Delayed administration of recombinant human erythropoietin reduces apoptosis and inflammation and promotes myelin repair and functional recovery following spinal cord compressive injury in rats

Abstract

Purpose: A previous study showed that a 1-h delay in treatment of thoracic spinal cord injury (SCI) with recombinant human erythropoietin (rhEPO) lacked neuroprotective efficacy. The aim of the present study was to reassess delayed administration of different doses of rhEPO on acute spinal cord compressive injury in rats.

Methods: The experiment was divided into first and second stages, which SCI rats were observed for 4 and 28 days, respectively. All rats were randomly divided into four groups at both stages: control group, and rhEPO-3,000U (Unit), rhEPO-4,000U and rhEPO-5,000U groups. SCI rats received rhEPO treatment at different time points. The primary indicators were locomotor recovery, histopathology, apoptotic index, inflammatory index, ultrastructural scoring system and volume of areas of demyelination.

Results: The most significant locomotor functional and histopathological improvements and the best myelin protection were observed after administration of 5,000 U/kg rhEPO. rhEPO at 3,000, 4,000 and 5,000 U/kg showed similar ultrastructural neuroprotection, as well as similar inhibition of apoptosis and regulation of inflammation.

Conclusion: Delayed administration of rhEPO can reduce apoptosis and inflammation, and promote myelin repair and functional recovery following spinal cord compressive injury in rats.

Keywords

Erythropoietin delayed treatment spinal cord injury myelin functional recovery rats

1 Introduction

Traumatic spinal cord injury (SCI) is a serious threat to patients due to high morbidity and mortality worldwide. In addition, SCI results in permanent and severe disability (Profyris et al., 2004). Many patients with paralysis associated with SCI need long-term care from their family or nurses, because they cannot take care of themselves. In addition, families have to bear the high cost of treatment, but it is usually difficult to achieve good clinical outcomes due to lack of effective therapies. How to improve function and enhance quality of life in patients with SCI has become a challenge for doctors. Methylprednisolone (MPS) has been regarded as the first choice and only available treatment for SCI patients over the past few decades and should be applied within 8 h after injury. MPS can attenuate progressive axon damage, reduce neuronal death, and inhibit microglial/macrophage accumulation, especially when initiated shortly after SCI (Zhang et al., 2015). The benefits of MPS therapy are due to inhibition of lipid peroxidation and inflammation (Hall & Braughler, 1986). However, possible severe side effects of high doses of MPS for SCI should be not ignored, such as gastrointestinal bleeding, aseptic necrosis of the femoral head, and intestinal perforation. Currently, MPS is no longer the standard of care and merely a treatment option, or, in some centers, abandoned altogether (Priestley et al., 2012).

Erythropoietin (EPO) is a glycoprotein cytokine that is produced mainly by the fetal liver and adult kidney and is involved in regulation of red blood cell production. EPO has recently received more attention from researchers, and hypoxia is important in inducing its expression. Hypoxic induction of EPO is up to150-fold higher than under normoxic conditions in the kidney, and is associated with an increase in the number of EPO-producing cells rather than the level of EPO production per cell (Eckardt et al., 1993; Obara et al., 2008). Currently, EPO and its analogs are used widely to treat some diseases in clinical practice, such as chronic kidney disease, anemia, cancer, chronic heart disease, and traumatic brain injury (Min et al., 2013). Recently, a retrospective study proposed that rhEPO could markedly improve neurological function without toxic or side effects in patients with acute SCI (Xiong et al., 2011).

EPO has more widespread actions in addition to regulating red blood cell production. Previous studies have shown that EPO inhibits oxidative stress (Ozturk et al., 2005), regulates inflammation (Chen et al., 2007), suppresses glutamate release (Kawakami et al., 2001), reduces severity of spinal cord edema (Jin et al., 2014), decreases lipid peroxidation (Solaroglu et al., 2003), and increases blood flow and tissue oxygenation (Cetin et al., 2006) in various animal models. All the above actions contribute to the neuroprotective functions of EPO through reducing apoptosis and necrosis. An electrophysiological study has proposed that the spinal cord pathways spared by EPO might contribute to improved neurotransmission, which in turn, might be responsible for locomotor recovery (Cerri et al., 2012). In spinal cord slice culture after lysolecithin-induced demyelination, EPO therapy enhanced remyelination by promoting oligodendrogenesis in association with elevated EPO receptor expression (Cho et al., 2012). EPO obviously improved motor performance as compared to MPS in a rat SCI model (Boran et al., 2005; Fumagalli et al., 2008).

Doses of 3,000 and 5,000 U/kg rhEPO have been widely used in animal experiments and have achieved good results (Cerri et al., 2012; Cetin et al., 2006; Jin et al., 2014; Kaptanoglu et al., 2004), however, there are no reports comparing the effects of the different doses on acute SCI. The animals with thoracic SCI received rhEPO treatment within a short time (up to 1 h) after injury (Cerri et al., 2012; Cetin et al., 2006; Jin et al., 2014; Kaptanoglu et al., 2004; Mann et al., 2008), but it is difficult for SCI patients to receive rhEPO treatment within 1 h after injury in clinical practice. Two previous studies found that rhEPO was ineffective when administered to rats with SCI (Mann et al., 2008; Pinzon et al., 2008), and delayed treatment of SCI with rhEPO did not show any neuroprotective activity (Mann et al., 2008). Therefore, a further systematic study is necessary to identify whether rhEPO therapy is effective, especially in animals with SCI that receive rhEPO treatment >1 h after injury. The dose of drug used in animal experiments can provide a reference value for its clinical application. Hence, further preclinical investigation is necessary to refine the treatment strategy of using rhEPO in acute thoracic SCI.

The aim of this study was to evaluate comprehensively and systemically the effectiveness of different doses of rhEPO, at 2 h after SCI, in a rat model of experimental acute thoracic spinal cord compression injury, through assessment of locomotor recovery, histopathology, apoptotic index, inflammatory index, volume of areas of demyelination regions, and ultrastructure. Transmission electron microscopy was used to observe the ultrastructure of the SCI after administration of rhEPO on day 4 post-injury. In addition, Luxol Fast Blue (LFB) staining was performed to observe demyelination on day 28post-injury.

2 Materials and methods

2.1 Animals

The study protocols conformed to the Guide for the Care and Use of Laboratory Animals from the National Institutes of Health and were approved by the Animal Care and Use Committee of Southern Medical University. A total of 56 adult male Sprague–Dawley rats (220–260 g) were purchased from the Animal Center of Sun Yat-Sen University. All rats were placed two per cage under temperature-controlled conditions with a 12-h light/dark cycle, and they had free access to tap water and food.

