A total of 50 adult (8-week-old) Sprague-Dawley male rats (230-250 g) were obtained from Hunan SJA Laboratory Animal Co., Ltd. (animal certificate no. 43004700000040). The animals were individually housed under a 12/12-h light/dark cycle at 22±2˚C with 55-65% humidity and fed with standard chow and water ad libitum. The rats were randomly allocated into a sham operation group (Sham), model group (SCI), treadmill training treatment group (TT), MP treatment group (MP) and MP + treadmill training treatment group (MP + TT) (n=10 rats in each group). One week before the SCI surgical procedure, an adaptive training was performed on a computer-controlled animal experiment platform (ZH-PT computer control experimental animal treadmill; Anhui Zhenghua Biologic Apparatus Facilities) in order for the rats to familiarize with the track environment to ensure a continuous run. Every morning and evening, each group was trained for 10 min using a treadmill speed of 20 m/min. All animal procedures strictly followed the Guide for the Care and Use of Laboratory Animals (23) and were approved by the Jiangxi University of Traditional Chinese Medicine Committee on Animal Research (approval no. JZLLSC2019-0147).

The SCI animal model was constructed as previously described, with several modifications (24). Briefly, rats were anesthetized with 5% chloral hydrate [300 mg/kg, intraperitoneal injection (i.p.); Shanghai yuanye Bio-Technology Co., Ltd.]. Using the T10 spinous process as the center, a 4.0 cm-longitudinal incision was performed to separate the muscles and fascia from both sides of the spine in order to expose the spinal dura mater. A T10 SCI was achieved by the impact of PinPoint^™ Precision Cortex Striker (diameter, 3.0 mm; Hatteras Instruments Inc.). The impact depth was 1.0 mm; the impact velocity was 2.0 m/sec and the residence time was 200 msec. The sham operation group was subjected to the same procedure described above, but without impact. Post-surgical anti-inflammatory therapy was provided by an intramuscular injection of penicillin and gentamicin sulfate (25 mg/kg every 12 h for 7 days). All rats were administered a daily analgesic, meloxicam (1 mg/kg, i.p.; Sigma-Aldrich; Merck KGaA), for the first 7 days. In case of urinary tract infection, the anti-inflammatory therapy was prolonged to 10 days. The rectal temperature was maintained at 37±1˚C using a thermostatic pad. The injured rats' bladders were manually emptied twice daily until spontaneous voiding occurred.

At 30 min after SCI, rats in the MP and MP + TT group received a bolus injection of MP sodium succinate (30 mg/kg; Solu-Medrol; cat no. S02340; Pfizer, Inc.) in 0.9% saline (0.5 ml) via tail vein injection. The rats in the Sham, TT and SCI groups were treated with an injection of 0.5 ml normal saline. A six-lane treadmill (ZH-PT computer control experimental animal treadmill; Anhui Zhenghua Biologic Apparatus Facilities) was used to train the animals and to obtain kinematic data. At 1 week before the SCI surgical procedure, an adaptive training was performed on the aforementioned six-lane treadmill for 6 days with a 10-min running period at 20 m/min each day and baseline gait measurements were taken. Two weeks after SCI, the TT and MP + TT groups commenced the training (Fig. 1). The treadmill speed was initially set at 5 m/min for 5 mins, twice/day, 6 days/week, and then the speed was increased by 1 m/min per week for 8 weeks. An excitatory stimulation at 0.05 mA for 1-2 sec was applied to encourage the rats to continue running for the duration of the training.

