Chondroitin Sulfate-Derived Disaccharide Protects Retinal Cells from Elevated Intraocular Pressure in Aged and Immunocompromised Rats

Abstract

purpose. A disaccharide (DS) derived from the naturally occurring compound chondroitin sulfate proteoglycan (CSPG) was recently shown to have neuroprotective activity. The authors examined the ability of this compound (CSPG-DS) to protect retinal ganglion cells (RGCs) from death caused by elevated intraocular pressure (IOP).

methods. With the use of chronic and acute models of elevated IOP, the authors examined the effects of CSPG-DS on RGC survival in adult (∼2 months old), aged (10–12 months old), and immunocompromised Lewis rats. Systemic, topical, and oral routes of administration were examined.

results. CSPG-DS protected RGCs from IOP-induced death. Treatment was effective in all three examined rat populations (normal adult, aged, and immunocompromised rats) and with all routes of administration, possibly in part through its control of microglial activity.

conclusions. Results point to the therapeutic potential of CSPG-DS for glaucoma, particularly in elderly populations for whom disease prevalence is high.

Glaucoma is characterized by slow, progressive degeneration of retinal ganglion cells (RGCs), causing gradual visual field loss and eventually leading to blindness. The primary cause of the disease is not yet known, and the factors contributing to its progression have not been fully characterized. Current treatment of patients with glaucoma is mostly limited to the reduction of elevated intraocular pressure (IOP), known to be one of the major factors responsible for disease progression. ¹ It is clear, however, that although a decrease in IOP can significantly reduce the loss of neurons, it does not necessarily halt the process of degeneration. ^{2 3 4 5 6 7 8}  

It was suggested by our group that the ongoing loss of neurons in glaucoma might be explained, at least in part, by secondary factors that emerge from injured tissue as a result of the degeneration of neuronal cells (RGCs and their axons) after the primary insult. ^{9 10 11} Thus, although the primary insult (for example, increased IOP) might not affect all axons and RGCs directly, it causes changes in the neuronal environment (neurotransmitter imbalance, depletion of growth factors, influx of calcium into the cells, and formation of free radicals), which in turn result in further loss of RGCs through secondary degeneration. ^{2 12 13}  

Studies in our laboratory have shown that boosting the immune response can provide a way to counteract the toxic effects of certain physiological compounds (such as glutamate and nitric oxide), whose presence in excessive amounts results in neuronal loss. ^{14 15} To be effective without causing harm, however, this immune response must be rigorously controlled. ^{16 17} A beneficial immune response is manifested by T cells that recognize specific autoantigens residing at the site of damage. ^{18 19} After an injury, T cells home to the site of injury where, on encountering their specific antigens, they become locally activated and can induce brain-resident microglia to adopt a phenotype that supports neural tissue by enabling the microglia to produce growth factors and to buffer threatening compounds. ²⁰  

CSPG-DS is a disaccharide (DS) derived from the extracellular matrix protein chondroitin sulfate proteoglycan (CSPG), which is expressed in the eye at early postnatal stages and gradually decreases in amount as the animal grows. CSPG contributes to formation of the neural network and maintenance of the interphotoreceptor matrix. ^{21 22} It also appears to play a role in glaucoma, as shown by evidence of increased levels of autoantibodies recognizing glycosaminoglycans in serum samples from patients with this disease. ²³ Moreover, elevation of intraocular pressure (IOP) reportedly leads to changes in the expression patterns of CSPG in the eye. ^{21 24 25} These changes might affect the susceptibility of remaining axons to pressure by changing the physical properties of their supporting tissues and altering the microenvironment of injured axons. ^{26 27}  

Although the role of intact CSPG in central nervous system (CNS) repair has been a focus of research in many laboratories and studies have shown that this compound impedes neural regrowth, ²⁸ little attention has been paid to its degradation products. Our group recently demonstrated that a CSPG-derived disaccharide, CSPG-DS, affects microglia in a way that enables them to acquire a protective phenotype without needing assistance from the adaptive arm of the immune system (T cells). Moreover, CSPG-DS can directly promote neuronal survival in several in vivo and ex vivo models by a mechanism associated with the activation of phosphokinase-C (PKC) and protein tyrosine kinase (PYK2) signaling pathways in those cells. ²⁹ The observed dual action of CSPG-DS on neurons and microglia encouraged us to examine its potential benefit in models of IOP-induced RGCs loss in adult, aging, and immunocompromised animals. 

