Systemic Administration of Connexin43 Mimetic Peptide Improves Functional Recovery after Traumatic Spinal Cord Injury in Adult Rats

Abstract

Blocking of Connexin43 hemichannels, the main gap junction protein located on astrocytes in the central nervous system, has been shown to reduce neural injury in a number of models. We demonstrated previously that local administration of a Connexin43 mimetic peptide, Peptide5, reduces secondary tissue damage after spinal cord injury (SCI). Here, we investigated whether acute systemic delivery of Peptide5 is also protective in a model of SCI. Rats were subjected to a mild spinal cord contusion using the Multicentre Animal Spinal Cord Injury Study impactor and were injected intraperitoneally with Peptide5 or a scrambled peptide immediately and at 2 h and 4 h post-injury. Rats were tested for locomotor recovery and pain hypersensitivity and euthanized at 8 h, 24 h, two weeks, or six weeks post-injury. Compared with control rats, Peptide5 treated rats showed significant improvement in hindlimb locomotor function between three and six weeks post-injury and reductions in at-level mechanical allodynia at weeks one and six post-injury. Immunohistochemistry showed that Peptide5 treatment led to a reduction in total Connexin43 and increased phosphorylated Connexin43 at 8 h compared with scrambled peptide. At two and six weeks, lesion size, the astrocytic and the activated macrophage, and/or microglial response were all decreased in the Peptide5 animals. In addition, neuronal cell numbers were higher in the Peptide5 animals compared with the scrambled peptide treated rats at two and six weeks. These results show for the first time that systemic administration of Peptide5 to block the pathological opening of Connexin43 hemichannels is a feasible treatment strategy in this setting, ameliorating the secondary SCI.

Introduction

Traumatic spinal cord injury (SCI) arising from an initial mechanical insult not only disrupts function of the injured segment but also leads to a series of secondary events resulting in an on-going spread of tissue damage. Secondary injuries include progressive ischemia with free radical formation and excitotoxicity,^1
–3 followed by inflammatory cell activation, demyelination, axonal degeneration, neuronal death, cavitation, reactive astrogliosis surrounding the lesion area, and both growth promoting and inhibitory factor release.^4

–7 Secondary injury limits the ability of the spinal cord for self-regeneration^,4,8
–10 and it is therefore an ideal candidate target for potential therapies.

Gap junctions are specialized intercellular connections found in a multitude of animal species that enable direct and facilitated passage of various ions and small cytoplasmic molecules between two adjacent cells.^11

–14 A gap junction consists of two hexameric hemichannels inserted in the cell membrane, and each hemichannel is composed of six oligomerized connexin proteins.¹⁵ Connexin hemichannels from adjacent cells are able to dock with each other to form functional gap junction channels.^16,17 In the spinal cord, Connexin43 is the major protein of both gap junctions and unopposed hemichannels on astrocytes and endothelial cells.

In pathological conditions, gap junction-mediated communication is involved in secondary cell death. Gap junctions have been considered to transmit neurotoxins including glutamate, nitric oxide, lactate, arachidonate, ammonia, and reactive oxygen species, as well as calcium waves from the initial injured cells to the adjacent healthy cells, known as the bystander effect.^18
–20 More recently, the opening of uncoupled hemichannels has been considered the principle perpetuator of secondary injury. Hemichannel opening facilitates the release of neurotoxins and adenosine triphosphate (ATP) and establishes calcium waves, and this can then affect surrounding intact cells.^21

–25

Connexin43 mRNA and protein levels are elevated after SCI^19,26

–29 and can be reduced using antisense oligodeoxynucleotides or gene knock-out to improve functional recovery.^26,30
–32 Mimetic peptides have also been used in experiments to modulate both gap junction coupling and hemichannel opening.^33

–37 These are short peptide sequences designed to match portions of the extracellular loops of the connexin proteins. Both gap junctions and hemichannels can be regulated by connexin mimetic peptides independently of each other depending on the concentration and incubation time.^28,36,38

A series of Connexin43 mimetic peptides has been designed against the two extracellular loops of Connexin43 protein, and one of these, referred to as Peptide5 (sequence VDCFLSRPETK), shows particular efficacy in regulating both Connexin43 gap junctions and hemichannels in a dose dependent manner²⁸ and improves cellular and functional recovery when delivered directly to a spinal cord contusion²⁹ without appearing to affect other connexin isoform gap junctions or hemichannels. Peptide5 has also been shown to be neuroprotective in retinal ischemia,^39
–41 cerebral ischemia,⁴² and an epileptiform lesion model.³⁶ Additional mimetic peptides against Connexin43 protein^43

–48 have also been used for other indications but not for SCI repair. It is also possible that other pharmacological agents may be developed to regulate downstream effects of Connexin43 hemichannel opening²⁴; for example, a P2X2 receptor agonist has been used to block ATP released through these hemichannels.⁴⁹

Importantly, it is feasible for Peptide5 to reach injury sites systemically where the capillary bed has been compromised in the injured spinal cord, thus offering a practical therapeutic approach for people with SCI. In the current study, we have focused our investigation on determining whether systemic delivery of Peptide5, rather than local delivery, influences Connexin43 expression and recovery after SCI. We demonstrate that Peptide5 also improves cellular and functional outcomes when delivered systemically, as well as reducing chronic pain in an in vivo contusion model of rat SCI.

Methods

Experimental groups

There were 104 adult female Sprague-Dawley rats weighing 250–300 g used (Animal Resource Centre, Perth, Australia). All experimental protocols and procedures used in the study were approved by the Animal Care and Ethics Committee of the University of Technology Sydney, in accordance with the guidelines of the National Health and Medical Research Council of Australia. Rats were anesthetized with 4% isofluorane in oxygen (1 L/min) and then maintained at 2% isofluorane in oxygen (1 L/min). They were then placed on a heated pad and their body temperature was monitored throughout the surgery.

Anesthetic (bupivacaine, 0.5%, 0.02 mL) was injected locally at all surgical sites before incisions were made. Skin and muscle layers were incised over the dorsal midline and retracted, and a T10 laminectomy was performed to expose the spinal cord. The Multicentre Animal Spinal Cord Injury Study impactor device was used to produce a mild contusion injury (6.25 mm drop, 10 g weight) to the cord at T10. The mild SCI was chosen for this study because it results in less tissue disruption compared with more severe injuries but still has measurable locomotor and sensory deficits. The surgical wound was then sutured closed in layers. Control rats (Shams) had laminectomies but were left uninjured.

Peptide treatment

The dose of peptide used was to give a blood concentration of 100 μM, as shown to be effective in our previous study.⁴¹ The following dosing regimen was used to maintain this concentration based on an expected half-life of the about 2 h in rat serum. The rats were administered intraperitoneally three doses of Peptide5 or scrambled peptide (SP) at 0 h (10 mg/kg, 0.5 mL), 2 h (5 mg/kg, 0.5 mL), and 4 h (2.5 mg/kg, 0.5 mL) after surgery. Peptide5 was designed to match a portion of the extracellular loop 2 of Connexin43 (sequence VDCFLSRPTEKT) as described previously.²⁸ A scrambled control peptide with the same amino acids but in a different order was designed as a control treatment to Peptide5 (sequence RFKPSLCTTDEV). Both peptides were solid phase synthesized by Mimotopes Pty Ltd., Australia, using Fmoc chemistries on a Protein Technology, Symphony instrument, purified by high-performance liquid chromatography and the structure confirmed by mass spectrometry analysis.

Post-operative care

Animals were housed individually for seven days post-surgery and then rehoused in groups of four. They were kept in a holding room with controlled temperature (22–25°C), humidity, and light cycle (12 h on/12 h off) and had access to food and water ad libitum. Post-operatively, animals were administered subcutaneously antibiotics (cephalothin sodium, 0.33 mg/kg,), analgesic (buprenorphine hydrochloride, 0.03 mg/kg), and supplementary fluid (Hartman replacement solution, 3.8 mL) twice daily for three days. Antibiotics were continued until urine was clear. Bladders were expressed manually twice daily for 1–2 weeks until the normal voiding response returned.

Behavioral assessments

After the surgical procedures, each animal was assessed for functional recovery using an open field and a horizontal ladder testing paradigm. Animals were tested at days 1, 3, 7, 14, 21, 28, 35, and 42 of the post-operative recovery period. Hindlimb locomotor function was assessed using the Basso-Beattie-Bresnahan (BBB) Locomotor Rating Scale for the open-field test⁵⁰ and by calculating the percentage of error hindlimb steps for the horizontal ladder test.⁵¹ All tests were digitally recorded, and assessments were performed by two independent reviewers who were blinded to the groups. The scores from the two reviewers had a high correlation (R² = 0.879) without statistically significant differences (p = 0.480), so the average scores were used for all further analysis.