Gait recording and analysis were performed as previously described (25). Briefly, a small animal gait analyzer for hind limb gait video was used (Clever Sys Inc.). Each rat's gait was recorded five times (20 sec each time), and three clear videos were selected for analysis. First, the speed of the running belt was set, so that the animals in the closed transparent tunnel containing the running tape were not running at their own speed. Next, a high-speed color camera was used to evaluate the ventral surface angle between the gait and the running tape. The video file was delivered to the meter connected to the computer, and TreadScan 4.0 system software (Clever Sys Inc.) was used for analysis. The angle of hind limb gait, hind limb swing time, rear foot pitch, stride time, right foot baseline and hind limb maximum lateral deviation were recorded. The recording and analysis were performed using a double-blinded method by two trained individuals. The basic data were recorded 1 day before SCI, and four rats in each group were used to record the gait at 2-8 weeks after SCI.

A section containing the spinal cord tissue was obtained as previously described (26). Briefly, after completing the behavioral assessment, rats were anesthetized with 5% chloral hydrate (300 mg/kg i.p.; Source Leaf Biological Technology) and subjected to transcardial perfusion with 0.9% saline followed by 4% paraformaldehyde (Sigma-Aldrich; Merck KGaA) in 0.1 M PBS (pH 7.4). After perfusion, the spine was removed, and the injured segment was collected (including ~1 cm of the spinal cord above and below the injury) and placed in 4% paraformaldehyde solution. The samples were fixed by incubation for 1 day at 4˚C and dehydrated with a gradient 0.1 mol/l cacodylic acid buffer containing 20 and 30% sucrose. Next, the samples were placed on a HM550 frozen microtome (Microm UK Ltd.), and 20 µm-thick sections were cut, dried at room temperature and stored at -80˚C. One slice showing the center of the impact (+0 mm) and approximately 3 and 5 mm from the impact was selected and stained with Harris' hematoxylin for 10 min and with eosin for 1 min at room temperature. (H&E; Sigma-Aldrich; Merck KGaA) and Luxol fast blue for 12 h at 37^°C (Sigma-Aldrich; Merck KGaA) for routine pathological examination. A total of 10 transversal 20-mm sections from each sample were processed to analyze the cavitation area, and the images were captured by an Olympus DP71 system with an inverted Olympus IX71 microscope (Olympus Corporation; magnification, x40). The myocardium area and the total area of the spinal cord were statistically analyzed with Image-Pro Plus (version 6.0; Media Cybernetics, Inc.). Following completion of sample acquisition, the rats were euthanized by cervical dislocation following anesthesia with intraperitoneal injection of 5% chloral hydrate (300 mg/kg).

Immunohistochemistry and image analysis were performed as previously described (27). Four replicate sections of the spinal cord damaged center (0 mm) and its front and back (3 and 5 mm) were subjected to immunohistochemistry using the SP-9000 Two-Step Immunohistochemistry kit (OriGene Technologies, Inc.). Briefly, tissue sections were fixed in cold acetone (Sigma-Aldrich; Merck KGaA) at 100% for 10 min at 4^°C and incubated in 3% deionized aqueous solution for 5-10 min at room temperature to eliminate endogenous peroxidase activity. Next, tissues were washed with 0.01 M PBS three times at 3 min each time. Subsequently, tissues were blocked with 10% normal goat serum (Beijing Solarbio Science & Technology Co., Ltd.) in PBS containing 0.2% Triton X-100 for 10-15 min at room temperature, and then washed with PBS. The sections were then incubated with primary antibodies against NgR (rabbit; 1:10,000; cat. no. AB15138; Merck KGaA) or CSPGs (rabbit; 1:10,000; cat. no. 980704W; Shanghai Tianyuan Biotechnology Co., Ltd.) for 3 h at room temperature. Subsequently, a biotinylated secondary antibody working fluid lgG (1:200; cat. no. SP-9000; OriGene Technologies, Inc.) in blocking solution (5% goat serum in PBS; Beijing Solarbio Science & Technology Co., Ltd.) was added, and the sections were incubated at 37˚C for 15 min. Subsequently, the sections were washed with 0.01 M PBS three times (5 min each time), HRP-labeled streptavidin working solution in blocking solution was added and the sections were incubated at 37˚C for 10-15 min. The sections were then washed again with 0.01 M PBS three times (5 min each time) and 3,3-diaminobenzidine was added to the slices, completely covering the tissue, to develop the color for 5-10 min at room temperature. Next, the tissue sections were placed in hematoxylin solution for 1-2 min for counterstaining at room temperature. The tissues were then subjected to gradient dehydration with 75, 80, 90, 95 and 100% ethanol solution, soaked for 2-3 min and submerged in xylene for ~3 min. Subsequently, 1-2 drops of neutral gum (Bioworld Technology, Inc.) were added to the coverslip for 2 min at room temperature and covered with a slide. The percentage of positive cells was calculated with Image-Pro Plus software (version 6.0; Media Cybernetics, Inc.) and ImageJ 1.45d image analysis system (National Institutes of Health) by using an inverted Olympus IX71 microscope (Olympus Corporation; magnification, x40), as previously described (26). Briefly, the fluoresced cell profiles were counted directly from the screen by placing marks of different colors onto positive and negative nuclei by clicking the mouse. ImageJ then automatically generated the immunohistochemistry index.

