Blocking Endothelin-B Receptors Rescues Retinal Ganglion Cells from Optic Nerve Injury through Suppression of Neuroinflammation

Abstract

Purpose.: The endothelins (ETs) cause reactive astrogliosis, which involves neuroinflammation and neurodegeneration in the central nervous system. The purpose of this study was to determine whether blocking the ET signals will protect retinal ganglion cells (RGCs) from optic nerve injury.

Methods.: We studied the effect of pretreatment with BQ-123, an antagonist of ETA receptors, and BQ-788, an antagonist of ETB receptors, on the survival of RGCs after the optic nerve of rats was crushed. We also performed immunohistological evaluations and real-time PCR of the crushed site to determine the expressions of the ET-1, CD68, GFAP, TNF-α, and iNOS genes in the neuroinflammation of the optic nerves.

Results.: The mRNA levels of the ETB receptors were upregulated (5.6-fold) on day 7 after crushing the optic nerves. Cells expressing ETB receptors were recruited mainly to the crushed site where the immunoreactivity to GFAP was weak. These cells were also immuunoreactive to ETs and CD68, a constitutive marker of microglia/macrophages. In the adjacent areas, immunoreactivity to GFAP was intense. Crushing the optic nerve increased the mRNA levels of ET-1 (4.5-fold), CD68 (87.5-fold), GFAP (2-fold), TNF-α (480-fold), and iNOS (6-fold) on day 7. Pretreatment with BQ-788 significantly suppressed the upregulation of these genes and loss of RGCs on day 7, whereas BQ-123 failed to protect the RGCs.

Conclusions.: These results suggest that the microglia/macrophages recruited to the crushed site are the possible cellular sources of the ETs, which caused reciprocal activation of astrocytes. Blocking the ETB receptors by BQ-788 rescued RGCs, most likely by attenuating neuroinflammatory events.

Introduction

The astrocytes are the main glial cells of the central nervous system (CNS), including the optic nerves. They play crucial roles in neurotransmission and in maintaining the extracellular environment and neuronal functions. Astrocytes are activated in some pathological conditions, and the reactive astrocytes undergo morphological transformation and proliferation. ¹ The important characteristic of reactive astrocytes is their hypertrophy, which is accomplished through the upregulation of intermediate filament protein and glial fibrillary acidic protein (GFAP). ²  

Reactive astrocytes express several cell adhesion molecules (e.g., intercellular adhesion molecule-1 and vascular cellular adhesion molecule-1) and they release several cytokines and growth factors. Some of these molecules participate in the repair processes under different types of stresses ³ ; however, excessive and sustained activation of astrocytes is associated with astrogliosis, which prevents axonal repair, leading to more neuronal damage. ⁴ In addition, reactive astrocytes in concert with recruited microglia/macrophages induce neuroinflammation with the formation of TNF-α and nitric oxide. ⁵ Thus, reactive astrogliosis is important pathophysiologically in the CNS, and changes in the level of GFAP can be used to evaluate the degree of reactive changes of astrocytes. ⁶  

The endothelins (ETs) are a family of 21 amino acid peptides with three isoforms, ET-1, ET-2, and ET-3. ET-1 is the most potent and long-acting vasoconstricting peptide ⁷ and is also related to the modulation of neuronal and glial activities in the CNS. ^{8 –10} ET-1 and its G-protein–coupled receptors, ETA and ETB, are abundantly expressed and widely distributed in ocular tissues, including the sensory retina and optic nerve, ^11,12 as they are in the brain. ¹³ Astrocytes and ET-1 are connected through an autocrine loop because astrocyes secrete ET-1. ^14,15 ET-1 has been shown to cause reactive astrogliosis after its infusion into the brain. ¹⁶ Furthermore, ET-1–induced astrogliosis has been linked to neurodegeneration in Alzheimer's disease, ¹⁷ and also after neurotrauma, ^18,19 neuroinflammation, ²⁰ and stroke. ²¹  

Some of the pathological roles of the ETs on ocular diseases have been presented. For example, Prasanna et al. ²² have shown that ET-1 stimulates astrocytes in the human optic nerve head and causes proliferation of astrocytes and alterations of the extracellular matrices. These changes are closely associated with glaucomatous optic neuropathy, and the interactions between ET-1 and optic nerve astrocytes play pivotal roles in the changes associated with glaucoma. ^{23 –25}  

Crushing the optic nerve is commonly used to study neurodegenerative process in the optic nerve and retina. A loss of retinal ganglion cells (RGCs) is known to take place after optic nerve injury through retrograde apoptosis. ^26,27 The neurodegenerative process shares some common pathways with glaucoma, as seen in the RGC loss in glaucomatous eyes. ²⁸ The ET-signaling pathways are activated in the retina of rats with experimental glaucoma, ^28,29 and also after crushing the RGC axons. ³⁰ Dysfunction of the RGC axons is clinically relevant in patients with optic neuritis and traumatic optic neuropathy. Thus, the interactions between ET-1 and the activated astrocytes after axonal injury are clinically important and need to be understood to determine the pathophysiology of these sight-threatening diseases. 

