TRIzol reagent was obtained from Invitrogen (Thermo Fisher Scientific, Waltham, MA, USA). Nogo-A (no. ab62024), NgR (no. ab26291) and RhoA (no. ab54835) antibodies were obtained from Abcam (Cambridge, MA, USA). ROCK, growth-associated protein 43 (GAP-43, no. 5307), β-actin (no. 3700) antibodies and anti-rabbit immunoglobulin (Ig) G (H+L), biotinylated secondary antibodies (no. 14708) were purchased from Cell Signaling Technology, Inc. (Beverly, MA, USA). Mouse proliferating cell nuclear antigen immunohistochemical (IHC) kits were purchased from Beijing Golden Bridge Biotechnology Co., Ltd. (Beijing, China). All other reagents used, unless otherwise stated, were obtained from Sigma-Aldrich (Merck KGaA, Darmstadt, Germany).

The experimental animal protocol of the present study was approved by the Ethics Committee of Fujian University of Traditional Chinese Medicine (Fuzhou, China; no. 2014028). All experiments were performed in strict accordance with the Guide for the Care and Use of Laboratory Animals published by the US. National Institutes of Health (18). A total of 108 male Sprague Dawley rats (weighing 220–250 g and aged 2.21±0.15 months) were purchased from Shanghai SLAC Laboratory Animal Co., Ltd. (Shanghai, China) and kept in a 23±2°C temperature- and 50±5% humidity-controlled room under a 12-h light/dark cycle. Rats had ad libitum access to food and water.

Rats (n=108) were randomly divided into three groups prior to middle cerebral artery occlusion (MCAO) surgery: The sham operation control group (SC; n=36), the ischemia control group (IC; n=36) and the EA group (EA; n=36). To establish the focal cerebral ischemia model, rats were subjected to MCAO, as described previously by Longa et al (19). Following fasting for 24 h with access to water, rats were anesthetized by intraperitoneal injection with 10% chloral hydrate (300 mg/kg, Beijing Golden Bridge Biotechnology Co., Ltd.). Throughout the surgical procedures, the rectal temperature of the rats was maintained at 37°C. In brief, a midline cervical incision was performed to expose and isolate the left common carotid artery (CCA), the left external carotid artery (ECA) and the internal carotid artery (ICA). An embolus was advanced through the ICA to the MCA (20±2 mm) until a mild resistance was encountered, thereby occluding the origin of the MCA. Subsequently, the cervical incision was closed with a silk suture. Rats in the sham group were treated similarly, however no ligations or occlusions were performed. Animals in each group were randomly assigned to three subgroups to be sacrificed at days 3, 7 or 14 after surgery.

In the EA group, EA treatment was performed at the Zusanli (ST36) and Quchi (LI11) acupoints on the right paralyzed limb using an EA stimulator instrument (Model G6805; Shanghai Huayi Co., Shanghai, China) following recovery from MCAO surgery. Two 30-gauge (0.3 mm diameter) stainless acupuncture needles were inserted at a depth of 2–3 mm into the aforementioned acupuncture points. The stimulation parameters were a dense disperse wave of 1 and 20 Hz (adjusted to the muscle twitch threshold), 30 min each time, once a day. Treatment was started the day after the operation and continued until the animals were sacrificed.

Neurological deficits were evaluated to confirm successful establishment of the rat model or cerebral ischemia injury. The severity of neurological deficits in the model animals was assessed at 2 and 24 h after surgery according to the four-point scoring system developed by Longa et al (19) and Bederson et al (20): 0, no apparent deficits; 1, failure to fully extend the right forepaw; 2, circling to the right; 3, falling to the right when walking; 4, loss of ability to walk. MCAO rats with neurological deficit scores of 1–3 were used for the experiments. Neurological deficits in each group of rats were re-evaluated at each time-point (day 3, 7 and 14 after surgery) prior to sacrifice.

Following cerebral ischemia injury for 72 h, the rats were sacrificed under deep anesthesia using 10% chloral hydrate (300 mg/kg) and transcardially perfused with 0.9% NaCl. The brains of all rats were frozen at −80°C in a freezer for 20 min and were cut into five coronal slices at a thickness of 2 mm/section. The fresh slices were placed in a 2% solution of 2,3,5-triphenyl-tetrazolium chloride (TTC) in phosphate-buffered saline (PBS; Hyclone; GE Healthcare, Logan, UT, USA) at 37°C for 20 min. The normal area of the brain was stained dark red based on intact mitochondrial function, whereas the infarct area remained unstained. Images of the stained slices were captured using a high-resolution digital camera (Canon SX20; Canon, Inc., Tokyo, Japan). A computerized image analysis system (Motic Med 6.0 System; Motic, Hong Kong, China) was used for quantification of the percentage of the total infarcted brain volume.