2.2 Experimental design

Studies were designed to evaluate comprehensively and systematically the effects of different doses of rhEPO on acute SCI through assessment of locomotor recovery, histopathology, apoptotic index, inflammatory index, ultrastructure, and volume of areas of demyelination. The experiment was divided into first and second stages. In the first stage, 32 rats were randomly divided into four groups of eight: control group (A), rhEPO-3,000U(Unit) group (B), rhEPO-4,000U group (C) and rhEPO-5,000U group (D). All rats with acute SCI were observed for 4 days after injury in order to assess locomotor recovery, histopathology, apoptotic index, inflammatory index, and ultrastructure. In the second stage, 24 rats were randomly divided into four groups of six: control group (Aa), rhEPO-3,000U group (Bb), rhEPO-4,000U group (Cc) and rhEPO-5,000U group (Dd). The 24 rats with acute SCI were observed for 28 days after injury in order to assess locomotor recovery and volume of areas of demyelination.

A. Control groups (A and Aa): the spinal cord was clamped for 1 min duration, and single dose 0.9% saline solution with infusion (5 ml/kg) was given at 2 h after injury by intraperitoneal (i.p.) injection. The same dose of saline solution was given to rats on days 1 and 3 after injury by i.p. injection. In addition, the rats in group Aa also received the same dose of saline solution on days 8, 15 and 22. For preventing infection of the surgical incision, penicillin [200,000 U/kg intramuscular (i.m.)] was given on day 1 after injury.

B. rhEPO-3,000U groups (B and Bb): the spinal cord was clamped for 1 min, and rhEPO (3,000 U/kg, i.p. infusion) was given at 2 h after injury. The same dose of rhEPO was given on days 1 and 3 after injury. In addition, the rats in group Bb also received the same dose of rhEPO on days 8, 15 and 22. For preventing infection of the surgical incision, penicillin (200,000 U/kg, i.m.) was given on day 1 after injury.

C. rhEPO-4,000U groups (C and Cc): the spinal cord was clamped for 1 min, and rhEPO (4,000 U/kg, i.p. infusion) was given at 2 h and days 1 and 3 after injury. In addition, the rats in group Cc also received the same dose of rhEPO on days 8, 15 and 22. Penicillin (200,000 U/kg, i.m.) was given on day 1 after injury in order to prevent infection of the surgical incision.

D. rhEPO-5,000U groups (D and Dd): the spinal cord was clamped for 1 min, and rhEPO (5,000 U/kg, i.p. infusion) was given to rats at 2 h and days 1 and 3 after injury. In addition, the rats in group Dd also received the same dose of rhEPO on days 8, 15 and 22. Penicillin (200,000 U/kg, i.m.) was given on day 1 after injury in order to prevent infection of the surgical incision.

2.3 Surgical procedures

Preoperatively, penicillin (200,000 U/kg) was given to animals by i.m. injection in order to prevent infection of the surgical incision. All animals were fasted for 12 h before surgery and humanely restrained. Animals were anesthetized with pentobarbital (70 mg/kg) by i.p. injection and then they were fixed in the prone position. Back hair on the surgical area was removed with scissors. Iodophor (3%) was used to disinfect the surgical area for infection prophylaxis. A midline incision of 2 cm was made from T8 to T12, the overlying musculature was separated laterally, and a complete T10 level laminectomy was performed using an operating microscope to expose the spinal cord. The spinal cord was subjected to extradural compression with a temporary aneurysm clip (70 g force; 65821T; Rebstock, Dürbheim, Germany) for 1 min to induce crush injury. The surgical site was closed using nondegradable sutures after removing the aneurysm clip, and then the closed skin incision was disinfected again by 3% iodophor. During the procedure, body temperature was maintained by exposing the animal to a heat lamp. After surgery, all rats received 2 ml 10% glucose solution immediately and were returned to their cages when they completely recovered from anesthesia. The rats underwent manual bladder evacuation, three times a day.

2.4 Locomotor recovery assessment

In the first stage, locomotor recovery was evaluated in an open field at 6 h, on days 1, 2, 3 and 4 after injury according to the Basso, Beattie, and Bresnahan (BBB) locomotion rating scale of 0 (complete paralysis) to 21 (normal locomotion) (Basso et al., 1995). In the second stage, all SCI rats were evaluated on days –1 (before surgery), 0, 2, 4, 7, 14, 21 and 28 using the BBB method. The BBB scale was developed based upon natural progression of locomotor recovery in rats with thoracic SCI. The scale grossly assessed hind limb movements, body weight support, fore limb-hind limb coordination, and whole-body movements. Each session was conducted at the same time (at 5 p.m. every day) under the same conditions for 5 min by two independent observers blinded to the experiments at different time points after injury.

2.5 Sample preparation and histopathology

After locomotor function had been tested using the BBB scale at day 4 post-injury in the first stage and at day 28 post-injury in the second stage, the rats were sacrificed using pentobarbital (107 mg/kg, i.p.). The rats in each group were perfused via left ventricular puncture with cold saline (4°C), followed by 4% neutral-buffered formalin, and then a 1.0-cm thoracic spinal cord segment was removed at 2 h after perfusion fixation (aneurysm clip caused the center of the injured spinal cords at least 2.2 mm, because the wide of the aneurysm clip baldes were 2.7 mm). Every tissue sample from the first stage was divided into two parts: about 50% was used for electron microscopy and stored in 2.5% glutaraldehyde and then placed for 24 h at 4°C; and the remainder was used for light microscopy, terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) and CD68 staining, and was stored overnight in 4% neutral-buffered formalin. After fixation, transverse sections of the spinal cord at the T10 level were embedded in paraffin, cut into 4-μm-thick sections, and stained with hematoxylin and eosin. Ischemic lesions were evaluated in a blinded manner by using light microscopy (Chongqing Optec Instrument Co., Ltd., China). The numbers of ischemic neurons with histological characteristics of abnormal structures [intrastoplasmic vacuolization, shrinkage and hyperdensity of the nuclei (pyknosis), perineuronal large halo, eosinophilic or dark, shrunken cytoplasm with indistinct neuronal processes] were counted in each section in four nonoverlapping high-power fields (HPFs) (at 400×magnification) in the gray matter (Grasso et al., 2002). Neuronal injury was evaluated using four randomly selected sections from each rat. Sections were assessed according to the presence and relative abundance of ischemic neurons in the ventral gray matter: mild injury (5% of motor neurons ischemic); moderate injury (5–20% neurons ischemic); and severe injury (20% of neurons affected) (Celik et al., 2002). Glial cells were also counted in each section in four nonoverlapping HPFs (at 400×magnification) in the gray matter. In addition, the samples from the second stage were observed by using LFB staining.