Rogers et al. ¹⁴ have shown that crushing the optic nerve enhanced the immunoreactivities to the ETB receptors and GFAP, leading to astrogliosis. Infusion of ET-1 into the optic nerve induced similar changes, and bosentan, a nonspecific antagonist to ET receptor, suppressed astrogliosis after crushing the optic nerve. ¹⁴ These results indicated that ET-1 is involved in the development of astrogliosis in the optic nerve and blocking the ET-signaling pathway is a promising method for neuroprotection. The cellular sources of ET-1 are still unclear, however, and the receptor mechanisms for inducing degeneration and protection remain to be determined. Recently, Howell et al. ³¹ showed that microglia/macrophages are recruited in the retinal nerve fiber layer and produce ET-2, which shares receptors with ET-1, in a mouse model of glaucoma. In addition, blocking the ET receptors with bosentan suppressed the loss of RGCs and their axons from glaucomatous damage, ³¹ whereas ET-2 was suggested to activate Müller cells and protect photoreceptors from various stresses. ³²  

Because the ETs are involved in the neurodegenerative process of the optic nerve, ^14,31 we hypothesized that blocking the ET-signaling pathway will increase the survival of the RGCs after the optic nerve is crushed. To test this hypothesis, we crushed the optic nerves of rats, and determined whether pretreatment of the nerves with BQ-123, antagonists of the ETA receptors, and BQ-788, an antagonist of the ETB receptors, would increase the survival of RGCs. Because reactive astrogliosis is known to recruit microglia/macrophages in cases of neurotrauma, ³³ we also performed immunohistological evaluations at the crushed site in the optic nerve and also in the retina to determine interaction between ET-1, reactive astrocytes, and the recruitment of microglia/macrophages. 

Materials and Methods

Animals

Nine-week-old male Wistar rats were purchased from Japan SLC (Shizuoka, Japan). The rats were housed in an air-conditioned room with a temperature of approximately 23°C, humidity of 60%, and on a 12:12 light:dark cycle. All animals were handled in accordance with the ARVO resolution for the Use of Animals in Ophthalmic and Vision Research. The experimental protocol was approved by the Committee of Animal Use and Care of the Osaka Medical College. A total of 149 adult rats were used. 

Chemicals

All chemicals were purchased from Sigma, Inc. (St. Louis, MO) unless otherwise noted. 

Optic Nerve Crush

Animals were anesthetized with an intraperitoneal injection of pentobarbital sodium. A skin incision was made along the superior orbital margin to expose the superior surface of the right eye. The superior rectus muscle was incised to expose the optic nerve, and the optic nerve was crushed with forceps 2 mm behind the eye for 10 seconds. ³⁴ Care was taken not to occlude the blood vessels and cause retinal ischemia. We confirmed that the retinal circulation was not blocked by indirect ophthalmoscopy. ³⁵ A sham operation was performed on the right eyes of other animals, and the optic nerve was exposed in the same way but not crushed, as in the experimental animals. The left eyes were not used as controls because it has been demonstrated that crushing one optic nerve affects the morphology of the contralateral retina. ³⁶  

In some animals, BQ-788 (molecular weight [MW]: 664) or BQ-123 (MW: 610) was used to block the ETB or ETA receptors, respectively. To do this, the optic nerves were exposed, and a small piece of sponge (Medical quick absorber; Inami, Tokyo, Japan) saturated with either BQ-788 (1.0 μM), BQ-123 (1.0 μM), or PBS was placed on the exposed optic nerves for 15 minutes and then removed. The sponge was placed over the entire exposed optic nerve (approximately 2 mm in length), including the scheduled site for crushing. This concentration of the blockers was chosen because BQ-788 at a dose of 1.0 μM suppresses proliferation of optic nerve head astrocytes caused by ET-1 at the most effective doses of 10 to 100 nM. ²² After this exposure, the optic nerves were crushed as described. We confirmed that when a sponge soaked with indocyanine green dye (MW: 775) was placed in the same way, the dye permeated through the entire crushed area of the optic nerve. 

A loss of RGCs is known to take place in a delayed fashion after crushing the optic nerve; the number of RGCs remains unchanged for 5 days and then abruptly decreases to 50% on day 7 and to less than 10% on day 14. ³⁵ Thus, the loss of RGCs was determined on day 7 after crushing the optic nerve, along with analyses for alterations of gene expression in the retina and optic nerve. 