The rats in each group (n=6) were humanely sacrificed at 3, 7 and 14 days after MCAO surgery. The tissue was subjected to routine paraffin embedding, sectioning at 5 µm thickness and antigen retrieval, followed by rinsing with 0.01 M PBS. Nogo-A, ROCK and GAP-43 levels were analyzed using an IHC assay kit (DAB kit-0017; MaixinBiotech, Co. Ltd., Fujian, China) according to the manufacturer's instructions. To block non-specific protein activity, the sections were incubated in 3% hydrogen peroxide and 10% normal goat serum (DAB kit-0017; MaixinBiotech, Co. Ltd., Fujian, China) for 10 min at 37°C. Subsequently, the sections were incubated overnight at 4°C with primary antibodies against Nogo-A (1:100 dilution), ROCK (1:100 dilution) and GAP-43 (1:200 dilution), followed by incubation with secondary biotinylated rabbit anti-mouse antibodies (1:100 dilution; DAB kit-0017; MaixinBiotech, Co. Ltd., Fujian, China). Following incubation with diaminobenzidine for 1 min, sections were washed with distilled water and dehydrated in a graded series of alcohol, cleared in xylene and mounted using xylene-based mounting medium. Images were captured on a microscope (Leica Microsystems, Wetzlar, Germany) and cells with positive staining were counted in four randomly selected microscopic fields at ×400 magnification. The positive expression rate was determined as the ratio of cells with brown staining and Image-Pro Plus 6.0 (Media Cybernetics, Inc., Rockville, MD, US) was used to count the number of positive cells.

The rats were sacrificed to obtain the left peri-infarct cerebral cortex from each group at each time-point (3, 7 and 14 days post-surgery; n=6 per group). Tissues were homogenized in RIPA Lysis Buffer (No. P0013, Beyotime Institute of Biotechnology, Haimen, China) and centrifuged at 12,000 × g for 15 min at 4°C. The supernatants were collected and frozen at −80°C prior to immunoblotting. The protein concentration of each homogenate was determined using a Bicinchoninic Acid Protein assay kit (No. CW0014, Beijing Comwin Biotech Co., Ltd., Beijing, China). A total of 50 µg protein per lane was separated by 10% SDS-PAGE. Subsequently, blots were transferred onto polyvinylidene difluoride membranes (Shanghai Bioscience Co., Shanghai, China) in a Tris-glycine transfer buffer and were blocked for 2 h with 5% non-fat dry milk at room temperature. Membranes were incubated with primary antibodies against Nogo-A (1:1,000 dilution), NgR (1:10,00 dilution), RhoA (1:1,000 dilution), ROCK (1:1,000 dilution), GAP-43 (1:5,000 dilution) and β-actin (1:5,000 dilution) overnight at 4°C, followed by incubation with the anti-rabbit IgG (H+L), biotinylated secondary antibodies (1:5,000 dilution) for 1 h at 37°C. The bands were visualized with the ECL Pico Western Blotting Substrate kit (Beijing Golden Bridge Biotechnology Co., Beijing, China), images were captured using a Bio-Rad ChemiDoc XRS+ (Bio-Rad Laboratories, Hercules, CA, USA) and densitometric quantification was performed using Quantity OneV4.62 (Bio-Rad Laboratories). The ratio of gray scale values of the target protein to the internal control was used to measure the relative amount of Raldh1 and Raldh2.

Rats were sacrificed to obtain the left peri-infarct cerebral cortex from each group each time-point (3, 7 and 14 days; n=6 per group). Total RNA was isolated with the TRIzol reagent according to the manufacturer's instructions. Primed RNA (1 µg) was reverse-transcribed with RevertAid™ First Strand cDNA Synthesis kit (5X Mix included Buffer, dNTP, HiScript® Reverse Transcriptase and RNase inhibitor; Fermentas; Thermo Fisher Scientific Inc.) according to the manufacturer's instructions. The reaction conditions for reverse-transcription were 50°C for 15 min, followed by 85°C for 2 min. cDNA was then amplified by PCR to determine the amount of Nogo-A, NgR, RhoA, ROCK and GAP-43 mRNA. β-actin was selected as a housekeeping gene. Primers were purchased from Shanghai BioSun Sci&Tech Co., Ltd. (Shanghai, China). Primer sequences were as follows: Nogo-A forward, 5′-AGGGATGTGCTGGCTGCTAG-3′ and reverse, 5′-GGTGCTTTCGGTTGCTGAGG-3′; NgR, forward, 5′-GGGCAACCTCACGCATCTCT-3′ and reverse, 5′-TCATGAGTCGGCCAAGGTCC-3′; RhoA forward, 5′-ACCAGTTCCCAGAGGTTTAT-3′ and reverse, 5′-TTTGGTCTTTGCTGAACACT-3′, ROCK forward, 5′-GATCCCCTGCAAAGTTTATT-3′ and reverse, 5′-AGCTTTTCCAAAAATGCAAA-3′; GAP-43 forward, 5′-TGCTGTGCTGTATGAGAAGAACC-3′ and reverse, 5′-GGCAACGTGGAAAGCCGTTT3′; β-actin forward, 5′-ACTGGCATTGTGATGGACTC-3′ and reverse, 5′-CAGCACTGTGTTGGCATAGA-3′. Following the addition of 2x AceTaq® Master Mix (no. P411, Vazyme biotech Co., Ltd.), which includes Buffer, dNTP and AceTaq® DNA Polymerase, a MasterCycler nexus PCR (No. 5331, Eppendorf Ltd., Hamburg, Germany) was used for the PCR reaction. The thermocycling conditions for real-time PCR were 95°C for 5 min, followed by 35 cycles of 95°C for 30 sec, 55°C for 30 sec and 72°C for 60 sec. Following 35 cycles, elongation was performed at 72°C for 7 min. PCR products were analyzed by gel electrophoresis (1.5% agarose) and a Gel Documentation system (Model Gel Doc 2000; Bio-Rad Laboratories, Inc.). Quantity One (Bio-Rad Laboratories) was used for quantification.

