Inhibiting High Mobility Group Box-1 Reduces Early Spinal Cord Edema and Attenuates Astrocyte Activation and Aquaporin-4 Expression after Spinal Cord Injury in Rats

Abstract

High mobility group box-1 (HMGB1) could function as an early trigger for pro-inflammatory activation after spinal cord injury (SCI). Spinal cord edema contributes to inflammatory response mechanisms and a poor clinical prognosis after SCI, for which efficient therapies targeting the specific molecules involved remain limited. This study was designed to evaluate the roles of HMGB1 on the regulation of early spinal cord edema, astrocyte activation, and aquaporin-4 (AQP4) expression in a rat SCI model. Adult female Sprague-Dawley rats underwent laminectomy at T10, and the SCI model was induced by a heavy falling object. After SCI, rats received ethyl pyruvate (EP) or glycyrrhizin (GL) via an intraperitoneal injection to inhibit HMGB1. The effects of HMGB1 inhibition on the early spinal cord edema, astrocyte activation (glial fibrillary acidic protein [GFAP] expression), and AQP4 expression after SCI (12 h–3 days) were analyzed. The results showed that EP or GL effectively inhibited HMGB1 expression in the spinal cord and HMGB1 levels in the serum of SCI rats. HMGB1 inhibition improved motor function, reduced spinal cord water content, and attenuated spinal cord edema in SCI rats. HMGB1 inhibition decreased SCI-associated GFAP and AQP4 overexpression in the spinal cord. Further, HMGB1 inhibition also repressed the activation of the toll-like receptor 4/myeloid differentiation primary response gene 88/nuclear factor-kappa B signaling pathway. These results implicate that HMGB1 inhibition improved locomotor function and reduced early spinal cord edema, which was associated with a downregulation of astrocyte activation (GFAP expression) and AQP4 expression in SCI rats.

Introduction

Spinal cord injury (SCI) is a devastating pathology, because it results frequently in permanent physical disability. It is well established that SCI involves both primary and secondary injuries. In the case of secondary injuries, which include edema, hemorrhage, and inflammation, these are considered to be reversible and are therefore frequently the focus of SCI treatment.^1,2 Spinal cord edema can result ultimately in a reduction in blood flow, which starves the spinal cord tissue of critical nutrients. Thus, spinal cord edema is often associated with poor functional recovery after SCI.^1,3,4 Currently, efficient therapies for SCI and spinal cord edema that target specific molecules remain limited.^5,6

Astrocytes are the most abundant type of glial cells in the central nervous system (CNS) and are known to undergo variable degrees of activation and accumulate the majority of edema fluid intracellularly. This represents a critical early component involved in spinal cord edema and has been demonstrated to generate the driving force for later vasogenic edema.^4,7

–10 As is well known, the expression of glial fibrillary acidic protein (GFAP), the astrocytic marker, could indicate the degree of astrocyte activation.^5,9,11 Interestingly, the expression of aquaporin-4 (AQP4), a spinal cord astrocyte water channel protein, has been shown to play a role in astrocyte swelling and spinal cord edema after spinal traumatic injury.^11
–13

Inflammation responses play an important role in the mediation of SCI cascades and are initiated by the detection of injury-associated molecules in local cells, including microglia and astrocytes.^14
–16 High mobility group box-1 (HMGB1) is a highly conserved nonhistone deoxyribonucleic acid (DNA)-binding protein, that has been defined recently as a key mediator in both inflammation and neuroinflammation.¹⁷ On introduction of pathogens or tissue injury, HMGB1 is secreted actively from reactive astrocytes and microglia and is released passively from necrotic cells into the extracellular milieu in the CNS.^14,18,19 Extracellular HMGB1 induces an inflammatory response in immune-competent cells, neurons, and astrocytes via activation of numerous receptors, such as toll-like receptor-4 (TLR4), TLR2, and receptor of advanced glycation end product (RAGE).^17,20,21 Binding of HMGB1 to these receptors results in the activation of a common signaling pathway that functions to promote activation of nuclear factor-kappa B (NF-κB) transcription factors. These transcription factors regulate numerous genes including pro-inflammatory cytokines and other mediators of inflammatory and immune responses.^2,15,22,23

After SCI, increases in both HMGB1 levels as well as its signaling were observed in the spinal cord. These are thought to play a critical role in the pro-inflammatory cascade that initiates the secondary injury.^{2,14,15,18,19,24} In the brain, it has been shown that HMGB1 levels and its signaling resulted in a significant increase in brain edema and astrocyte swelling after injury and disease.^25

–28 Interestingly, it has been shown that an inhibitor of HMGB1 resulted in a reduction in the brain water content in rats.²⁹ Until now, however, the role of HMGB1 on spinal cord edema and astrocyte activation and swelling after SCI has not yet been completely characterized, and potential mechanisms for its regulation remain unclear.

The purpose of this study was to understand whether HMGB1 is an important candidate in the regulation of early spinal cord edema, astrocyte activation (GFAP expression), and AQP4 expression in a rat SCI model. We studied HMGB1 expression, motor function recovery, spinal cord edema, astrocytes activation, and AQP4 expression levels using a rat SCI model. We observed the effect of HMGB1 inhibition using ethyl pyruvate (EP)²³ or glycyrrhizin (GL)^30,31 on HMGB1 expression, motor function recovery, spinal cord edema, astrocytes activation, and AQP4 expression of SCI rats. In addition, we examined alterations in TLR4, myeloid differentiation primary response gene 88 (MyD88), and nucleus NF-κB levels after SCI with either EP or GL treatment, because these proteins have been reported to regulate astrocyte swelling and tissue edema in the CNS after a HMGB1 challenge.^{24

–28,32} We demonstrated that HMGB1 may play a role in the induction of early spinal cord edema, astrocyte activation (GFAP expression), and AQP4 expression after SCI in rats, while HMGB1 inhibition resulted in a reduction in early spinal cord edema and was found to benefit motor function recovery of rats.

Methods

Animals

A total of 276 pathogen-free adult female Sprague-Dawley (SD) rats (RRID: RGD_5508397), aged 8 to 10 weeks with the body weight 220 to 250 g, were obtained from the Beijing Haidian Thriving Experimental Animal Farms (Certificate No, SCXK-(Jun) 2012-0004, Beijing, China). Rats were housed at a temperature of 25°C with 40–50% humidity and a 12 h light/dark cycle. Food and water intake were supplied to the animals ad libitum. All rats were acclimated for at least one week before the surgical procedure. The management and experimental protocols of the animals in this study were approved by the Animal Care and Use Committee of Shanxi Medical University, and were conducted following recommendations provided by the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals.

Animal model of SCI

The rat SCI model was established according to a modified Allen's Weight-Dropping procedure.^3,11 At two days after estrus, female SD rats were anesthetized using an intraperitoneal administration of 10% chloral hydrate (400 mg/kg) followed by disinfection of the T10 spinous process. Laminectomy was performed at the T10 level to expose the spinal cord. A weight-drop impactor was used with an impact rod weighing 10 g (bottom diameter of 2 mm), with a 25 mm height of fall and an injury energy of 25 g·cm, resulting in a moderate injury to the T10 spinal cord. Immediately after SCI, the spinal cord surface exhibited signs of congestion.

Rats that quickly retracted their lower limbs, wagged their tails, and had flaccid paralysis were included in the study. Rats that underwent laminectomy, but not SCI, comprised the Sham group. After operation, rats were housed separately in individual cages, with sodium penicillin (400,000 U/rat) administered subcutaneously for three days. Urination was aided twice daily for a total of 7–10 days until micturition reflex was recovered.

Experimental protocol

First, 12 normal rats and 36 SCI rats were utilized. The HMGB1 expression in the spinal cord, HMGB1 levels in the serum, and spinal cord water content of rats were analyzed in normal rats, SCI 12 h rats, SCI one day rats, and SCI three day rats. SCI rats underwent laminectomy followed by SCI and received 0.9% saline via an intraperitoneal injection immediately following injury and daily after the injury. The experimental protocols were shown in Fig. 1.

FIG. 1.