2.6 TUNEL staining

Paraffin-embedded tissue samples of the spinal cord were sectioned at 4-μm thickness with a microtome (RM2016; Leica, Mannheim, Germany) and four sections per rat were selected from serial sections in the same rat, including the 5th, 10th, 15th and 20th sections, which were scored for apoptotic cells by TUNEL. After deparaffinization and hydration, sections were treated with 20 μg/ml proteinase K (Sangon Biotech Co. Ltd., Shanghai, China) in 0.1 M Tris buffer (pH 7.8) at 37°C for 20 min to strip excess protein from the tissue sections. Sections were washed twice for 5 min in phosphate-buffered saline. An In Situ Cell Death Detection Kit POD (Roche Diagnostics, Basle, Swit) was used according to the manufacturer’s protocol. The sections were incubated with Cy3-labeled Goat Anti-Rabbit IgG(H + L) (1:200; Abcam, Cambridge, UK) at 37°C for 30 min. Finally, the sections were counterstained with 2-(4-amidinophenyl)-6-indolecarbamidine dihydrochloride (DAPI; Beyotime Biotech, Jiangsu, China) solution at room temperature for 10 min. Every section was studied under an inverted fluorescence microscope (DMI6000B; Leica). The total number of TUNEL-positive cells was determined by three people blinded to the treatment, including a pathologist not participating in this study. TUNEL-positive cells were counted in the sections through the epicenter of the injury. The extent of spinal damage was evaluated by apoptotic index, which was the average percentage of TUNEL-positive cells in each section counted in five nonoverlapping HPFs (at 400×magnification) in the gray matter.

2.7 CD68 staining

The paraffin-embedded spinal cord tissue samples were sectioned at 4-μm thickness with a microtome (RM2016; Leica) and four sections per rat were selected from serial sections in the same rat, including the 6th, 11th, 16th and 21st sections, which were scored for inflammatory cells by CD68. After deparaffinization and graded alcohol dehydration, sections were boiled (in a 500-W microwave oven) for 25 min in an all-purpose powerful antigen retrieval solution (Beyotime Biotech). Endogenous peroxidase was inhibited with 3% H₂O₂ in methanol for 1 h. Samples were incubated with 5% bovine serum albumin (Sigma, St Louis, MO, USA) to block non-specific binding of immunoglobulins at room temperature for 20 min, and incubated with the primary mouse anti-rat monoclonal antibody CD68 (1:500; Abcam) for reactive macrophages or microglia overnight at 4°C. Sections were incubated with Cy3-labeled Goat Anti-Rabbit IgG(H + L) (1:200; Abcam) at 37°C for 30 min. Every section was studied under an inverted fluorescence microscope (DMI6000B; Leica). The total number of CD68-positive cells was determined by three people blinded to the treatment, including a pathologist not participating in this study. The numbers of positively stained cells were counted in the sections through the epicenter of the injury. The degree of SCI was evaluated by inflammatory index, which was the average percentage of CD68-positive cells in each section counted in five nonoverlapping HPFs (at 400×magnification) in the gray matter.

2.8 Electron microscopy

The tissues used for transmission electron microscopy were obtained from the injury site in 1-mm samples as described by Kaptanoglu et al. (2002). First, the samples were kept in 2.5% phosphate-buffered glutaraldehyde and 2% paraformaldehyde solution for 24 h, post-fixed with phosphate-buffered 1% osmium tetroxide for 1 h, and dehydrated in a graded series of alcohol. Then, the samples were embedded in araldite. After araldite embedding, 1–2-μm semithin sections were obtained with an LKB NOVA ultratome (Bormma, Sweden), stained with toluidine blue, and observed using a light microscope. Finally, the same ultratome was used to obtain 60–90-nm-thick sections, which were contrast stained with uranyl acetate and lead citrate, and imaged using a Tecnai G2 Spirit (FEI Company, Hillsboro, OR, USA) transmission electron microscope. According to an ultrastructural scoring system for SCI (Kaptanoglu et al., 2002), the samples were evaluated by two independent observers blinded to this experiment. The detailed assessment criteria are shown in Table 1 (Kaptanoglu et al., 2002).

2.9 LFB staining

At 28 days post-injury, rats were sacrificed, and the tissues were dissected and processed as stated previously. For myelin quantification, LFB staining was performed on 10-μm-thick cross-sections. Paraffin sections were processed according to the myelin sheath Fast Blue staining kit protocol (Shanghai biological technology development co., LTD, China). Sections were stained at room temperature in 0.1% LFB in acidified 95% ethanol for 20 h. Differentiation and counterstaining were carried out with 0.01% Li₂CO₃. For quantitative analysis of the total lesion volume in whole spinal cords, sections were observed using an Olympus BX-51 light microscope (Olympus, Tokyo, Japan) connected with a computer screen and then analyzed with Image-Pro Plus 6.0 analysis software (Media Cybernetics, Silver Spring, MD, USA). An unbiased estimation of the percentage of lesion cavity (or volume of demyelination) was calculated as previously described (Yin et al., 2013; Yu et al., 2008). The percentage total volume of the injured tissue was calculated by dividing the total lesion volume by the total spinal cord volume (Yin et al., 2013; Yu et al., 2008).

2.10 Statistical analysis

SPSS version 19.0 (SPSS Inc., Chicago, IL, USA) was used for statistical analysis. The data were expressed as means ± standard deviation (SD) except for histopathological neuronal ischemic grade. Statistically significant differences between data were evaluated by one-way analysis of variance (ANOVA) followed by post-hoc least significant difference (LSD) test. For comparison between BBB scores of groups at different time points, the data were analyzed by two-way ANOVA with repeated measures and post-hoc LSD test. For nonparametric data, the Kruskal–Wallis test was used for comparing differences between groups. When ANOVA showed significant differences, the post-hoc multiple comparison was performed using the Nemenyi test. P < 0.05 was accepted as statistically significant.

3 Results

3.1 Locomotor recovery

The BBB score of every rat was 21 before surgery. Immediately after injury, the BBB score of every rat was approximately 0, which meant that the spinal cord crush injury model was successful.