Mechanical hypersensitivity tests were conducted a week before the surgical injury to determine the baseline level and then once weekly after surgery, for up to six weeks. The rats were placed on an elevated wire mesh platform in individual Perspex cubicles and were allowed to acclimatize to the pain testing apparatus for 40 min in a quiet and temperature controlled room. Mechanical hypersensitivity was measured at level (injury site) and below level (hindpaws) using a series of calibrated von Frey filaments with incremental stiffness ranging from 0.02 g to 60 g. The von Frey filaments were applied serially according to the up-down method, and the 50% withdrawal threshold was calculated, as described previously.⁵²

At-level pain was tested via application of the von Frey filaments, three times on both the medial and lateral side, approximately 3 cm away from the injury site. Below-level pain was tested via application of the von Frey filaments to the plantar surface of both hindpaws. Positive pain responses included licking, flinching, vocalization, and lifting of the leg in the case of below-level pain. Responses that occurred independent of the filament were not recorded.

Immunohistochemistry

At 8 h (n = 8), 24 h (n = 32), two weeks (n = 32), and six weeks (n = 32) after surgery, animals were euthanized with sodium pentobarbital intraperitoneally (100 mg/kg) and perfused intracardially with 200 mL of heparinized saline followed by 400 mL 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4). Approximately 20 mm of spinal cord centered at T10 was dissected, removed, and post-fixed in 4% paraformaldehyde overnight and then transferred to 30% sucrose. Spinal cords were sectioned horizontally (from the dorsal to ventral side) on a cryostat at 15 μm onto Matsunami Platinum PLC-14 and 2% gelatine-coated glass slides alternatively. Sections were mounted in a series of 10 so that each section on a slide was at 150 μm intervals. Mayer's hematoxylin and eosin (H&E) staining was performed to assess tissue morphology and determine the lesion area.

Immunohistochemistry was performed by first incubating slides in 5% normal goat serum (NGS) in phosphate buffered saline with Triton X-100 at pH 7.4 (PBST) for 30 min. Before the primary antibody incubation, heat-induced antigen retrieval was performed in citrate buffer (10 mM citric acid, pH 6.0) for 10 min in a pressure cooker for NeuN staining instead of NGS blocking. Primary antibodies were diluted in phosphate buffer with 5% normal goat serum (PBG) and incubated overnight at 4°C.

For the early time points (8 h and 24 h), primary antibodies used were rabbit anti-Connexin43 (1:2000, Sigma-Aldrich, Germany) to show levels of Connexin43 and rabbit anti-phosphorylated Connexin43 (1:100, Santa Cruz, Texas). This second antibody recognizes Connexin43 phosphorylation on Ser368, which has been shown to change Connexin43 hemichannels from an open to a closed state.^29,53 For two and six weeks, primary antibodies used were rabbit anti-glial fibrillary acidic protein (GFAP) (1:1000, Dako, Denmark) to label astrocytes, mouse anti-neuronal nuclei (NeuN) (1:1000, Millipore, Germany) to label neurons, mouse anti-CD68 (ED1) (1:2000, AbD Serotec, UK), and rabbit anti-ionized calcium-binding adapter molecule 1 (Iba1) (1:1000, Abcam, UK) to demonstrate activated microglia and macrophages.

Slides were then washed in PBST and incubated with goat anti-rabbit Alexa Fluor 488 (Connexin43, GFAP and Iba1; 1:200, Life Technologies), goat anti-rabbit Alexa Fluor 568 (phosphorylated Connexin43; 1:200, Life Technologies), goat anti-mouse Alexa Fluor 488 (NeuN; 1:200, Life Technologies), or goat anti-mouse Alexa Fluor 568 (ED1; 1:200, Life Technologies) in PBG for 2 h at room temperature. Slides were washed with PBST and counterstained by Hoechst (1:5000, Invitrogen) for 10 min. Primary antibody was omitted from negative controls. All slides were coverslipped in fluorescent mounting medium (Dako, Denmark).

Image analysis

Imaging was performed using an Olympus BH-2 light microscope with a PixeLINK Pl-A662 camera (H&E staining) or Olympus BX-51 light microscope with an Olympus U-RFL-T fluorescence burner and appropriate filters (immunohistochemistry staining). Injury on H&E stained sections was observed as hemorrhage, axonal swellings, cells with no nuclei or pyknotic nuclei, areas of cellular debris, or cavity formation. Sections were imaged at 150 μm intervals through the dorsal-ventral plane, and the injury area was measured on each section using Image J software. The lesion size was then calculated for each spinal cord as an approximate volume using these measurements at the lesion center and 3.5 and 7 mm distal to the lesion center.

Spinal cord sections were aligned using the central canal as an anatomical landmark for immunohistochemistry analysis. Images were taken adjacent to the lesion and at both 3.5 and 7 mm distal to the lesion center in both rostral and caudal directions.

Image analysis was performed using ImageJ software (National Institutes of Health, version 2X). Analysis of Connexin43, p-Connexin43, and astrogliosis was undertaken by measuring the staining intensity (mean grey scale value) in 10x images of midline sections. All slides were stained at the same time, and the images used for quantification were taken at the same microscope settings. In addition, the background black balance was normalized for each image to minimize any variation between slides.

For activated microglia quantification, all cells were identified with nuclear Hoechst stain and co-labeled with both ED1 (phagocytic macrophages) and Iba1 (microglia), counted manually in 40x images of midline sections, and reported as number of cells/field of view. Neurons were identified using NeuN and counted manually in 40x images of midline sections and in sections at 300 μm dorsal and ventral to the midline and are reported as number of cells/field of view.

Statistics

Statistical analysis was performed using Graphpad Prism (GraphPad Software, Inc, version 6). Data are expressed as mean ± standard error of the mean and analyzed using unpaired t tests, two-way analysis of variance (ANOVA) for the spinalized rats only with “treatment” and “time” as the two factors, two-way ANOVA at the different time points with “treatment” and “injury status (sham or SCI)” as the two factors. Finally, all the behavior data were analyzed using two-way ANOVA with treatment and injury status as the two factors, and using t tests to compare the two groups at each time point. Differences were considered to be statistically significant at p < 0.05.

Results

Surgery

All surgical procedures and intraperitoneal peptide injections were conducted without incident. None of the rats undergoing sham surgeries showed any indications of SCI or any post-operative complications. The impactor parameters for the SCI rats in the Peptide5 (n = 28) and SP (n = 28) groups were not significantly different (p > 0.05) for impact velocity (0.319 ± 0.004 m/sec vs. 0.319 ± 0.04 m/sec), compression (1.502 ± 0.025 mm vs. 1.505 ± 0.025 mm), or compression rate (0.281 ± 0.004 m/sec vs. 0.281 ± 0.004 m/sec). After SCI, rats showed immediate loss of locomotor function that gradually improved over the study period consistent with a mild SCI. Bladder function and weight gain returned to normal within one week to 10 days for all rats.

Connexin43 and phosphorylated Connexin43 at 8 h post-injury

Analysis of Connexin43 protein and phosphorylated Connexin43 protein (closed hemichannels) was undertaken at 8 h post-injury using immunohistochemistry on the midline sections. Connexin43 expression, as measured by gray scale values of immunohistochemistry labeling intensity, were lower in the Peptide5 (n = 4) treated animals compared with the SP (n = 4) control group in both gray (63.35 ± 6.68 vs. 49.51 ± 2.69) (p < 0.01) and white matter (46.91 ± 5.43 vs. 28.41 ± 2.57) (p < 0.001) (Fig. 1A, 1B). In contrast, there was an increase in the level of Connexin43 phosphorylation at 8 h after injury in the Peptide5 treated animals (n = 4) compared with the animals receiving SP control treatment (n = 4) with an increase in overall labeling intensity from 30.93 ± 2.53 to 48.61 ± 1.71 in gray matter (p < 0.001) and from 15.75 ± 1.96 to 24.90 ± 1.28 in white matter (p < 0.001) (Fig. 1C, 1D).

FIG. 1.

Changes in Connexin43 and phosphorylated Connexin43-Ser368 levels at 8 h after spinal cord injury (SCI). (A) At 8 h after SCI, images of Connexin43 immunofluorescence (green) at the lesion center in Peptide5 (P5) or scrambled peptide (SP) treated animals. Scale bar = 10 μm. (B) Immunohistochemistry analysis showing changes in Connexin43 levels in both gray (GM) and white matter (WM). Systemic administration of P5 limited the increase in post-injury Connexin43 protein level (t test, **p < 0.01, ***p < 0.001). (C) At 8 h after SCI, images of Connexin43 phosphorylation at residue Ser368 immunofluorescence (red) at the lesion center in P5 or SP treated animals. Scale bar = 10 μm. (D) Immunohistochemistry analysis showing changes in Connexin43 phosphorylation at residue Ser368. Systemic administration of P5 significantly increased phosphorylation of Connexin43 at residue Ser368 (t test, ***p < 0.001). GSV, gray scale value. Color image is available online at www.liebertpub.com/neu

Lesion size

There was no histological evidence of any tissue damage in the sham animals at any time point. For the SCI groups, there was an obvious lesion apparent from 24 h post-injury that increased in size and changed in morphology over the six week period. At the earlier time points, there was more hemorrhage visible, and over time this changed to include more cell loss and increased cellular debris before development of a clearly defined central cavity. Lesion size was significantly reduced by the systemic administration of Peptide5 in comparison with the SP control treatment at all time points post-injury.