Experimental protocols to analyze high mobility group box-1 (HMGB1) expression in the spinal cord, HMGB1 levels in the serum and spinal cord water content in normal rats, spinal cord injury (SCI) in 12 h rats, SCI one day rats, and SCI three day rats. *The protein and serum samples in SCI 12 h rats, SCI on day rats, and SCI three day rats were reused to analyze the effects of HMGB1 inhibition on HMGB1 expression in the spinal cord and HMGB1 levels in the serum of SCI rats in following experiments. # The values of spinal cord water content in SCI 12h rats, SCI 1d rats and SCI 3d rats were reused to analyze the effects of HMGB1 inhibition on spinal cord water content of SCI rats in following experiments.

Then, other experiments were performed based on Sham group, SCI group, SCI + EP group, and SCI + GL group in this study. Sixty-six rats that received laminectomy alone comprised the Sham group. A total of 198 SCI model rats were randomized as randomization table into three groups: (1) SCI group: 66 rats underwent laminectomy followed by SCI and received 0.9% saline via an intraperitoneal injection immediately after injury. Rats then continued to receive this injection daily after the injury. (2) SCI + EP group: 66 rats underwent laminectomy followed by SCI and received a 50 mg/kg dose of EP (diluted in 0.9% saline) via an intraperitoneal injection immediately after injury. Rats then continued to receive this injection daily after the injury. (3) SCI + GL group: 66 rats underwent laminectomy followed by SCI and received a 100 mg/kg dose of GL (diluted in 0.9% saline) via an intraperitoneal injection immediately after injury. Rats then continued to receive this injection daily after the injury.

The EP or GL dosage used in the study was based on previous experiments in which EP or GL effectively inhibited HMGB1 expression in the spinal cord and HMGB1 levels in the serum of SCI rats and had no side effect for rats. These experimental protocols were shown in Fig. 2.

FIG. 2.

Experimental protocols to analyze the effects of high mobility group box-1 (HMGB1) inhibition using ethyl pyruvate (EP) or glycyrrhizin (GL) on spinal cord HMGB1 expression, HMGB1 levels in the serum, rats motor function by the Basso, Beattie, and Bresnahan (BBB) locomotor rating scale and the inclined plane test, spinal cord water content, spinal cord edema using magnetic resonance imaging (MRI), and spinal cord expression levels of glial fibrillary acidic protein (GFAP), aquaporin-4 (AQP4), toll-like receptor-4 (TLR4), myeloid differentiation primary response gene 88 (MyD88) as well as nuclear factor-kappa B (NF-κB) in spinal cord injury (SCI) rats.

Experimenters were blinded when the rat SCI model was established, the experiments were performed, and the images and values were analyzed. One rat in the Sham group that had delayed lower limbs paralysis after operation and three rats that died after SCI were removed, and other rats that underwent the experiment were supplemented.

Special chemicals, antibodies, and other material/tools

Special chemicals: ethyl pyruvate (Sigma-Aldrich Co., Cat# E47808, Saint Louis, MO); glycyrrhizin (EKEAR, Cat# C0225, Shanghai, China); phosphatase and protease inhibitors (Beyotime, Cat#S1873, Shanghai, China); BCA protein Assay Kit (Beyotime, Cat# P0012S); protein extraction kit (Beyotime, Cat# P0033); enhanced chemiluminescence (ECL, Beyotime, Cat# P0018); enzyme-linked immunosorbent assay (ELISA) kit (Xitang, Cat# F15870, Shanghai, China); bovine serum albumin (BSA, Boster, Cat# 11D21C, Wuhan, China); streptavidin-biotin complex (Sloarbio, Cat#P1400, Beijing, China); 3, 3′-diaminobenzidine (DAB, ZSGB-Biology, Cat# ZLI-9018, Beijing, China); hematoxylin staining (ZSGB-Biology, Cat# AR0005).

Antibodies: rabbit polyclonal anti-HMGB1 antibody (Abcam, Cat# ab18256, RRID:AB_444360, Cambridge, UK); mouse monoclonal anti-GFAP antibody (Cell Signaling Technology, Cat# 3670, RRID:AB_561049, Boston, MA); rabbit polyclonal anti-AQP4 antibody (Abcam, Cat# ab46182, RRID: AB_955676); mouse monoclonal anti-TLR4 antibody (Novus, Cat# 76B357.1, RRID: AB_839000, Littleton, CO); rabbit polyclonal anti-MyD88 antibody (Abcam, Cat# ab131071, RRID: AB_11156885); rabbit monoclonal anti-NF-κB antibody (Cell Signaling Technology, Cat# 4764, RRID:AB_823578); mouse monoclonal anti-GAPDH antibody (Beyotime, Cat# AF0006, RRID: AB_2715590); mouse monoclonal anti-Histone H3 antibody (Beyotime, Cat# AF0009, RRID: AB_2715593); horseradish peroxidase-conjugated goat anti-rabbit secondary antibody (absin, Cat# abs20002A, RRID: AB 2716554, Shanghai, China); horseradish peroxidase-conjugated goat anti-mouse secondary antibody (absin, Cat# abs20001A, RRID: AB 2716555).

Other material/tools: GE Discovery MR750w (Siemens Medical Solutions, CT); Onis 2.4 Free Edition software (DigitalCore, http://www.onis-viewer.com); Image Pro plus 6.0 software (Media Cybernetics, http://image-pro-plus.updatestar.com); Polyvinylidene fluoride membrane (Millipore Corp. Billerica, MA); Quantity One software (Bio-Rad, http://www.bio-rad.com, RRID: SCR_014280); light microscopy (Olympus, Cat# CKX41SF, Tokyo, Japan); SPSS 18.0 program (IBM, https://www.ibm.com).

Evaluation of motor function

The Basso, Beattie, Bresnahan (BBB) rating scale and inclined plane test were used to evaluate locomotor function in rats. Rats were placed in an open field to assess the recovery of motion dysfunction using the 21-point BBB rating scale.³³ For the inclined plane test, a rat was placed on an adjustable inclined plane, and the maximum angle at which the plane could be inclined where the rat was able to maintain a stable position for at least 5 sec was determined.³⁴ The BBB scores and inclined plane test were evaluated by two laboratory personnel who were blinded to the treatments. The average score and angle were recorded. Interobserver reliability was determined to be good (κ = 0.76∼0.91).

Determination of spinal cord water content

The wet weight/dry weight method was used to determine the water content of the spinal cord tissue. Spinal cord segments measuring 10 mm in length were centered at the injury epicenter and were collected after rats were anesthetized via an intraperitoneal injection of 10% choral hydrate. The wet weight of the injured spinal cord was measured using a fine electronic balance. The spinal cord segments were then dried at 80°C for 48 h to determine the dry weight. The water content in spinal cord tissue was calculated using the following equation: (wet weight - dry weight)/ wet weight × 100%.

Evaluation of spinal cord edema by magnetic resonance imaging (MRI)

The spinal cord edema was evaluated by MRI.^35,36 MRI experiments were performed on a 3T clinical dedicated head MR scanner (GE Discovery MR750w) with a HD wrist Array coil for transmitting and receiving the MRI signal. Rats were anesthetized using an intraperitoneal injection of 10% choral hydrate and placed in the supine position. No corrections for motion artifacts from spontaneous respiration or blood flow were applied. Two-dimensional axial and sagittal images of the spine were acquired. T1-weighted spin echo axial images were acquired with the following sequence parameters: repetition time (TR) of 589 msec, echo time (TE) of 22.4 msec, image matrix of 224 × 224, field of view (FOV) of 6 × 6 cm, and number of acquisitions (NA) of 12 times. The acquired images had an in-plane resolution of 268 μm with a section thickness of 1.5 mm and spacing is 0.1 mm between sections. The total acquisition time with five averages and seven slices in two concatenations was 8 min and 53 sec.

T2-weighted turbo spin echo axial sequences were acquired with the following sequence parameters: TR of 2902 msec, TE of 85 msec, image matrix 224 × 224, FOV of 6 × 6 cm, and NA of 12 times. The acquired images had an in-plane resolution of 268 μm with a section thickness of 1.5 mm and spacing is 0.1mm between sections. The total acquisition time with five averages and seven slices in two concatenations was 8 min and 14 sec.

Proton density-weighted turbo spin echo axial images were acquired with a TR of 1500 msec, TE of 35 msec, image matrix 224 × 224, FOV of 6 × 6 cm, and NA of 12 times. The images had an in-plane resolution of 277 μm with a section thickness of 1.5 mm. The total acquisition time with five averages and seven slices was 7 min and 15 sec.