Materials and Methods

Animals

Inbred male Lewis rats were supplied by the Animal Breeding Center at the Weizmann Institute of Science. For the studies on immunocompromised animals, we used Sprague-Dawley rats that underwent thymectomy at birth and therefore lacked mature T cells. The rats were maintained in a light- and temperature-controlled room and were matched for age and weight before each experiment. All were handled according to the regulations formulated by the Institutional Animal Care and Use Committee and in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 

Induction of Chronically High Intraocular Pressure

Blockage of aqueous outflow causes an increase in IOP, which results in RGC death. ^{30 31} Increased IOP was achieved in the right eyes of deeply anesthetized rats (ketamine hydrochloride 50 mg/kg, xylazine hydrochloride 0.5 mg/kg; injected intramuscularly) by blocking of the aqueous outflow in that eye with 80 to 120 applications of blue-green argon laser irradiation from a Haag-Streit slit lamp. The laser beam, which was directed at three of the four episcleral veins and at 270° of the limbal plexus, was applied with a power of 1 W for 0.2 second, producing a spot size of 100 μm at the episcleral veins and 50 μm at the limbal plexus. At a second laser session, 1 week later, the same parameters were used except that the spot size was 100 μm for all applications, and this time the radiation was directed toward all four episcleral veins and 360° of the limbal plexus. ³²  

Induction of Acute High Intraocular Pressure

Acute increases in IOP were induced in the right eyes of deeply anesthetized rats by insertion of a 30-gauge needle connected to a polyethylene tube and a bag of normal saline (0.9%) infusion. The infusion bag was placed 1 m above each rat’s head, creating closed-loop circulation. The induction process was maintained for exactly 1 hour, after which the needle was removed from the eye and the hole sealed itself. Within 24 hours of termination, IOP values returned to normal (18.21 ± 1.26 mm Hg). 

Anatomic Assessment of Retinal Damage Caused by the Increase in IOP

To count RGCs in each of the examined groups, we used the hydrophilic neurotracer dye dextran tetramethylrhodamine (rhodamine dextran; Molecular Probes, Eugene, OR) 21 days after the first laser session in all experiments except when we examined the effects of CSPG-DS over a period of 8 weeks, in which case the dye was applied 7 weeks after the first laser session. In all cases, the optic nerve was completely transected 2 or 3 mm proximal to the globe, and the dye was then applied directly into the intra-orbital portion of the exposed nerve. Rats were humanely killed 24 hours later, and their eyes were excised. Retinas were detached from the eyes, prepared as flattened whole-mounts in 4% paraformaldehyde solution, and examined for labeled ganglion cells by fluorescence microscopy. RGCs were labeled after rather than before IOP was elevated because our experience had shown that dye administered after IOP elevation labels only surviving RGCs. In addition, because CSPG-DS affects immune components, ^{29 33} applying dye before IOP was elevated would have resulted in preconditioned activation of the immune cells. 

Because the distribution of RGCs throughout the retina is not uniform, not all regions of the retina can reliably demonstrate differences in RGC survival. ^{11 34} Thus, for example, in the retinal region over a distance of 0 to 800 μm from the optic disk, the RGC population is dense and cannot be accurately counted. In the region furthest from the optic disk (the band of retinal tissue 3200–4000 μm from the optic disk) there is significant variability in RGC density. In contrast, RGC density in the band between 800 and 3200 μm from the optic disk is fairly consistent; this region can therefore be considered suitable for counting of RGCs, as also demonstrated in a recent study showing that damage to RGCs in the laser-induced model of elevated IOP is not uniform. ³⁵ Counting in the entire retinal area can provide the most precise results. ³⁵ It is possible to consistently count, as we did in our study, only fields located within the same band (800–1600 μm). Counting in this way does not produce the absolute RGC numbers but reliably represents differences among treatments to show the least variability within each treatment group. ¹¹ Results presented throughout our study are the mean ± SD RGCs/mm² in the region at a distance of approximately 800 to 1600 μm from the optic disk. This method was further validated throughout this study by adoption of different counting procedures using the same retinas. 