Longitudinal H&E stained sections including the lesion are shown for two weeks (Fig. 2A) and six weeks (Fig. 2B) after SCI. There were significant differences between treatment groups (p < 0.001) but not over time using two-way ANOVA tests. At two weeks post-SCI, the lesion was 6.72 ± 1.24 mm³ in the SP control group (n = 8), while the Peptide5 treatment (n = 8) was able to limit the lesion size to 2.58 ± 0.89 mm³ (p < 0.01) (Fig. 2C). By six weeks post-injury, the lesion was found to be 7.55 ± 1.55 mm³ in the SP treated animals (n = 8) while the lesion in the animals receiving Peptide5 treatment (n = 8) was limited to 3.60 ± 0.30 mm³ (p < 0.01) (Fig. 2C).

FIG. 2.

Longitudinal sections at the midline level of the spinal cord stained with Mayer hematoxylin and eosin showing the lesion at (A) two weeks and (B) six weeks after contusion spinal cord injury (SCI). There were no lesions seen in the sham groups. Scale bar = 1 mm. (C) Graph showing that lesion size was reduced in the spinal cord injured rats by the Peptide5 (P5) treatment compared with the scrambled peptide (SP) at both two weeks and six weeks post-injury (two-way analysis of variance **p < 0.01 for treatment only). There were no differences seen in either treatment group with time. Color image is available online at www.liebertpub.com/neu

Astrogliosis (GFAP immunohistochemistry)

Astrogliosis (astrocyte activation) is known to be a key response after SCI and ultimately results in the formation of a glial scar starting from approximately one week. GFAP immunohistochemistry sections at the midline level were used to evaluate the effect of Peptide5 on astrogliosis after SCI (Fig. 3A for representative sections at two weeks post-surgery). The GFAP staining intensity at the lesion edge was quantified in mean gray scale value for statistical analysis (Fig. 3B, 3C).

FIG. 3.

Glial fibrillary acidic protein (GFAP) expression at the lesion edge after contusion spinal cord injury (SCI). (A) Representative images taken at two weeks post-injury showing GFAP immunofluorescence adjacent to the lesion in sham and injury animals treated with Peptide5 (P5) or scrambled peptide (SP). Scale bar = 50 μm. Quantification of GFAP immunofluorescence intensity in mean gray scale value (GSV) at (B) two weeks and (C) six weeks. The level of GFAP immunofluorescence intensity was increased after SCI, and the systemic administration of P5 led to a significant reduction in the level of GFAP immunofluorescence intensity adjacent to the lesion. Two-way analysis of variance ***p < 0.001 for treatment and injury status. Color image is available online at www.liebertpub.com/neu

The normal patterns and intensity of astrocyte staining were seen in the sham spinal cords, and these measurements were used as the baseline for comparison for injury and for treatment. The level of GFAP immunoreactivity in the SP control animals at two and six weeks post-injury (n = 8 at each time point) was increased 2.1 and 3.3 times, respectively, compared with the sham groups (n = 8 at each time point) (two weeks, p < 0.001; six weeks, p < 0.001 for injury status). The Peptide5 treatment resulted in a significantly reduced level of GFAP immunoreactivity compared with SP treated animals with SCI at both post-injury time points (n = 8 at each time point) (two weeks, p < 0.001; six weeks, p < 0.001 for treatment) with levels of 1.5 and 2.4 fold decrease compared with that seen in sham animals, respectively (n = 8 at each time point).

Microglial/macrophage activation (Iba1 & ED1 immunohistochemistry)

One of the events that occurs during secondary SCI is the activation of microglial cells and the recruitment and activation of macrophages. These cells are involved in the inflammatory cascades that follow injury and are generally thought to be detrimental to recovery. Activated microglia and/or macrophages were identified using Hoechst (nuclear staining) and double labeling with Iba1 and ED1 (Fig. 4A for representative sections at two weeks post-surgery). Because activated microglia and macrophages share similar cellular markers and switch their genotypes and phenotypes frequently, it is difficult to distinguish between them. We therefore considered the Iba1⁺/ED1⁺ cells in the injured spinal cords as activated microglia and/or macrophages.

FIG. 4.

Activation of microglia and/or macrophages at the lesion edge after contusion spinal cord injury (SCI). (A) Representative images taken at two weeks post-injury showing co-localization of Iba1 (green) and ED1 (red) adjacent to the lesion in sham and injury animals treated with Peptide5 (P5) or scrambled peptide (SP). Scale bar = 10 μm. Quantification of Iba1⁺/ED1⁺ microglia and/or macrophages (arrows) at (B) two weeks and (C) six weeks. The number of activated microglia and/or macrophages was increased after SCI, and the systemic administration of P5 led to a significant reduction in the number of activated microglia and/or macrophages adjacent to the lesion (two-way analysis of variance **p < 0.01 ***p < 0.001 for treatment and injury status).

The activated microglia and/or macrophages were observed in all areas of both gray and white matters, and the quantification of midline sections was conducted at the lesion edge where the cells double labeled with Iba1 and ED1 appeared in the greatest number (Fig. 4B and C). There were no Iba1⁺/ED1⁺ cells in any of the sham surgery spinal cords. In the SCI rats treated with SP (n = 8 at each time point), the number of Iba1⁺/ED1⁺ cells increased to 95 ± 6 cells/field of view at two weeks (p < 0.001) and 68 ± 1 cells/field of view at six weeks (p < 0.001 for injury status). Peptide5 treatment (n = 8 at each time point) reduced the number of Iba1⁺/ED1⁺ cells at both two weeks (20 ± 4 cells/field of view, p < 0.001) and at six weeks (24 ± 1 cells/field of view, p < 0.001 for treatment).

Neuronal survival (NeuN immunohistochemistry)

A significant loss of neuronal cells during the secondary phase of SCI is known to be a consequence of inflammatory responses including microglial/macrophage activation. Because Peptide5 treatment via systemic delivery resulted in a reduced level of activated microglia and/or macrophages in the secondary phase, it was important to determine whether neuronal survival was enhanced; for this, we used NeuN immunohistochemistry (Fig. 5A for representative sections at two weeks post-surgery). The surviving neuronal nuclei were identified by NeuN positive staining and then quantified at 7 mm to the lesion center at the midline level (Fig. 5B, 5C). There were no significant differences between the cell counts at 7 mm rostral and caudal to the lesion center using the paired t test; therefore, the mean of rostral and caudal measurements for each sample was used for further analysis.

FIG. 5.

Surviving neuronal cells after contusion spinal cord injury (SCI). (A) Representative images taken at two weeks post-injury showing NeuN immunofluorescence 7 mm rostral to the lesion at the midline level in sham and injury animals treated with Peptide5 (P5) or scrambled peptide (SP) treated control. Scale bar = 10 μm. Quantification of NeuN⁺ neurons (arrows) at (B) two weeks and (C) six weeks. The number of spared neurons was decreased after injury, but the systemic administration of P5 led to a significant increase in the number of spared neurons 7 mm from the lesion (two-way analysis of variance ***p < 0.001 for treatment and injury status). Color image is available online at www.liebertpub.com/neu

Neuronal cell number and distribution was not altered over time in the sham animals (36 ± 0 cells/field of view). As anticipated, the number of neurons per field of view in the SP treated animals (n = 8 at each time point) significantly decreased to 15 ± 1 cells/field of view at two weeks (p < 0.001) and 11 ± 1 cells/field of view by six weeks (p < 0.001 for injury status) post-injury. In Peptide5 treated animals (n = 8 at each time point), this number was retained at 32 ± 1 cells/field of view at two weeks (p < 0.001) and 24 ± 0 cells/field of view by six weeks (p < 0.001 for treatment) post-injury compared with the injured animals that received SP control treatment (n = 8 at each time point).

The number of neurons in the dorsal horn showed a similar pattern to that at the midline level, but the neurons in the ventral horns did not decrease significantly after this mild contusion injury at the dorsal level of the spinal cord (data not shown). Therefore, the Peptide5 treatment via systemic delivery was able to promote neuronal survival during the secondary phase of SCI.

Motor function (open field and error ladder)

We used two behavioral assessments—open field and horizontal ladder tests—to determine whether Peptide5 treatment led to an improvement in functional recovery. The animals with SCI treated with Peptide5 (n = 8) showed a significant improvement in hindlimb locomotor function in both tests from week three compared with the injury animals treated with SP (n = 8) (Fig. 6A, 6B). There was no deficit in hindlimb locomotion in any of the sham animals regardless of treatment (n = 8 for each treatment) for both behavioral assessments.

FIG. 6.

