Adult male Sprague-Dawley rats (8–10 weeks old; 180–220 g) were obtained from the Laboratory Animal Center of Xuzhou Medical University, (license no. SCXK2007-2005; Xuzhou, China), where an SPF level laboratory was founded, as authorized by the Jiangsu province government. All animals were maintained at a constant temperature of 25±2°C under a 12/12 h light/dark cycle. Rats were allowed free access to food and water ad libitum. Animal experiments were conducted in accordance to the principles provided by National Institute of Health (NIH) Guideline for the Care and use of Laboratory Animals. The approval to proceed with this experiment was issued by the Animal Ethics Committee of Xuzhou Medical University which also conforms to the Guidelines for Ethical Conduct in the Care and Use of Animals. All efforts were made to prevent unavoidable pain and distress when the approved endpoint is reached, no animal death occurred during this study. Euthanasia should result in rapid loss of consciousness, followed by respiratory and cardiac arrest and ultimate loss of all brain function. Death was confirmed after euthanasia and prior to disposal of the animal. Each rat was weighed weekly, while the measurement of blood glucose was performed using a glucometer via the tail vein (Nanjing Jianqiao Medical Device Co. Ltd.).

GbE was used in the current study and is an extract of dried Ginkgo biloba leaves. GbE was obtained from Shaanxi Huike Botanical Development Co., Ltd. (Purity, >98%; cat. no. HK20121201). For administration in vivo, GbE was dissolved in 1% CMC-Na at concentrations of 10, 20 and 40 mg/ml. Streptozotocin (STZ; cat. no. 0130) was purchased from Sigma-Aldrich; Merck KGaA. Nimodipine (cat. no. 120554), which was used as a positive control, was purchased from Yabao Pharmaceutical Group Co., Ltd. and suspended in 1% CMC-Na solution.

The rats that were fasted overnight were subjected to a single intraperitoneal injection of 60 mg/kg STZ that was freshly dissolved in 0.1 mol/l cold citrate buffer at pH 4.3. Age-matched normal rats were injected with an equal volume of citrate buffer alone. Blood glucose was measured a period of one week after STZ injection. Rats with fasting blood glucose of ≥13.88 mmol/l were considered diabetic and included in the present study (30).

Normal and diabetic rats were placed in the supine position on a heated pad and had a body temperature of 36.5–37.5°C, which was monitored using a rectal thermometer. After being anesthetized with an intraperitoneal injection of 10% chloral hydrate (300 mg/kg) (33,34), rats were subjected to 30 min of middle cerebral artery (MCA) occlusion followed by 24 h of reperfusion. No signs of pain and peritonitis were observed following administration of 10% chloral hydrate. Sham operation and transient middle cerebral artery occlusion were performed as previously described (35). The right common carotid artery (CCA), external carotid artery, and internal carotid artery (ICA) were isolated. A nylon filament, which was purchased from Beijing Sunbio Biotech Co., Ltd., was subsequently introduced into the CCA lumen and gently advanced to the ICA until a slight resistance was felt.

A total of eighty rats were randomly divided into eight groups: The Con + sham group were nondiabetic and sham-operated rats (Con + sham group; n=10); Con + ischemia-reperfusion (I/R) rats were nondiabetic and received I/R injury (Con + I/R group; n=10); diabetic and sham-operated rats were in the STZ + sham group (STZ + sham group, n=10); diabetic rats treated with ischemia-reperfusion injury were the STZ + I/R group (STZ + I/R group; n=10). These aforementioned groups were administered the same volume of 1% CMC solution for a period of three weeks. Before ischemia-reperfusion injury, the diabetic rats were administrated intragastrically with 50, 100 and 200 mg/kg GbE for the GL group (low dose; n=10), the GM group (moderate dose; n=10) and the GH group (high dose; n=10), respectively. Diabetic rats that received nimodipine prior to ischemia-reperfusion injury were the positive control group (Nimo group; n=10; 5 mg/kg/day; intragastrically). All groups were administered with the corresponding agents for three weeks before induction of ischemia. All rats were anesthetized with pentobarbital (45 mg/kg; intraperitoneal injection) and sacrificed using cervical dislocation. Rat brains and blood samples were collected. The experimental protocol is presented in Fig. 1.