T2-weighted turbo spin echo sagittal sequences were acquired with the following sequence parameters: TR of 2000 msec, TE of 85 msec, image matrix 288 × 192, FOV of 8 × 8 cm, and NA of 14 times. The acquired images had an in-plane resolution of 277 μm with a section thickness of 1.2 mm and spacing is 0.1 mm between sections. The total acquisition time with one average and five slices in two concatenations was 8 min and 4 sec. Onis 2.4 Free Edition software was used for image processing.

The relatively long TE in axial and sagittal T2-weighted MRI were chosen to optimize detection of edema as the sequel of SCI. Quantitative analysis of axial MRI was performed using Image Pro plus 6.0 software. Data were analyzed blind to treatment condition. Seven consecutive T1- and PD-weighted MRI images and seven consecutive T2-weighted MRI around the injury epicenter were overlaid, and regions of interest were traced manually using Image Pro plus 6.0 software.

The images provided the most detail to clearly determine areas of the whole cord in the combination of T1- and PD-weighted MRI images and areas of hyperintense signal in the T2-weighted MRI (commonly thought to reflect edema) within the lesioned cord. Pixel counts were converted into area units (mm²) by scaling with the in-plane pixel size. Volume measurements (mm³) were obtained by adding the individual slice areas and multiplying by the slice thickness. The whole cord volume, the cord volume containing hyperintense pixels, and the volume of one slice at the level of lesion epicenter containing hyperintense pixels were measured.

Western blot analysis

Immediately after anesthesia by intraperitoneal injection of 10% choral hydrate, the 10 mm long segments of damaged spinal cords centered at the injury epicenter were dissociated, harvested at 4°C, and stored at −80°C in liquid nitrogen. Spinal cord tissue was then homogenized completely on ice in lysis buffer supplemented with phosphatase and protease inhibitors. Lysates were centrifuged at 14,000 × g for 10 min at 4°C. The lysate protein concentration was then determined using a BCA Protein Assay Kit.

To evaluate nuclear NF-κB expression, nuclear extracts were prepared successively using a protein extraction kit. Protein samples (20 μg per lane) were then subjected to electrophoresis on a 10% sodium dodecyl sulfate polyacrylamide gels, followed by transfer to polyvinylidene fluoride membrane. After blocking the membrane in 5% nonfat milk at 37°C for 2 h, membranes were incubated with primary antibody overnight at 4°C: anti-HMGB1 (1:1000), anti-GFAP (1:1000), anti-AQP4 (1:800), anti-TLR4 (1:800), anti-MyD88 ( 1:800), anti-NF-κB (1:800), anti-GAPDH (1:1000), or anti-Histone H3 (1:1000). Membranes were then washed, followed by incubation with the horseradish peroxidase-conjugated goat anti-rabbit or goat anti-mouse secondary antibody diluted at 1:5000 at room temperature for 2 h. Bands were visualized using the ECL technique.

Quantification was performed using Quantity One software. The expression levels of protein, with an exception of NF-κB, were normalized to that of GAPDH. The expression level of nuclear NF-κB was normalized to that of Histone H3. The band density for the normal group or Sham group was set at 1.0, and all band density values were normalized by the value of the normal group or Sham group. The Western blot data were represented as the fold change relative to the normal group or Sham group.

ELISA

Serum HMGB1 levels in SCI rats were measured by ELISA. Rat serum was sampled, and HMGB1 concentrations were determined using an ELISA kit according to the protocol. Absorbance at 450 nm was measured using a microplate reader.

Immunohistochemistry

Rats were anesthetized using an intraperitoneal injection of 10% choral hydrate. Rats were then rapidly perfused transcardially with saline followed by fixation with 4% paraformaldehyde in 0.1 M phosphate buffered saline (PBS). Spinal cord segments 10 mm in length centered at the injury epicenter were removed and fixed in 4% paraformaldehyde/PBS at 4°C for 24 h. Fixed spinal cords were then dehydrated and paraffin-embedded. Transverse tissue sections (3 μm) were sectioned using the microtome from 3 mm above the SCI epicenter. Paraffin sections were deparaffinized in xylene, rehydrated in gradient ethanol, and treated with 3% H₂O₂ at room temperature for 10 min.

Antigen retrieval was performed using 0.01 M citrate buffer (pH 6.0) heated to 95°C for 3 min, followed by 55°C for 3 min. Sections were then blocked in 5% BSA at 37°C for 30 min. Sections were then incubated with the following primary antibodies at 4°C overnight: anti-GFAP (1:100), anti-AQP4 (1:100), anti-TLR4 (1:200), or anti-MyD88 (1:200). Sections were then rinsed in PBS, followed by incubation with the respective secondary antibody at 37°C for 30 min. Sections were then washed and incubated with StreptAvidin-Biotin Complex at 37°C for 30 min. The reaction product was visualized by incubating samples with DAB for 5 min.

Hematoxylin staining was performed for 8 min to reveal cell nuclei. Sections were then dehydrated, mounted, sealed, and visualized using light microscopy (at 40 × , 100 × , and 400 × magnification). Immunohistochemical assays were performed using Image Pro plus 6.0 software. A total of five different visual fields were selected randomly, and immunoreactive cells were counted. The mean optical density values of immunoreactive protein in the spinal cord were measured.

Statistical analysis

All values are represented as the mean ± standard deviation (SD), and statistical analysis was performed using the SPSS 18.0 program. Intergroup data were compared using one-way analysis of variance (ANOVA) followed by the Dunnett post hoc test. For BBB rating scale and inclined plane test, statistical analysis was based on two-way repeated measures ANOVA with the Tukey post hoc test. A p value <0.05 was considered statistically significant.

Results

SCI increases HMGB1 expression in spinal cord, HMGB1 levels in the serum and spinal cord water content of rats

To evaluate the effects of SCI on HMGB1 expression in the spinal cord and HMGB1 serum levels in rats, we performed Western blot and ELISA analyses. Western blot analysis demonstrated that HMGB1 expression in the spinal cord was significantly increased from 12 h to three days after SCI compared with normal rats (p < 0.05). The value was observed to peak at the one day time point (p < 0.05, Fig. 3A). As shown in Fig. 3B, the levels of circulating of HMGB1 were found to increase at 12 h, one day, and three days after SCI (p < 0.05).

FIG. 3.

Effects of spinal cord injury (SCI) on high mobility group box-1 (HMGB1) expression in spinal cord, HMGB1 levels in the serum and spinal cord water content of rats. (A) Western blot analysis demonstrated that HMGB1 levels in spinal cord from 12 h to 3 days after SCI were significantly increased compared with normal rats. Values are means ± standard deviation (SD). n = 6. #p < 0.05 vs. normal; *p < 0.05 vs. SCI one day. (B) HMGB1 levels in serum of rats, evaluated by enzyme-linked immunosorbent assay, were found to increase at 12 h, one day, and three days after SCI. Means ± SD. n = 6. #p < 0.05 vs. normal; *p < 0.05 vs. SCI one day. (C) The water content in spinal cord of rats harvested at 12 h, one day, and three days after SCI were significantly increased compared with that of the normal rats, with a peak increase at one day after SCI. Means ± SD. n = 6. #p < 0.05 vs. normal; *p < 0.05 vs. SCI one day. GAPDH, glyceraldehyde 3-phosphate dehydrogenase.

Spinal cord water content was determined by measuring the wet weight/dry weight of spinal cord. Spinal cord water content was measured at different time points after SCI (12 h–3 days). We found that the spinal cord water content after SCI was significantly increased compared with that of the normal rats, with a peak increase observed at one day after SCI (p < 0.05, Fig. 3C).

EP or GL effectively inhibits HMGB1 expression in spinal cord and HMGB1 levels in the serum of SCI rats

To determine the effects of EP and GL on HMGB1 expression in the spinal cord and HMGB1 serum levels in rats after SCI, we performed Western blot and ELISA analyses. Western blot analysis demonstrated that after administration of EP, the increased HMGB1 expression in the spinal cord was found to be remarkably suppressed in the SCI + EP group compared with those in the SCI group at 12 h, one day, and three days after SCI (p < 0.05), but the HMGB1 expression in the spinal cord in the SCI + EP group was increased compared with those in the Sham group at 12 h and one day after SCI (p < 0.05). GL was also found to suppress significantly the increased HMGB1 expression in the spinal cord of rats with SCI compared with those in the SCI group at 12 h, one day, and three days after SCI (p < 0.05), but the HMGB1 expression in the SCI + GL group was increased compared with those in the Sham group (p < 0.05) (Fig. 4A).