Behavioral assessments of hindlimb locomotor function after contusion spinal cord injury (SCI). (A) The Basso-Beattie-Bresnahan (BBB) scores are shown for the Peptide5 (P5) treated animals compared with the scrambled peptide (SP) treated animals. There were significant differences seen in BBB scores using two-way analysis of variance (ANOVA) for both the treatment (p < 0.001) and for time (p < 0.001). P5 treated rats showed improvement in BBB scores from week 3 to week 6 (*p < 0.05, ***p < 0.001). (B) The percentage of hindlimb error steps is shown for the P5 treated animals compared with the SP treated animals. There were significant differences seen in the percentage of errors using two-way ANOVA for both the treatment (p < 0.001) and for time (p < 0.001). P5 treated rats showed improvement in stepping from week 3 to week 6 (*p < 0.05, **p < 0.01). (C) Long-term effects of P5 on mechanical pain hypersensitivity after contusion SCI on at-level pain. Mechanical withdrawal thresholds are shown for P5 treated animals (compared with the SP treated animals). There were significant differences seen in the mechanical withdrawal thresholds using two-way ANOVA for both the treatment (p < 0.05) and for time (p < 0.001). P5 treated rats showed improvement at week 1 and week 6 (*p < 0.05, **p < 0.01). (D) Long-term effects of P5 on mechanical pain hypersensitivity after contusion SCI on below-level pain. There were no significant differences in paw withdrawal thresholds between P5 and SP treated animals.

In the open field test, injury groups displayed improvements in the BBB scores from day one to week six (SCI/Peptide5, 1.44 ± 0.50 to 15.81 ± 0.76, p < 0.01; SCI/SP, 1.00 ± 0.67 to 12.00 ± 0.41, p < 0.01) as expected after a mild contusion SCI (Fig. 6A), but Peptide5 treated injury animals showed improved hindlimb locomotor function when compared with the SCI animals treated with SP with significant differences observed from week three (13.56 ± 1.14 vs. 11.00 ± 0.57, p < 0.05) to week six (15.81 ± 0.76 vs. 12.00 ± 0.41, p < 0.001) at which time point behavioral assessment ceased.

Similar results were obtained with a horizontal ladder test (Fig. 6B). All injury animals with Peptide5 or SP treatments exhibited a reduction in the percentages of error steps from week one to six (SCI/Peptide5, 47.20 ± 11.68% to 9.64 ± 1.66%, p < 0.001; SCI/SP, 61.09 ± 10.02% to 32.04 ± 6.58%, p < 0.001). The injury animals treated with Peptide5, however, displayed fewer hindlimb error step percentages compared with those that received the SP control treatment with significant differences seen from week three (21.65 ± 4.25% vs. 49.72 ± 9.42%, p < 0.05) to week six (9.64 ± 1.66% vs. 32.04 ± 6.58%, p < 0.01). Differences in the percentages of error steps at weeks one and two were seen between Peptide5 and SP treated injury animals; however, these differences were not statistically significant (p > 0.05).

Sensory function (mechanical allodynia)

Spinal astrogliosis is well regarded as a contributor to the development of neuropathic pain states via the release of small excitatory molecules through opened hemichannels after SCI.⁵⁴ Therefore, we investigated whether treatment with Peptide5 alleviates mechanical pain hypersensitivity after SCI. At-level hypersensitivity demonstrated by reduced withdrawal thresholds after the injury developed in rats with SCI. Compared with SP treated controls (n = 16), treatment with Peptide5 (n = 16) significantly improved at-level pain with a 41% increase in the withdrawal threshold at week one (p < 0.01) and a 27% increase in the withdrawal threshold at week 6 (n = 8) (p < 0.05) after SCI (Fig. 6C). Below-level mechanical pain hypersensitivity was also measured on the hindpaws (Fig. 6D). There was no significant difference in paw withdrawal thresholds between the treatment groups during the course of the six weeks, however (p > 0.05) (Fig. 6D).

Discussion

We have demonstrated previously that the local application of Peptide5 at a low concentration within the first 24 h after traumatic SCI reduces tissue swelling, lesion spread, inflammation, astrogliosis, neuronal cell death, and functional deficits in ex vivo and in vivo models.^28,29 Acute administration of Peptide5 at low doses has also been shown to be neuroprotective in models of retinal and cerebral ischemia.^{39
–41,55
–57} The local administration of Peptide5, however, is not practical for use in spinal patients because this would necessitate a highly invasive surgical procedure to expose the injury area or involve another procedure such as inserting an intrathecal catheter.

Upregulation of Connexin43 has been demonstrated early after SCI.^19,27,29 Therefore, application of Peptide5 should occur within 8 h, before maximal upregulation of Connexin43 protein, to be most effective. Unfortunately, many patients with SCI arrive in emergency departments with traumatic co-morbidities or life-threatening injuries and a timely spinal cord procedure is not always possible. If the acute clinical management of a patient with SCI includes an open surgical procedure such as decompression or stabilization, there may be an avenue for application of either an Connexin43 antisense oligodeoxynucleotide in a topical gel²⁶ or a direct application of Peptide5.²⁹ Unfortunately, in most cases, these procedures do not occur early enough for effective regulation of Connexin43 protein.

Because mimetic peptides have been shown to permeate the lesion site of spinal cords through the disrupted blood spinal cord barrier,⁵⁸ vascular administration of Peptide5 is a suitable approach for early delivery to the lesion. Our findings demonstrate the neuroprotective effects of Peptide5 via this systemic delivery is in agreement with our previous studies using direct peptide administration.^28,29

Effect of Peptide5 on Connexin43 protein in the spinal cord

The increase in total Connexin43 protein level after injury was suppressed with administration of Peptide5 and phosphorylated Connexin43 was higher in the spinal cords at 8 h after Peptide5 administration. There is substantial evidence to indicate an increase of Connexin43 protein after brain ischemia^59

–63 and SCI.^19,26

–29 Down-regulation of connexin or blocking gap junction channels using gap junction blockers has been shown to be beneficial.^26,63
–65 Other studies using gene knockout models have demonstrated conversely an increase in lesion size resulting from complete loss of Connexin43 stopping normal gap junctional communication necessary for spatial buffering and neuronal survival at the later stages of recovery.^66
–68

The exact mechanisms of Peptide5 are still being elucidated, but at the concentration used in this study, it is acting to block Connexin43 hemichannels rather than affecting intact gap junctions.²⁸ It is thus logical to assume that Peptide5 is acting by blocking Connexin43 hemichannels leading to a reduced propagation of neurotoxins to the surrounding healthy cells and extracellular matrix.^{29,40,41,55,69}

Neither the sham animals treated with Peptide5 nor the animals treated with SP showed any evidence of adverse effects in this study or in other studies using Peptide5.^41,57 We do not think that Peptide5 will affect the uninjured tissue nor should it act globally on gap junctions in other organs at the concentration we are administering. Certainly, higher doses may prove deleterious, and this has been shown to be the case in a fetal model of ischemia in sheep.⁶² While long-term effects are unknown at this stage and future studies will have to determine whether Peptide5 treatment has any effects on the levels of other connexins in the injured spinal cord, only transient block is required to obtain the improved outcomes described.

Targeted tissue effects of Peptide5 in injured spinal cord tissue

Injured spinal cords treated with Peptide5 had smaller lesions, decreased astrogliosis, less inflammatory cell infiltration and activation, and increased neuron survival, all indicating reduced effects of secondary injury and lesion spread. There is increasing evidence to suggest that the uncontrolled opening of Connexin43 hemichannels causes release of neurotoxins and ATP from activated microglia and/or macrophages, thus propagating secondary injury.^22

–25,37 In addition, many of these neurotoxins including glutamate and reactive oxygen species have been shown to increase the opening of Connexin43 hemichannels,^{21,22,70
–72} thus forming a positive feedback loop. We show here that blocking Connexin43 hemichannels with Peptide5 halts the amplification of injury and reduces a number of downstream effects that would normally be part of the secondary injury cascade, including astrogliosis.