Tissue Processing

Rats were deeply anesthetized with an intraperitoneal injection of pentobarbital sodium and perfused through the heart with saline followed by 4% paraformaldehyde (PFA) in 0.1 M phosphate buffer, pH 7.4. After removal of skull and cerebral hemispheres, the optic nerves and the eyes were carefully removed and postfixed in 4% PFA in PBS overnight. These tissues were used for immunohistochemistry. After washing with PBS, the tissues were immersed in 30% sucrose overnight at 4°C and then embedded in ornithine carbamoyltransferase compound (BDH Laboratory Supplies, Poole, UK). Then, 14-μm-thick frozen sections were cut with a cryostat. After blocking with 10% normal goat or donkey serum plus 2% BSA in Tris-buffered saline (TBS), the sections of the optic nerves were incubated with primary antibodies of rabbit anti-endothelins (ETs;1:500; Peninsula Laboratories, San Carlos, CA), rabbit anti-ETA receptor antibodies (1:500; Millipore, Temecula, CA), rabbit anti-ETB receptors antibody (1:500; Millipore), mouse monoclonal anti-GFAP antibody (1:200; Sigma), mouse anti-CD68 antibody (1:500; Serotec, Oxford, UK), or rabbit anti-proliferating cell nuclear antigen (PCNA) antibody (1:200, Santa Cruz Biotechnology, Santa Cruz, CA) overnight at 4°C. In addition, sheep anti-ETB receptor antibody (1:200, Abcam, Cambridge, MA) was used for double labeling of ETB receptors and ETs. The sections were then incubated for 2 hours at room temperature in Alexa 594, Alexa 488 (Invitrogen, Carlsbad, CA), or FITC conjugated to the appropriate secondary antibodies diluted by 1:500. Control staining for ETA and ETB receptors was performed with primary antibody preabsorbed with recombinant ETA and ETB receptors.  

We also examined whether neuroinflammatory events occurred in the retina. For this, 14-μm-thick frozen sections were cut with a cryostat through the optic nerve. The retinal sections were incubated with primary antibodies of mouse anti-CD68 antibody (1:500; Serotec) overnight at 4°C, followed by incubation in FITC-conjugated goat anti-mouse IgG antibody for 2 hours at room temperature. 

Labeling Retinal Ganglion Cells

The number of RGCs surviving 7 days after crushing the optic nerves was determined. The retinas were carefully removed from the eyes of rats as described in detail by Winkler. ³⁷ In brief, rats were euthanized by CO₂, and the globe was proptosed by placing forceps around the optic nerve just behind the eyeball. The globe was transected along the equator and the cornea and lens were removed. The retina was detached from the pigment epithelium by pressing upward with the forceps and removed by cutting its attachment to the optic nerve head. The isolated retina was placed in PBS solution immediately, and any vitreous remaining on the isolated retina was carefully removed. 

The retinas were then flat mounted, sandwiched between nylon mesh sheets, and fixed in 4% PFA in PBS overnight at 4°C. After washing in PBS and blocking in PBS containing 1.0% BS and 0.3% triton X-100, the retinas were incubated with Alexa 488-conjugated mouse monoclonal neuron-specific class III beta-tubulin (Tuj 1) antibody (Covance, Princeton, NJ) (1:500). Tuj 1 is a specific marker for RGCs, ^38,39 and the sections were placed in the same medium overnight at 4°C, washed with PBS, and coverslipped. 

To determine the number of RGCs, the stained flat mounts were photographed through a fluorescent microscope (BZ 9000; Keyence, Osaka, Japan). Eight areas (0.48 × 0.48 mm²) from the four quadrants at a distance of 1.0 and 1.5 mm from the margin of the optic disc were photographed, and all Tuj 1–positive cells in an area of 0.2 × 0.2 mm² at the center of each image were counted using National Institutes of Health Image J. 

The mean density of the RGC/mm² was calculated and loss of RGCs was determined by comparing the density in the retinas with optic nerve crush with that of retinas from untouched optic nerves (n = 8 each). The number of RGCs was counted by an observer masked to whether it was from an experimental or control animal. 

Quantitative RT-PCR Analysis

We determined the changes in the expression of several genes in the optic nerves by real-time PCR on day 7. These were the ET-1, ETA, and ETB receptors and CD68 genes. We also determined the expressions of GFAP, TNF-α, and inducible nitric oxide synthase (iNOS). In addition, we examined whether some of these genes were also expressed in the retina. To rule out the possibility of retinal ischemia after crushing the optic nerve, the expression of the HIF-1α gene in the retina was examined on days 2 and 7.  

The animals were killed, and approximately 4 mm of the optic nerves, including the crush site, were collected. Retinas were also obtained as described. The optic nerves and retinas were homogenized in lysis buffer, and the RNA was extracted using the RNeasy plus mini kit (QIAGEN, Valencia, CA). The RNA concentration and purity were calculated from the absorbance at 260/280 nm. The RNAs were reverse transcribed with PrimeScript reverse transcriptase reagent (Takara; Ohtsu, Shiga, Japan). 

The cDNA was used for quantitative real-time PCR amplification with the TaqMan Gene Expression Assays for target genes (Applied Biosystems, Foster City, CA). Rat TaqMan Gene Assays for ET1 Rn00561129_m1, endothelin receptor type A (ETA-R) Rn00561137_m1, endothelin receptor type B (ETB-R) Rn00569139_m1, CD68 Rn01495634_g1, TNF-α Rn01525859_g1, GFAP Rn00566603_m1, nitric oxide synthase 2 (NOS2) Rn00561646_m1, and HIF-1α Rn00577560_m1 were used. Amplicons were detected using the relevant probes tagged with MGB quencher and FAM dye. TaqMan rat 18S rRNA control expression assays (Applied Biosystems) were used as the reference genes. 

