Adult male Sprague-Dawley (SD) rats weighing 200–250 g were provided by Shanghai Laboratory Animal Co., Ltd. (SLAC), Shanghai, China (license no. SCXK 2014-007) for this study. All rats were maintained in standard conditions under constant temperature (23±2°C), humidity (60–70%) and 12-h light/dark cycle with ad libitum access to food and water. All experiments were strictly conducted in accordance with the guidelines of the National Institutes of Health (NIH), and the care and use of the rats were carried out following protocols approved by the Animal Care and Use Committee of Fujian University of Traditional Chinese Medicine (FUTCM).

Rats were randomly divided into the sham operation group (n=16) and the ischemia group (n=54). Rats were anesthetized with 1.5% isoflurane in 68.5% N₂O and 30% O₂. The model of cerebral ischemia was performed by middle cerebral artery occlusion (MCAO) on animals in the ischemia group, as described previously (22). Briefly, a 2-cm midline incision was made on the neck, and the left common, external and internal carotid arteries were carefully exposed and identified. The external and common carotid arteries were ligated, and then 4-0 nylon filament (Jia Ling embolus; Jia Ling Biological Technology Ltd., Guangzhou, China) with a heat-blunted tip was introduced into the internal carotid artery from the common carotid artery to block the origin of the MCA for 2 h. Animals in the sham group had the same surgical procedures but received no occlusion on the artery. The rats were reperfused by withdrawing the filament. The MCAO surgery and sham-operation were carried out and the body temperature of the sham-operated or operated rats was maintained at 37±0.5°C during the surgical procedure (23,24). The Zea-Longa score conducted in a blinded manner at 24 h was made to exclude the rats of failed ischemia and death (25). Using a random number table, the rest of the rats in the ischemia group were randomly assigned into the MCAO group (MCAO), the EA treatment group (EA) and the non-acupoint EA treatment group (non-EA) (n=16 each group).

The groups received EA or non-EA treatment at 24 h after reperfusion and continued treatment for a consecutive 7 days by using an electrical stimulator (Model G6805; SMIF, Shanghai, China). Meanwhile, the Sham and MCAO rats received the same holding but were not subjected to EA or non-EA. A stainless steel needle (0.3 mm in diameter and 30 mm in length) was inserted into the right unilateral paralyzed limb of Zusanli (ST36, located on the anterior lateral side of the leg close to the anterior crest of the tibia) and Quchi (LI11, located at the midpoint between the lateral end of the transverse cubical crease) at a depth of 5 mm. Additionally, the non-acupoint EA treatments were performed by stimulating the distal non-acupoints. One of the non-acupoints is ~3 cm distal from the ST36 acupoint toward the tail and opposite to the knee joint. It is located over the semitendinosus muscle at 5 mm from the tail base. The other one is ~3 cm distal from the LI11 acupoint toward the shoulder joint, located over the biceps brachii inside of the elbow. The parameters of the EA stimulation consisted of a dense disperse wave of 2/20 Hz, 30 min each time, once a day (26).

Neurological function assessments based on a 5-point scale (25) and modified neurological severity scores (mNSS) were performed to evaluate the degree of neurological deficits at 24 h and day 7 after surgery. Behavioral test was performed by the Zea-Longa 5-point scale. To assess neurologic function, mNSS is recognized as a crucial mean, which focuses on overall performance of motion, sensation, balance function and nerve reflex (27,28).

The infarct volumes were detected by T2-weighted imaging (T2WI) at day 7 after reperfusion. Before the scans, animals were anesthetized with 3% isoflurane for 5 min. Rats were placed into a holder compatible with MRI acquisition systems and body temperature was maintained at 37±0.5°C. Assessments were performed by 7.0 T MRI (Pharmascan, microMRI; Bruker, Karlsruhe, Germany), and T2WI spin-echo sequence was used with the parameters: echo time/repetition time, 35/4200 msec; field of view, 32×32 mm; slice thickness, 0.8 mm; slices, 24; matrix, 256×256; average, 4.

Infract regions were defined on the region of increased signal in the ipsilateral hemisphere by using ImageJ software (NIH, Bethesda, MD, USA). The total volume of the infarcted region was calculated by multiplying the total region of the delimited region by the slice thickness.

CatWalk XT9 (Noldus Information Technology, Wageningen, The Netherlands), a real-time and quantitative gait analysis system (29,30) was used for this study in which animals cross a 7-cm wide glass walkway illuminated by green light from the side in a darkened room. When the paws contact the glass floor, the light is scattered and captured by a digital high-speed camera which is then recorded and measured with CatWalk software 10.0. For each animal, we performed analysis of a package of footprint parameters included base of support, stride length, step sequence, print width, swing speed, cadence, average speed and duration (31). The rats were trained for one week, 6 times a day (32). The continual running without any stopovers for the entire distance and 3 valid result recordings are a basic requirement (33). Animals received gait assessment at day 7 after EA treatment.