Normally, astrogliosis will commence in the rat spinal cord from about three days after injury and lead to glial scar formation, a major physical and chemical barrier to neural regeneration.^73

–77 The physical barrier acts to prevent remyelination and regrowth of axons,^78
–80 while the chemical barrier adds to the nerve regrowth inhibitory microenvironment by producing a number of molecules such as chondroitin sulfate proteoglycans, tenascins, and bone morphogenetic proteins.^{4,81

–86} The reactive astrocytes in astrogliosis are believed to be adverse for neuroregeneration by inhibiting axonal regeneration and oligodendrocyte differentiation from precursor cells, leading to remyelination failure, as well as generating and secreting nitric oxide, a neurotoxin that damages local neural cells contributing to the secondary injury.^{74,84,87
–89}

Phagocytes found at the lesion area of CNS injury arise from two sources: resident microglia proliferating in situ and infiltrating macrophages recruited from the bloodstream. Both become activated and remain for several weeks after injury in response to the pro-inflammatory environment at the site of injury.^90

–93 With the systemic administration of Peptide5, there was a decrease in the number of activated microglia and/or macrophages present in or adjacent to lesions. Studies have shown that activated microglia and/or macrophages are involved in an acute inflammatory response to neurotrauma, such as synthesis and release of cytokines and chemokines.^94

–99 These pro-inflammatory cytokines, such as tumor necrosis factor-α (TNF-α), interleukin (IL)-1ß, and IL-6 are believed to cause neuronal and glial toxicity.^1,99

–102

The activated microglia and/or macrophages have been reported to increase Connexin43 expression and gap junction coupling in response to a trauma in the CNS.¹⁰³ The decrease in activated microglia and/or macrophages in Peptide5 treated spinal cords may not only be a downstream consequence of a decrease in astrocytic Connexin43, but may also result from Connexin43 hemichannel closure blocking pro-inflammatory signaling molecule release from the activated microglia and/or macrophages themselves.

Death of neuronal cells after CNS injury is thought to be a consequence of a complex cascade of events involving uncontrolled opening of Connexin43 hemichannels, spread of inflammation, and release of neurotoxic molecules.^12,22,104 There is growing evidence to suggest that a number of neurotoxins are synthesized and released from the activated microglia and/or macrophages in response to injury, including glutamate, nitric oxide, pro-inflammatory cytokines, ATP, and prostaglandins.^{31,101,102,104
–106} These neurotoxic molecules enhance the opening of astrocytic Connexin43 hemichannels, resulting in a further release of neurotoxins from the injured astrocytes.^107,108 Thus, there is a positive feedback loop developed in the lesion area to decrease the neuroprotective functions of astrogliosis, leading to an increase of neuronal cell death via the activation of amyloid-β peptide^{25

–35} and pannexin 1 hemichannel pathways.^107,109

Behavioral improvements

Most importantly, behavioral assessments over a six-week period after injury showed significant improvement in hindlimb locomotor ability in the Peptide5 treated animals. Given the time frame of the current study, we believe this represents a reduction in tissue damage that translates directly to a reduction in the degree of functional deficit seen after injury. By three weeks after injury, the Peptide5 treated animals reached an average BBB score of 14, which corresponds with consistent weight-supported plantar steps. This is a significant improvement in rat locomotor function and affords the rat increased mobility and activity compared with a control injury group BBB score of less than 9, where rats have no weight bearing stepping at all.⁵⁰ In the error ladder test, we also saw improvements in function with the Peptide5 treatment that were indistinguishable from the sham operated rats.

At-level pain after SCI has been reported previously to be the most common of neuropathic pain types (41% of subjects), with an early onset and pain described as severe or excruciating.^110,111 In the current study, we demonstrated that Peptide5 treatment was able to alleviate at-level pain during the early (one week) and later (six week) stages of recovery after SCI, confirming the role of connexin hemichannels in the development of neuropathic pain.¹¹²

Treatment with Peptide5 had no effect on paw withdrawal threshold in response to mechanical stimuli. It is noteworthy that interpretation of spinal-mediated reflex response may be problematic when assessing below-level SCI pain because of the development of hyperactive reflex circuitries.¹¹³ Hindpaw stimulation caused increased innate reflex responses in the absence of increased brainstem responses, thus reflecting a spasticity syndrome rather than pain-like behavior.¹¹³ In addition, in humans, below-level pain is reported to occur months to years after SCI, in contrast to the days/weeks for the development of at-level pain.¹¹¹ Although promising, further studies are required to determine the effects of Peptide5 treatment on the full spectrum of SCI-induced pain behaviors.

The findings of the current study provide strong in vivo data supporting the benefits of systemic delivery of the Connexin43 mimetic peptide, Peptide5, in modulating Connexin43 hemichannels as a feasible approach for acute SCI. These findings support the need to further investigate the timing of administration, the optimal dose, pharmacokinetics, and administration route for humans and to further evaluate the mode of action of Peptide5 to begin translation to clinical trials.

Footnotes

Acknowledgments

This work was supported financially by the Spinal Cord Injury Network, Australia and New Zealand and the CatWalk Spinal Cord Injury Trust. We acknowledge the University of Technology Sydney for the financial support in the form of postgraduate research scholarships for Yilin Mao. The authors would like to thank staff at the Ernst Facility, University of Technology Sydney, for their care of the experimental animals.

Author Disclosure Statement

Prof. Colin Green is a founding scientist of CoDa Therapeutics, Inc. (USA) that holds the intellectual property rights to Peptide5. For the remaining authors, no competing financial interests exist.

References

Hausmann

O.N.

(2003). Post-traumatic inflammation following spinal cord injury. Spinal Cord, 41, 369–378.

Norenberg

M.D.

, Smith

, and Marcillo

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

Fleming

J.C.

, Norenberg

M.D.

, Ramsay

D.A.

, Dekaban

G.A.

, Marcillo

A.E.

, Saenz

A.D.

, Pasquale-Styles

, Dietrich

W.D.

, and Weaver

L.C.

(2006). The cellular inflammatory response in human spinal cords after injury. Brain, 129, 3249–3269.

Fitch

M.T.

, and Silver

(2008). CNS injury, glial scars, and inflammation: inhibitory extracellular matrices and regeneration failure. Exp. Neurol., 209, 294–301.

Dusart

, and Schwab

M.E.

(1994). Secondary cell death and the inflammatory reaction after dorsal hemisection of the rat spinal cord. Eur. J. Neurosci., 6, 712–724.

Carlson

S.L.

, Parrish

M.E.

, Springer

J.E.

, Doty

, and Dossett

(1998). Acute inflammatory response in spinal cord following impact injury. Exp. Neurol., 151, 77–88.

Kyritsis

, Kizil

, and Brand

(2014). Neuroinflammation and central nervous system regeneration in vertebrates. Trends Cell Biol. 24, 128–135.

Horn

K.P.

, Busch

S.A.

, Hawthorne

A.L.

, Van Rooijen

, and Silver

(2008). Another barrier to regeneration in the CNS: activated macrophages induce extensive retraction of dystrophic axons through direct physical interactions. J. Neurosci., 28, 9330–9341.

Brambilla

, Bracchi-Ricard

, Hu

W.H.

, Frydel

, Bramwell

, Karmally

, Green

E.J.

, and Bethea

J.R.

(2005). Inhibition of astroglial nuclear factor κB reduces inflammation and improves functional recovery after spinal cord injury. J. Exp. Med., 202, 145–156.

10.

Shechter

, London

, Varol

, Raposo

, Cusimano

, Yovel

, Rolls

, Mack

, Pluchino

, and Martino

(2009). Infiltrating blood-derived macrophages are vital cells playing an anti-inflammatory role in recovery from spinal cord injury in mice. PLoS Med. 6, e1000113.

11.

Goodenough

D.A.

, Goliger

J.A.

, and Paul

D.L.

(1996). Connexins, connexons, and intercellular communication. Ann. Rev. Biochem., 65, 475–502.

12.

Kang

, Kang

, Lovatt

, Torres

, Zhao

, Lin

, and Nedergaard

(2008). Connexin 43 hemichannels are permeable to ATP. J. Neurosci., 28, 4702–4711.

13.

Söhl

, and Willecke

(2003). An update on connexin genes and their nomenclature in mouse and man. Cell Commun. Adhes., 10, 173–180.

14.

Söhl

, Maxeiner

, and Willecke

(2005). Expression and functions of neuronal gap junctions. Nat. Rev. Neurosci., 6, 191–200.

15.

Orthmann-Murphy

J.L.

, Abrams

C.K.

, and Scherer

S.S.

(2008). Gap junctions couple astrocytes and oligodendrocytes. J. Mol. Neurosci., 35, 101–116.

16.

Contreras

J.E.

, Sáez

J.C.

, Bukauskas

F.F.

, and Bennett

M.V.

(2003). Gating and regulation of connexin 43 (Cx43) hemichannels. Proc. Natl. Acad. Sci. U.S.A., 100, 11388–11393.

17.

Retamal

M.A.

, Schalper

K.A.

, Shoji

K.F.

, Bennett

M.V.

, and Sáez

J.C.

(2007). Opening of connexin 43 hemichannels is increased by lowering intracellular redox potential. Proc. Natl. Acad. Sci. U.S.A., 104, 8322–8327.

18.

Lin

J.H.

, Weigel

, Cotrina

M.L.

, Liu

, Bueno

, Hansen

A.J.

, Hansen

T.W.

, Goldman

, and Nedergaard

(1998). Gap-junction-mediated propagation and amplification of cell injury. Nat. Neurosci., 1, 494–500.

19.

Theriault

, Frankenstein

U.N.

, Hertzberg

E.L.

, and Nagy

J.I.

(1997). Connexin43 and astrocytic gap junctions in the rat spinal cord after acute compression injury. J. Comp. Neurol., 382, 199–214.

20.