Real-time PCR was performed in Premix Ex Taq (Perfect Real Time; Takara). All reactions were run on a Thermal Cycler Dice Real time system TP870 (Takara) with the following cycling parameters: 95°C for 30 seconds followed by 40 cycles of 95°C for 5 seconds and 60°C for 30 seconds. A standard curve of cycle thresholds using serial dilutions of cDNA samples was made. Target gene values were normalized to the relative amounts of 18S rRNA. 

Statistical Analyses

The data are expressed as means ± SEMs unless otherwise noted. Statistical analysis was done by one-way ANOVA, and if a significant change was detected, then the Scheffe post hoc test was done for statistical comparisons among groups. When comparisons were made between 2 groups, Student's t-tests were used. The level of significance was set at P < 0.05. 

Results

Changes in Expression of ETA and ETB Receptors in Optic Nerve after Optic Nerve Crushing

The optic nerves were examined by real-time PCR on day 7 after being crushed (n = 6) or sham operated (sham control, n = 4) to detect changes in the mRNA levels of ETA and ETB receptors. The mRNA level of each gene was normalized to that of 18S rRNA, a housekeeping gene (Fig. 1A). 

Figure 1. 

 

Changes in the expression of ETA and ETB receptors after crushing the optic nerve. (A) The mRNA levels of ETA and ETB receptors in the optic nerves were normalized to that of 18S rRNA, and comparisons were made between sham-operated animals (n = 4) and crushed optic nerve animals (n = 6). Data are shown as fold changes in the mRNA expression to the sham control. Both receptors are significantly upregulated but the increase of ETB receptors is greater. (B) Representative photographs of the results of immunohistochemistry for ETA and ETB receptors in the crushed optic nerves from three independent samples. Cells stained by nuclear DAPI are concentrated at the crushed site (arrows). Immunoreactivity to ETB receptor is intensified at the crushed site, whereas the staining pattern of the ETA receptors is uneven. Immunoreactivities to ETA and ETB receptors are mostly negative in the absorption control. ETA/B receptor staining: rabbit anti-ETA /B receptors (primary) and Alexa 594-conjugated goat anti-rabbit IgG (secondary antibodies). Bar = 100 μm.

The relative expression levels of the mRNA of the ETA and ETB receptors to that of 18S rRNA in the sham control optic nerves were not significantly different with the levels of 3.0% ± 0.6% for the ETA and 3.0% ± 0.8% for the ETB receptors (mean ± SEM, n = 4). These levels showed a 1.9-fold increase for the ETA receptors (P = 0.03) and 5.6-fold increase for the ETB receptors (P = 0.005, t-test) relative to the sham control after crushing the optic nerve. Both increases were significant but the increase was greater (P = 0.004; t-test) for the ETB receptors than the ETA receptors. Thus, we focused our study on the effects of blocking the ETB receptors. 

Immunohistochemistry

Immunohistological analyses were used to determine where the genes were expressed after the optic nerves were crushed. Representative photographs of immunostained sections for ETA and ETB receptors after day 7 are shown in Figure 1B. In agreement with the results of real-time PCR, immunoreactivity to ETB receptors was more intense than that to ETA receptors around the site of the crushed optic nerve (Fig. 1B). 

We then sought to determine the effects of blocking the ETB receptors on the expression of GFAP and ETs. Representative photomicrographs of the immunohistochemical staining for the ETB receptors and GFAP obtained from the crushed and sham-operated optic nerves on day 7 are shown in Figure 2. At the crushed site with prior treatment with PBS (crush [PBS]), immunoreactivity to the ETB receptors was clearly increased compared with the sham control. On the other hand, immunoreactivity to GFAP was only weakly present at the crushed site, whereas its activity was intensified in the area surrounding the lesion compared with the sham-control (Fig. 2A). 

Figure 2. 

 

Immunohistochemistry for ETB receptors and GFAP in the optic nerve from sham control and experimental animals after crushing the optic nerves. Representative photographs from three independent samples for each condition are presented. (A) Crushing the optic nerve intensified the immunoreactivity to ETB receptor at the crushed site (crush [PBS]; arrows) compared with the sham control. In contrast to the staining pattern of the ETB receptor, immunoreactivity to GFAP is faint at the crushed site (arrows) but is increased in the area surrounding the site (crush [PBS]). Pretreatment with BQ-788 attenuates the immunoreactivities to both the ETB receptor at the crushed site (arrows) and the GFAP in the area surrounding the crushed site (crush [BQ-788]). Similar effects are detected by pretreatment with BQ-123 (crush [BQ-123]). A total of 12 rats were used in this analyses. ETB receptor staining: rabbit anti-ETB receptors (primary) and Alexa 594-conjugated goat anti-rabbit IgG (secondary antibodies); GFAP staining: mouse monoclonal anti-GFAP (primary) and FITC-conjugated goat anti-mouse IgG (secondary antibodies). Bar = 100 μm. (B) Higher magnification of the crushed site (area B). Cells that have infiltrated into the crushed site are stained by DAPI (blue) and are also positively stained by antibodies to ETB receptor (red). The merged images of DAPI and ETB receptors were made by Photoshop (Adobe Systems Inc., San Jose, CA). Note the high incidence of colocalization of nuclear staining with DAPI and ETB receptors (center panels) at the crushed site where a dense cellular infiltration of cells expressing ETB receptors is seen, and GFAP immunoreactivity is almost absent. In the area surrounding the crushed site, immunoreactivity to GFAP (green) is intensified (right panels) where an immersion of cells between the fibrils can be seen. Bar = 100 μm. (C) Higher-magnification view of the crushed site (area C) stained with antibodies to rabbit anti-ETB receptors (Alexa 594-conjugated goat anti-rabbit IgG secondary antibody) and mouse anti-CD68 (FITC-conjugated goat anti-mouse IgG secondary antibody). Some of the cells that have infiltrated the crushed site are positively stained with both CD68 (green) and ETB receptor (red) antibodies. Bar = 100 μm. (D) Higher-magnification view of the edge of the crushed site from another optic nerve with prior treatment with PBS. The sample was stained with antibodies to mouse anti-GFAP (FITC-conjugated goat anti-mouse IgG secondary antibody) and rabbit anti-PCNA (Alexa 594-conjugated goat anti-rabbit IgG secondary antibody). Some of the GFAP-positive cells (green) are also positively stained with PCNA (red). Bar = 100 μm.

