The experimental procedure used in the present study was approved by the Ethics Committee for Animal Experimentation of Sun Yat-sen University (Guangzhou, China) and was performed according to the Guidelines for Animal Experimentation of Sun Yat-sen University. All efforts were made to minimize animal suffering and the number of animals used. Male specific pathogen-free (SPF) Sprague-Dawley rats (n=160), weighing 310±30 g, were provided by the Experimental Animal Center of Sun Yat-sen University and housed under controlled conditions with a 12 hour light/dark cycle, temperature of 24±1°C and humidity of 50±5% for at least 1 week prior to drug treatment or surgery. The rats were allowed free access to a standard rodent diet and tap water.

The rats were anesthetized with an intraperitoneal injection of sodium pentobarbital (3%; Sigma-Aldrich, St. Louis, MO, USA) at a dose of 30 mg/kg. Core body temperature was monitored using a rectal probe (Zhengzhou Haorunqi Electronic Sci-Tech Co., Ltd., Henan, China) and were maintained at 37±0.5°C using a heating lamp and a heating pad (Zhengzhou Haorunqi Electronic Sci-Tech Co., Ltd.). The arterial blood gases, pH, PaO₂, PaCO₂ and blood pressure were closely monitored via catheterizing the right femoral artery using a RM-6240BD physiological minitoring system (Chengdu Instrument Factory, Sichuan, China). MCAO was achieved using the Intraluminal Filament method, as previously described (24). Following exposure of the external carotid artery, the internal carotid artery (ICA) and the pterygopalatine artery of the ICA, a piece of monofilament nylon suture (diameter, 280 µm; Shadong Biotek, Beijing, China), with its tip rounded by gentle heating (diameter, 380±20 µm), was introduced via the lumen of the left external carotid artery stump and left ICA to embed into the left anterior cerebral artery, resulting in the occlusion of the right middle cerebral artery at its origin. Following surgery, the rats were transferred to their cage, in which the temperature was maintained at 37°C until the animals were completely conscious. Sham-operated rats (S group) were manipulated in the same way, however the ICA was not occluded.

Subsequent to the regaining consciousness, neurological deficits were preliminarily determined using a modified Bederson's scoring system (25,26) as follows: 0, no observable deficit; 1, forelimb flexion; 2, forelimb flexion with decreased resistance to lateral push; 3, unidirectional circling; 4, unidirectional circling with decreased level of consciousness. Rats with a score of 2–3 were selected for use in the subsequent experiments.

A total of 128 eligible rats with successful MCAO were randomly assigned into either the electroacupuncture treatment group (EA group) or control group (M group). In the EA group rats, stainless acupuncture needles (diameter of 0.3 mm) were inserted into the acupuncture points Baihui (DU20) and Shuigou (DU26) at a depth of 2–3 mm. Stimulation was then generated by the EA apparatus (Model G6805; SMIF, Shanghai, China), and the stimulation parameters were set as follows: Disperse wave, 4 and 20 Hz; electric current, 1–2 mA; voltage, 2–4 mV; 15 min of each treatment; once a day. EA treatment was performed at 72 h following surgery and continued until the animals were sacrificed for tissue preparation. The rats in the M group were subjected to the same manipulation procedure without any electric stimulation.

Behavioral assessments of the rats in the two groups were performed by an investigator in a blinded-manner. The assessments consisted of the modified Neurological Severity Score (mNSS) and Morris water maze. All measurements were completed in a dedicated behavioral investigation facility during this interval in order to minimize the environmental impact associated with transfer between home cage and the test arenas.

The mNSS test was performed at 3, 7, 14 and 21 days of EA or control treatment. Table I shows the mNSS scores (27). Neurological function was graded on a scale of 0 to 18, with 0 indicating a normal score and 18 indicating the maximal deficit score. The mNSS is a composite of motor, sensory, reflex and balance tests. In the severity scores of injury, 1 score point is awarded for the inability to perform the test or for the lack of a reflex; thus, the higher the score, the more severe the injury (27).