Scemes

, Suadicani

S.O.

, and Spray

D.C.

(2000). Intercellular communication in spinal cord astrocytes: fine tuning between gap junctions and P2 nucleotide receptors in calcium wave propagation. J. Neurosci., 20, 1435–1445.

21.

Thompson

R.J.

, Zhou

, and MacVicar

B.A.

(2006). Ischemia opens neuronal gap junction hemichannels. Science, 312, 924–927.

22.

Z.C.

, Wyeth

M.S.

, Baltan-Tekkok

, and Ransom

B.R.

(2003). Functional hemichannels in astrocytes: a novel mechanism of glutamate release. J. Neurosci., 23, 3588–3596.

23.

Orellana

J.A.

, Hernández

D.E.

, Ezan

, Velarde

, Bennett

M.V.

, Giaume

, and Sáez

J.C.

(2010). Hypoxia in high glucose followed by reoxygenation in normal glucose reduces the viability of cortical astrocytes through increased permeability of connexin 43 hemichannels. Glia, 58, 329–343.

24.

Decrock

, De Bock

, Wang

, Bultynck

, Giaume

, Naus

C.C.

, Green

C.R.

, and Leybaert

(2015). Connexin and pannexin signaling pathways, an architectural blueprint for CNS physiology and pathology?. Cell. Mol. Life Sci., 72, 2823–2851.

25.

Davidson

J.O.

, Green

C.R.

, Bennet

and Gunn

A.J

(2015). Battle of the hemichannels—connexins and pannexins in ischemic brain injury. Int. J. Dev. Neurosci., 45, 66–74.

26.

Cronin

, Anderson

P.N.

, Cook

J.E.

, Green

C.R.

, and Becker

D.L.

(2008). Blocking connexin43 expression reduces inflammation and improves functional recovery after spinal cord injury. Mol. Cell. Neurosci., 39, 152–160.

27.

Lee

I.H.

, Lindqvist

, Kiehn

, Widenfalk

, and Olson

(2005). Glial and neuronal connexin expression patterns in the rat spinal cord during development and following injury. J. Comp. Neurol., 489, 1–10.

28.

O'Carroll

S.J.

, Alkadhi

, Nicholson

L.F.

, and Green

C.R.

(2008). Connexin43 mimetic peptides reduce swelling, astrogliosis, and neuronal cell death after spinal cord injury. Cell Commun. Adhes., 15, 27–42.

29.

O'Carroll

S.J.

, Gorrie

C.A.

, Velamoor

, Green

C.R.

, and Nicholson

L.F.

(2013). Connexin43 mimetic peptide is neuroprotective and improves function following spinal cord injury. Neurosci. Res., 75, 256–267.

30.

Green

C.R.

, Law

L.Y.

, Lin

J.S.

, and Becker

D.L.

(2001). Spatiotemporal depletion of connexins using antisense oligonucleotides, in: Methods in Molecular Biology: Connexin Methods and Protocols. Bruzzone

, Giaume

(eds). Humana Press: Totowa, NJ., pps. 175–185.

31.

Huang

, Han

, Li

, Lam

, Peng

, Lou

, Torres

, Yang

, Garre

J.M.

, Tian

G.F.

, Bennett

M.V.

, Nedergaard

, and Takano

(2012). Critical role of connexin 43 in secondary expansion of traumatic spinal cord injury. J. Neurosci, 32, 3333–3338.

32.

Cronin

, Anderson

P.N.

, Green

C.R.

, and Becker

D.L.

(2006). Antisense delivery and protein knockdown within the intact central nervous system. Front. Biosci., 11, 2967–2975.

33.

Berthoud

V.M.

, Beyer

E.C.

, and Seul

K.H.

(2000). Peptide inhibitors of intercellular communication. Am. J. Physiol. Lung Cell. Mol. Physiol., 279, L619–L622.

34.

Boitano

, and Evans

W.H.

(2000). Connexin mimetic peptides reversibly inhibit Ca2+ signaling through gap junctions in airway cells. Am. J. Physiol. Lung Cell. Mol. Physiol., 279, L623–L630.

35.

Martin

P.E.

, Wall

, and Griffith

T.M.

(2005). Effects of connexin‐mimetic peptides on gap junction functionality and connexin expression in cultured vascular cells. Br. J. Pharmacol., 144, 617–627.

36.

Yoon

J.J.

, Green

C.R.

, O'Carroll

S.J.

, and Nicholson

L.F.

(2010). Dose-dependent protective effect of connexin43 mimetic peptide against neurodegeneration in an ex vivo model of epileptiform lesion. Epilepsy Res. 92, 153–162.

37.

Vuyst

, E., Decrock

, De

Bock

, M., Yamasaki

, Naus

C.C.

, Evans

W.H.

, and Leybaert

(2007). Connexin hemichannels and gap junction channels are differentially influenced by lipopolysaccharide and basic fibroblast growth factor. Mol. Biol. Cell, 18, 34–46.

38.

Leybaert

, Braet

, Vandamme

, Cabooter

, Martin

P.E.

and Evans

W.H.

(2003). Connexin channels, connexin mimetic peptides and ATP release. Cell Commun. Adhes., 10, 251–257.

39.

Chen

Y.S.

, Green

C.R.

, Teague

, Perrett

, Danesh-Meyer

H.V.

, Toth

, and Rupenthal

I.D.

(2015). Intravitreal injection of lipoamino acid-modified connexin43 mimetic peptide enhances neuroprotection after retinal ischemia. Drug Deliv. Transl Res., 5, 480–488.

40.

Chen

Y.S.

, Green

C.R.

, Wang

, Danesh-Meyer

H.V.

, and Rupenthal

I.D.

(2014). Sustained intravitreal delivery of connexin43 mimetic peptide by poly (D, L-lactide-co-glycolide) acid micro- and nanoparticles—Closing the gap in retinal ischaemia. Eur. J. Pharm. Biopharm., 95, 378–386.

41.

Danesh-Meyer

H.V.

, Kerr

N.M.

, Zhang

, Eady

E.K.

, O'Carroll

S.J.

, Nicholson

L.F.

, Johnson

C.S.

, and Green

C.R.

(2012). Connexin43 mimetic peptide reduces vascular leak and retinal ganglion cell death following retinal ischaemia. Brain, 135, 506–520.

42.

Davidson

J.O.

, Green

C.R.

, Nicholson

L.F.

, Bennet

, and Gunn

A.J.

(2012). Deleterious effects of high dose connexin 43 mimetic peptide infusion after cerebral ischaemia in near-term fetal sheep. Int. J. Mol. Sci., 13, 6303–6319.

43.

Iyyathurai

, D'hondt

, Wang

, De Bock

, Himpens

, Retamal

M.A.

, Stehberg

, Leybaert

, and Bultynck

(2013). Peptides and peptide-derived molecules targeting the intracellular domains of Cx43: gap junctions versus hemichannels. Neuropharmacology, 75, 491–505.

44.

Calder

B.W.

, Rhett

J.M.

, Bainbridge

, Fann

S.A.

, Gourdie

R.G.

, and Yost

M.J.

(2015). Inhibition of connexin 43 hemichannel-mediated ATP release attenuates early inflammation during the foreign body response. Tissue Eng. Part A, 21, 1752–1762.

45.

Abbott

N.J.

(2002). Astrocyte–endothelial interactions and blood–brain barrier permeability. J. Anat., 200, 629–638.

46.

Abudara

, Bechberger

, Freitas-Andrade

, De Bock

, Wang

, Bultynck

, Naus

C.C.

, Leybaert

, and Giaume

(2014). The connexin43 mimetic peptide Gap19 inhibits hemichannels without altering gap junctional communication in astrocytes. Front. Cell. Neurosci. 8, 306.

47.

Hawat

, Benderdour

, Rousseau

, and Baroudi

(2010). Connexin 43 mimetic peptide Gap26 confers protection to intact heart against myocardial ischemia injury. Pflügers Arch. 460, 583–592.

48.

Pollok

, Pfeiffer

A.C.

, Lobmann

, Wright

C.S.

, Moll

, Martin

P.E.

, and Brandner

J.M.

(2011). Connexin 43 mimetic peptide Gap27 reveals potential differences in the role of Cx43 in wound repair between diabetic and non‐diabetic cells. J. Cell. Mol. Med., 15, 861–873.

49.

Peng

, Cotrina

M.L.

, Han

, Yu

, Bekar

, Blum

, Takano

, Tian

G.F.

, Goldman

S.A.

, and Nedergaard

(2009). Systemic administration of an antagonist of the ATP-sensitive receptor P2X7 improves recovery after spinal cord injury. Proc. Natl. Acad. Sci. U.S.A., 106, 12489–12493.

50.

Basso

D.M.

, Beattie

M.S.

, and Bresnahan

J.C.

(1996). Graded histological and locomotor outcomes after spinal cord contusion using the NYU weight-drop device versus M.transection. Exp. Neurol., 139, 244–256.

51.

