using Tukey’s post hoc method. The number of experimental repeats for each sample was 3. Pairwise comparisons in multiple groups was conducted with the Student-Newman-Keuls method. P<0.05 was considered to indicate a statistically significant difference.

A total of 24 h following ischemia/reperfusion, the neuroprotective effect of Cur was examined by evaluating neurological deficit scores and brain water content. As presented in Fig. 1, compared with the IR group, the IC group can significantly alleviate nerve damage symptoms (P<0.05), most notably in the IC group of 300 mg/kg.

A total of 24 h following ischemia/reperfusion, the effect of Cur was examined by investigating brain water content. The results demonstrated that Cur can reduce cerebral edema. As presented in Fig. 2, IC groups can significantly reduce the brain water content (P<0.05), particularly the IC group at a dose of 300 mg/kg. Through these two experiments, it was also demonstrated that neurological deficit scores and brain water content were decreased in a dose-dependent manner. In addition, only the IC group of 300 mg/kg were used in the rest of the experiments.

A total of 24 h following ischemia/reperfusion, the infract volume was measured. As presented in Fig. 3, compared with the IR group, the IC group (IC, 300 mg/kg) displayed significantly decreased pale-colored regions (P<0.01).

A total of 24 h following ischemia/reperfusion, changes in neurons were observed using TEM. As presented in Fig. 4, TEM of the SC group (magnification, ×6,000) displayed normal neuronal structure and no significant alterations. However TEM of the IR group (×6,000) revealed serious neuronal damage in the CA1 area, exhibiting nucleus electron density decreases, dissolution, cavitation, mitochondrial swelling and endoplasmic reticulum. Compared with the IR group, the IC group (300 mg/kg; magnification, ×6,000) exhibited less severe changes.

A total of 24 h following ischemia/reperfusion, the IR group displayed abnormal cell structures and morphology. Specifically, cell body and nucleus condensation, stained nucleoli, and gaps around the cells were observed. As presented in Fig. 5, compared with the IR group, the IC group (300 mg/kg) significantly reduced abnormal cells (P<0.01).

A total of 24 h following ischemia/reperfusion, the TUNEL method was utilized to detect neuronal apoptosis of hippocampal CA1. In hippocampal CA1, normal cells were stained blue, but apoptotic cell nuclei were stained brown. As presented in Fig. 6, in the SC group, almost no TUNEL-positive cells were present in the hippocampal CA1 region. Compared with the SC group, the number of apoptotic cells in the IR group was significantly increased. However the IC group (300 mg/kg) displayed a significantly lower number of apoptotic cells (P<0.01).

The expression of p-MEK, MEK, p-ERK, ERK, p-CREB, CREB, Bcl-2 and Bax were detected by western blotting 24 h following reperfusion in vivo. As presented in Fig. 7, it was demonstrated that the IC group (300 mg/kg) increased the expression of p-ERK, p-CREB, Bcl-2 and reduced the levels of Bax significantly (P<0.01).

Astrocytes were identified by immunofluorescence staining with astrocyte-specific marker GFAP antibody. Under a fluorescent microscope, a notable green fluorescence was visible in the cell cytoplasm and projections, and nuclei appeared blue by DAPI staining. Furthermore, cytons were large and irregular with few cell projections of GFAP-positive cells. These characteristics were consistent with the morphology of astrocytes and distribution characteristics of GFAP, demonstrating that astrocytes were present (Fig. 8).

A total of 24 h following oxygen-glucose deprivation/reoxygenation, the effect of Cur on the viability of astrocytes was examined. As presented in Fig. 9. compared with the OGD/Reoxy group, the OGD/Reoxy + C groups significantly increased the astrocytes’ viability (P<0.01), particularly the OGD/Reoxy + C group of 20 µmol/l.

Following oxygen-glucose deprivation/reoxygenation for 24 h, the effect of Cur on LDH leakage rates was examined. As presented in Fig. 10, compared with the OGD/Reoxy group, the OGD/Reoxy + C groups significantly reduced the LDH leakage rate (P<0.01), particularly the OGD/Reoxy + C group of 20 µmol/l. Therefore, only C-treatment group of 20 µmol/l was used in other experiments.

The expression of p-MEK, MEK, p-ERK, ERK, p-CREB, CREB, Bcl-2 and Bax were detected by western blotting for 24 h following reperfusion in vitro. As presented in Fig. 11, the IC groups (20 µmol/l) increased the expression of p-MEK, MEK, p-ERK, ERK, p-CREB, CREB, Bcl-2 and reduced the Bax levels significantly compared with the IR group (P<0.01). These results were consistent with the experimental results in vivo.