FIG. 4.

Effects of ethyl pyruvate (EP) or glycyrrhizin (GL) on the inhibition of high mobility group box-1 (HMGB1) expression in rats with spinal cord injury (SCI). (A) Western blot analysis demonstrated that after administration of EP or GL, the increased HMGB1 levels in spinal cord were found to be remarkably suppressed at 12 h, one day, and three days after SCI. The HMGB1 expression in the spinal cord in the SCI + EP group was increased compared with those in the Sham group at 12 h and one day after SCI. The HMGB1 expression in the SCI + GL group was increased compared with those in the Sham group at 12 h, one day, and three days after SCI. Values are means ± standard deviation (SD). n = 6. *p < 0.05 vs. SCI group; #p < 0.05 vs. Sham group. (B) Enzyme-linked immunosorbent assay results showed that the increased levels of circulating HMGB1 were found to be significantly attenuated by EP or GL at 12 h, one day, and three days after SCI. The serum HMGB1 levels in SCI + EP group and SCI + GL group were increased compared with those in the Sham group at 12 h, 1 day, and three days after SCI. Means ± SD. n = 6. *p < 0.05 vs. SCI group; #p < 0.05 vs. Sham group. GAPDH, glyceraldehyde 3-phosphate dehydrogenase.

We next performed an ELISA to measure serum HMGB1 levels. As shown in Fig. 4B, the increased levels of circulating HMGB1 were found to be significantly attenuated by EP or GL compared with those in the SCI group at 12 h, one day, and three days after SCI (p < 0.05); however, the serum HMGB1 levels in the SCI + EP group and SCI + GL group were increased compared with those in the Sham group (p < 0.05).

Inhibiting HMGB1 improves motor function of rats with SCI

The hindlimb motor function of animals was evaluated using a BBB rating scale and inclined plane test, respectively. The pre-operative BBB scores of the rats were measured to be 21, and these BBB scores were found to drop to 0 after SCI. Rats in the SCI group, SCI + EP group, and SCI + GL group continued to show significantly worse performance compared with those in the Sham group from 12 h to 14 days after SCI (p < 0.05). The BBB scores of rats in the SCI + EP group and SCI + GL group were found to be significantly higher compared with those in the SCI group from seven days after SCI (p < 0.05) (Fig. 5A).

FIG. 5.

Effects of high mobility group box-1 (HMGB1) inhibition using ethyl pyruvate (EP) or glycyrrhizin (GL) on the improvement of motor function of spinal cord injury (SCI) rats. (A) The hind-limb motor function of rats from each group was evaluated using a Basso, Beattie, and Bresnahan locomotor rating scale (BBB rating scale). The pre-operative BBB scores of the rats were 21, and these BBB scores were found to drop to 0 after SCI. The BBB scores of rats in the SCI + EP group and SCI + GL group were found to be significantly higher compared with those in the SCI group from seven days after SCI. Values are means ± standard deviation (SD). n = 6. *p < 0.05 vs. SCI group. (B) The hindlimb motor function of rats from each group was evaluated using an inclined plane test. The average angles in the SCI group, SCI + EP group, and SCI + GL group continued to exhibit significantly lower values compared with those in the Sham group. In the SCI + EP group, the average angle was measured to be significantly greater than that in the SCI group from one day after SCI. In the SCI + GL group, the average angle was found to be significantly greater than that of the SCI group from three days after SCI. Means ± SD. n = 6. *p < 0.05 vs. SCI group. Black solid circle: Sham group, red open circle: SCI group, blue diamond: SCI + EP group, and green triangle: SCI + GL group. Color image is available online at www-liebertpub-com.web.bisu.edu.cn/neu

For the inclined plane test, the mean angles in all groups were measured to be approximately 60 degrees before SCI. The average angles in the SCI group, SCI + EP group, and SCI + GL group continued to exhibit significantly lower values compared with those in the Sham group during the period of observation (p < 0.05). In the SCI + EP group, the average angle was measured to be significantly greater than that in the SCI group from one day after SCI (p < 0.05). In the SCI + GL group, the average angle was found to be significantly greater than that of the SCI group from three days after SCI (p < 0.05) (Fig. 5B).

Inhibiting HMGB1 reduces spinal cord water content of rats with SCI

Spinal cord water content was determined by measuring the wet weight/dry weight of spinal cord. We found that HMGB1 inhibition in rats with EP exhibited a reduction in spinal cord water content compared with those in the SCI group at 12 h, one day, and three days after SCI (p < 0.05). HMGB1 inhibition in rats with GL, compared with those in the SCI group, exhibited a reduction in spinal cord water content at one day and three days after SCI (p < 0.05) (Fig. 6).

FIG. 6.

Effects of high mobility group box-1 (HMGB1) inhibition using ethyl pyruvate (EP) or glycyrrhizin (GL) on the spinal cord water content in rats after spinal cord injury (SCI). Compared with those in the SCI group, HMGB1 inhibition in rats with EP exhibited a reduction in spinal cord water content at 12 h, one day, and three days after SCI. HMGB1 inhibition in rats with GL exhibited a reduction in spinal cord water content at one day and three days after SCI. Values are means ± standard deviation. n = 6. *p < 0.05 vs. SCI group.

Inhibiting HMGB1 attenuates spinal cord edema of rats with SCI

We investigated spinal cord edema of SCI rats using the MRI method. The sagittal and axial T2-weighted MRI showed that no spinal cord edema was observed in the Sham group, while extensive edema was observed in the spinal cord of rats at one day after SCI. Treatment with an HMGB1 inhibitor, EP or GL, was found to reduce spinal cord edema of SCI rats (Fig. 7A, 7B).

FIG. 7.

Effects of inhibiting high mobility group box-1 (HMGB1) using ethyl pyruvate (EP) or glycyrrhizin (GL) on the spinal cord edema of spinal cord injury (SCI) rats. The spinal cord edema of rats at one day after SCI was evaluated by magnetic resonance imaging (MRI). The sagittal (A) and axial (B) T2-weighted MRI showed that no edema was observed in the Sham group, while extensive edema was observed in the spinal cord of rats at one day after SCI. Treatment with an HMGB1 inhibitor, EP or GL, was found to reduce spinal cord edema of SCI rats. (B) The tracing method was used for quantification of spinal cord volume in axial T1- and PD-weighted MRI images, and volume of hyperintense signal in axial T2-weighted MRI. (C) The whole cord volume in T1- and PD-weighted MRI images was significantly increased compared with the Sham group at one day after SCI. The increased volume was significantly attenuated by EP or GL treatment compared with that in the SCI group. Values are means ± standard deviation (SD). n = 6. *p < 0.05 vs. SCI group. (D) The cord volume containing hyperintense pixels in T2-weighted MRI was significantly increased at one day after SCI. The increased volume was significantly attenuated by EP or GL treatment compared with that in the SCI group. Means ± SD. n = 6. *P < 0.05 vs. SCI group. (E) The volume of one slice at the level of lesion epicenter of cord containing hyperintense pixels in T2-weighted MRI was significantly increased at 1 day following SCI. The increased volume was significantly attenuated by EP or GL treatment compared to that in the SCI group. Means ± SD. n = 6. *p < 0.05 vs. SCI group. Color image is available online at www-liebertpub-com.web.bisu.edu.cn/neu

Quantitative analysis of axial MRI demonstrated that the whole cord volume in T1- and PD-weighted MRI images, the cord volume containing hyperintense pixels in T2-weighted MRI, and the volume of one slice at the level of lesion epicenter containing hyperintense pixels in T2-weighted MRI were significantly increased compared with those in the Sham group at one day after SCI (p < 0.05). These increased volumes were found to be significantly attenuated by EP or GL treatment in rats compared with those in the SCI group at one day after SCI (p < 0.05). (Fig. 7B, 7C, 7D, 7E)

Effects of inhibiting HMGB1 on the GFAP expression of spinal cord astrocytes in rats with SCI

We examined the effect of HMGB1 inhibition on spinal cord GFAP expression, a specific astrocytic marker. GFAP immunohistochemistry results are shown in Fig. 8A, 8B, 8C. The number of GFAP positive cells and the mean optical density values of GFAP were both found to be significantly increased in the SCI group compared with those of the Sham group at one day after SCI (p < 0.05). In the SCI + EP group and SCI + GL group, the number of GFAP positive cells and the mean optical density values of GFAP were dramatically decreased at one day after SCI compared with those in the SCI group (p < 0.05). In addition, Western blot analysis also demonstrated significantly increased GFAP protein levels in the spinal cord of the SCI group at one day after SCI (p < 0.05). This increase, however, was found to be significantly attenuated by EP or GL treatment (p < 0.05) (Fig. 8D).