Neurological function was evaluated 2 and 24 h after reperfusion by an investigator blinded to the study groups: 0, no deficit; 1, failure to extend right forelimb while the tail was pulled; 2, spontaneous circling or walking to the contralateral side; 3, stumble only when stimulated with a depressed level of consciousness; 4, unresponsive to stimulation.

2,3,5-Triphenyltetrazolium chloride (TTC) staining was performed according to previous descriptions (36) for the evaluation of the infarct volume in experimental ischemic stroke. TTC stained the normal cerebral areas deep red without any effect on the infarct tissue, which enables identification of the healthy regions from the infarcted areas. A total of 5 rats from each group were used for infarct volume measurement (n=5 per group). Rats were euthanized and brains were removed immediately. Brain samples were placed in a brain matrix and sliced into 2 mm sections. The slices were incubated in a 2% solution of TTC (cat. no. 129K1867V; Sigma-Aldrich; Merck KGaA) at 37°C for 30 min, then fixed in a 4% buffered paraformaldehyde solution and scanned using a scanner (EPSON Perfection V33). The infarct area and the hemisphere area of each section were traced and quantified using ImageJ software (National Institutes of Health) and expressed as the percent of infarct area in the whole brain.

The amount of serum S100B protein was detected using a commercially available ELISA kit (cat. no. 1302271; Shanghai Bio-Tech Co., Ltd.), according to the manufacturer's protocol, and expressed as ng/ml.

The right striatum (50 mg) was homogenized with 450 µl 0.9% NaCl and centrifuged at 4°C at 12,000 × g for 15 min. A total of 20 µl 6 mol/l NaOH was then added to 100 µl supernatant in an Eppendorf tube and the sample was incubated in a water bath at 60°C for 30 min. The hydrolyzed sample was acidified with 50 µl 35% (v/v) perchloric acid. The resulting suspension was then mixed on a vortex for 30 sec and centrifuged at 12,000 × g for 10 min. A total of 200 µl top clear supernatant was transferred to a 1.5 ml Eppendorf tube. The resultant supernatant was mixed with 20 µl 2,4-dinitrophenylhydrazine solution (5 mmol/l in 2 mol/l HCl; pH=0.09) and incubated at room temperature for 30 min. After derivatization, samples were filtered through a 0.22 µm filter. Aliquots of 50 µl were injected into a HPLC system (37), in which an Agilent Zorbax SB-C18 column (250×4.6 mm; Agilent Technologies, Inc.) was used. The mobile phase was acetonitrile-distilled water (38:62, v/v) containing 0.2% (v/v) acetic acid at a flow rate of 1.0 ml/min, and the wavelength of the UV detector was set at 310 nm. The level of striatum malondialdehyde (MDA) was expressed as µmol/g protein, and protein concentration was determined using a bicinchoninic acid (BCA) assay.

The level of GSH was measured as previously described by Liu et al (38). The compound 3-carboxy-4-nitrophenyl disulfide can react with sulfhydryl compounds (including GSH) and form a yellow compound with a strong absorption at 420 nm. The measurement of GSH was performed using a commercial kit (cat. no. A006-1; Nanjing Jiancheng Bioengineering Institute), and the level of striatum GSH was expressed as mg/g protein.

SOD is an important antioxidative enzyme, and, in the current study, its activity was determined according to the method of Sun et al (39). This method uses the inhibition of nitroblue tetrazolium reduction by the xanthine and xanthine oxidase system as a superoxide generator. SOD activity was subsequently measured at 550 nm by the degree of inhibition using a commercial kit (cat. no. A001-1; Nanjing Jiancheng Bioengineering Institute). A total of one unit of enzyme was defined as the amount of enzyme required at an inhibitory rate of 50%. The activity of SOD was expressed as units/mg protein.