Motor function was also evaluated by Rota-rod test (Ugo Basile, Varese, Italy) which is widely used in ischemic animal research (34–36). The animals had 3 sessions of training at 4 rpm for 3 days in the pre-training period. After EA treatment, the time to fall on the accelerating rod at 4–40 rpm over a period of 4 min was recorded (37). Each trial was repeated three times.

¹⁸F-fluorodeoxyglucose (FDG)-PET (MILabs BV, Utrecht, The Netherlands) examination was conducted at day 7 after EA treatment to measure glucose metabolic activity of the brain. The rats were sent to the small animal PET/CT laboratory after 12 h of fasting with only water, and were then subjected to the following procedure. i) The rats received gas (3.5% isoflurane, 400 ml/min) anesthesia in the anesthesia induction chamber. ii) ¹⁸F-FDG (1 mCi; PET-CT Center of Nuclear Medicine of Fujian Province Hospital, China) was injected via the tail vein. iii) The rats were allowed to rest for 30 min in a dark room without noise disturbance. iv) Thirty minutes later, the rats again were gas anesthetized, and were subjected to a PET scan until fully anesthetized. v) Rats underwent PET scanning on a VECT or +/CT scanner with constant anesthesia (2.0% isoflurane, 250 ml/min). After setting the scanning parameters (55 kV voltage, 615 µA intensity), spiral CT detection was performed for the first 10 min, and PET scanning was then conducted for 15 min. Any signs of distress include the breathing of the rats were monitored while the image was acquired.

All PET images were corrected for emission scatter and attenuation by CT, and the images were reconstructed with the pixel-based ordered subset expectation maximization (POSEM) with 15 iterations and 32 subsets. The PET data were displayed and analyzed by PMOD software (PMOD Group, Zurich, Switzerland) which drew the region of interest (ROI) in the cerebral hemisphere. The relative metabolic activity, standardized uptake value (SUV), was measured in the regions of interest. The extent of metabolism was expressed as the percentage obtained through the formula: Metabolic SUV right/metabolic SUV left × 100%.

After EA treatment, all rats were euthanized. Then the brains were rapidly removed from the skull, followed by dissention into the left caudate putamen (CPu), motor cortex (MCTX), somatosensory cortex (SCTX) tissue and frozen in liquid nitrogen. The brain samples were stored at −80°C. Samples were homogenized in lysis buffer (RIPA); after centrifugation, a BCA (Thermo Fisher Scientific, Waltham, MA, USA) assay was used to determine protein concentrations. An equal amount of protein was subjected to electrophoresis on a 5–12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gel (Bio-Rad, Hercules, CA, USA) and subsequently transferred to polyvinylidene difluoride membranes (0.45 µm; Millipore, Billerica, MA, USA). Total AMPKα (t-AMPKα, 1:1,000; no. orb14583; Biorbyt, San Francisco, CA, USA), phosphorylated AMPKα (p-AMPKα, Thr172, 1:1000; no. 2531; Cell Signaling Technology, Danvers, MA, USA) were the primary antibodies, and β-actin (1:8,000; Sigma, St. Louis, MO, USA) was used as the loading control. Blots were maintained at room temperature for 1 h in 5% BSA and primary antibody incubation was carried out overnight at 4°C. Incubation with the secondary antibody (1:5,000, goat anti-rabbit IgG; Chemicon, Temecula, CA, USA) was carried out for 2 h, and the ECL detection kit was used for signal detection.

All data are expressed as the mean ± SEM or mean ± SD in each group. Multiple comparisons were carried out with one-way ANOVA, and comparison of two groups was carried out with the LSD test or Mann-Whitney U test. P-value of <0.05 or <0.01 was indicative of statistical significance.

As shown in Fig. 1, compared with the Sham group, the rats with reperfusion had significantly higher scores in terms of the neurological deficit according to the 5-point scale and the mNSS score (P<0.01), whereas no significant differences were noted among the MCAO group, the EA group and the non-EA group (P>0.05). However, the neurological deficits were improved after EA treatment at day 7 (Fig. 1) (P<0.05).

As shown in Fig. 2, the cerebral infarct volume was significantly reduced in the EA group compared with that in the MCAO group and non-EA group at day 7 after EA treatment (P<0.05). These results suggest that EA treatment attenuated the cerebral infarct lesion.