The Morris water maze test was used to measure spatial learning and memory at days 7, 14 and 21 EA or control treatment, as previously described (28). Each rat was placed individually in a black water tank (136 cm in diameter and 60 cm in height) in a well-lit room and filled with water (50 cm height at 27°C). The tank was visually separated into four quadrants and, in the center of one quadrant, a plexiglass platform (diameter, 10 cm; Zhengzhou Haorunqi Electronic Sci-Tech Co., Ltd.) was hidden 2 cm below the waterline. Numerous extra-maze cues surrounding the maze were fixed at specific locations and were visible to the rats. Following a single habituation trial, the rats were trained over 4 days and received four trials each day. On each trial, the rat had 120 sec to escape to the submerged platform; rats that failed to escape were led to the platform and remained there for 30 sec prior to being removed from the maze and dried off. On the fifth day, a probe trial was performed following removal of the platform. The latency of quadrant search, path length and swim speed were measured by a TSE video tracking system (TSE Systems, Bad Homburg, Germany) interfaced to a computer on days 3, 7, 14 and 21 of treatment.

In order to track the fate of proliferating cells, the rats were administered with intraperitoneal (i.p.) injections of 75 mg/kg body weight bromodeoxyuridine (BrdU; Sigma-Aldrich) dissolved in phosphate-buffered saline (PBS; Whiga Technology Co., Ltd., Guangzhou, Guangdong, China) twice a day for 5 days prior to sacrifice. At days 3, 7, 14 and 21 of EA or control treatment, rats (n=5 per experiment subgroup per time-point) were sacrificed through deep anesthetization with 45 mg/kg sodium pentobarbital and perfused intracardially with cold PBS and 4% buffered paraformaldehyde. Brains were paraffin-embedded (Whiga Technology Co., Ltd.) and sliced into 5-µm thick coronal sections, starting at 3 mm posterior to the anterior pole. The sections were then incubated with the mixture of monoclonal mouse antibodies (1:400) for BrdU (cat. no. MCA2060T; Oxford Biotech, Oxford, UK) with polyclonal mouse antibodies (1:400) for neuronal nuclei (NeuN; cat. no. SAB4300883; Sigma-Aldrich) or polyclonal mouse antibodies (1:10,000) for glial fibrillary acidic protein (GFAP; cat. no. SAB1405864; Sigma-Aldrich) overnight at 4°C, followed by a mixture of fluorescein isothiocyanate-conjugated goat anti-rabbit immunoglobulin (Ig)G and cyanine dye (Cy)3-conjugated goat anti-mouse IgG (1:100; Jackson Laboratory, Bar Harbor, ME, USA). The sections were washed in Tris-buffered saline (TBS; 0.2% Triton X-100 in PBS; Whiga Technology Co., Ltd.) and mounted with Vectashield mounting medium (Vector Laboratories, Burlingame, CA, USA). Confocal images were captured using an LSM710 confocal spectral microscope (magnification, x400; ZEISS, Oberkochen, Germany). Image analysis was performed using Leica QWin software (Leica Microsystems, Weltzlar, Germany). The number of cells showing double immunostaining was estimated by counting cells in five random fields of the peri-necrotic cortex in five non-continuous sections.