A total of 5 rats from each group were used for western blot analysis (n=5 per group). After weighing the rats, the right hippocampus were dissected and homogenized using a sonicator with six-fold volumes (w/v) of 50 mmol/l Tris buffer (pH=7.4) containing 150 mmol/l NaCl, 1 mmol/l EDTA, 1 mmol/l PMSF, 1 mmol/l Na₃VO₄, 2 mmol/l DTT and 50 mmol/l NaF, in an ice-cold bath. Samples were left at 4°C for at least 30 min, the homogenates were then centrifuged at 4°C at 10,000 × g for 15 min and the supernatant was collected and denatured in SDS. The protein concentration in the supernatant was determined using a BCA protein assay kit (Thermo Fisher Scientific, Inc.).

The same amount of protein (80 µg) was electrophoresed on 8% SDS-PAGE and transferred to a PVDF membrane. The membranes were incubated overnight at 4°C with primary antibodies, including Akt Rabbit monoclonal antibody (1:1,000; cat. no. 4691; Cell Signaling Technology, Inc.), p-Akt (Ser473) Rabbit monoclonal antibody (1:1,000; cat. no. 4060; Cell Signaling Technology, Inc.), mTOR Rabbit monoclonal antibody (1:1,000; cat. no. 2983; Cell Signaling Technology, Inc.), p-mTOR (Ser2448) Rabbit monoclonal antibody (1:1,000; cat. no. 5536; Cell Signaling Technology, Inc.), GLT1 Rabbit polyclonal antibody (1:1,000; cat. no. ab106289; Abcam), and β-actin Rabbit polyclonal antibody (1:1,000; cat. no. AP0060; Bioworld Technology, Inc.), respectively. The membranes were washed and incubated with alkaline phosphatase-conjugated IgG (1:10,000; cat. no. E030220-02; EarthOx Life Sciences) at room temperature for 2 h before being exposed to BCIP/NBT alkaline phosphatase color developing reagent (cat. no. C3206; Beyotime Institute of Biotechnology) for 15 min. Western blot density was measured using Image J software (Rawak Software, Inc.) and normalized using β-actin as an internal control.

All data in the different experimental groups were expressed as the mean ± SD. Data were analyzed using GraphPad Prism (version 5.0; GraphPad Software, Inc.). A comparison between groups was conducted using one-way ANOVA, followed Tukey's multiple comparisons tests. P<0.05 and P<0.01 were considered to indicate a statistically significant difference.

The body weights of rats were measured after STZ injection (day 0) and on days 7, 14, 21 and 28. As presented in Fig. 2, the original body weights of eight groups were ~200 g. After STZ injection, the body weights of diabetic rats (STZ + sham group and STZ + I/R group) remained unchanged, whereas a marked increase was observed in nondiabetic rats (Con + sham group and Con + I/R group) during the 4 consecutive weeks (P<0.01). GbE (50, 100 and 200 mg/kg/day) was administered intragastrically once per day from day 7 to day 28, and there were no significant differences among the groups at each time point.

Fasting blood glucose levels were measured on days 7 (after a week of STZ injection) and 28 (after 3 weeks of consecutive administration). On days 7 and 28, the fasting blood glucose levels in diabetic rats was significantly increased compared with nondiabetic rats (P<0.01; Table I). No significant differences in fasting blood glucose levels were observed in the three GbE groups and the Nimo group.

Neurological deficits were evaluated 2 and 24 h after reperfusion. Compared with the 2 h time point after reperfusion, I/R injury in nondiabetic rats (Con + I/R group) had severe to mild neurological deficits (P<0.01), whereas I/R injury in diabetic rats (STZ + I/R group) resulted in severe to very severe neurological deficits 24 h after reperfusion (P<0.01), and sham-operated animals did not exhibit any deficits (Table II). The Nimo and three GbE dose groups had significantly improved neurological scores at 24 h of reperfusion (P<0.01; Table II).

