Male Sprague-Dawley rats, n = 48, weighing 180~220 g (Suzhou Industrial Park Airmate Technology Company, Suzhou, China), were kept in clean cages and housed in a room at a temperature of 21–25 °C and humidity of 60–65% in a natural light/dark cycle. All rats underwent a 12-h fast prior to surgery, with free access to water. Randomly and equitably, the rats were separated into four groups: Normal, Sham operation, Model, and EA. In accordance with the recommendations and protocols for the use of laboratory animals in China, all procedures were conducted in accordance with animal welfare standards. All surgical procedures were conducted under anesthesia with sodium pentobarbital, and every attempt was made to reduce animal suffering.

As reported previously, the right middle cerebral artery embolism model was generated utilizing the thread-embolization method [6, 7]. In brief, rats were anesthetized via intraperitoneal injection of sodium pentobarbital (30 mg/kg, AM00469, Beijing Chemical Reagent Company, Beijing, China) and were placed supine on the operating table with their necks shaved and disinfected. A median incision was made to expose and isolate the right common carotid artery (CCA), the external carotid artery (ECA), and the extracranial segment of the internal carotid artery (ICA). ECA and the proximal end of CCA were ligated. A tiny incision was made close to the bifurcation of the right CCA in order to insert a 0.26 mm nylon thread to a depth of 18–19 mm. The wound was sutured layer by layer with surgical sutures. At the same time, the Sham operation group, the procedure was the same except that CCA, ECA and ICA on the right side were only separated bluntly without filament insertion and artery occlusion, there was no intervention in the normal group.

One week after the successful modeling, the EA group treatment was based on the experimental animal acupoint map formulated by the Experimental Acupuncture Research Association of the Chinese Acupuncture and Moxibustion Society. A 30-ga 1-inch needle was used to puncture the “Baihui” point obliquely, 1 mm into the scalp, and 2 mm perpendicularly into the left “Zusanli” point. The needles were connected to a Huatuo electronic acupuncture instrument (model SDZ-II, Suzhou Medical Supplies Factory Co., Ltd., Suzhou, China), and low-frequency (2 Hz) current stimulation was performed using 2 V intensity, 1 ms wave width, and 30 min duration. Stimulation was performed once a day for 21 consecutive days. No intervention procedures were conducted on the other 3 groups.

At the end of EA treatment (on the 28th day of the experiment), intraperitoneal injections of 1% sodium pentobarbital (30 mg/kg) were used to anesthetize the rats. Immunohistochemical tests were performed using six rats per group. After thoracotomy, saline was perfused into the left ventricle-ascending aorta, followed by 4% paraformaldehyde. Following perfusion, the brain was prepared by fixing it with 4% formaldehyde and embedding it in paraffin. Another 6 rats were used for Western blot experiments. The frontal cortex around the right ischemic focus was quickly separated in a low-temperature environment, placed in a cryotube, and kept for later use in a –80 °C freezer.

To examine the morphology of the neurons in the rat frontal cortex, the Nissl-staining method was used. Tissue was obtained from the area surrounding the frontal cortex’s ischemia region according to a stereotaxic map of the rat brain [7] and sliced at a thickness of 5 $\mu$m. The slices were routinely deparaffinized, stained in 0.1% Tar Violet Stain Solution for 5 minutes in a 37 °C incubator, rinsed in distilled water, decolored for 3 minutes in 70% ethanol, bathed in 95% ethanol once, and bathed in 100% ethanol three times for 3 minutes each. Xylene Ⅰ and Ⅱ were added, respectively, and kept for 5 min. Slices were then sealed and observed under the microscope using the Express C image analysis system (BX51, OLYMPUS, Tokyo, Japan).

The right frontal cortex’s Gas7- and NGF-positive neurons were identified using immunohistochemical techniques. Slices were prepared from the frontal cortex, sectioned at 5 $\mu$m, and were subjected to the following sequence of steps: (a) Slices were immersed in 3% H${}_2$O${}_2$ at room temperature for 10 min. (b) They were then incubated in 5% BSA (AR0004, Wuhan Boster Biotechnology Co., Ltd., Wuhan, China) at room temperature for 60 min without washing. (c) Primary antibodies anti-Gas7 (1:100, sc-34364, Santa Cruz Biotechnology, Inc., Dallas, Texas, USA) and anti-NGF (1:150, BM4100, Wuhan Boster Biotechnology Co., Ltd.) were added and incubated in the refrigerator at 4 °C (18 h) overnight. (d) Rabbit anti-goat IgG (1:100, SA2003, Wuhan Boster Biotechnology Co., Ltd.) or biotinylated goat anti-rabbit IgG (1:100, SA2002, Wuhan Boster Biotechnology Co., Ltd.) were added for 60 min of incubation at room temperature. (e) Streptavidin-biotin-peroxidase complex (1:100, SA1022, Wuhan Boster Biotechndogy Co., Ltd.) was added for a 60-min incubation, and (f) Diaminobenzidine (PA110, Tiangen Biochemical Technology, Beijing, China) was used for staining. After the above steps, slices were thoroughly washed in 0.02 mol/L phosphate buffer (pH 7.2–7.6) 3 times, for 5 min each. After conventional dehydration and transparentizing, neutral gums were used to seal the slices for light-microscope observation. In each group, the same coronal section was selected to count the number of Gas7 and NGF positive cells in the same area of the frontal cortex under a 200-X field of view.