Metz

G.A.

, and Whishaw

I.Q.

(2002). Cortical and subcortical lesions impair skilled walking in the ladder rung walking test: a new task to evaluate fore-and hindlimb stepping, placing, and co-ordination. J. Neurosci. Methods, 115, 169–179.

52.

Chaplan

S.R.

, Bach

F.W.

, Pogrel

J.W.

, Chung

J.M.

, and Yaksh

T.L.

(1994). Quantitative assessment of tactile allodynia in the rat paw. J. Neurosci. Methods, 53, 55–63.

53.

Faigle

, Seessle

, Zug

, El Kasmi

K.C.

, and Eltzschig

H.K.

(2008). ATP release from vascular endothelia occurs across Cx43 hemichannels and is attenuated during hypoxia. PLoS One 3, e2801.

54.

Tonkin

R.S.

, Mao

, O'Carroll

S.J.

, Nicholson

L.F.

, Green

C.R.

, Gorrie

C.A.

, and Moalem-Taylor

(2015). Gap junction proteins and their role in spinal cord injury. Front. Mol. Neurosci. 7, 102.

55.

Davidson

J.O.

, Green

C.R.

, Nicholson

L.F.

, Bennet

, and Gunn

A.J.

(2013). Connexin hemichannel blockade is neuroprotective after, but not during, global cerebral ischemia in near-term fetal sheep. Exp. Neurol., 248, 301–308.

56.

Davidson

J.O.

, Drury

P.P.

, Green

C.R.

, Nicholson

L.F.

, Bennet

, and Gunn

A.J.

(2014). Connexin hemichannel blockade is neuroprotective after asphyxia in preterm fetal sheep. PLoS One 9, e96558.

57.

Davidson

J.O.

, Green

C.R.

, Nicholson

L.F.

, O'Carroll

S.J.

, Fraser

, Bennet

, and Gunn

A.J.

(2012). Connexin hemichannel blockade improves outcomes in a model of fetal ischemia. Ann. Neurol., 71, 121–132.

58.

Han

, Arnold

S.A.

, Sithu

S.D.

, Mahoney

E.T.

, Geralds

J.T.

, Tran

, Benton

R.L.

, Maddie

M.A.

, D'Souza

S.E.

, Whittemore

S.R.

, and Hagg

(2010). Rescuing vasculature with intravenous angiopoietin-1 and αvβ3 integrin peptide is protective after spinal cord injury. Brain, 133, 1026–1042.

59.

Budd

S.L.

, and Lipton

S.A.

(1998). Calcium tsunamis: do astrocytes transmit cell death messages via gap junctions during ischemia?. Nat. Neurosci., 1, 431–432.

60.

Contreras

J.E.

, Sánchez

H.A.

, Véliz

L.P.

, Bukauskas

F.F.

, Bennett

M.V.

, and Sáez

J.C.

(2004). Role of connexin-based gap junction channels and hemichannels in ischemia-induced cell death in nervous tissue. Brain Res. Brain Res. Rev., 47, 290–303.

61.

Haupt

, Witte

O.W.

, and Frahm

(2007). Up-regulation of Connexin43 in the glial scar following photothrombotic ischemic injury. Mol. Cell. Neurosci., 35, 89–99.

62.

Davidson

J.O.

, Green

C.R.

, Bennet

, Nicholson

L.F.

, Danesh-Meyer

, O'Carroll

S.J.

, and Gunn

A.J.

(2013). A key role for connexin hemichannels in spreading ischemic brain injury. Curr. Drug Targets, 14, 36–46.

63.

Frantseva

M.V.

, Kokarovtseva

, and Velazquez

J.L.

(2002). Ischemia-induced brain damage depends on specific gap-junctional coupling. J. Cereb. Blood Flow Metab., 22, 453–462.

64.

Roh

D.H.

, Yoon

S.Y.

, Seo

H.S.

, Kang

S.Y.

, Han

H.J.

, Beitz

A.J.

, and Lee

J.H.

(2010). Intrathecal injection of carbenoxolone, a gap junction decoupler, attenuates the induction of below-level neuropathic pain after spinal cord injury in rats. Exp. Neurol., 224, 123–132.

65.

Frantseva

M.V.

, Kokarovtseva

, Naus

C.G.

, Carlen

P.L.

, MacFabe

, and Perez Velazquez

J.L.

(2002). Specific gap junctions enhance the neuronal vulnerability to brain traumatic injury. J. Neurosci., 22, 644–653.

66.

Nakase

, Söhl

, Theis

, Willecke

, and Naus

C.C.

(2004). Increased apoptosis and inflammation after focal brain ischemia in mice lacking connexin43 in astrocytes. Am. J. Pathol., 164, 2067–2075.

67.

Takahashi

, Vargas

, and Wilcox

(2010). Increased coupling and altered glutamate transport currents in astrocytes following kainic-acid-induced status epilepticus. Neurobiol. Dis., 40, 573–585.

68.

Siushansian

, Bechberger

J.F.

, Cechetto

D.F.

, Hachinski

V.C.

, and Naus

C.C.

(2001). Connexin43 null mutation increases infarct size after stroke. J. Comp. Neurol., 440, 387–394.

69.

Contreras

J.E.

, Sánchez

H.A.

, Eugenín

E.A.

, Speidel

, Theis

, Willecke

, Bukauskas

F.F.

, Bennett

M.V.

, and Sáez

J.C.

(2002). Metabolic inhibition induces opening of unapposed connexin 43 gap junction hemichannels and reduces gap junctional communication in cortical astrocytes in culture. Proc. Natl. Acad. Sci. U.S.A., 99, 495–500.

70.

Retamal

M.A.

, Cortés

C.J.

, Reuss

, Bennett

M.V.

, and Sáez

J.C.

(2006). S-nitrosylation and permeation through connexin 43 hemichannels in astrocytes: induction by oxidant stress and reversal by reducing agents. Proc. Natl. Acad. Sci. U.S.A., 103, 4475–4480.

71.

Perez Velazquez

J.L.

, Kokarovtseva

, Sarbaziha

, Jeyapalan

, and Leshchenko

(2006). Role of gap junctional coupling in astrocytic networks in the determination of global ischaemia‐induced oxidative stress and hippocampal damage. Eur. J. Neurosci., 23, 1–10.

72.

Ramachandran

, Xie

L.H.

, John

S.A.

, Subramaniam

, and Lal

(2007). A novel role for connexin hemichannel in oxidative stress and smoking-induced cell injury. PLoS One, 2, e712.

73.

Frisén

, Johansson

C.B.

, Török

, Risling

, and Lendahl

(1995). Rapid, widespread, and longlasting induction of nestin contributes to the generation of glial scar tissue after CNS injury. J. Cell Biol., 131, 453–464.

74.

Kozlova

E.N.

(2003). Differentiation and migration of astrocytes in the spinal cord following dorsal root injury in the adult rat. Eur. J. Neurosci., 17, 782–790.

75.

Renault-Mihara

, Okada

, Shibata

, Nakamura

, Toyama

, and Okano

(2008). Spinal cord injury: emerging beneficial role of reactive astrocytes' migration. Int. J. Biochem. Cell Biol., 40, 1649–1653.

76.

Tysseling-Mattiace

V.M.

, Sahni

, Niece

K.L.

, Birch

, Czeisler

, Fehlings

M.G.

, Stupp

S.I.

, and Kessler

J.A.

(2008). Self-assembling nanofibers inhibit glial scar formation and promote axon elongation after spinal cord injury. J. Neurosci., 28, 3814–3823.

77.

Mao

, Mathews

, and Gorrie

C.A.

(2016). Temporal response of endogenous neural progenitor cells following injury to the adult rat spinal cord. Front Cell. Neurosci., 10, 58.

78.

Sharma

, Selzer

M.E.

, and Li

(2012). Scar-mediated inhibition and CSPG receptors in the CNS. Exp. Neurol., 237, 370–378.

79.

Fawcett

J.W.

(2006). The glial response to injury and its role in the inhibition of CNS repair. Adv. Exp. Med. Biol., 557, 11–24.

80.

Lane

M.A.

, Truettner

J.S.

, Brunschwig

J.P.

, Gomez

, Bunge

M.B.

, Dietrich

W.D.

, Dziegielewska

K.M.

, Ek

C.J.

, Vandeberg

J.L.

, and Saunders

N.R.

(2007). Age‐related differences in the local cellular and molecular responses to injury in developing spinal cord of the opossum, Monodelphis domestica. Eur. J. Neurosci., 25, 1725–1742.

81.

Jones

L.L.

, Margolis

R.U.

, and Tuszynski

M.H.

(2003). The chondroitin sulfate proteoglycans neurocan, brevican, phosphacan, and versican are differentially regulated following spinal cord injury. Exp. Neurol., 182, 399–411.

82.

Barkho

B.Z.

, Song

, Aimone

J.B.

, Smrt

R.D.

, Kuwabara

, Nakashima

, Gage

F.H.

