Male Sprague-Dawley rats (n=122, weight, 250–280 g; age, 60 days) were purchased from the Experimental Animal Centre of Central South University and were employed in the present study. These rats were housed in cages at room temperature and had free access to food and water with 12-h dark/light cycles. All experiments were performed according to guidelines for the care and use of animals and were approved by the animal ethics committees of Central South University.

Focal cerebral ischaemia was induced by middle cerebral artery occlusion (MCAO) according to Hirayama and Koizumi (19). In brief, rats were anesthetized by intraperitoneal injection with 10% chloral hydrate (300 mg/kg body weight). A mid-line incision in the neck of the rat was made to expose the common and external carotid arteries and ligated them. Specifically, the internal carotid artery was closed by an artery clamp. Subsequently, a 6-0 silicon rubber-coated nylon monofilament (RWD Life Science Co., Ltd.) was inserted into the right internal carotid artery. Insertion was stopped after occlusion for 90 min. The animals were reanaesthetized and reperfused by withdrawal of the monofilament. After surgery, the rats were kept for ~2 h in a warm box heated by lamps to maintain body temperature. Finally, the common carotid artery was permanently ligated, and the neck wound was closed.

All rats with neurological deficits were randomly divided into the following four groups 24 h post-MCAO: i) MCAO, where rats were housed in standard cages after MCAO (n=6); ii) willed movement (WM), where rats were housed in a ladder-containing cage 24 h after MCAO surgery and had to climb the ladder or walls of the apparatus to reach food and water (n=6); iii) environmental modification (EM), where rats were housed in the same environment as the WM group but could get water and food without climbing the ladder (n=6); iv) common rehabilitation (CR), where rats were housed in the rolling-grid training cage 24 h after MCAO surgery and were trained twice a day (9:00 am and 4:00 pm) for 15 min each time (n=6).

Neurological evaluation was performed on days 0, 3, 7, 15 and 30 after excision using rats with a modified neurological deficit score (20). The rats were trained for 15 days before sacrifice, and those that died during training were excluded. Specifically, A total of 50 rats did not recover from anaesthesia or died after the MCAO operation. All rats were randomly allocated to groups for training, as aforementioned. Neurological scores were defined as follows: 0, no deficit; 1, flexion of the contralateral forelimb upon lifting the entire animal by the tail; 2, decrease in thrust towards the contralateral plane; and 3, circling to the contralateral side.

The cerebral infarction area was examined in six animals per group (n=24 for each replicate, the experiments were performed three times) for 15 days post-training. Rats were anaesthetized by intraperitoneal injection with 1% sodium pentobarbital (40 mg/kg) and decapitated. The whole brains were frozen at −20°C for 20 min. The tissues were cut into 2-mm-thick coronal sections and then stained with 2% TTC solution at 37°C for 30 min. The stained slides were immersed with 4% paraformaldehyde at 4°C overnight. The infarct area on each TTC-stained section was measured using ImageTool 2.0 software (University of Texas Health Science Center) and calculated as the infarct area × thickness (2 mm).

Human umbilical vein endothelial cells (HUVECs) were purchased from American Type Culture Collection and grown in vascular cell basal medium (Gibco; Thermo Fisher Scientific, Inc.) supplemented with 5 ng/ml VEGF (Sigma-Aldrich; Merck KGaA) at 37°C, 5% CO₂. For primary cerebral endothelial cell isolation (3), brains from rats of each group 30 days post-training were removed and then stored in Dulbecco's Modified Eagle's Medium/F12 (Gibco; Thermo Fisher Scientific, Inc.) until further use. The tissues were washed with PBS several times and minced and incubated with Liberase Blendzyme 2 (Roche Diagnostics GmbH) at a concentration of 0.625 Wu/ml at 37°C for 2 h. Subsequently, the cell suspension was washed with PBS and centrifuged at 228 × g for 5 min at 25°C. Subsequently, the cells were plated in dishes coated with collagen type I (BD Biosciences) and grown in Endothelial Cell Growth Basal Medium (EBM) complete medium (Lonza Group Ltd). Cells were cultured at 37°C, 5% CO₂. After 24 h, non-adherent cells were removed and seeded into new collagen type I-coated dish with EBM media. The cultured cells were passaged at a ratio of 1:4 every 14 days. To mimic focal ischemia in vitro, HUVEC were treated with 500 µM CoCl₂ (Shanghai Aladdin Biochemical Technology Co., Ltd.) treatment for 24 h.

The isolated primary cerebral endothelial cells from all four groups were examined via MTT analysis. The cells were counted and plated into 96-well plates for 24 h at 37°C. Then 0.1 mg/ml MTT was added to cells at 37°C for 3 h. The cells were lysed in DMSO at room temperature for 30 min. Finally, the absorbance was measured at 490 nm by a microplate reader (Bio-Tek China).

To examine the ability of endothelial tube formation in vitro, 24-well plates were coated with 10 µl of Matrigel (BD Biosciences) as previously described (21) and 2×10⁴ HUVECs or primary cerebral endothelial cell were seeded. The tube length was measured using MetaMorph Microscopy Automation and Image Analysis Software (version 1.0; Molecular Devices, LLC). The images were taken and processed by an inverted microscope. In total, >5 randomly selected fields of view were analyzed.