Examinations of the crushed site at higher magnification showed colocalization of DAPI in the nuclei and immunoreactivity to ETB receptors (Fig. 2B). This suggested that an infiltration of cells with ETB receptors occurred at the site of the lesion. Around the crushed site, the immunoreactivity to GFAP was intensified, indicating hypertrophy/proliferation of astrocytes (Fig. 2B). In addition, some of these cells with ETB receptors at the crushed site were stained positively with CD68, a constitutive marker expressed on macrophages and microglia ⁴⁰ (Fig. 2C). 

To determine whether the increased expression of GFAP was due to astrocytic hypertrophy or proliferation, double labeling of GFAP and PCNA was performed. We detected some but not all GFAP-positive cells that also expressed PCNA, a proliferation marker (Fig. 2D). Thus, increased GFAP expression seen around the crushed site was due to both hypertrophy and proliferation of astrocytes. 

When we exposed the optic nerves to BQ-788 before crushing (Fig. 2A, crush [BQ-788]), the immunoreactivity to GFAP was less intense than prior treatment with PBS. Similar effects were obtained by pretreatment with BQ-123 (Fig. 2A, crush [BQ-123]). 

The alterations in the immunoreactivities to ETs and GFAP after optic nerve injury are shown in Figure 3. Immunoreactivity to the ETs was increased mainly at the crushed site on day 7, which spread to the area around the lesion with increasing intensities on day 14 after crushing the optic nerves (Fig. 3A; crush [PBS]). Double labeling with nuclear DAPI staining and immunoreactivity to ETs showed that they were colocalized (Fig. 3B). Around these cells expressing ETs, the immunoreactivity to GFAP was intensified. This interaction was similar between cells immunopositive to ETB receptors and GFAP (Fig. 2B). Indeed, a high incidence of double labeling with antibodies to ETB receptors and ETs was found (Fig. 3C). This supported the idea that these ETB receptor–positive cells have an ability to secrete ETs. In addition, immunoreactivity to CD68 was colocalized with ETs (Fig. 3D), suggesting that microglia/macrophages are possible cellular sources of the ETs. 

Figure 3. 

 

Immunohistochemistry for ETs, GFAP, and ETB receptors in the optic nerves of the sham controls and experimental animals with a crushed optic nerve. Representative pictures from three independent samples for each condition are presented. (A) Changes in the expression of ETs on days 7 and 14 after crushing the optic nerve. The sections were stained with rabbit anti-ETs and Alexa 594–conjugated goat anti-rabbit IgG secondary antibodies. Immunoreactivity to ETs (red) is expressed mainly at the crushed site on day 7 (arrows), which spreads and intensifies on day 14 (crush [PBS]) compared with the sham controls. Pretreatment with BQ-788 suppresses the increased immunoreactivity to ETs (crush [BQ-788]; crush site is indicated by arrows). Pretreatment with BQ-123 also suppresses the increase but the effects are less intense than those by BQ-788 (crush [BQ-123]). A total of 24 rats were used in this analysis. (B) Higher magnification view of the crushed site (area B) stained with antibodies to rabbit anti-ETs and mouse monoclonal anti-GFAP. Cells that have infiltrated into the crushed site are stained by DAPI (blue) and are also positively stained by antibodies to ETs (Alexa 594). Merged image by Photoshop indicates high incidence of colocalization of nuclear staining with DAPI and ETs (left panels). At the crushed site where a dense cellular infiltration with the expression of ETs can be seen, GFAP immunoreactivity is very weak. In the area surrounding the crushed site, immunoreactivity to GFAP (FITC) is increased as shown in Figure 2B (right panels). (C) Higher-magnification view of the crushed site (area C) stained with antibodies to sheep anti-ETB receptors and rabbit anti-ETs. Some cells that have infiltrated the crushed site are positively stained with both antibodies to ETB receptors (Alexa 488-conjugated donkey anti-sheep IgG) and ETs (Alexa 594-conjugated goat anti-rabbit IgG). (D) Double labeling of other crushed optic nerve sections stained with antibodies to CD68 and ETs. The migratory cells that are concentrated at the crushed site (arrows) are positively stained with CD68 (green) at high density. Some of these CD68-positive cells at the crushed site are also stained with ETs (red). Bar = 100 μm.

