A total of 32 male Sprague-Dawley rats (n=8 per group, 200–220 g, 8–10 weeks old) purchased from Guangdong Medical Laboratory Animal Center (Foshan, China) were used in the current study. These rats were individually housed under a constant temperature (23±2°C) and maintained in a 12-h light/dark cycle with free access to food and water. Rats were anesthetized intraperitoneally with 300 mg/kg chloral hydrate (TargetMol, Boston, MA, USA) prior to transplantation of NSCs. The hair on the thighs of the rats was removed and each animal was placed in the prostrate position on an operating table. A 2 cm skin incision parallel to the femur and inferior at 1 cm was made on the right thigh of the rat to expose the sciatic nerve. The nerve was cut off at 1.5 cm, piercing the piriformis, and a neurological defect of 1 cm was made. The proximal nerve was ligated using a 7-0 noninvasive micro stitch and the distal end was placed aside. The wound was washed with 0.1% povidone-iodine and sutured. All animal experiments were conducted in accordance with the guidelines developed by the National Institutes of Health (13) and approved by the Institutional Animal Care and Use Committee of Peking University Shenzhen Hospital (permit no. 09–215).

The ultrasound system used in the current study included an arbitrary AFG3102 waveform generator (Tektronix, Inc., Beaverton, OR, USA), an AR150A100B RF power amplifier (AR, Inc., Souderton, PA, USA) and a 1.0 MHz unfocused single-element transducer (Panametrics, Inc., Waltham, MA, USA). The parameters of ultrasonic intensity and repetition frequency were optional.

Hippocampi from 7 embryonic day (E)14 Sprague-Dawley rats purchased from Guangdong Medical Laboratory Animal Center (Foshan, China; permit no. SCXK2013-0002) were isolated in a biological safety cabinet and placed in a flask for primary NSC culture. Hippocampal tissues were sheared into 1 mm³ sized tissue blocks. 0.25% trypsin (EMD Millipore, Billerica, MA, USA; 1:250) was used to digest the tissue block at room temperature for 20 min. The digested tissue solution was collected and placed in a 15 ml centrifuge tube, and then Dulbecco's modified Eagle's medium (DMEM)/F12 (1:1 v/v, Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) containing fetal bovine serum (Gibco; Thermo Fisher Scientific, Inc.) was added to stop the digestion. Thereafter, the solution was centrifuged at 450 × g at room temperature for 5 min. Then, cells were harvested and cultured in DMEM/F12 (1:1 v/v) containing 2% B27 (Gibco; Thermo Fisher Scientific, Inc.) and 1% N2 (Gibco; Thermo Fisher Scientific, Inc.) supplements, 0.5 mM L-glutamine (Hyclone; GE Healthcare, Logan, UT, USA), 0.5 mM non-essential amino acids, 20 ng/ml basic fibroblast growth factor (Promega Corp., Madison, WI, USA), 50 IU/ml penicillin (Hyclone; GE Healthcare) and 50 µg/ml streptomycin (Hyclone; GE Healthcare). The cells were cultured in an incubator containing 5% CO₂ at 37°C. The culture medium was replaced half every 3 days, and the cells were passaged after 7 days of culture. Cells were observed and used at day 20.

A total of 200 µl MBs (3×10⁸/ml; Bracco Spa, Milan, Italy) with a mean size of 2–5 µm were mixed with 20 µl pEGFP-NT-3 recombinant plasmid (1 µg/µl) (Corning Incorporated, Corning, NY, USA) in 24-well plates at room temperature for 30 min. The treatments used in each group were as follows: i) In the control group 500 µl NSCs (3×10⁵/ml) were incubated in complete medium (mentioned above) with no supplements; ii) in the NSC+MB+pEGFP-NT-3 group the number of NSCs following trypsin digestion were counted (500 µl; 3×10⁵/ml) and mixed with 20 µl pEGFP-NT-3 recombinant plasmid (1 µg/µl) and 200 µl MBs; iii) in NSC+MB group the number of NSCs following trypsin digestion were counted (500 µl; 3×10⁵/ml) and mixed with 200 µl MBs; and iv) in the NSC+MB+NT-3 group the number of NSCs following trypsin digestion were counted (500 µl; 3×10⁵/ml) and mixed with 200 µl MBs and NT-3 protein (PeproTech, Inc., Rocky Hill, NJ, USA) (100 µl; 50 ng/ml) in place of the plasmid. The ultrasound probe was then placed below each well to sonicate the mixture for the transfection of NT-3 using the following parameters: Ultrasonic intensity, 1.5 W/cm²; sonication time, 60 sec; duty cycle, 25%; and MB concentration, 3×10⁸/ml. Cells were then incubated with 5% CO₂ at 37°C for 48 h.