Total RNA was extracted using TRIzol (Thermo Fisher Scientific, Inc.). Briefly, the cells were lysed with TRIzol buffer, and then 200 µl chloroform (100%) was added to the mixture, which was vortexed for 15 sec at 25°C. The resulting solution was centrifuged at 17,366 × g for 15 min at 4°C. The supernatant was harvested and mixed with an equivalent volume of isopropanol. The resultant mixture was centrifuged at 17,366 × g for 10 min at 4°C. The supernatant was removed and 75% ethanol was added to wash the pellet and centrifuged at 17,366 × g for 10 min at 4°C. Ethanol was discarded and the pellet was left to dry, then used 20–30 µl DEPC-treated H₂O (Invitrogen; Thermo Fisher Scientific, Inc.) to elute the RNA pellet.

Total RNA (1 µg) was used to perform reverse transcription according to the manufacturer's protocol (PrimeScript Kit; Takara Bio, Inc.). For quantitative PCR, SYBR Green (QuantiFast SYBR Green PCR Kit; Qiagen GmbH) was used as the probe dye and the signal was detected according to the manufacturer's protocol. GAPDH was used as the internal control. The following primers were used: HIF-1α-Forward, CGCAAGTCCTCAAAGCACAG and HIF-1α-Reverse, TCTGTTTGGTGAGGCTGTCC; VEGF-Forward, TTTCTGCTGTCTTGGGTGCA and VEGF-Reverse, CCAGGGTCTCGATTGGATGG; Ang-1-Forward, GCCTGATCTTACACGGTGCT and Ang-1-Reverse, GCCTGATCTTACACGGTGCT; MMP2-Forward, GATCTACTCAGCCAGCACCC and MMP2-Reverse, ACGACGGCATCCAGGTTATC; MMP9-Forward, TACTCGACCTGTACCAGCGA and MMP9-Reverse, ATGCCATTCACGTCGTCCTT; GAPDH-Forward, GAGTCAACGGATTTGGTCGT and GAPDH-Reverse, TTGATTTTGGAGGGATCTCG. The thermocycling conditions were the following: Initial PCR activation step at 95°C for 15 min, followed by 40 cycles of denaturation at 94°C for 15 sec, annealing at 53°C for 30 sec, and elongation at 72°C for 30 sec. Cq values were used for quantification using the 2^−∆∆Cq method (22).

In total, 1×10⁶ cells of each group were harvested and washed with 1×PBS once. Subsequently, 2×SDS loading buffer (Dalian Meilun Biotech Co., Ltd.) was used to lyse cells. The lysates were boiled at 95°C for 10 min and centrifuged at 16,172 × g for 1 min at 4°C. Approximately 50 µg of total protein was loaded onto a 10% SDS-polyacrylamide gel and resolved at 120 V for 0.5–1 h at 25°C. Subsequently, proteins were transferred onto polyvinylidene fluoride membranes at 300 mA for 2–3 h at 4°C. The membrane was blocked with 5% non-fat milk in 1×TBS with 0.1% Tween (TBST) for 1 h at 25°C, then incubated with the corresponding primary antibodies at 4°C overnight. The following day, the membrane was washed with 1×TBST 3 times and incubated with secondary antibodies at room temperature for 1 h. Finally, the membrane was incubated with enhanced chemiluminescence reagent (7Sea Biotech) and exposed using the Bio-Rad ChemiDoc Touch Imaging System. The following antibodies were used in the present study: Anti-HIF-1α (1:1,000; cat. no. ab51608; Abcam), anti-VEGF (1:1,000; cat. no. ab52917; Abcam), anti-Ang-1 (1:1,000; cat. no. ab95230; Abcam), anti-MMP-2 (1:1,000; cat. no. 87809; Cell Signaling Technology), anti-MMP-9 (1:1,000; cat. no. 3852; Cell Signaling Technology Inc.). The secondary antibody horseradish peroxidase-conjugated goat anti-rabbit immunoglobulin G (cat. no. A0208; 1:5,000; Beyotime Institute of Biotechnology) was incubated with the membranes at room temperature for 1 h.

SiRNAs (scrambled and siHIF-1α) were synthesized by GenePharma and diluted to 20 µM as stock. To transfect cells, siRNAs were diluted to 50 nM and were mixed with RNAiMAX reagent (Thermo Fisher Scientific, Inc.) for 20 min at 25°C, and the mixture was added to cultured cells. The sequence of siHIF-1α was GCTCCCAATGTCGGAGTTT. Cells were harvested 48 h after transfection.

GraphPad Prism (version 6.0; GraphPad Software, Inc.) was used for statistical analysis. Each experiment was performed three times, and all values are presented as the mean ± standard deviation. Comparison of two groups was performed using the two-tailed unpaired Student's t-test. Comparison of multiple groups was performed using one-way analysis of variance followed by Tukey's post-hoc test. P<0.05 was considered to indicate a statistically significant difference.