To detect CSPG and NgR gene expression, reverse transcription-quantitative PCR (RT-qPCR) was performed as previously described (28). Rats were sacrificed at 8 weeks after the initial injury. Four rats in each group were selected to analyze CSPGs and NgR gene expression. For this purpose, the spinal cord was collected at ~1 cm before and after the center of the spinal cord damage. Total RNA extraction and cDNA synthesis were performed as previously described (29). Briefly, total RNA from each individual sample was isolated using the RNeasy Protect Mini kit (Qiagen, Inc.). RNA purity and quantity were assessed using SmartSpec Plus (Bio-Rad Laboratories, Inc.). cDNA synthesis was performed using the iScript cDNA Synthesis kit (Bio-Rad Laboratories, Inc.) in accordance with the manufacturer's instructions. RT-qPCR analysis was performed using the iQ5 instrument (Bio-Rad Laboratories, Inc.) and detected with SYBR Green (BioChain Institute, Inc.). The reaction mixtures contained diluted cDNA, Super Mix (Invitrogen; Thermo Fisher Scientific, Inc.), PCR primers (Table I), and nuclease-free water to a final volume of 20 µl. The following thermocycling conditions were used for the qPCR: Initial denaturation at 95˚C for 30 sec, followed by 40 cycles of 95˚C for 15 sec and 60˚C for 35 sec. Relative quantification of gene expression was calculated using the 2^-ΔΔCq method (30). The primers were obtained and synthesized by Invitrogen (Thermo Fisher Scientific, Inc.). β-actin was used as the internal reference (28).

SPSS software 16.0.2 (SPSS, Inc.) was used for statistical analysis. All data were expressed as the mean ± SD. Comparisons between groups were analyzed using one-way ANOVA followed by Tukey's post hoc test. P<0.05 was considered to indicate a statistically significant difference.

To investigate whether MP and TT combination improved the recovery of motor function in rats with SCI, the locomotor function of each group was compared using the TreadScan analysis system. The motor function of the affected hind limbs was evaluated at the beginning of the 2-week period post-SCI and weekly thereafter for 8 weeks (Fig. 2). TreadScan analysis revealed that the combined treatment of MP and TT resulted in a significantly higher gait angle compared with the MP and TT groups (P<0.01; Fig. 2A). At 2-3 weeks post-injury, MP + TT treatment exerted a significant decrease in the distance between the bilateral hind legs and the body's horizontal axis compared with the effects of MP and TT treatment (P<0.05), and the difference was more significant at 4-8 weeks post-SCI (P<0.01; Fig. 2B). The time required for the rats under MP + TT treatment to perform the walk step test was significantly lower compared with rats under MP and TT treatment at 2-4 weeks post-SCI (P<0.05), and this difference was more significant at 5-8 weeks post-injury (P<0.01; Fig. 2C). The hind limb swing time of the rats subjected to MP + TT treatment was significantly lower compared with rats under MP and TT treatment at 2-4 weeks post-SCI (P<0.01), and a significant difference between the two groups was still present at 6-8 weeks post-injury (P<0.05; Fig. 2D). Rats under MP + TT treatment at 6-7 weeks post-SCI showed a significant reduction in rear foot pitch compared with rats under MP and TT treatment (P<0.01; Fig. 2E). The distance between the right foot and the baselinein the MP + TT group was significantly smaller than that observed in the MP group at 2-4 weeks after SCI (P<0.01), and this difference was still significant at 8 weeks after injury (P<0.05; Fig. 2F). These results suggested that MP + TT treatment was more effective than MP and TT alone in promoting the recovery of hind limb motor function in rats with SCI.