In the retinas excised 21 days after the first laser session, RGCs within each retina were counted in an area measuring either 300 μm² or 640 μm². Counting was performed with the use of a fluorescence microscope (Carl Zeiss, Oberkochen, Germany). 

In the experiment in which CSPG-DS effects were examined for 8 weeks, RGCs were counted according to two independent counting methods so that the numbers could be compared. We covered 50% of the retinal area within the specified band (RGCs were sampled from an area measuring 2400 μm² located 800-1600 μm from the optic disk). 

To ensure objectivity of the RGC counts, an independent investigator randomized the retinas before they were counted, and retinal identities were not disclosed until all RGC counts were completed. Comparison of the two counting methods is depicted (see 1 2 3 4 5 6 Figure 7 ). In the first method, an observer manually counted the number of surviving RGCs; identities of the retinas were masked. In the second, a different observer recorded the number of surviving RGCs calibrated automatically (ImagePro; Media Cybernetics, Silver Spring, MD). In both counting methods, an different independent observer masked the identity of the patients. To determine the effects of CSPG-DS on RGCs, cell densities in each experimental subgroup were compared with those in the corresponding control subgroup treated with phosphate-buffered saline (PBS; Sigma-Aldrich, St. Louis, MO). 

CSPG-DS Administration

The CSPG-DS used in this study was the six-sulfated disaccharide C-4170 (α-ΔUA-(1Þ3)-GalNAc-6S [ΔUA = 4-deoxy-L-threo-hex-4 enopyranosyluronic acid; GalNAc = N-acetyl-D-galactosamine]; 6S = 6-sulfate [Sigma-Aldrich, Steinheim, Germany]). Purity of the disaccharide was analyzed by liquid chromatography/mass spectrometry (LC/MS) and was further verified with MS/MS analysis (Applicable Research Center, Ben-Gurion University of the Negev, Beer Sheba, Israel). As a control we used the nonsulfated disaccharide C-3920 (α-ΔUA-(1⇒3)-GalNAc; Sigma-Aldrich), whose effect is mild compared with that of CSPG-DS. ²⁹ Disaccharides were dissolved in PBS and injected intravenously (25 μg/rat) at different times after the primary insult. Because only approximately 10% of topically applied agents enter the blood, ³⁶ we applied eye drops at a dosage equivalent to 10 times the dosage we found to be optimal when administered intravenously. Five drops (total, 250 μg) were applied topically, one drop every 5 minutes. For oral administration, we fed each rat with 100 or 250 μg CSPG-DS using an 18-gauge curved feeding needle attached to a syringe. Oral treatment was applied according to the same regimen as the intravenous treatment. When the effect of oral administration was examined over a period of 8 weeks, each rat had been fed once a week with 100 μg CSPG-DS. 

Measurement of IOP

To obtain accurate pressure measurements, we determined the pressure in the laser-treated eye with an application tonometer (Tono-Pen XL; Automated Ophthalmics, Ellicott City, MD). On average, 10 measurements were recorded. In the indicated experiments, measurements were obtained 14 days and 21 days after the first laser treatment. Untreated contralateral eyes served as controls. 