Prior treatment with BQ-788 again suppressed the increase in the immunoreactivity to ETs after crushing the optic nerves (Fig. 3A, crush [BQ-788]). Prior treatment with BQ-123 also suppressed the increased immunoreactivity to ETs (Fig. 3A, crush [BQ-123]), but the effects seemed to be less than that to BQ-788. 

After crushing the optic nerve, some CD68-positive cells in the retina were found in the internal retina (Fig. 4); however, the migratory cells were apparently much scarcer than those seen at the crush site. Prior treatment with either BQ-788 or BQ-123 suppressed the expression of CD68-positive cells in the retina to some extent (Fig. 4). 

Figure 4. 

 

Immunohistochemistry for CD68 expression in the retina on day 7 after crushing the optic nerve. Representative photographs from three independent samples for each condition are presented. Compared with the sham control (A), some CD68-positive cells (arrows) are presented in the internal retina on day 7 after crushing the optic nerve (B); however, the migratory cells are much fewer than those seen in the optic nerve. Pretreatment with BQ-788 (C) or BQ-123 (D) suppresses the expression of CD68 cells to some extent. GCL, ganglion cell layer; IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer. Bar = 10 μm.

For each immunohistological examination described above, three rats were used for each condition (e.g., sham control, crushed with prior BQ-788, crushed with prior BQ-123, and crushed with prior PBS on days 7 and 14). We observed similar findings as described from two additional samples for each group. Thus, a total of 36 rats were used in these immunohistological studies. 

Alterations in Expression of Genes for ET-1, CD68, GFAP, TNF-α, and iNOS from Optic Nerves

We then determined the effects of prior treatment with BQ-123 and BQ-788 on the expressions of the ET-1 and GFAP genes quantitatively by real-time PCR. The neuroinflammatory events included proliferation and activation of microglia/macrophages, upregulation of cytokine production, and nitrosative stress. Thus, alterations of CD68, a constitutive marker of microglia/macrophage, TNF-α, and iNOS were also examined. Our findings on the mRNA levels of these genes in the optic nerves on day 7 after crushing the optic nerves and normalized to the 18S rRNA levels are shown in Figure 5. The levels are relative to that of the sham controls (n = 6 to 8 in each condition). 

Figure 5. 

 

Changes in the mRNA levels of ET-1 (A), CD68 (B), GFAP (C), TNF-α (D), and iNOS (E) in the optic nerve, and CD68 (F), TNF-α (G), iNOS (H), and HIF-1α (I) in the retina. Data are shown as fold changes (mean ± SE) to the sham control in the mRNA expressions. (A–E) RNA was extracted from 4 mm of the optic nerves including the crushed site. All genes are significantly increased over the sham control after crushing the optic nerves. Pretreatment with BQ-788 suppresses the increase significantly, and the effects were consistently higher than those of BQ-123 (n = 6–8 in each condition, for a total of 30 rats used). (F–I) RNA was extracted from the retinas after crushing the optic nerves. The CD68 gene was significantly upregulated (F), but the increases of TNF-α (G) and iNOS (H) were not significant in the retina on day 7 after crushing the optic nerve. No increase in HIF-1α gene was detected on days 2 and 7 (n = 6–8 in each condition, for a total of 38 rats used).

The mRNA level of ET-1 had a 4.5-fold increase over the control level after the optic nerve was crushed with prior treatment with PBS. This increase was suppressed by prior treatment with either BQ-788 (P = 0.004, Scheffe) or BQ-123 (P = 0.048), and the levels remained at 1.4-fold and 2.5-fold increases, respectively (Fig. 5A). Similarly, the mRNA levels of CD68 were increased (P < 0.0001, Scheffe) to 87.5 times as high as those of the sham controls after crushing the optic nerve. This increase was significantly suppressed by either treatment with BQ-788 or BQ-123 (P < 0.01, Scheffe; Fig. 5B). 

Crushing the optic nerves caused a 2-fold increase in the expression of the mRNA of GFAP from the sham control animals on day 7. Both BQ-123 and BQ-788 suppressed the increase significantly when they were exposed to the optic nerves before crushing (P < 0.001, Scheffe; Fig. 5C). 

Because TNF-α is a neurodestructive inflammatory cytokine, its level after crushing the optic nerve was determined. There was a 480-fold increase in the mRNA levels of TNF-α after the optic nerve was crushed. This increase was significantly suppressed by prior treatment with BQ-788 (P = 0.04, Scheffe), but the increase was not significantly affected by prior BQ-123 (P = 0.11, Scheffe; Fig. 5D). Similar changes were also observed in the expression of iNOS. Crushing the optic nerves caused a 6-fold increase of the mRNA levels of iNOS (P = 0.04, Scheffe). This increase was significantly suppressed by BQ-788 (P = 0.046, Scheffe), but not significantly by BQ-123 (P = 0.25, Scheffe; Fig. 5E). 