Following evaluation of neurological deficits, rats (n=6 per experiment group per time-point) were sacrificed with an overdose of sodium pentobarbital at days 3, 7, 14 and 21 of EA or control treatment, as described above. Total RNA was extracted from the perinecrotic cortex using TRIzol reagent (Invitrogen Life Technologies, Carlsbad, CA, USA) according to the manufacturer's instructions. RNA (2 µg) was then reverse transcribed into first-strand cDNA using M-MLV Reverse Transcriptase (Promega Corp., Madison, WI, USA) according to the manufacturer's instructions. Notch1, Hes1 and GAPDH were amplified by qPCR using the following primers: Notch1, forward 5′-TCGTGCTCCTGTTCTTTGTG-3′ and reverse 5′-GGGTTCTCTCCGCTTCTTCT-3′; Hes1, forward 5′-GCTGGAGAAGGCAGACATTC-3′ and reverse 5′-GTCACCTCGTTCATGCACTC-3′; and GAPDH, forward 5′-TGCCACTCAGAAGACTGTGG-3′ and reverse, 5′-TTCAGCTCTGGGATGACCTT-3′ (BGI Tech, Guangdong, China). Gene-specific amplification was performed in an ABI 7500 real-time PCR system (Life Technologies, Carlsbad, CA, USA) using 15 µl PCR mix containing 0.5 µl cDNA, 7.5 µl 2x SYBR Green master mix (Invitrogen Life Technologies) and 200 nM of the appropriate primers. The mix was preheated at 95°C for 30 sec and amplified in 45 cycles of 95°C for 5 sec and 60°C for 34 sec. The resolution curve was measured at 95°C for 15 sec, 60°C for 15 sec and 95°C for 15 sec. The threshold cycle (Ct) value of each sample was calculated and the relative expression of Notch1 and Hes1 mRNA was normalized to the GAPDH value (2^−ΔCt method) (29).

All values are expressed as mean ± standard deviation. A repeated measurement analysis of variance was performed to analyze differences in the behavioral assessment data and relative mRNA expression levels. Analysis of variance and the Least Significant Difference test were used to further compare the differences at each time-point. SPSS 17.0 software (SPSS, Inc., Chicago, IL, USA) was used for all statistical analyses. P<0.05 was considered to indicate a statistically significant difference between values.

In order to examine the effects of EA treatment on the functional recovery of MCAO rats, the mNSS was used to assess neurological deficits. The mNSS of the rats in the S group were 0 at all time-points. As shown in Fig. 1A, the rats in the EA group had a significantly lower mNSS compared with that in the M group at 14 (5.33±1.22 and 7.00±0.87, respectively; P<0.05) and 21 (2.78±0.67 and 4.78±0.67, respectively; P<0.05) days of treatment. Compared with those in the M group, rats in EA group only had significant improvement in the motor test score (P<0.05; Fig. 1B) but not in scores of other tests at 14 and 21 days of treatment. The cognitive recovery of MCAO rats was further assessed using the Morris water maze test. As shown in Fig. 1C, rats in the EA group had significantly decreased escape latency compared with those in the M group at 14 (90.63±2.62 and 98.13±2.90, respectively; P<0.05) and 21 (84.63±1.92 and 94.13±1.46, respectively; P<0.05) days of treatment. These results suggested that EA treatment effectively improved the neurological dysfunction of cerebral ischemia in rats, promoting nerve functional recovery.

BrdU/GFAP and BrdU/NeuN sdouble-labeling immunofluorescence was performed to evaluate the proliferation and differentiation of NSCs following MCAO in each group. As shown in Figs. 2 and 3, MCAO rats had significantly higher frequencies of BrdU⁺/GFAP⁺ and BrdU⁺/NeuN⁺ cells in the DG area compared with those in the S group (P<0.05), indicating that the proliferation and differentiation of NSCs was stimulated by ischemic stress. In addition, EA treatment significantly elevated the frequencies of BrdU⁺/GFAP⁺ cells at 3, 7 and 14 days of treatment compared with the control treatment (P<0.05), indicating the enhanced effects of EA treatment on the differentiation of NSCs into glia cells. Furthermore, EA treatment enhanced the neuronal differentiation of NSCs, as indicated by increased frequencies of BrdU⁺/NeuN⁺ cells in the EA group, compared with that of the M group at 21 days of treatment (P<0.05).

In order to examine the underlying molecular mechanisms by which EA treatment promoted the proliferation and differentiation of NSCs, the expression of Notch1 and its target molecule, Hes1, were analyzed by RT-qPCR. The results revealed that MCAO stress significantly stimulated the expression of Notch1 and Hes1 in the DG area compared with that of the S group at 3, 7 and 14 days of treatment (P<0.05; Fig. 4).