The neurological defect score in the MCAO, CR, WM and EM groups was measured first (n=6 per group). The results showed that rats in the CR, WM and EM groups exhibited a significantly lower score compared with the MCAO group. However, the groups CR, WM, and EM did not display significant differences from each other (Fig. 1).

To investigate whether willed movement therapy decreased the focal ischaemia damage, TTC staining of cerebral tissue was performed to assess the infarctions. Representative tissues correspond to different coronary sections per group as presented in Fig. 2A. The unstained areas in Fig. 2A are infarcted tissues, and the red areas are normal tissues (Fig. 2A). In the four groups, different volumes of infarcted tissue areas were observed. Compared with the MCAO group, the CR, WM and EM groups displayed statistically significant decreases in the volume of cerebral infarcted tissue (Fig. 2B). The volume of cerebral infarction tissue in the WM group was not significantly lower than the other two training groups (Fig. 2B). In addition, there was no significant difference between CR and EM group.

To investigate the molecular mechanism by which the three forms of exercise improved neurological rehabilitation, primary cerebral endothelial cells were isolated from rats subjected to CR, WM and EM and cultured. Subsequently, the MTT method was performed to examine cell viability in the MCAO, CR, WM and EM groups. The endothelial cells of the MCAO group showed the lowest viability (Fig. 3A).

To investigate whether these therapies improved tube formation of cerebral endothelial cells, cells of the CR, WM and EM groups possessed a stronger ability for tube formation (evaluated by total tube length) in vitro compared with the MCAO group. Cells of the CR, WM and EM groups showed no significant differences from each other (Fig. 3B). The tube length of vessels formed in the MCAO group was half of that in the other three groups, while no length disparity appeared between the WM, EM and CR groups (Fig. 3C). Taken together, these results demonstrated that CR, EM, and WM increased cell viability and enhanced angiogenesis of cerebral endothelial cells compared with the MCAO group.

As angiogenesis has been associated with some classical genes (12), and angiogenesis-associated gene expression levels in the MCAO, CR, WM and EM groups were investigated next. RT-qPCR was employed to detect the mRNA expression levels of VEGF, Ang-1, MMP-2 and MMP-9 in cerebral endothelial cells of all four groups. The CR, EM and WM groups showed elevated mRNA expression levels of all four genes compared with MCAO (Fig. 4A). Additionally, the protein expression levels of these four angiogenesis genes were also investigated. Notably, the protein expression levels of these genes only increased in the WM group compared with the MCAO group, whereas the CR and EM groups showed no difference in their expression levels (Fig. 4B), suggesting that the protein expression level was regulated at the post-transcriptional level.

HIF-1α has an important role in angiogenesis and endothelial cell survival (18). The expression levels of HIF-1α in endothelial cells of the MCAO, CR, WM and EM groups were investigated. RT-qPCR and western blot analyses were performed to determine the HIF-1α mRNA and protein expression levels, respectively. HIF-1α mRNA and protein expression levels were higher in endothelial cells of the CR, WM and EM groups compared with the MCAO (Fig. 4C and D). In conclusion, the present study suggested that CR, WM and EM may promote cell proliferation and angiogenesis and upregulation of angiogenesis-associated genes and HIF-1α. However, the direct effects of CR, WM and EM on angiogenesis and cell proliferation require further investigation.

To mimic focal ischaemia in vitro, we generated a HUVEC-based hypoxia model using CoCl₂ treatment in culture. First, whether the expression of VEGF, Ang-1, MMP-2 and MMP-9 was affected in hypoxic cells was investigated by RT-qPCR and western blot analysis. It was found that these genes were induced by hypoxia (Fig. 5A and B). HIF-1α was gradually upregulated when cells were treated by hypoxia and peaked at 12 h after hypoxia treatment (Fig. 5C and D). Taken together, the present results suggested that hypoxia upregulated the expression levels of HIF-1α and its downstream genes.

To confirm the role of HIF-1α in angiogenesis of cerebral endothelial cells, HIF-1α-knockdown HUVECs using siRNA against HIF-1α was constructed. RT-qPCR analysis showed the mRNA expression levels of HIF-1α was notably reduced in HIF-1α-knockdown HUVECs compared with the negative control (Fig. 6A). Subsequently, the effect of HIF-1α on hypoxia-induced angiogenesis was investigated. The results demonstrated that angiogenesis was potentiated in cells under hypoxic conditions. However, HIF-1α depletion reduced the in vitro tube formation in cells under hypoxia and significantly shortened the tube length (Fig. 6B). Together, these data demonstrated that HIF-1α is essential for hypoxia-enhanced angiogenesis of cerebral endothelial cells. Finally, whether HIF-1α depletion impacted on angiogenesis-associated gene expression levels under hypoxic conditions was investigated. The results showed that hypoxia induced the expression of VEGF, Ang-1, MMP-2 and MMP-9 at the mRNA and protein levels, whereas HIF-1α knockdown reversed the increased expression of these genes (Fig. 6C and D). Collectively, these findings demonstrated that HIF-1α serves a key role in in vitro tube formation of endothelial cells under hypoxic conditions, but this needs to be verified with future studies.