Using the Western blot technique, the expression of Gas7 and NGF proteins in the right frontal cortex was determined. The frontal cortex proteins of each group were extracted and quantified, followed by: (a) Gel electrophoresis; (b) Film transfer; (c) Shaking in 5% skimmed milk at room temperature for 1–2 h; (d) Incubation with TBST-diluted Gas7 primary antibody (1:1000, sc-34364, Santa Cruz Biotechnology, Inc.) and NGF (1:500, BM4100, Wuhan Boster Biotechndogy Co., Ltd.) or glyceraldehyde-3-phosphate dehydrogenase (GAPDH) secondary antibody (1:1000, AF7021, Affinity Biosciences, Cincinnati, OH, USA); and (e) Shaking at room temperature for 2 h. At the end of each of the above steps, samples were thoroughly washed 3 times with 0.01 mol/L PBS for 10 min each. Enhanced chemiluminescence (ECL) was used for imaging. After scanning the film, the obtained images were measured in the gel imaging analysis system to measure the average absorbance values of Gas7, NGF, and GAPDH protein bands in the right frontal cortex of each group of rats to obtain the relative expression.

All data are expressed using the mean and standard deviation format. The statistical analysis was conducted using SPSS 16.0. (IBM Corp., Chicago, IL, USA) ANOVA with one-way analysis of variance was utilized for comparison, and the least significance difference (LSD) test was employed for further comparison between two groups. p 0.05 was used to represent a statistically significant difference.

In the Normal and Sham groups, Nissl-positive neurons were densely distributed in the PFC; the nuclei were centered, the nucleoli were visible, and the cell structures were entire and well-defined. The distribution of Nissl-positive neurons in the Model group was sparse, showing large intracellular vacuoles, shifted nuclei, and unclear nucleoli. In the Model group, there were fewer Nissl-positive neurons than in the Normal group. The neuron cell body morphology of the EA group was comparable to that of the Normal group. Significantly more Nissl-positive neurons were seen in the EA group than in the Model group (see Fig. 1).

Fig. 1.

Effects of EA on histopathological changes in right PFC of MCAO rats. (A) Representative pictures of Nissl staining results; $\times$200; the scale bar (lower right corner) equals to 50 $\mu$m. (B) Histogram of comparison of Nissl staining results; ${}^{\mathrm{\#}\mathrm{\#}}$p 0.01, compared with sham group; **p 0.01, compared with model group; F value = 78.390, df = 23, n = 6/group.

Gas7 and NGF immunoreactive proteins were detected in the PFC of each group, and the positive cells were brownish yellow, mainly colored in the cytoplasm and cell membrane. In the Normal group, the quantity of Gas7 and NGF-positive cells was low, and their distribution was sparse. The number of Gas7 and NGF-positive cells did not differ substantially between the Sham and Normal groups, whereas the number of Gas 7Gas7 and NGF-positive cells in the Model group was significantly bigger than either. Significantly more Gas7 and NGF-positive cells were seen in the EA group than in the Model group (see Fig. 2).

Fig. 2.

Effect of EA on Gas7 and NGF protein expression was examined through immunohistochemical staining in the right PFC. (A) Representative pictures of Gas7 immunohistochemical staining results; $\times$200; the scale bar (lower right corner) equals to 50 $\mu$m. (B) Histogram of Comparison of Gas7 Immunohistochemical Staining Results; ${}^{\mathrm{\#}\mathrm{\#}}$p 0.01, compared with sham group; **p 0.01, compared with model group; F value = 147.985, df = 23, n = 6/group. (C) Representative pictures of NGF immunohistochemical staining results; $\times$200; the scale bar (lower right corner) equals to 50 $\mu$m. (D) Histogram of Comparison of NGF Immunohistochemical Staining Results; ${}^{\mathrm{\#}\mathrm{\#}}$p 0.01, compared with sham group; **p 0.01, compared with model group; F value = 82.779, df = 23, n = 6/group.

Western blot results showed that Gas7 and NGF proteins were expressed at different levels in the PFC of rats in each group. The expressions of Gas7 and NGF proteins in the PFC of rats in each group were normalized by GAPDH. In the Model group, the expression of Gas7 and NGF proteins in the right (ischemic side) PFC was considerably higher than in the Normal group. In comparison to the Model group, the EA group’s PFC on the ischemia side had considerably higher expression levels of the proteins Gas7 and NGF. The right PFC of the Sham surgery group did not statistically express more Gas7 and NGF proteins than did the Normal group (see Fig. 3).

Fig. 3.

Effect of EA on Gas7 and NGF protein expression was examined through Western blot. (A) Western blot experiment representative band. (B) Histogram of Gas7 protein expression level comparison; ${}^{\mathrm{\#}\mathrm{\#}}$p 0.01, compared with sham group; **p 0.01, compared with model group; F value = 74.369, df = 23, n = 6/group. (C) Histogram of NGF protein expression level comparison; ${}^{\mathrm{\#}\mathrm{\#}}$p 0.01, compared with sham group; **p 0.01, compared with model group; F value = 74.921, df = 23, n = 6/group.