In the first stage, BBB scores of different groups are shown in Fig. 1. No significant differences were observed in the BBB scores among the four groups at 6 h and 1 day post-injury (p_6h = 0.601; p_1d = 0.113). The rhEPO-treated groups showed significantly improved motor function as compared with Group A on days 2 (P < 0.001), 3 (P < 0.001) and 4 (P < 0.001). On day 2, although the mean BBB score of Group D was higher than that of Groups B and C, it was only significant between Groups B and D (P = 0.024). This means that 5,000 U/kg rhEPO promoted neurological recovery in the early stage. On days 3 and 4, Group D had significantly improved motor function compared with Groups B (P₃ = 0.003; P₄ < 0.001) and C (P₃ = 0.021; P₄ < 0.001). No significant differences were found at days 3 and 4 between Groups B and C. The results suggested that rhEPO significantly improved motor function after injury and the most significant improvement was observed after administration of 5,000 U/kg rhEPO.

In the second stage, BBB scores of different groups are shown in Fig. 2. There was a significant difference in the mean BBB score among the groups (P < 0.001) at 4, 7, 14, 21 and 28 days post-injury, and the BBB score in the different groups gradually increased over time. Importantly, the rhEPO-5,000U-treated group had significantly improved motor function compared with Group Aa on days 4 (P < 0.001), 7 (P < 0.001), 14 (P < 0.001), 21 (P < 0.001) and 28 (P < 0.001); Group Bb on days 4 (P = 0.001), 7 (P < 0.001), 14 (P = 0.01), 21 (P = 0.002) and 28 (P = 0.001); and Group Cc on days 4 (P = 0.003), 7 (P < 0.001), 14 (P = 0.025), 21 (P = 0.021) and 28 (P = 0.007). We also found that Groups Bb and Cc had significantly improved motor function as compared with Group Aa at 4, 7, 14, 21 and 28 days post-injury (P < 0.05). However, there was no significant difference between Groups Bb and Cc (P₄ = 0.498; P₇ = 1; P₁₄ = 0.690; P₂₁ = 0.275; P₂₈ = 0.379). We also found that rhEPO significantly improved motor function after injury and the most significant improvement was observed after administration of 5,000 U/kg rhEPO.

The results of the two stages were similar, and the rhEPO-5,000U-treated group had the most significant improvement in locomotor function.

3.2 Histopathological evaluation

Various degrees of neuronal ischemia are observed in the experimental and control groups (Fig. 3A) and a high frequency of necrotic neurons was found in Group A, in which most neurons were ischemic. A large number of inflammatory cells were also found in Group A (Fig. 3A1), whereas only a few were observed in the experimental groups (Fig. 3A2–A4). When Groups B–D were compared with Group A, there was a significant decrease in ischemic damage (P_B < 0.05; P_C < 0.05; P_D < 0.01). The degree of neuronal ischemia in Group D was significantly lower than that in the other experimental groups (P_B < 0.05; P_C < 0.05). However, there was no significant difference between Groups B and C. Samples from Group D had only mild histological evidence of injury (Fig. 3A4), whereas moderate histological evidence of injury was often observed in Groups B and C (Fig. 3A2–A3). The differences in ischemic neurons in all the groups are shown in Table 2.

The numbers of glial cells and ischemic/necrotic neurons are shown in Fig. 3B and 3C, respectively. The rhEPO-treated groups showed significant inhibition of glial cell proliferation as compared with control group (P < 0.05). When Group C was compared with Groups B and D, there were no significant differences (P_b = 0.484; P_d = 0.109). Group D showed significantly greater inhibition of glial cell proliferation as compared with Group B (P = 0.025). In terms of ischemic/necrotic neurons, the rhEPO-treated groups showed preferential protection of neurons as compared with the control group (P < 0.05). When Group D was compared with Groups B and C, there was only a significant difference between groups D and B (P = 0.001). This indicates that rhEPO-5,000U-treated group preferentially protected the neurons, therefore promoting locomotor recovery. However, there was no significant difference between Groups B and C (P = 0.089).

In summary, our results suggested that rhEPO significantly improved neural damage in rats after SCI, and the most significant histopathological improvement was observed in Group D. Groups B and C showed similar protection of neurons and inhibition of glial proliferation.

3.3 Apoptotic index

Cells in Group A had a higher apoptotic index than the other groups had (Fig. 4). In Groups B–D, the apoptotic index was significantly decreased compared with that in Group A (P_B < 0.05; P_C < 0.01; P_D < 0.01). The apoptotic index in Group B did not differ significantly from that in Groups C and D. The apoptotic index in Group C did not differ significantly from that in Group D, although the mean index in Group C was higher than that in Group D. The results suggested that rhEPO treatment following SCI reduced apoptotic cell death in the injured spinal cord tissue.

3.4 Inflammatory index

Cells in Group A had a higher inflammatory index than the other groups had (Fig. 5). In Group A, the inflammatory index was significantly increased compared with that in Group B (P < 0.05), C (P < 0.05) and D (P < 0.05). In Group B, the inflammatory index did not differ significantly from that in Groups C and D. The inflammatory index in Group C was not significantly higher than that in Group D. The results suggested that administration of rhEPO following SCI inhibited inflammation in the injured spinal cord tissue.

3.5 Ultrastructural findings

In Group A, there was obvious ultrastructural damage of the spinal cord. Evaluation of ultrastructural changes was based on general neural score (GNS) and subcellular changes. Electron microscopic images of the groups are shown in Figs. 6 –9.

GNS: the score for Group A was significantly different from that of Groups B–D (P < 0.001). All treatment groups showed significant neuroprotection but there was no significant difference between Groups B, C and D. Results of ultrastructural scoring are shown in Fig. 10 according to GNS of the groups.

Intracytoplasmic edema: in the control group, intracytoplasmic edema was increased to significant levels, when compared with Groups B–D (P < 0.001). The rhEPO-treated groups showed improved intracytoplasmic edema but there was no significant difference among the groups.

Nucleus: in Group A, nuclear damage was significantly increased compared with Groups B–D (P < 0.001). There was no significant difference among Groups B, C and D, which meant that the rhEPO-treated groups had similar nuclear protection.

Axonal myelin: in Group A, nuclear damage was significantly increased compared with that in Groups B–D (P < 0.001). The rhEPO-treated groups showed similar axonal myelin protection, but there was no significant difference among the three groups.