, and Zhao

(2006). Identification of astrocyte-expressed factors that modulate neural stem/progenitor cell differentiation. Stem Cells Dev. 15, 407–421.

83.

Karimi-Abdolrezaee

, and Billakanti

(2012). Reactive astrogliosis after spinal cord injury—beneficial and detrimental effects. Mol. Neurobiol., 46, 251–264.

84.

Wang

, Cheng

, He

, Zheng

, Kim

D.H.

, Whittemore

S.R.

, and Cao

Q.L.

(2011). Astrocytes from the contused spinal cord inhibit oligodendrocyte differentiation of adult oligodendrocyte precursor cells by increasing the expression of bone morphogenetic proteins. J. Neurosci., 31, 6053–6058.

85.

Mizuno

, Warita

, Aoki

, and Itoyama

(2008). Accumulation of chondroitin sulfate proteoglycans in the microenvironment of spinal motor neurons in amyotrophic lateral sclerosis transgenic rats. J. Neurosci. Res., 86, 2512–2523.

86.

Sofroniew

M.V.

(2009). Molecular dissection of reactive astrogliosis and glial scar formation. Trends Neurosci. 32, 638–647.

87.

, Zhou

, Luo

, Lin

, Wang

, Li

, Bian

, Li

, Wan

, Yu

, and Feng

(2010). Glial scar and neuroregeneration: histological, functional, and magnetic resonance imaging analysis in chronic spinal cord injury. J. Neurosurg. Spine, 13, 169–180.

88.

Ikeda

, and Murase

(2004). Glial nitric oxide-mediated long-term presynaptic facilitation revealed by optical imaging in rat spinal dorsal horn. J. Neurosci., 24, 9888–9896.

89.

Barnabe-Heider

, Goritz

, Sabelstrom

, Takebayashi

, Pfrieger

F.W.

, Meletis

, and Frisen

(2010). Origin of new glial cells in intact and injured adult spinal cord. Cell Stem Cell, 7, 470–482.

90.

Hinkerohe

, Smikalla

, Haghikia

, Heupel

, Haase

C.G.

, Dermietzel

, and Faustmann

P.M.

(2005). Effects of cytokines on microglial phenotypes and astroglial coupling in an inflammatory coculture model. Glia, 52, 85–97.

91.

Kreutzberg

G.W.

(1996). Microglia: a sensor for pathological events in the CNS. Trends Neurosci. 19, 312–318.

92.

Loane

D.J.

, and Byrnes

K.R.

(2010). Role of microglia in neurotrauma. Neurotherapeutics, 7, 366–377.

93.

Amat

J.A.

, Ishiguro

, Nakamura

, and Norton

W.T.

(1996). Phenotypic diversity and kinetics of proliferating microglia and astrocytes following cortical stab wounds. Glia, 16, 368–382.

94.

Donnelly

D.J.

, and Popovich

P.G.

(2008). Inflammation and its role in neuroprotection, axonal regeneration and functional recovery after spinal cord injury. Exp. Neurol., 209, 378–388.

95.

Fitch

M.T.

, Doller

, Combs

C.K.

, Landreth

G.E.

, and Silver

(1999). Cellular and molecular mechanisms of glial scarring and progressive cavitation: in vivo and in vitro analysis of inflammation-induced secondary injury after CNS trauma. J. Neurosci., 19, 8182–8198.

96.

Klusman

, and Schwab

M.E.

(1997). Effects of pro-inflammatory cytokines in experimental spinal cord injury. Brain Res. 762, 173–184.

97.

Yan

, Li

, Kim

G.M.

, Xu

, Hsu

C.Y.

, and Xu

X.M.

(2001). Cellular localization of tumor necrosis factor-α following acute spinal cord injury in adult rats. J. Neurotrauma, 18, 563–568.

98.

Trivedi

, Olivas

A.D.

, and Noble-Haeusslein

L.J.

(2006). Inflammation and spinal cord injury: infiltrating leukocytes as determinants of injury and repair processes. Clin. Neurosci. Res., 6, 283–292.

99.

Pineau

, and Lacroix

(2007). Proinflammatory cytokine synthesis in the injured mouse spinal cord: multiphasic expression pattern and identification of the cell types involved. J. Comp. Neurol., 500, 267–285.

100.

Flaris

N.A.

, Densmore

T.L.

, Molleston

M.C.

, and Hickey

W.F.

(1993). Characterization of microglia and macrophages in the central nervous system of rats: definition of the differential expression of molecules using standard and novel monoclonal antibodies in normal CNS and in four models of parenchymal reaction. Glia, 7, 34–40.

101.

Yang

, Blumbergs

P.C.

, Jones

N.R.

, Manavis

, Sarvestani

G.T.

, and Ghabriel

M.N.

(2004). Early expression and cellular localization of proinflammatory cytokines interleukin-1β, interleukin-6, and tumor necrosis factor-α in human traumatic spinal cord injury. Spine, 29, 966–971.

102.

Yang

, Jones

N.R.

, Blumbergs

P.C.

, Van Den

Heuvel

, C., Moore

E.J.

, Manavis

, Sarvestani

G.T.

, and Ghabriel

M.N.

(2005). Severity-dependent expression of pro-inflammatory cytokines in traumatic spinal cord injury in the rat. J. Clin. Neurisci., 12, 276–284.

103.

Eugenín

E.A.

, Eckardt

, Theis

, Willecke

, Bennett

M.V.

and Sáez

J.C.

(2001). Microglia at brain stab wounds express connexin 43 and in vitro form functional gap junctions after treatment with interferon-γ and tumor necrosis factor-α. Proc. Natl. Acad. Sci. U.S.A., 98, 4190–4195.

104.

Takeuchi

, Jin

, Wang

, Zhang

, Kawanokuchi

, Kuno

, Sonobe

, Mizuno

, and Suzumura

(2006). Tumor necrosis factor-α induces neurotoxicity via glutamate release from hemichannels of activated microglia in an autocrine manner. J. Biol. Chem., 281, 21362–21368.

105.

Morita

, Saruta

, Kozuka

, Okubo

, Itakura

, Takahashi

, and Kudo

(2007). Dual regulation of astrocyte gap junction hemichannels by growth factors and a pro‐inflammatory cytokine via the mitogen‐activated protein kinase cascade. Glia, 55, 508–515.

106.

Retamal

M.A.

, Froger

, Palacios-Prado

, Ezan

, Sáez

P.J.

, Sáez

J.C.

, and Giaume

(2007). Cx43 hemichannels and gap junction channels in astrocytes are regulated oppositely by proinflammatory cytokines released from activated microglia. J. Neurosci., 27, 13781–13792.

107.

Orellana

J.A.

, Froger

, Ezan

, Jiang

J.X.

, Bennett

M.V.

, Naus

C.C.

, Giaume

, and Sáez

J.C.

(2011). ATP and glutamate released via astroglial connexin 43 hemichannels mediate neuronal death through activation of pannexin 1 hemichannels. J. Neurochem., 118, 826–840.

108.

Orellana

J.A.

, Sáez

P.J.

, Shoji

K.F.

, Schalper

K.A.

, Palacios–Prado

, Velarde

, Giaume

, Bennett

M.V.

, and Sáez

J.C.

(2009). Modulation of brain hemichannels and gap junction channels by pro-inflammatory agents and their possible role in neurodegeneration. Antioxid. Redox Signal., 11, 369–399.

109.

Orellana

J.A.

, Shoji

K.F.

, Abudara

, Ezan

, Amigou

, Sáez

P.J.

, Jiang

J.X.

, Naus

C.C.

, Sáez

J.C.

, and Giaume

(2011). Amyloid β-induced death in neurons involves glial and neuronal hemichannels. J. Neurosci., 31, 4962–4977.

110.

Nashold

B.S.

Jr.

Paraplegia and pain in: Nashold

B.S.

Jr , Ovelman-Levitt

. (eds.). Advances in Pain Research and Therapy. Deafferentation Pain Syndromes: Pathophysiology and Treatment. Raven Press: New York, 1991, pps. 301–319.

111.

Siddall

P.J.

, McClelland

J.M.

, Rutkowski

S.B.

, and Cousins

M.J.

(2003). A longitudinal study of the prevalence and characteristics of pain in the first 5 years following spinal cord injury. Pain, 103, 249–257.

112.

Lee-Kubli

C.A.

, Ingves

, Henry

K.W.

, Shiao

, Collyer

, Tuszynski

M.H.

, and Campana

W.M.

(2016). Analysis of the behavioral, cellular and molecular characteristics of pain in severe rodent spinal cord injury. Exp. Neurol., 278, 91–104.

113.

Baastrup

, Maersk-Moller

C.C.

, Nyengaard

J.R.

, Jensen

T.S.

, and Finnerup

N.B.

(2010). Spinal-, brainstem- and cerebrally mediated responses at- and below-level of a spinal cord contusion in rats: evaluation of pain-like behavior. Pain, 151, 670–679.