To determine the extent of tissue destruction in each group, H&E and Luxol solid blue staining were used. The results of H&E staining in Fig. 3 showed that the SCI, TT, MP + TT and MP groups had different degrees of tissue loss and cyst appearance at 8 weeks after injury. Although the MP + TT group showed fibrous scar formation in the periphery of the lesion, the total glial cell proliferation, tissue gap and cyst area were significantly smaller than those observed in the MP and TT groups.

The results of Luxol solid blue staining in Fig. 4 showed no significant difference in residual myelin area among the groups at ±5 mm from the injury center at 8 weeks after injury (Fig. 4B and F). At the injury center, the residual myelin area in the MP + TT group (0.353±0.033) was significantly higher compared with the MP group (0.221±0.067) (P<0.05; Fig. 4D). In addition, a significant difference in residual myelin area between the MP + TT group and the MP group was observed at ±3 mm from the injury center (P<0.01; Fig. 4C and E).

To investigate whether the combined treatment of MP and TT affected nerve regeneration in rats with SCI, NgR and CSPG gene and protein expression were detected in each group. NgR-positive areas were mainly concentrated around the lesion center. With the increase in distance from the lesion center, the expression was gradually decreased (Fig. S1). NgR expression in the Sham group was significantly lower compared with the MP and TT groups (P<0.01; Fig. 5B-E). In addition, the positive expression of NgR in other areas in the SCI group was significantly higher than that displayed by the Sham group (P<0.01). NgR expression in the MP + TT group was significantly lower compared with the MP group at 0 and ±3 mm from the injury center (P<0.01; Fig. 5C-E), and the difference was more evident at ±5 mm from the injury center (P<0.01; Fig. 5B and F).

The results of CSPG immunohistochemical staining are shown in Fig. 6. The positive CSPG areas in each group were mainly concentrated in the center of the injury, spreading to both sides (Fig. S2). CSPG expression in the MP + TT and Sham groups was not significant at ±5 mm from the injury center compared with the MP group (Fig. 6B and F). At the center of the injury (0 mm) and at ±3 mm from this point, CSPG expression in the MP + TT and Sham groups was significantly lower compared with the MP and TT groups (P<0.01) (Fig. 6C-E).

To further confirm that the MP and TT combination therapy inhibited the expression of NgR and CSPGs, the mRNA expression of these molecules in the different treatment groups was evaluated by fluorescence qPCR. The NgR mRNA expression in the MP + TT group (0.467±0.150) and in the Sham group (0.231±0.102) was significantly lower compared with the MP and TT groups (P<0.01; Fig. 7A). The NgR mRNA expression in the SCI group (1.413±0.110) was significantly higher compared with the MP and TT groups (P<0.05; Fig. 7A). Similarly, CSPG mRNA expression in the MP + TT (0.798±0.012) and Sham (0.798±0.012) groups was significantly lower compared with the MP group (P<0.05; Fig. 7B). The CSPG mRNA expression in the SCI group (1.364±0.259) was significantly higher compared with the MP group (P<0.05; Fig. 7B).