Axonal score: in Group A, axonal scores increased significantly compared to those in Groups B–D (P < 0.001). The best results were obtained from Groups B–D, but there was no significant difference among the three groups.

Mitochondrion score: in Group A, mitochondrion scores increased significantly compared to those in Groups B–D (P < 0.001). The best results were obtained from Groups B–D but there was no significant difference among the three groups.

3.6 LFB staining in the second stage

LFB staining and volume of areas of demyelination are shown in Fig. 11. Pathology in tissue cross-sections was identified by the following characteristics: areas of normal myelin appeared bright blue, and demyelinated tissue appeared blanched. The volume of areas of demyelination (% of total volume) in the different groups was as follows: Group Aa (22.77 ± 1.78), Group Bb (19.51 ± 2.88), Group Cc (17.20 ± 2.77) and Group Dd (14.86 ± 1.37). There was a large unstained area at the epicenter of the spinal cord lesion in the control group. There was a significant deference among the groups for the volume of areas of demyelination (P < 0.001). rhEPO treatment significantly reduced the volume of the area of demyelination as compared with the control group (P_Bb < 0.023; P_Cc < 0.001; P_Dd < 0.001). There was no significant difference between Groups Bb and Cc (P = 0.096) and Groups Dd and Cc (P = 0.093), although the volume of demyelination in Group Dd was lower than that of Group Cc. However, there was a significant difference between Groups Bb and Dd (P = 0.002), which meant that 5,000 U/kg rhEPO significantly decreased the volume of the area of demyelination compared with 3,000 U/kg rhEPO. Overall, rhEPO significantly reduced demyelination, and 4,000 and 5,000 U/kg rhEPO showed similar protection of myelin, and similar results were found between the rhEPO-4,000U and rhEPO-3,000U groups. However, the rhEPO-5,000U-treated group showed greater improvement of demyelination than the rhEPO-3000-treated group showed. Thus, the best myelin protection was observed in Group Dd.

4 Discussion

EPO is a glycoprotein cytokine produced mainly by the fetal liver and adult kidney, which is involved in regulation of red blood cell production. EPO exerts its effects by interaction with EPO receptor (EPO-R), which belongs to the type I family of single-transmembrane cytokine receptors (Kasper, 2003) and is localized on capillaries within the white matter and on bodies and proximal dendrites of motor neurons of the ventral horn (Arishima et al., 2006; Sekiguchi et al., 2003). Expression of EPO and EPO-R in the nervous system is regulated by tissue hypoxia, and proinflammatory cytokines also induce EPO-R expression, but suppress endogenous EPO production (Nagai et al., 2001). This means that production of EPO-R rapidly increases after SCI, whereas endogenous EPO expression is inhibited. EPO also enhances differentiation of neural progenitor cells expressing a high level of EPO-R toward the oligodendrocyte-lineage cells, through activation of cyclin E and Janus kinase 2 pathways (Cho et al., 2012). Therefore, high expression of EPO-R in SCI provides an opportunity for application of exogenous EPO for SCI.

Although the pathogenesis of SCI remains unclear, it is usually divided into the acute phase and secondary phase. The acute phase is caused by direction disruption of tissues. It frequently results in limited neuronal death surrounding the lesion epicenter, and damage to axons and blood vessels at the site of injury that causes vasoconstriction, hemorrhage and ischemia (Tator, 1995). The secondary phase involves a cascade of secondary damage, including fluid–electrolyte imbalance, calcium-mediated cellular injury, regional blood flow alterations, free radical generation, glutamate-induced excitoxicity, disturbances in mitochondrial function, proinflammatory cytokine production, apoptotic cell death, and lipid peroxidation. All above the pathogenesis can cause nerve cell injury or death, leading to neurological dysfunction. Currently, rhEPO is regarded as a promising agent for SCI and many studies have evaluated the effects of rhEPO administration on rat spinal cords, and satisfactory results have been achieved in different experimental models (Cerri et al., 2012; Cetin et al., 2006; Jin et al., 2014; Kaptanoglu et al., 2004). The present study showed that rhEPO significantly improved motor function as compared to the control group. Although the animals received rhEPO treatment at 2 h after injury in this study, our results were similar to previous studies in which animals received rhEPO within 1 h after SCI (Cerri et al., 2012; Cetin et al., 2006; Jin et al., 2014; Kaptanoglu et al., 2004). The most significant improvement in locomotor function was observed in the rhEPO-5,000U group, as measured by the BBB scale. Jin et al. (2014) reported that treatment of SCI with 5,000 U/kg rhEPO decreased the severity of the locomotor deficit, spinal cord edema, and apoptosis. They also demonstrated that rhEPO induced a cytoprotective response mediated by Nrf-2 (nuclear factor erythroid 2-related factor 2), and that the Nrf2 signaling pathway had antioxidant and anti-inflammatory roles in the injured spinal cord. Using an electrophysiological test to evaluate the degree of ascending and descending transmission through the damaged cord after SCI, Cerri et al. (2012) found that spinal cord pathways improved by rhEPO contributed to the improvement in transmission. Nerve growth factor (NGF) and platelet-derived growth factor (PDGF)-B can act as protective agents through rescuing neuronal cells after injury of specific neuronal pathways (Deller et al., 2006). A recent study reported that administration of rhEPO markedly increased NGF expression in a rat model of traumatic SCI (Fumagalli et al., 2008). Hong et al. (2011) have reported that rhEPO repairs injured tissue and recovers neurological function via increased expression of PDGF-B in SCI. This means that rhEPO decreases the extent of SCI through upregulating expression of NGF and PDGF-B. rhEPO can exert neuroprotection in SCI through a series of mechanisms that promote the functional rehabilitation of nerve injury.