FIG. 8.

Effects of inhibiting high mobility group box-1 (HMGB1) using ethyl pyruvate (EP) or glycyrrhizin (GL) on glial fibrillary acidic protein (GFAP) expression in the spinal cord of rats at one day after spinal cord injury (SCI). (A) Immunohistochemistry results of GFAP in spinal cord (40 × , 100 × , and 400 × magnification, bar equal 100 μm). (B) The number of GFAP positive cells was evaluated by immunohistochemistry. Values are means ± standard deviation (SD). n = 6. *P < 0.05 vs. SCI group. (C) The mean optical density values of GFAP were evaluated by immunohistochemistry. Means ± SD. n = 6. *P < 0.05 vs. SCI group. The number of GFAP positive cells and the mean optical density values of GFAP were both found to be increased significantly in the SCI group compared with those of the Sham group at one day after SCI. In the SCI + EP group and SCI + GL group, the number of GFAP positive cells and the mean optical density values of GFAP were dramatically decreased at one day after SCI compared with those of the SCI group. (D) GFAP protein levels in spinal cord were evaluated by Western blot. The increased GFAP protein level in the spinal cord at one day after SCI was found to be significantly attenuated by EP or GL treatment. Means ± SD. n = 6. *p < 0.05 vs. SCI group. Color image is available online at www-liebertpub-com.web.bisu.edu.cn/neu

Inhibiting HMGB1 reduces the AQP4 expression of spinal cord in rats with SCI

Immunohistochemistry and Western blot analyses were performed to measure AQP4 expression levels in the spinal cord after SCI. AQP4 immunohistochemistry results showed that AQP4 overexpression induced by SCI was present primarily in the gray matter and in astrocytes. We observed a significant upregulation in the number of AQP4 positive cells as well as the mean optical density values of AQP4 in the SCI group compared with those in the Sham group at one day after SCI (p < 0.05). This was shown to be able to be reduced significantly by EP or GL treatment (p < 0.05) (Fig. 9A, 9B, 9C). Western blot analysis demonstrated that AQP4 protein levels in the spinal cord were increased significantly in the SCI group at one day after SCI (p < 0.05), and that this increase could be attenuated significantly with EP or GL treatment (p < 0.05, Fig. 9D).

FIG. 9.

Effects of inhibiting high mobility group box-1 (HMGB1) using ethyl pyruvate (EP) or glycyrrhizin (GL) on the aquaporin-4 (AQP4) expression in spinal cord of rats at one day after spinal cord injury (SCI). (A) Immunohistochemistry results of AQP4 in spinal cord (40 × , 100 × , and 400 × magnification, bar equal 100 μm). (B) The number of AQP4 positive cells was evaluated by immunohistochemistry. Values are means ± standard deviation (SD). n = 6. *p < 0.05 vs. SCI group. (C) The mean optical density values of AQP4 were evaluated by immunohistochemistry. Means ± SD. n = 6. *p < 0.05 vs. SCI group. The number of AQP4 positive cells and the mean optical density values of AQP4 were both found to be increased significantly in the SCI group compared with those of the Sham group at one day after SCI. In the SCI + EP group and SCI + GL group, the number of AQP4 positive cells and the mean optical density values of AQP4 were dramatically decreased at one day after SCI compared with those of the SCI group. (D) AQP4 protein levels in spinal cord were evaluated by Western blot. The increased AQP4 protein level in the spinal cord at one day after SCI was found to be significantly attenuated by EP or GL treatment. Means ± SD. n = 6. *p < 0.05 vs. SCI group. Color image is available online at www-liebertpub-com.web.bisu.edu.cn/neu

Inhibiting HMGB1 reduces the TLR4 and MyD88 expression of spinal cord in rats with SCI

It has been suggested that the TLR4/MyD88 signaling pathway could be activated by HMGB1 after injury or disease in the CNS.^16,17 We performed immunohistochemistry and Western blot analyses to determine the TLR4 and MyD88 expression levels in the spinal cord in each group. In the SCI group, both the number of positive cells as well as the mean optical density values of TLR4 were found to be significantly increased compared with those in the Sham group at one day after SCI (p < 0.05). These were observed to be significantly decreased, however, in the SCI + EP group and the SCI + GL group compared with the SCI group (p < 0.05, Fig. 10A, 10B, 10C). Western blot analysis demonstrated a significantly higher TLR4 protein level in the SCI group compared with that of the Sham group at one day after SCI (p < 0.05), which was found to be reduced by EP or GL administration (p < 0.05, Fig. 10D).

FIG. 10.

Effects of inhibiting high mobility group box-1 (HMGB1) using ethyl pyruvate (EP) or glycyrrhizin (GL) on toll-like receptor 4 (TLR4) expression in the spinal cord of rats at one day after spinal cord injury (SCI). (A) Immunohistochemistry results of TLR4 staining in spinal cord (40 × , 100 × , and 400 × magnification, bar equal 100 μm). (B) The number of TLR4 positive cells was evaluated by immunohistochemistry. Values are means ± standard deviation (SD). n = 6. *p < 0.05 vs. SCI group. (C) The mean optical density values of TLR4 were evaluated by immunohistochemistry. Means ± SD. n = 6. *p < 0.05 vs. SCI group. In the SCI group, both the number of positive cells as well as the mean optical density values of TLR4 were found to be significantly increased compared with those of the Sham group at one day after SCI. These were observed to be significantly decreased in the SCI + EP group and the SCI + GL group, however, compared with the SCI group. (D) TLR4 protein levels in spinal cord were evaluated by Western blot. The increased TLR4 protein level in the spinal cord at one day after SCI was found to be significantly attenuated by EP or GL treatment. Means ± SD. n = 6. *p < 0.05 vs. SCI group. Color image is available online at www-liebertpub-com.web.bisu.edu.cn/neu

Immunohistochemistry demonstrated the presence of a small number of MyD88 positive cells as well as some mean optical density values of MyD88 in the Sham group, which was found to increase on SCI, and reverse with EP or GL treatment at one day after SCI (p < 0.05, Fig. 11A, 11B, 11C. We found that MyD88 expression levels were upregulated at one day after SCI, and EP or GL was found to significantly decrease MyD88 expression levels, as shown by Western blot analysis (p < 0.05, Fig. 11D). TLR4 and MyD88 overexpression after SCI were found to be reduced by EP or GL, indicating that HMGB1 inhibition could function to repress the activation of this pathway. These results suggest a potential mechanism underlying spinal cord edema induced by HMGB1 after SCI.

FIG. 11.

Effects of inhibiting high mobility group box-1 (HMGB1) using ethyl pyruvate (EP) or glycyrrhizin (GL) on myeloid differentiation primary response gene 88 (MyD88) expression in the spinal cord of rats at one day after spinal cord injury (SCI). (A) Immunohistochemistry results of MyD88 staining in the spinal cord (40 × , 100 × , and 400 × magnification, bar equal 100 μm). (B) The number of MyD88 positive cells was evaluated by immunohistochemistry. Values are means ± standard deviation (SD). n = 6. *p < 0.05 vs. SCI group. (C) The mean optical density values of MyD88 were evaluated by immunohistochemistry. Means ± SD. n = 6. *p < 0.05 vs. SCI group. In the SCI group, both the number of positive cells as well as the mean optical density values of MyD88 were found to be significantly increased compared with those of the Sham group at one day after SCI. These were observed to be significantly decreased, however, in the SCI + EP group and the SCI + GL group compared with the SCI group. (D) MyD88 protein levels in the spinal cord were evaluated by Western blot. MyD88 expression levels were upregulated at one day after SCI, and EP or GL was found to significantly decrease MyD88 expression levels. Means ± SD. n = 6. *p < 0.05 vs. SCI group. Color image is available online at www-liebertpub-com.web.bisu.edu.cn/neu

Inhibiting HMGB1 reduces the nucleus NF-κB expression of spinal cord in rats with SCI

We also studied the effect of HMGB1 inhibition on the spinal cord expression of nuclear NF-κB after SCI using Western blot analysis (Fig. 12). The result demonstrated that SCI resulted in a significant increase in nuclear NF-κB levels compared with that of the Sham group at one day after SCI (p < 0.05), while treatment with EP or GL was found to significantly attenuate the increased nuclear NF-κB levels (p < 0.05).