Alterations in Expression of Genes for CD68, TNF-α, and iNOS from Retinas

Crushing the optic nerve caused a modest but significant (P = 0.0003, Scheffe) increase of mRNA levels of CD68 (3.7-fold increase of the sham control, Fig. 5F); however, BQ-788 failed to suppress the increase significantly (P = 0.49, Scheffe; Fig. 5F). The mRNA levels of the TNF-α (1.4-fold; P = 0.28) and iNOS (1.9-fold; P = 0.46, Scheffe) genes were not upregulated significantly after crushing the optic nerve (Figs. 5G, 5H). 

Importantly, because HIF-1α did not increase on either day 2 or 7 in the retina (Fig. 5I), retinal ischemia secondary to crushing the optic nerve seemed to be unlikely the cause of the changes (n = 6–8 in each condition). 

Decrease in Number of RGCs after Optic Nerve Crush

Representative photomicrographs of flat-mounted retinas taken approximately 1.5 mm from the optic disc margin are shown in Figure 6. The RGCs are stained green with Alexa 488-conjugated Tuj 1, and the number of RGCs can be seen to be fewer after crushing the optic nerve (Fig. 6B) than in the sham control (Fig. 6A). 

Figure 6. 

 

Representative photomicrographs of flat mounted retinas stained with Alexa 488-conjugated Tuj 1 antibody (A–D). Retinas from sham control (A), crushed optic nerves pretreated with PBS (B), BQ-788 (C), and BQ-123 (D). Density of retinal ganglion cells (RGCs/mm²) was quantified as a whole (E), at 1.0 mm (F), and 1.5 mm from the optic disc margin (G). Pretreatment with BQ-788 had a significant protective effect on the RGCs day 7 after crushing the optic nerves (n = 8 each, total number = 32).

The mean number (± SEM) of RGCs stained by TUJ-1 antibody was 1928.4 ± 125.3/mm² in sham-operated rats (n = 8), which decreased to 841.8 ± 98.9/mm² (n = 8) on day 7 after the optic nerve was crushed with prior treatment of PBS (Fig. 6E). This reduction was significantly (P = 0.03, Scheffe) reduced to 1251.6 ± 85.0/mm² (n = 8) when BQ-788 was applied before the optic nerve was crushed (Figs. 6C, 6E). On the other hand, the number of RGCs was 916.8 ± 69.6/mm² (n = 8) after prior treatment by BQ-123 (P = 0.9, Scheffe; Figs. 6D, 6E). 

When we compared the effects of BQ-788 on the loss of RGCs at a distance of 1.0 and 1.5 mm from the margin of the optic disc, BQ-788 had a greater protective effect at 1.5 mm (P = 0.02, Fig. 6G) than those at 1.0 mm (P = 0.049, Scheffe; Fig. 6F). 

Discussion

Our results showed that there was a recruitment of ETB receptor–positive cells including microglia/macrophages to the site of the crushed optic nerve where the immunoreactivities to ETs were intensified. Immunoreactivity to GFAP was essentially lost at the lesion, whereas its activity was increased surrounding the crushed site. Pretreatment with BQ-788, a blocker of ETB receptor, suppressed the increase in the mRNA levels of ET-1 and genes related to neuroinflammation that were expressed after crushing the optic nerves. Furthermore, the blocking ETB receptors had a protective effect on the RGCs. 

Recruitment of microglia/macrophages occurs at traumatic, ischemic, and neurodegenerative sites in the CNS. ^41,42 This was also the case in a mouse model of glaucoma. ³¹ The inflammatory cells were recruited to the optic nerve fiber layer of the retina, the primary site of the disease process. These findings indicated that these inflammatory cells may play a crucial role in the local neurodegenerative events. 

Our immunohistological results showed that the intense immunoreactivity to ETs and ETB receptors was restricted to the crushed site on day 7, whereas immunoreactivity to GFAP was almost lost. Later, immunoreactivity to ETs was found around the crushed lesion on day 14. Thus, although astrocytes have an ability to secrete ET-1, ^43,44 it seems reasonable to consider that these ETB receptor–positive cells primarily secreted ETs at the crushed site, which may induce a reciprocal hypertrophy/proliferation of the astrocytes surrounding the lesion. The presence of cells double positive to GFAP and PCNA suggested that some astrocytes would undergo proliferation around the crushed site. Because some of these cells immunopositive to ETB receptors were also immunoreactive to CD68, microglia/macrophages recruited to the crushed site were possible cellular sources of ETs, as these migratory cells can secrete ETs. ^45,46  