Lipid peroxidation plays an important role in the secondary phase of SCI. Increased lipid peroxidation causes disruption of endothelial cell membranes and control of hemorrhage after trauma is worsened after SCI (Yazihan et al., 2008). After lipid peroxidation, lipid hydrolysis leads to prostaglandin and leukotriene formation. Increases in leukotrienes and prostaglandins cause inflammation and increase vascular permeability that aggravates spinal cord edema and ischemia. Inflammatory cells such as neutrophils, macrophages, and resident microglia are attracted by cell death and exacerbate injury through releasing proinflammatory cytokines at the site of injury (Norenberg et al., 2004). Kaptanoglu et al. (2004) reported that 5,000 IU/kg EPO markedly inhibited lipid peroxidation after acute SCI in rats. Previous studies have proposed that rhEPO administration inhibits inflammation (Chen et al., 2007) and proinflammatory cytokines, reduces microglial infiltration, attenuates scar formation, and sustains neurological improvement following SCI (Gorio et al., 2005). We found similar results, and the rhEPO-treated groups had significantly decreased inflammation as compared with the control group. It is possible that rhEPO inhibition of inflammation is not dose related, because there was no significant difference among the three doses of rhEPO. Recent studies have suggested that rhEPO alleviates spinal cord edema (Jin et al., 2014; Vitellaro-Zuccarello et al., 2008) by regulating expression of aquaporin 4 (astrocytic water channel) (Vitellaro-Zuccarello et al., 2008). Decreased spinal cord edema reduces vascular compression and improves microcirculation, which in turn, increase blood flow and tissue oxygenation, further promoting regeneration of the spinal cord. Administration of rhEPO prevents the apoptosis between the acute and subacute stages of SCI (Arishima et al., 2006). Our data confirmed the above conclusion, because we found that rhEPO significantly decreased apoptotic index at 4 days post-injury in the experimental groups. rhEPO also regulates blood vessel function and repairs severe blood vessel damage after SCI (Cetin et al., 2006). It has been demonstrated that rhEPO enhances VEGF secretion in neural progenitor cells by activation of the PI3K/Akt and ERK1/2 signaling pathways. Neural progenitor cells treated with rhEPO increase VEGF receptor 2 expression in cerebral endothelial cells, which along with VEGF secreted by neural progenitor cells, promotes angiogenesis (Wang et al., 2008). Thus, it is possible that rhEPO protects blood vessels in SCI through repairing the endothelial cells and promoting angiogenesis by the aforementioned mechanism.

In this study, we administered 3,000, 4,000 and 5,000 U/kg rhEPO intraperitoneally after causing crush injury to the spinal cord by applying an aneurysm clip for about 1 min. Our data showed that rhEPO significantly decreased the ischemic damage histopathologically as compared to that in the control group. When the rhEPO-5,000U group was compared with the rhEPO-3,000U and rhEPO-4,000U groups, histopathologically striking improvements were observed. However, rhEPO at 3,000 and 4,000 U/kg exerted similar histopathologically improvements. To confirm the above conclusion, glial cells and neurons were counted, and we found that rhEPO markedly inhibited glial cell proliferation and protected neurons. We also found that 5,000 U/kg rhEPO preferentially inhibited glial cell proliferation and protected neurons. Celik et al. (2002) reported that different doses of EPO (350, 800 and 1,000 U/kg) improved histopathology of rats with neural damage after SCI. Similarly, Cetin et al. found that the most significant histopathological improvements were observed after administration of MPS+rhEPO 3,000 U and rhEPO 3,000 U (Cetin et al., 2006). Our results were similar to the above two studies. However, one difference was that we found that delayed treatment of SCI with rhEPO improved histopathology of rats, and the most significant improvements were observed in the rhEPO-5,000U group.

In this pilot study, we used transmission electron microscopy to observe the ultrastructural changes in the nerve cells on day 4 after SCI. We found that the control group had significantly increased ultrastructural damage of the spinal cord as compared with the rhEPO-treated groups, but there was no significant difference among the three experimental groups. The rhEPO treatment groups also had improved intracytoplasmic edema and less damage to the nuclei, axonal myelin, axons and mitochondria. The ultrastructural results confirmed the neuroprotective effects of rhEPO on SCI. A similar study also used electron microscopy to investigate the effectiveness of rhEPO in attenuating the severity of experimental SCI (Kaptanoglu et al., 2004). They found that 1,000 and 5,000 U/kg rhEPO exerted similar ultrastructural neuroprotection in the acute stage (SCI rats were sacrificed at 2 h after injury). Our study suggested that 3,000, 4,000 and 5,000 U/kg rhEPO had similar ultrastructural neuroprotection in the subacute stage. Therefore, it is possible that administration of rhEPO protects the ultrastructure of the spinal cord between the acute and subacute stages.

The axons of many nerve fibers are wrapped in a white insulating sheath called myelin, and myelination is performed by oligodendrocytes in the central nervous system (CNS). Previous studies have found that demyelination (loss of myelin) and dysmyelination (abnormal myelination) are important contributors to behavioral deficits associated with CNS disorders (Blight, 1983; Bunge et al., 1993), and persistent demyelination can lead to further axonal loss (Grigoriadis et al., 2004). Cho et al. (2012) found that EPO promoted oligodendrogenesis and myelin repair following lysolecithin-induced injury in spinal cord slice culture. EPO also induced expression of myelin genes in oligodendrocytes through the classical EPO-R, and that study demonstrated that EPO-R mediated nerve repair through combining with EPO (Cervellini et al., 2013). Our study showed that rhEPO treatment obviously attenuated demyelination as compared with the control group. Our results are similar to the previous study of Cho et al., (2012). In our study, the best myelin protection was found after administration of 5,000 U/kg rhEPO, which was in accordance with the BBB score of SCI rats at 28 days. Our study confirmed that rhEPO attenuated demyelination in vivo in SCI rats, so rhEPO can improve behavioral deficits. It is possible that rhEPO attenuates demyelination in vivo through inducing expression of myelin genes in oligodendrocytes.

We reported that delayed treatment of acute thoracic SCI with rhEPO had neuroprotective effects when administrated 2 h after injury, although Mann and coworkers reported that delayed treatment of SCI with 5,000 U/kg rhEPO lacked neuroprotective efficacy (Mann et al., 2008). The time of rhEPO differed between the two studies. In our study, rats received rhEPO treatment at 2 h, 1 day and 3 days during the 4 days of the first stage, and in the second stage, they received rhEPO at 2 h, and days 1, 3, 15 and 22 during the 28 days study. In contrast, in the previous study, rhEPO was administered only at 1 h post-injury over 42 days. It is well known that the effectiveness of the drug is correlated with dose and frequency of administration. Other factors may also have interfered with the results, including the degree of SCI, sex, body weight of rats, etc. So a further study is necessary to establish whether these factors interfered with the results. We comprehensively and systematically evaluated the effects of rhEPO administration on rat acute spinal cord injury, through assessment of locomotor recovery, histopathology, apoptotic index, inflammatory index, ultrastructure, and volume of areas of demyelination. Our data confirmed that administration of rhEPO reduced apoptosis, regulated inflammation, improved histopathology and motor function, and promoted myelin repair and regeneration of the spinal cord. We first used transmission electron microscopy to observe the ultrastructure of the SCI after administration of rhEPO on day 4 post-injury. Currently, the therapeutic time window to achieve the neuroprotective efficacy of rhEPO remains unclear. This issue will be explored in our future research.