FIG. 12.

Effects of inhibiting high mobility group box-1 (HMGB1) using ethyl pyruvate (EP) or glycyrrhizin (GL) on nuclear factor-kappa B (NF-κB) expression in the spinal cord. The nuclear NF-κB expression in the spinal cord of rats at one day after spinal cord injury (SCI) was evaluated by Western blot. The result demonstrated that SCI resulted in a significant increase in nuclear NF-κB levels, while treatment with EP or GL was found to significantly attenuate the increased nuclear NF-κB levels. Values are means ± standard deviation (SD). n = 6. *p < 0.05 vs. SCI group.

Discussion

This study demonstrated that SCI resulted in an enhancement in HMGB1 release from spinal cord tissue, which was accompanied by increased neurological deficits, spinal cord water content, spinal cord edema, and GFAP and AQP4 levels in the spinal cord of SD rats. Inhibition of HMGB1 using EP or GL administered via an intraperitoneal route after SCI was found to moderately improve locomotor dysfunction, reduce early spinal cord edema, and repress GFAP and AQP4 overexpression. Further experiments demonstrated this protective effect by showing that HMGB1 inhibition could be associated with inhibition of the activation of the TLR4, MyD88, and NF-κB signaling pathway, which were shown to be up-regulated after SCI.

HMGB1 is a nonhistone DNA-binding protein that has been known to play an intracellular role in transcriptional regulation and an extracellular role serving as a cytokine-like mediator of inflammation.³⁷ In a CNS injury, HMGB1 could function as an early trigger for pro-inflammatory activation by creating edema, disrupting the blood-spinal cord barrier, and infiltrating inflammatory cells, which would correlate with poor neurological outcomes.^{18,23,26,38,39} After SCI in humans, the maximum plasma HMGB1 level in persons with acute SCI was shown to be greater than nine-fold higher compared with that in an uninjured person.¹⁴ In rats, HMGB1 was shown to be released after SCI, with the spinal cord cells, astrocytes, microglia, and necrotic cells acting as the primary sources of HMGB1.^15,18,19 Similarly, we demonstrated elevated HMGB1 expression levels in the spinal cord and serum of rats after SCI in our study.

Spinal cord edema after SCI is most prominent in the gray matter of the spinal cord that is closely associated with poor neurological outcomes of SCI.^3,9,11,40,41 It is known that spinal cord edema is largely because of the inflammatory response after SCI. HMGB1, an inflammatory cytokine, is closely related with spinal cord edema after SCI.^2,15,18 Accordingly, our study demonstrated that spinal cord edema occurred from 12 h to three days in rats after SCI, peaking at one day. We also showed that HMGB1 expression levels also increased between 12 h to three days.

The mechanisms of spinal cord edema formation after SCI are cytotoxic edema and vasogenic edema.¹³ Astrocyte activation and swelling are thought to play a major role in cytotoxic edema formation^5,9,42 and are associated with increased expression levels of the AQP4 water channel.^12,43,44 AQP4 overexpression in the early stage of SCI could contribute to the transfer of circulatory water into the spinal parenchyma, allowing the development of cytotoxic edema and exacerbation of spinal cord edema.^3,9,11,13,44 As demonstrated in this study, GFAP and AQP4 expression in the spinal cord was markedly upregulated in rats after SCI.

HMGB1 inhibition in the CNS can be performed using an anti-HMGB1–neutralizing mAb, EP or GL, and other methods.^{23,30,31,38,45,46} As reported, both EP and GL are effective HMGB1 inhibitors in the CNS injury of rats.^23,30,38 In our study, we used either EP or GL via an intraperitoneal injection to rats after SCI to inhibit HMGB1 activity. Using this method, we observed significant HMGB1 inhibition effects. We did not attempt to compare the effects of HMGB1 inhibition between EP and GL in this study, however.

HMGB1 inhibition could function to improve the motor impairments induced by injury or other diseases in the CNS. Su and associates²³ reported that EP could inhibit HMGB1 expression and thereby provide neuroprotective effects after traumatic brain injury in rats. Huang and colleagues³⁸ demonstrated that inhibition of HMGB1 activity using GL resulted in alleviation of the aggravated damage of stroke in hyperglycemic rats. Sun and coworkers⁴⁶ showed that GL was a confirmed natural inhibitor of HMGB1 and functioned to alleviate early brain injuries after subarachnoid hemorrhage. Further, several studies have demonstrated that the improvement of motor impairments observed after HMGB1 inhibition, induced by either traumatic brain injury or intracerebral hemorrhage-induced injury in rats, was consistent with the reduction in brain edema.^31,45

In the case of SCI, HMGB1 inhibition could also provide this neuroprotective effect.^14,30 In our study, we found that HMGB1 inhibition proved effective in improving locomotor function and also attenuated SCI-induced spinal cord edema in rats. We believed that the improvement in BBB scores and inclined plane test observed in the SCI + EP group and SCI + GL group compared with those observed in the SCI group, at least in part, was associated with the reduction of spinal cord edema in those groups. We do note that the improvement in locomotor function observed with HMGB1 inhibition after SCI could also be related to the inhibition of other inflammatory effects of HMGB1.^23,30

AQP4 allows bidirectional transmembrane water flow and is suggested to be critically involved in the formation as well as clearance of edema in the CNS.⁴⁷ There is a close association between the formation of edema and increased AQP4 expression in astrocytes. Several studies have demonstrated that the reduction of cerebral edema was combined by the decreased AQP4 overexpression in astrocytes.^11,48
–50 In the spinal cord, inhibition or downregulation of AQP4 overexpression, which often accompanies decreased GFAP levels, could result in reduced spinal cord edema after SCI.^3,9

Based on immunohistochemistry and Western blot analyses, we found that HMGB1 inhibition resulted in reduced spinal cord edema after SCI, while at the same time decreasing GFAP and AQP4 overexpression. Therefore, we hypothesized that the reduction of spinal cord edema observed with HMGB1 inhibition after SCI may be associated with decreased astrocyte activation and AQP4 overexpression in the spinal cord.

We have further examined the activation of some signaling pathway after reduced spinal cord edema and AQP4 overexpression by HMGB1 inhibition. After SCI, administration of EP or GL was found to markedly reduce spinal cord edema and AQP4 overexpression and significantly suppress the up-regulation of HMGB1, TLR4, MyD88, and nuclear NF-κB expression at one day after SCI. This could suggest that one possible signaling pathway in the complex interacts with multiple signaling networks.

TLR4 is one receptor in the CNS that HMGB1 signaling relies on^23,51 and is known to be expressed in astrocytes.¹⁶ HMGB1 interacts with TLR4, and TLR4 is known to signal through adaptor protein pathways such as the MyD88 pathway. This leads to activation of NF-κB and results in the production of pro-inflammatory cytokines or adjusting effector protein.^23,27,30,52 In another of our studies, we have confirmed that HMGB1 inhibition reduces cell swelling and attenuates AQP4 expression in cultured rat spinal cord astrocytes in vitro after oxygen-glucose deprivation and reoxygenation, which is mediated through HMGB1/TLR4/MyD88/NF-κB signaling and in an interleukin-6-dependent manner.⁵³

Combined with our results, we hypothesized that after SCI, HMGB1 inhibition could function to reduce the expression level of HMGB1, likely via extracellular HMGB1, and may further decrease interactions between HMGB1 and its receptor, TLR4. It could then play a role in the TLR4/MyD88 signaling pathway, which converges on NF-κB and exerts a comprehensive impact on the reduction in AQP4 expression and spinal cord edema.