The effects of GbE on rat infarct volume were investigated using TTC staining. No lesion was observed in sham-operated groups. Con + I/R rats that were subjected to 30 min MCAO followed by 24 h reperfusion presented smaller infarct volumes of 9.80±1.48%. The STZ + I/R group exhibited markedly increased infarct volume percentages (41.34±7.88%) compared with the Con + I/R group (P<0.01; Fig. 3B). Intermediate and high doses of GbE, and nimodipine markedly reduced the infarct volume (P<0.01; Fig. 3B). However, the low dose of GbE had little effect on it.

S100 calcium-binding protein B (S100B), which is a biomarker of traumatic brain injury, is primarily expressed in the central nervous system by astrocytes (40). Ischemia is associated with the increased expression of S100B, which may be released from damaged astrocytes (41). The level of serum S100B is an indicator of brain injury following stroke (42). Therefore, the concentration of S100B was examined using an ELISA to investigate the neuroprotective effect of GbE. As presented in Fig. 4, in diabetic rats, I/R injury significantly increased the level of serum S100B (P<0.01). STZ-induced diabetic increased S100B level (P<0.01) in STZ + I/R group compared with the nondiabetic rats with I/R injury (Con + I/R group). However, pretreatment with nimodipine and GbE significantly decreased S100B compared with the STZ + I/R group (P<0.01).

Increasing doses of GbE resulted in a stepped decrease of infarct volume and S100B level. The moderate dose of GbE (100 mg/kg, GM) was subsequently used for the following experiments.

One potential mechanism for diabetes and its complications is oxidative stress (43). To investigate whether the neuroprotective effect of GbE was associated with the decrease in oxidative stress levels, three associated molecules were examined, including MDA, GSH and SOD. As presented in Fig. 5, in nondiabetic rats, compared with the Con + sham group, I/R injury significantly decreased GSH level (P<0.05). In diabetic rats, I/R injury significantly caused oxidative stress damage, and this was indicated by increased MDA level (P<0.01), and decreased GSH (P<0.05) and decreased SOD activity (P<0.05). Compared with nondiabetic rats with I/R injury (Con + I/R group), STZ-induced diabetic rats exhibited increased MDA level (P<0.01). However, GbE pretreatment significantly suppressed MDA level (P<0.01), and inhibited the decrease of GSH (P<0.05) and SOD (P<0.01).

It was previously revealed that GbE pretreatment improved neurological deficits, reduced the infarct volume and relieved oxidative stress following cerebral I/R injury in diabetic rats. To elucidate the mechanism by which GbE ameliorated neuronal damage, western blot analysis was used to identify associated protein expression. The results indicated that in diabetic rats, I/R injury decreased the expression of p-Akt (P<0.01; Fig. 6A). Compared with nondiabetic rats with I/R injury (Con + I/R group), STZ-induced diabetic decreased p-Akt/Akt ratio (P<0.01) in the STZ + I/R group. However, in the nimodipine and GbE-pretreated groups, the ratio of p-Akt/Akt was significantly increased compared with the STZ + I/R group (P<0.01; Fig. 6A). Variation in the p-mTOR/mTOR ratio was consistent with the change in the p-Akt/Akt ratio (Fig. 6B).

A recent study has revealed that glutamate uptake exhibits a protective function in hippocampal astrocytes (44). Therefore, in the current study, the hippocampus was collected to elucidate the possible mechanism underlying the neuroprotective effect of GbE against injury. The results indicated that whether in nondiabetic or diabetic rats, I/R injury decreased the expression of GLT1 (P<0.01 or P<0.05; Fig. 6C). Compared with nondiabetic rats with I/R injury (Con + I/R group), STZ-induced diabetic rats exhibited decreased GLT1 expression (P<0.01) in the STZ + I/R group. However, in the nimodipine and GbE-pretreated groups, the expression of GLT1 was significantly increased compared with the STZ + I/R group (P<0.01; Fig. 6C).