In conclusion, our study suggests that delayed treatment of acute thoracic SCI with rhEPO exerts neuroprotection. The most significant locomotor functional and histopathological improvements and the best myelin protection were observed after administration of 5,000 U/kg rhEPO, but 3,000, 4,000 and 5,000 U/kg rhEPO showed similar ultrastructural neuroprotection, and they had similar inhibition of apoptosis and regulation of inflammation. In brief, delayed administration of different doses of EPO can reduce apoptosis and inflammation, and promote myelin repair and functional recovery following spinal cord compressive injury in rats.

Footnotes

Acknowledgments

This work was supported by Science and Technology Planning Project of Guangdong Province (2011B031300005) and Guangzhou Haizhu District(2011-YL-01). We thank Dr. Jia-Gen Chen and Dr. Yan-Yang Jia for the technical assistance.

References

Arishima

, Setoguchi

, Yamaura

, Yone

, & Komiya

(2006). Preventive effect of erythropoietin on spinal cord cell apoptosis following acute traumatic injury in rats. Spine (Phila Pa 1976), 31, 2432–2438.

Basso

D.M.

, Beattie

M.S.

, & Bresnahan

J.C.

(1995). A sensitive and reliable locomotor rating scale for open field testing in rats. Journal of neurotrauma, 12(1), 1–21.

Blight

A.R.

(1983). Cellular morphology of chronic spinal cord injury in the cat: Analysis of myelinated axons by line-sampling. Neuroscience, 10, 521–543.

Boran

B.O.

, Colak

, & Kutlay

(2005). Erythropoietin enhances neurological recovery after experimental spinal cord injury. Restor Neurol Neurosci, 23, 341–345.

Bunge

R.P.

, Puckett

W.R.

, Becerra

J.L.

, Marcillo

, & Quencer

R.M.

(1993). Observations on the pathology of human spinal cord injury. A review and classification of 22 new cases with details from a case of chronic cord compression with extensive focal demyelination. Adv Neurol, 59, 75–89.

Kasper

(2003). Erythropoietin. In: Thomson

AW.

, Lotze

MT.

(eds) The cytokine handbook, I, 4th edn.Elsevier, London, pp. 149–166.

Tator

C.H.

(1995). Update on the pathophysiology and pathology of acute spinal cord injury. Brain Pathology, 5(4), 407–413.

Celik

, Gokmen

, Erbayraktar

, Akhisaroglu

, Konakc

, Ulukus

, Genc

, Sagiroglu

, Cerami

, & Brines

(2002). Erythropoietin prevents motor neuron apoptosis and neurologic disability in experimental spinal cord ischemic injury. Proc Natl Acad Sci U S A, 99, 2258–2263.

Cerri

, Montagna

, Madaschi

, Merli

, Borroni

, Baldissera

, & Gorio

(2012). Erythropoietin effect on sensorimotor recovery after contusive spinal cord injury: An electrophysiological study in rats. Neuroscience, 219, 290–301.

10.

Cervellini

, Annenkov

, Brenton

, Chernajovsky

, Ghezzi

, & Mengozzi

(2013). Erythropoietin (EPO) increases myelin gene expression in CG oligodendrocyte cells through the classical EPO receptor. Molecular Medicine, 19, 223–229.

11.

Cetin

, Nas

, Buyukbayram

, Ceviz

, & Olmez

(2006). The effects of systemically administered methylprednisolone and recombinant human erythropoietin after acute spinal cord compressive injury in rats. European Spine Journal, 15, 1539–1544.

12.

Chen

, Shi

J.X.

, Hang

C.H.

, Xie

, Liu

, & Liu

(2007). Inhibitory effect on cerebral inflammatory agents that accompany traumatic brain injury in a rat model: A potential neuroprotective mechanism of recombinant human erythropoietin (rhEPO). Neuroscience Letters, 425, 177–182.

13.

Cho

Y.K.

, Kim

, Park

, Sim

J.H.

, Won

Y.J.

, Hwang

C.H.

, Yoo

J.Y.

, & Hong

H.N.

(2012). Erythropoietin promotes oligodendrogenesis and myelin repair following lysolecithin-induced injury in spinal cord slice culture. Biochem Biophys Res Commun, 417, 753–759.

14.

Deller

, Haas

C.A.

, Freiman

T.M.

, Phinney

, Jucker

, & Frotscher

(2006). Lesion-induced axonal sprouting in the central nervous system. Advances in Experimental Medicine and Biology, 557, 101–121.

15.

Eckardt

K.U.

, Koury

S.T.

, Tan

C.C.

, Schuster

S.J.

, Kaissling

, Ratcliffe

P.J.

, & Kurtz

(1993). Distribution of erythropoietin producing cells in rat kidneys during hypoxic hypoxia. Kidney International, 43, 815–823.

16.

Fumagalli

, Madaschi

, Brenna

, Caffino

, Marfia

, Di Giulio

A.M.

, Racagni

, & Gorio

(2008). Single exposure to erythropoietin modulates Nerve Growth Factor expression in the spinal cord following traumatic injury: Comparison with methylprednisolone. European Journal of Pharmacology, 578, 19–27.

17.

Grasso

, Buemi

, Alafaci

, Sfacteria

, Passalacqua

, Sturiale

, Calapai

, De Vico

, Piedimonte

, Salpietro

F.M.

, & Tomasello

(2002). Beneficial effects of systemic administration of recombinant human erythropoietin in rabbits subjected to subarachnoid hemorrhage. Proc Natl Acad Sci U S A, 99, 5627–5631.

18.

Gorio

, Madaschi

, Di Stefano

, Carelli

, Di Giulio

A.M.

, De Biasi

, Coleman

, Cerami

, & Brines

(2005). Methylprednisolone neutralizes the beneficial effects of erythropoietin in experimental spinal cord injury. Proc Natl Acad Sci U S A, 102, 16379–16384.

19.

Grigoriadis

, Ben-Hur

, Karussis

, & Milonas

(2004). Axonal damage in multiple sclerosis: A complex issue in a complex disease. Clin Neurol Neurosurg, 106, 211–217.

20.

Hall

E.D.

, & Braughler

J.M.

(1986). Role of lipid peroxidation in post-traumatic spinal cord degeneration: A review. Cent Nerv Syst Trauma, 3, 281–294.

21.