We note that there exist some limitations to this study. First, although we found that HMGB1 inhibition resulted in a reduction in the early spinal cord edema and a better locomotor function after SCI in rats, the observation time was relatively short. A longer observation (more than three days after SCI) may need to be performed in future experiments. Second, both cytotoxic edema and vasogenic edema are important for the development of spinal cord edema after SCI. In the present study, we observed the effects of HMGB1 inhibition on cytotoxic edema. We think that inhibiting HMGB1 may have some effects on vasogenic edema in the spinal cord after SCI, especially on the blood-spinal cord barrier. Further studies should be performed to evaluate the effects of HMGB1 inhibition on vasogenic edema after SCI. Third, the reducing of spinal cord edema and downregulation of astrocyte AQP4 expression by HMGB1 inhibition could be associative with the repression of activation of TLR4/MyD88/NF-κB signaling in SCI rats, but the correlative-causality should be the target of further investigation in the future.

Conclusion

We demonstrated that HMGB1 inhibition resulted in reducing early spinal cord edema and functioned to improve locomotor function. This was associated with a downregulation of astrocyte activation (GFAP expression) and AQP4 expression in SCI rats. HMGB1 plays a critical role in this process and has the potential to become the target of studies and treatment plans for SCI.

Footnotes

Acknowledgments

This study was supported by grant from the National Natural Science Foundation of China (grant numbers: 81401028), the Youth Science and Technology Research Foundation of Shanxi (grant numbers: 2015021201), and the Doctoral Foundation of Shanxi Medical University (grant numbers: 03201422).

Author Disclosure Statement

No competing financial interests exist.

References

Leonard

A.V.

, Thornton

, and Vink

(2015). The relative contribution of edema and hemorrhage to raised intrathecal pressure after traumatic spinal cord injury. J. Neurotrauma, 32, 397–402.

Kang

, Hai

, Yang

, Liang

, and Gao

C.J.

(2015). Hyperbaric oxygen intervention reduces secondary spinal cord injury in rats via regulation of HMGB1/TLR4/NF-κB signaling pathway. Int. J. Clin. Exp. Pathol. 8, 1141–1153.

A.M.

, Li

J.J.

, Sun

, Yang

D.G.

, Yang

M.L.

, Du

L.J.

, Gu

, Gao

, Li

, Chu

H.Y.

, Zhang

, and Gao

L.J.

(2015). Myelotomy reduces spinal cord edema and inhibits aquaporin-4 and aquaporin-9 expression in rats with spinal cord injury. Spinal Cord, 53, 98–102.

Fan

Z.K.

, Wang

Y.F.

, Cao

, Zhang

M.C.

, Zhang

, Lv

, Lu

, and Zhang

Y.Q.

(2010). The effect of aminoguanidine on compression spinal cord injury in rats. Brain Res. 1342, 1–10.

Liu

, Wang

, Yang

, Liu

, Zhou

, Hou

, and Xiang

(2015). Anti-edema effect of melatonin on spinal cord injury in rats. Biomed. Pap. Med. Fac. Univ. Palacky, Olomouc, Czech Repub. 159, 220–226.

Zhou

, Wang

, Xue

, Xie

, Huang

, Chen

, Gong

, and Zeng

(2013). Beneficial effects of hydrogen-rich saline against spinal cord ischemia-reperfusion injury in rabbits. Brain Res. 1517, 150–160.

Xiao

and Hu

(2014). Involvement of aquaporin 4 in astrocyte function and neuropsychiatric disorders. CNS Neurosci. Ther. 20, 385–390.

Thrane

A.S.

, Rangroo Thrane

, and Nedergaard

(2014). Drowning stars: reassessing the role of astrocytes in brain edema. Trends Neurosci. 37, 620–628.

, Zhu

, Diao

, Tao

, Yuan

, and Xiong

X.C.

(2013). Anti-edema effect of epigallocatechin gallate on spinal cord injury in rats. Brain Res. 1527, 40–46.

10.

Stokum

J.A.

, Gerzanich

, and Simard

J.M.

(2016). Molecular pathophysiology of cerebral edema. J. Cereb. Blood Flow Metab. 36, 513–538.

11.

, Wang

, Xu

, Zhuang

, Gong

, and Yan

(2014). Curcumin improves the recovery of motor function and reduces spinal cord edema in a rat acute spinal cord injury model by inhibiting the JAK/STAT signaling pathway. Acta Histochem. 116, 1331–1336.

12.

Oklinski

M.K.

, Lim

J.S.

, Choi

H.J.

, Oklinska

, Skowronski

M.T.

, and Kwon

T.H.

(2014). Immunolocalization of water channel proteins AQP1 and AQP4 in rat spinal cord. J. Histochem. Cytochem. 62, 598–611.

13.

Saadoun

and Papadopoulos

M.C.

(2010). Aquaporin-4 in brain and spinal cord oedema. Neuroscience, 168, 1036–1046.

14.

Papatheodorou

, Stein

, Bank

, Sison

C.P.

, Gibbs

, Davies

, and Bloom

(2017). High-Mobility Group Box 1 (HMGB1) is elevated systemically in persons with acute or chronic traumatic spinal cord injury. J. Neurotrauma, 34, 746–754.

15.

Yang

, Liu

, Zhou

, Wang

, Gao

, and Su

(2013). Hyperbaric oxygen alleviates experimental (spinal cord) injury by down-regulating HMGB1/ NF-κB expression. Spine, 38, E1641–E1648.

16.

Wang

Y.C.

, Lin

, and Yang

Q.W.

(2011). Toll-like receptors in cerebral ischemic inflammatory injury. J. Neuroinflammation, 8, 134.

17.

Fang

, Schachner

, and Shen

Y.Q.

(2012). HMGB1 in development and diseases of the central nervous system. Mol. Neurobiol. 45, 499–506.

18.

Chen

K.B.

, Uchida

, Nakajima

, Yayama

, Hirai

, Rodriguez Guerrero

, Kobayashi

, Ma

W.Y.

, Liu

S.Y.

, Zhu

, and Baba

(2011). High-mobility group box-1 and its receptors contribute to proinflammatory response in the acute phase of spinal cord injury in rats. Spine, 36, 2122–2129.

19.

Zhu

X.L.

, Chen

, Wang

, Li

, Huo

, Wang

, Min

Y.Y.

, Su

B.X.

, and Pei

J.M.

(2017). Electroacupuncture pretreatment attenuates spinal cord ischemia-reperfusion injury via inhibition of high-mobility group box 1 production in a LXA4 receptor-dependent manner. Brain Res. 1659, 113–120.

20.

Park

J.S.

, Svetkauskaite

, He

, Kim

J.Y.

, Strassheim

, Ishizaka

, and Abraham

(2004). Involvement of toll-like receptors 2 and 4 in cellular activation by high mobility group box 1 protein. J. Biol. Chem. 279, 7370–7377.

21.

Hori

, Brett

, Slattery

, Cao

, Zhang

, Chen

J.X.

, Nagashima

, Lundh

E.R.

, Vijay

, Nitecki

, et al. (1995). The receptor for advanced glycation end products (RAGE) is a cellular binding site for amphoterin. Mediation of neurite outgrowth and co-expression of rage and amphoterin in the developing nervous system. J. Biol. Chem. 270, 25752–25761.

22.

Malek

, Borowicz

K.K.

, Jargiello

, and Czuczwar

S.J.

(2007). Role of nuclear factor kappaB in the central nervous system. Pharmacol. Rep. 59, 25–33.

23.

, Wang

, Zhao

, Pan

, and Mao

(2011). Beneficial effects of ethyl pyruvate through inhibiting high-mobility group box 1 expression and TLR4/NF-kappaB pathway after traumatic brain injury in the rat. Mediators Inflamm. 2011, 807142.

24.

, Jin

, Zhu

, Wang

, Yuan

, Jiang

, Liang

, and Ma

(2015). Curcumin modulates TLR4/NF-kappaB inflammatory signaling pathway following traumatic spinal cord injury in rats. J. Spinal Cord Med. 38, 199–206.

25.

Ohnishi

, Monda

, Takemoto

, Fujimoto

, Sugitani

, Iwamura

, Hiroyasu

, and Inoue

(2014). High-mobility group box 1 up-regulates aquaporin 4 expression via microglia-astrocyte interaction. Neurochemistry Int. 75, 32–38.

26.

Laird

M.D.

, Shields

J.S.

, Sukumari-Ramesh

, Kimbler

D.E.

, Fessler

R.D.

, Shakir

, Youssef

, Yanasak

, Vender

J.R.

, and Dhandapani

K.M.