NSCs were collected and centrifuged at 450 × g at room temperature for 5 min following sonication. A microsyringe at a depth of 4 mm was then used to vertically inject 100 µl NSC (1×10⁷/ml) suspension into the upper third of the right tibialis anterior muscle in the belly of the rats (n=8 in each group). The needle was retained in the muscle for 10 min and 200,000 units penicillin was then intramuscularly injected into the left thigh.

After being sacrificed, the bilateral tibialis anterior muscle was dissected completely from the origin to the insertion and weighed immediately with an electron scale with 0.001 g precision. The muscle wet weight ratio of affected-side/health-side was compared between groups to exclude the individual differences in the weight of the tibialis anterior muscle.

All rats were sacrificed 6 weeks following transplantation for histological evaluation. The entirety of the bilateral tibialis anterior muscle with 2 mm deep peroneal nerve muscular branches of the anterior tibial muscle were immediately removed and samples were weighed using an analytical balance (accuracy to 0.001 g). The muscle from the right thigh was then divided into four parts: The tissue block on the right side of the nerve entering point otherwise known as muscle hilus; the tissue block on the left side of the muscle hilus; and two tissue blocks (1×1×2 mm) around the muscle hilus.

The tissue blocks on the right side of the muscle hilus were fixed in 10% buffered neutral formalin at room temperature for 12 hand a series of sections (5 µm) were cut for H&E staining at room temperature for 15 min. Three fields of view of each section were randomly captured at a magnification of ×200 (Olympus BX53, Olympus Corporation, Tokyo, Japan) and analyzed using ImageJ software (version 1.48, National Institutes of Health, Bethesda, MD, USA). Four muscle fibers were randomly selected and their cross-sectional areas were measured.

AChE staining was used to assess the number of motor endplates in all groups. The tissue blocks on the left side of the muscle hilus were embedded in optimum cutting temperature compound for cryosection at −20°C for 10 min. The serial sections were frozen, repeatedly washed with PBS, and incubated in AChE incubation solution (Beijing Leagene Biotech Co., Ltd., Beijing, China) at 4°C for 12 h and washed again with PBS. The washed sections were then mounted with neutral gum. Three sections were randomly selected from each rat. Two fields of view of each section were randomly selected and observed using an Olympus BX53 microscope at a magnification of ×100. The number of motor endplates was counted.

Gold chloride staining was used to estimate the shape and number of motor endplates. The tissue blocks around the muscle hilus (1×1×2 mm) were fixed in 20% formic acid solution at room temperature for 5 h and then stained with 1% gold chloride until the color changed to a brown-yellow shade at room temperature for 1 h. The tissue blocks were then washed three times with distilled water and restored in 20% formic acid solution until the color changed to a chocolate brown. The tissue was then washed twice with distilled water and stored in glycerin until it softened. Following this, the tissue samples were mounted on glass slides. Two fields of view of each section were randomly selected and observed using an Olympus BX53 microscope at a magnification of ×100.

TEM was used to evaluate the ultrastructural basis of the attenuation of denervated muscle atrophy. The tissue blocks around the muscle hilus (1×1×2 mm) were fixed by perfusion using 2.5% paraformaldehyde and 1.5% glutaraldehyde in 0.1 M phosphate buffer (pH 7.2) at room temperature for 24 h and then post-fixed in a solution of 1% osmium tetroxide and 1.5% potassium ferrocyanine (1 h, 4°C). The tissues were then dehydrated with increasing ethanol (25, 50, 70, 90 and 100%) for 5 min each. The samples were sectioned (50 nm) and observed using a LIBRA 200 FE microscope (Zeiss GmbH, Jena, Germany).