Hong

, Hong

, Chen

, Wang

, & Hong

(2011). Investigation of the protective effect of erythropoietin on spinal cord injury in rats. Experimental and Therapeutic Medicine, 2, 837–841.

22.

Jin

, Ming

, Hou

, Zhu

, Yuan

, Wang

, Ni

, Jiang

, Wang

, & Liang

(2014). Protective effects of erythropoietin in traumatic spinal cord injury by inducing the Nrf2 signaling pathway activation. J Trauma Acute Care Surg, 76, 1228–1234.

23.

Kaptanoglu

, Palaoglu

, Surucu

H.S.

, Hayran

, & Beskonakli

(2002). Ultrastructural scoring of graded acute spinal cord injury in the rat. Journal of Neurosurgery, 97, 49–56.

24.

Kaptanoglu

, Solaroglu

, Okutan

, Surucu

H.S.

, Akbiyik

, & Beskonakli

(2004). Erythropoietin exerts neuroprotection after acute spinal cord injury in rats: Effect on lipid peroxidation and early ultrastructural findings. Neurosurgical Review, 27, 113–120.

25.

Kawakami

, Sekiguchi

, Sato

, Kozaki

, & Takahashi

(2001). Erythropoietin receptor-mediated inhibition of exocytotic glutamate release confers neuroprotection during chemical ischemia. Journal of Biological Chemistry, 276, 39469–39475.

26.

Mann

, Lee

J.H.

, Liu

, Stammers

A.M.

, Sohn

H.M.

, Tetzlaff

, & Kwon

B.K.

(2008). Delayed treatment of spinal cord injury with erythropoietin or darbepoetin–a lack of neuroprotective efficacy in a contusion model of cord injury. Experimental Neurology, 211, 34–40.

27.

Min

, Song

, Lee

J.H.

, Kang

M.S.

, Jang

S.J.

, Kim

S.H.

, & Kim

(2013). Allogenic umbilical cord blood therapy combined with erythropoietin for patients with severe traumatic brain injury: Three case reports. Restor Neurol Neurosci, 31, 397–410.

28.

Nagai

, Nakagawa

, Choi

H.B.

, Hatori

, Kobayashi

, & Kim

S.U.

(2001). Erythropoietin and erythropoietin receptors in human CNS neurons, astrocytes, microglia, and oligodendrocytes grown in culture. J Neuropathol Exp Neurol, 60, 386–392.

29.

Norenberg

M.D.

, Smith

, & Marcillo

(2004). The pathology of human spinal cord injury: Defining the problems. J Neurotrauma, 21, 429–440.

30.

Obara

, Suzuki

, Kim

, Nagasawa

, Imagawa

, & Yamamoto

(2008). Repression via the GATA box is essential for tissue-specific erythropoietin gene expression. Blood, 111(10), 5223–5232.

31.

Ozturk

, Demirbilek

, Kadir

B.A.

, Saricicek

, Gulec

, Akyol

, & Ozcan

E.M.

(2005). Antioxidant properties of propofol and erythropoietin after closed head injury in rats. Prog Neuropsychopharmacol Biol Psychiatry, 29, 922–927.

32.

Pinzon

, Marcillo

, Pabon

, Bramlett

H.M.

, Bunge

M.B.

, & Dietrich

W.D.

(2008). A re-assessment of erythropoietin as a neuroprotective agent following rat spinal cord compression or contusion injury. Experimental Neurology, 213, 129–136.

33.

Priestley

J.V.

, Michael-Titus

A.T.

, & Tetzlaff

(2012). Limiting spinal cord injury by pharmacological intervention. Handb Clin Neurol, 109, 463–484.

34.

Profyris

, Cheema

S.S.

, Zang

, Azari

M.F.

, Boyle

, & Petratos

(2004). Degenerative and regenerative mechanisms governing spinal cord injury. Neurobiology of Disease, 15, 415–436.

35.

Sekiguchi

, Kikuchi

, Myers

R.R.

, & Campana

W.M.

(1976). ISSLS prize winner: Erythropoietin inhibits spinal neuronal apoptosis and pain following nerve root crush. Spine (Phila Pa), 28, 2577–2584.

36.

Solaroglu

, Solaroglu

, Kaptanoglu

, Dede

, Haberal

, Beskonakli

, & Kilinc

(2003). Erythropoietin prevents ischemia-reperfusion from inducing oxidative damage in fetal rat brain. Childs Nerv Syst, 19, 19–22.

37.

Vitellaro-Zuccarello

, Mazzetti

, Madaschi

, Bosisio

, Fontana

, Gorio

, & De Biasi

(2008). Chronic erythropoietin-mediated effects on the expression of astrocyte markers in a rat model of contusive spinal cord injury. Neuroscience, 151, 452–466.

38.

Wang

, Chopp

, Gregg

S.R.

, Zhang

R.L.

, Teng

, Jiang

, Feng

, & Zhang

Z.G.

(2008). Neural progenitor cells treated with EPO induce angiogenesis through the production of VEGF. J Cereb Blood Flow Metab, 28, 1361–1368.

39.

Xiong

, Chen

, Yu

, Liu

, Zeng

, & Li

(2011). Neuroprotection of erythropoietin and methylprednisolone against spinal cord ischemia-reperfusion injury. J Huazhong Univ Sci Technolog Med Sci, 31, 652–656.

40.

Yazihan

, Uzuner

, Salman

, Vural

, Koken

, & Arslantas

(2008). Erythropoietin improves oxidative stress following spinal cord trauma in rats. of the Care of the Injured. Injury-International Journal, 39, 1408–1413.

41.

Yin

, Sun

, Li

, Zhang

, Cui

, Deng

, Xie

, Xiang

, & Zou

(2013). Effects of combining methylprednisolone with rolipram on functional recovery in adult rats following spinal cord injury. Neurochemistry International, 62, 903–912.

42.

, Huang

, Zou

, Yu

, Wang

, Xu

, Liu

, Xu

X.M.

, & Lu

P.H.

(2008). Immunization with recombinant Nogo-66 receptor (NgR) promotes axonal regeneration and recovery of function after spinal cord injury in rats. Neurobiology of Disease, 32, 535–542.

43.

Zhang

, Zhang

, Ji

, Pang

, Ju

, Zhang

, Li

, Zhang

, He

, Gan

W.B.

, & Tang

(2015). Two-photon microscopy as a tool to investigate the therapeutic time window of methylprednisolone in a mouse spinal cord injury model. Restor Neurol Neurosci.