(2014). High mobility group box protein-1 promotes cerebral edema after traumatic brain injury via activation of toll-like receptor 4. Glia, 62, 26–38.

27.

Jayakumar

A.R.

, Tong

X.Y.

, Curtis

K.M.

, Ruiz-Cordero

, Abreu

M.T.

, and Norenberg

M.D.

(2014). Increased toll-like receptor 4 in cerebral endothelial cells contributes to the astrocyte swelling and brain edema in acute hepatic encephalopathy. J. Neurochem. 128, 890–903.

28.

Jayakumar

A.R.

, Tong

X.Y.

, Ruiz-Cordero

, Bregy

, Bethea

J.R.

, Bramlett

H.M.

, and Norenberg

M.D.

(2014). Activation of NF-kappaB mediates astrocyte swelling and brain edema in traumatic brain injury. J. Neurotrauma, 31, 1249–1257.

29.

Zhang

, Takahashi

H.K.

, Liu

, Wake

, Liu

, Maruo

, Date

, Yoshino

, Ohtsuka

, Mori

, and Nishibori

(2011). Anti-high mobility group box-1 monoclonal antibody protects the blood-brain barrier from ischemia-induced disruption in rats. Stroke, 42, 1420–1428.

30.

Gong

, Yuan

L.B.

, Hu

, Wu

, Yin

, Hou

J.L.

, Liu

Y.H.

, and Zhou

L.S.

(2012). Glycyrrhizin attenuates rat ischemic spinal cord injury by suppressing inflammatory cytokines and HMGB1. Acta Pharmacol. Sin. 33, 11–18.

31.

Ohnishi

, Katsuki

, Fukutomi

, Takahashi

, Motomura

, Fukunaga

, Matsuoka

, Isohama

, Izumi

, Kume

, Inoue

, and Akaike

(2011). HMGB1 inhibitor glycyrrhizin attenuates intracerebral hemorrhage-induced injury in rats. Neuropharmacology, 61, 975–980.

32.

, Wu

, Yang

, He

, Liu

, Jin

, and Yuan

(2016). TLR4-mediated NF-kappaB signaling pathway mediates HMGB1-induced pancreatic injury in mice with severe acute pancreatitis. Int. J. Mol. Med. 37, 99–107.

33.

Basso

D.M.

, Beattie

M.S.

, and Bresnahan

J.C.

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

34.

Rivlin

A.S.

and Tator

C.H.

(1977). Objective clinical assessment of motor function after experimental spinal cord injury in the rat. J. Neurosurg. 47, 577–581.

35.

Sandner

, Pillai

D.R.

, Heidemann

R.M.

, Schuierer

, Mueller

M.F.

, Bogdahn

, Schlachetzki

, and Weidner

(2009). In vivo high-resolution imaging of the injured rat spinal cord using a 3.0T clinical MR scanner. J. Magn. Reson. Imaging, 29, 725–730.

36.

Nout

Y.S.

, Miha

, Tovar

C.A.

, Schmalbrock

, Bresnahan

J.C.

, and Beattie

M.S.

(2009). Hypertonic saline attenuates cord swelling and edema in experimental spinal cord injury: a study utilizing magnetic resonance imaging. Crit. Care Med. 37, 2160–2166.

37.

Lee

S.A.

, Kwak

M.S.

, Kim

, and Shin

J.S.

(2014). The role of high mobility group box 1 in innate immunity. Yonsei Med. J. 55, 1165–1176.

38.

Huang

, Liu

, Yang

, Chen

, Eunice

, and Yuan

(2013). Acute hyperglycemia worsens ischemic stroke-induced brain damage via high mobility group box-1 in rats. Brain Res. 1535, 148–155.

39.

Zheng

, Liu

, Wang

, Tang

, Dong

, Zhou

, and Zhong

(2016). Ethanol extracts from Portulaca oleracea L. attenuated ischemia/reperfusion induced rat neural injury through inhibition of HMGB1 induced inflammation. Am. J. Transl. Res. 8, 5016–5024.

40.

Oshio

, Binder

D.K.

, Yang

, Schecter

, Verkman

A.S.

, and Manley

G.T.

(2004). Expression of aquaporin water channels in mouse spinal cord. Neuroscience, 127, 685–693.

41.

X.Q.

, Fang

, Tan

W.F.

, Wang

Z.L.

, Sun

X.J.

, Zhang

Z.L.

, and Ma

(2016). miR-320a affects spinal cord edema through negatively regulating aquaporin-1 of blood-spinal cord barrier during bimodal stage after ischemia reperfusion injury in rats. BMC Neurosci. 17, 10.

42.

Dibaj

, Kaiser

, Hirrlinger

, Kirchhoff

, and Neusch

(2007). Kir4.1 channels regulate swelling of astroglial processes in experimental spinal cord edema. J. Neurochem. 103, 2620–2628.

43.

Rao

K.V.

, Reddy

P.V.

, Curtis

K.M.

, and Norenberg

M.D.

(2011). Aquaporin-4 expression in cultured astrocytes after fluid percussion injury. J. Neurotrauma, 28, 371–381.

44.

Hemley

S.J.

, Bilston

L.E.

, Cheng

, Chan

J.N.

, and Stoodley

M.A.

(2013). Aquaporin-4 expression in post-traumatic syringomyelia. J. Neurotrauma, 30, 1457–1467.

45.

Okuma

, Liu

, Wake

, Zhang

, Maruo

, Date

, Yoshino

, Ohtsuka

, Otani

, Tomura

, Shima

, Yamamoto

, Takahashi

H.K.

, Mori

, and Nishibori

(2012). Anti-high mobility group box-1 antibody therapy for traumatic brain injury. Ann.Neurol. 72, 373–384.

46.

Sun

, Wang

, Li

, Hu

Y.C.

, Li

, Zhu

J.H.

, Zhou

, and Hang

C.H.

(2013). Glycyrrhizic acid confers neuroprotection after subarachnoid hemorrhage via inhibition of high mobility group box-1 protein: a hypothesis for novel therapy of subarachnoid hemorrhage. Med. Hypotheses, 81, 681–685.

47.

Iacovetta

, Rudloff

, and Kirby

(2012). The role of aquaporin 4 in the brain. Vet. Clin. Pathol. 41, 32–44.

48.

Rama Rao

K.V.

, Verkman

A.S.

, Curtis

K.M.

, and Norenberg

M.D.

(2014). Aquaporin-4 deletion in mice reduces encephalopathy and brain edema in experimental acute liver failure. Neurobiol. Dis. 63, 222–228.

49.

Nito

, Kamada

, Endo

, Narasimhan

, Lee

Y.S.

, and Chan

P.H.

(2012). Involvement of mitogen-activated protein kinase pathways in expression of the water channel protein aquaporin-4 after ischemia in rat cortical astrocytes. J. Neurotrauma, 29, 2404–2412.

50.

Thrane

A.S.

, Rappold

P.M.

, Fujita

, Torres

, Bekar

L.K.

, Takano

, Peng

, Wang

, Rangroo Thrane

, Enger

, Haj-Yasein

N.N.

, Skare

, Holen

, Klungland

, Ottersen

O.P.

, Nedergaard

, and Nagelhus

E.A.

(2011). Critical role of aquaporin-4 (AQP4) in astrocytic Ca2+ signaling events elicited by cerebral edema. Proc. Natl. Acad. Sci. U. S. A. 108, 846–851.

51.

Y.Q.

, Chen

Y.R.

, Leng

Y.F.

, and Wu

Z.W.

(2014). Tanshinone IIA Downregulates HMGB1 and TLR4 Expression in a spinal nerve ligation model of neuropathic pain. Evid. Based Complement Alternat. Med. 2014, 639563.

52.

Wang

, Zhang

, Liu

, Yang

, Cui

, and Li

(2010). Atorvastatin protects rat brains against permanent focal ischemia and downregulates HMGB1, HMGB1 receptors (RAGE and TLR4), NF-kappaB expression. Neurosci.Lett. 471, 152–156.

53.

Sun

, Li

, Ma

, Feng

, Song

, Lv

, and He

(2017). Inhibition of HMGB1 reduces rat spinal cord astrocytic swelling and AQP4 expression after oxygen-glucose deprivation and reoxygenation via TLR4 and NF-κB signaling in an IL-6-dependent manner. J. Neuroinflammation, 14, 231.