It has recently been shown that ETs can modulate neuroinflammation associated with reactive astrocytes in the CNS. ^47,48 Inflammatory cells, including microglia/macrophages, are believed to be responsible for the breakdown of the myelin sheath in multiple sclerosis, where reactive astrocytes release chemokines, such as monocyte chemoattractant protein-1, and attract myelin-degrading phagocytes in active demyelinating lesions. ⁴⁹ Activation of the ETB receptors of astrocytes stimulates immunoresponses through chemokine production ⁵⁰ and enhances the release of nitric oxide and IL-6. ⁵¹ Thus, because of the ETs' proinflammatory roles, inhibition of ETB receptors may have some beneficial effects on anti-inflammation in the CNS. ⁵²  

In addition to ET-1, upregulations of the genes of CD68, GFAP, TNF-α, and iNOS were found in the crushed optic nerves on day 7. These findings may reflect the neuroinflammatory events related to activation of microglia/macrophages and astrocytes. Treatment with BQ-788, blocker of ETB receptors, significantly inhibited the upregulation of these genes. The inhibitory effect was consistently higher after BQ-788 than after BQ-123, blocker of ETA receptors, along with the higher protective effects of BQ-788 than BQ-123 on RGCs. BQ-788 was found to suppress the upregulation of TNF-α and iNOS significantly after crushing the optic nerve, whereas BQ-123 failed to suppress these enhancements. Thus, blocking the ETB receptors by BQ-788 might have rescued the RGCs through suppression of ET-1 and the cytotoxic molecules from microglia/macrophages and astrocytes. 

In spite of the prominent upregulation of the ETB receptors, treatment with either BQ-788 or BQ-123 significantly suppressed the upregulation of GFAP gene expression in the crushed optic nerves. Thus, both receptors were involved in the ET-1–induced hypertrophy/proliferation of astrocytes in the optic nerve. These findings were somewhat unexpected, because the roles of ET-1 on cortical astrocytes were almost exclusively mediated through ETB receptors. ^16,53,54 On the other hand, these findings were in good agreement with the findings that showed that ET-1 induced proliferation of rat optic nerve head astrocytes through both ETA and ETB receptors. ⁵⁵ Because BQ-123 did not protect the RGCs in spite of the nearly complete suppression of GFAP upregulation, astrocytic hypertrophy/proliferation alone was not responsible for the pathogenesis but inflammatory events including production of cytotoxic molecules from microglia/macrophages and astrocytes must have also played crucial roles in the loss of RGCs after crushing the optic nerve. This idea is consistent with the immunohistological findings with exclusive dominant expression of ETB receptors in the cells recruited at the crushed site. Thus, we suggest the proinflammatory roles of ETs, including the upregulation of TNF-α and iNOS genes, were mainly mediated by ETB receptors. 

Crushing the optic nerve also caused inflammatory events in the retina, as evidenced by increased mRNA expression of CD68; however, the increase of protein (immunohistochemistry) and mRNA levels of CD68 was less intense than that in the optic nerve. In addition, the increase of the mRNA levels of TNF-α and iNOS genes were not significant in the retina. Thus, BQ-788 primarily suppressed the inflammatory events in the optic nerve. 

There are limitations of this study. We did not determine the changes in the protein levels of the genes in the optic nerves. In addition, the expression levels and time course of cytokines, chemokines, and the adhesion molecules were not determined. Further studies are required to clarify these issues, including how ET-1 involves cytokine network in neuroinflammation and neurodegeneration in the optic nerve. 

In conclusion, ET-1 is involved in the neuroinflammation and neurodegeneration after the optic nerve is crushed. The proposed ET-signaling pathways are summarized in Figure 7. Crushing the optic nerve induces a migration of microglia/macrophages, and these cells secrete ETs, causing reciprocal activation of astrocytes. The astrocytes become reactive and secrete proinflammatory and cytotoxic molecules such as ETs, TNF-α, and nitric oxide in concert with microglia/macrophages. Blocking the ETB receptors rescued the RGCs most likely by attenuating these neuroinflammatory events. These findings should be considered when new therapeutic strategies against ETB receptors for optic nerve diseases are being developed. 

Figure 7. 

 

Schematic presentation of possible roles of ETs on neuroinflammation and damages of RGCs and their axons. Recruited ETB-R–positive cells, including microglia/macrophages secrete ETs (1), which causes reciprocal activation of astrocytes (2) and their hypertrophy/proliferation (3). Activated astrocytes in concert with microglia/macrophages secrete proinflammatory molecules (ET-1, TNF-α, nitric oxide, and so forth), forming autocrine and paracrine loop for glial activation (4). These cytotoxic molecules enhance the loss of RGCs and their axons (5). Our results suggest both BQ-788 and BQ-123 suppress hypertrophy/proliferation of astrocytes; however, the proinflammatory roles of ET-1 are mainly mediated through ETB-R as evidenced by higher inhibitory effects of BQ-788 on TNF-α and iNOS expression.

Acknowledgments

The authors thank Duco Hamasaki, PhD, Bascom Palmer Eye Institute, University of Miami School of Medicine, for editing this manuscript. 

Footnotes

 Supported in part by Japanese Ministry of Education Grant 225919739 (HO) and an Osaka Eye Bank Association Fund grant (HO).

Footnotes

 Disclosure: M. Tonari, None; T. Kurimoto, None; T. Horie, None; T. Sugiyama, None; T. Ikeda, None; H. Oku, None