SPSS 18.0 for Windows (SPSS Inc., Chicago, IL, USA) was used for statistical analysis. Values are expressed as the mean ± standard deviation. Differences between the two groups were compared using the independent-samples Student's t-test. For multiple comparisons of quantitative data, one-way analysis of variance was used. If the data fulfilled the criteria of homogeneity of variance, Fisher's least significant difference method was applied; otherwise, Tamhane's method was applied. P<0.05 was considered to indicate a statistically significant difference.

The effect of EA on neurological function following cerebral ischemic injury was examined by neurological deficit scoring. As presented in Table I, there was no manifestation of neurological deficits observed in the SC group at each time-point, however rats in the IC and EA groups exhibited marked neurological deficits. The results indicated that successful cerebral ischemic models were established and rats in the SC group did not exhibit any indication of cerebral injury. There were no significant differences between the neurological deficit scores of the IC and EA groups at 2 and 24 h following cerebral ischemic injury (P>0.05). However, the scores in the EA group were significantly lower than those in the IC group at 3, 7 and 14 days (P<0.01 or P<0.05; Table I) and gradually decreased over time. These results suggest that EA effectively improved the neurological dysfunction caused by cerebral ischemia and promoted the functional recovery of nerves.

To further investigate the therapeutic efficacy of EA treatment against cerebral ischemia injury, cerebral infarct volumes were evaluated using TTC staining. Sections of the SC group were stained red, whereas the unstained infarct brain area was visible on the left side in the IC and EA groups (Fig. 1A). Furthermore, EA treatment at the Quchi and Zusanli acupoints significantly reduced cerebral infarct volumes (P<0.05; Fig. 1B).

Axonal regeneration can be identified by the elevated expression of GAP-43. GAP-43 is a major protein of axonal growth cones and has been implicated in the mechanism of axonal regeneration (21). To identify whether the expression of various axonal and myelin markers in the peri-infarct cortex were affected by EA treatment, western blotting, RT-qPCR and IHC analyses of GAP-43 in the peri-infarct cortex were performed. Compared with the SC group, expression of GAP-43 mRNA and protein increased following cerebral ischemia in the IC and EA groups (P<0.05). Furthermore, EA treatment significantly enhanced the expression of GAP-43 protein and mRNA on days 3, 7 and 14 post-ischemia, compared with the IC group (P<0.05; Fig. 2). Immunohistochemical staining indicated that only few GAP-43-positive cells were detected in the SC group, whereas the number of GAP-43-positive cells was increased in the IC and EA groups (Fig. 3). The percentage of GAP-43-positive cells in the SC group at day 3 was 1.04±0.07%, while those in the IC and EA groups were significantly higher (P<0.05), at 11.32±1.85 and 16.51±2.17%, respectively. This was also the case at days 7 and 14 (P<0.05). Furthermore, compared with the IC group, the number of GAP-43-positive cells was significantly increased in the EA group at days 3, 7 and 14 (P<0.05, Fig. 3).

To investigate the mechanism of axonal regeneration with EA treatment, the expression of the vital target genes of the Nogo-A signaling pathway, Nogo-A, NgR, RhoA and ROCK, was investigated. The expression of Nogo-A, NgR, RhoA and ROCK mRNA and protein was significantly increased in the IC group compared with the SC group at on days 3, 7 and 14 (all P<0.05). However, expression of Nogo-A, NgR, RhoA and ROCK mRNA and protein were significantly inhibited compared with the IC group, following EA treatment (all P<0.05; Fig. 4). To confirm these results, Nogo-A and ROCK were detected in the peri-infarct cortex tissue by immunohistochemistry. As expected, few Nogo-A- and ROCK-positive cells were observed in the SC group (Fig. 5). By contrast, in the IC group, Nogo-A and ROCK expression significantly increased on day 3 (13.24±1.74 and 12.27±1.98%, respectively; P<0.05), peaked on day 7 (18.13±2.06 and 18.41±2.31%, respectively; P<0.05), and then decreased slightly but was still maintained at a high level on day 14 (14.84±1.67 and 15.27±1.50%, respectively; P<0.05; Fig. 5). However, compared with the IC group, Nogo-A and ROCK expression significantly decreased in the EA group (P<0.05; Fig. 5), suggesting that EA treatment weakened the inhibition of axonal growth following cerebral ischemia.