Data are expressed as the mean ± standard error of the mean. The experiments were repeated three times. One-way analysis of variance followed by Tukey's post hoc test was used to compare the means between groups using SPSS (version 17.0; SPSS, Inc., Chicago, IL, USA). P<0.05 was considered to indicate a statistically significant difference.

The affected-side/health-side ratio of the muscle wet weight was compared between groups to exclude the individual differences in the weight of the tibialis anterior muscle (Fig. 1). The NSC+MB+pEGFP-NT-3 group had the highest ratio, whereas the control group had the lowest. The ratios in the NSC+MB+pEGFP-NT-3, NSC+MB and NSC+MB+NT-3 groups were significantly increased compared with the control (P<0.05). A pairwise comparison also identified that the affected-side/health-side ratio of the muscle wet weight was significantly increased in the NSC+MB+pEGFP-NT-3 group compared with the NSC+MB group (P<0.05).

Following H&E staining of the tibialis anterior muscle, visual observation of the sections demonstrated that the control group had the smallest cross-sectional area of muscle fibers, followed by the NSC+MB and NSC+MB+NT-3 groups, while the NSC+MB+pEGFP-NT-3 group had the largest cross-sectional area (Fig. 2). The comparison made using ImageJ software revealed significant differences in the cross-sectional area of the muscle fiber among the groups (Fig. 3). Pairwise comparison demonstrated that the cross sectional area of the muscle fibers in the NSC+MB+pEGFP-NT-3, NSC+MB and NSC+MB+NT-3 groups was significantly increased compared with the control (P<0.05). In addition, the cross sectional area of the muscles fibers in the NSC+MB and NSC+MB+NT-3 groups was significantly decreased compared with the NSC+MB+pEGFP-NT-3 group (P<0.05). However, the difference in the cross-sectional area of the muscle fibers between the NSC+MB and NSC+MB+NT-3 groups was not significant.

AChE staining was used to determine the number of motor endplates in each group (Fig. 4). The lowest number was recorded in the control. Pairwise comparison demonstrated that the number of motor endplates in the NSC+MB+pEGFP-NT-3 group was significantly increased compared with the NSC+MB group, and the number in the control was significantly decreased compared with the NSC+MB and NSC+MB+NT-3 groups (all P<0.05; Fig. 5). However, the difference in the number of motor endplates between the NSC+MB and NSC+MB+NT-3 groups was not significantly different (Fig. 5).

Normal (control group) motor endplates stained with gold chloride exhibited a horseshoe or oval shape and clear color, whereas normal muscle fibers were irregularly shaped (Fig. 6). The motor endplates in the experimental groups (NSC+MB and NSC+MB+NT-3 groups) were fewer, smaller, unevenly dyed and irregular. In addition, atrophy manifestations and fibrous connective tissue increased to different levels. The motor endplates in the control, NSC+MB and NSC+MB+NT-3 groups exhibited unclearly stained areas with an irregular shape and were markedly decreased compared with the NSC+MB+pEGFP-NT-3 group, in which motor endplates exhibited a more regular shape.

The muscle fibers in the NSC+MB+pEGFP-NT-3 group differed slightly compared with the normal (control group) ultrastructure (Fig. 7). The Z lines, dark band and M line were clear, and their shapes were normal. In addition, muscle fiber atrophy and a slightly loose arrangement were observed, and the mitochondrial cristas were shorter. In the NSC+MB+NT-3 and NSC+MB groups, the Z line was slightly blurred, the dark bands and M lines were not clear, the muscle fibers were not neatly arranged. In the muscle fibers of the control group severely abnormal morphology was observed, the Z line was blurred and discontinuous, and the dark bands and M lines were unclear. In the NSC+MB and MSC+MB+NT-3 groups, the motor endplate synaptic cleft was shortened and the secondary folds became shallow or disappeared (Fig. 8). In the NSC+MB+pEGFP-NT-3 group, the secondary folds became slightly shallow and the number of synaptic vesicles in the motor endplates was increased. There were no clear motor endplates present in the control group.