Immunohistochemistry

At the indicated time after the first laser session, rats were humanely killed and their eyes were excised, rinsed in PBS, and transferred to 30% sucrose for 3 days at 4°C. Eyes were embedded in compound (Tissue-Tek; Miles, Elkhart, IN) and placed in liquid nitrogen. Transverse cryosections (10 μm) were collected onto gelatin-coated slides and dried at room temperature. They were then fixed in absolute ethanol for 10 minutes at room temperature, washed twice in double-distilled water, and incubated for 3 minutes with 0.05% Tween-20 in PBS. To increase permeability and reduce nonspecific staining, the sections were blocked with a solution containing 10% normal fetal calf serum (FCS; Biological Industries, Beit Ha-Emek, Israel). As primary antibody, we used monoclonal antibodies to major histocompatibility complex (MHC) class II proteins (anti-MHC class II antibodies, 1:50; Serotec, Oxford, UK) and applied them to the tissue sections for 1 hour at room temperature in a humidified chamber. Sections were rinsed three times with Tris-buffered saline (0.05% Tween-20 in PBS) and then incubated with the appropriate fluorescein-conjugated secondary antibody (FITC, 1:200; Serotec) in blocking solution (10% FCS) for 1 hour at room temperature. Control sections (not treated with primary antibodies) were used to distinguish specific staining from nonspecific antibody binding or autofluorescence components (such as lipofuscin) in the eye. Slides were mounted in antifading oil, covered with coverslips, and examined under a laser scanning confocal microscope (LSM510; Zeiss). Sections that contained more than 10 stained cells per slice were considered positively stained. This threshold was defined to allow a clear distinction between values in the two groups. Positively stained sections in each group were counted and expressed as percentages of the total number of sections examined. Eyes from at least three rats in each group were examined. 

Preparation and Culturing of Microglia

Microglia were purified from the brains of newborn (day 0) Lewis rats. Brains were stripped of their meninges and minced with scissors under a microscope (Stemi DV4; Zeiss) in medium (Leibovitz-15; Biological Industries). After trypsinization (0.5% trypsin, 10 minutes, 37°C, 5% CO₂), the tissue was triturated. The cell suspension was washed in culture medium for glial cells (DMEM supplemented with 10% FCS, l-glutamine [1 mM], sodium pyruvate [1 mM], penicillin [100 U/mL], and streptomycin [100 mg/mL]) and was cultured at 37°C, 5% CO₂ in 75-cm² tissue-culture flasks (Falcon; BD Biosciences, San Diego, CA) coated with poly-d-lysine (PDL; 10 mg/mL; Sigma-Aldrich). Half the medium was changed after 6 hours in culture and every second day thereafter starting on day 2, for a total culture time of 10 to 14 days. Microglia were shaken off the primary mixed brain glial cell cultures (150 rpm, 37°C, 4 hours) with maximum yields between days 10 and 14. They were then seeded on a PDL-coated culture flask (1 hour, 37°C, 5% CO₂), and nonadherent cells were rinsed off and grown in culture medium for microglia (RPMI-1640 medium [Sigma-Aldrich] supplemented with 10% FCS, l-glutamine [1 mM], sodium pyruvate [1 mM], β-mercaptoethanol [50 mM], penicillin [100 U/mL], and streptomycin [100 mg/mL]). 

RNA Purification, RT-PCR, and cDNA Synthesis

When indicated, CSPG-DS (50 μg/mL) or lipopolysaccharide (LPS; 100 ng/mL) was added to the microglial culture for 12 hours. Cells were lysed with TRI reagent (MRC, Cincinnati, OH), and total cellular RNA was purified from lysates using the RNeasy kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. RNA was converted to cDNA using SuperScript II (Promega, Madison, WI), as instructed by the manufacturer. We assayed the expression of specific mRNAs using reverse transcription-PCR (RT-PCR) with selected gene-specific primers pairs: β-actin, sense 5-TTGTAACCAACTGGGACGATATGG, antisense 5-GATCT TGATCTTCATGGTGCTAGG (target size, 764 bp); tumor necrosis factor (TNF)-α, sense 5-AGGAGGCGCTCCCCAAAAAGATGGG, antisense 5-GTACATGGGCTCATA CCAGGGCTTG (target size, 551 bp); and CTIIA, sense 5-TTTCAGGGCCTCCTTGAGTGAC-3, antisense 5-AGCAAGGGAGTAGCC ATGCG-3. RT reactions were carried out with 1 μg cDNA and 35 nmol each primer (ReadyMix PCR Master Mix; Abgene, Epson, UK) in 30-μL reaction mixtures. PCR reactions were carried out in a PCR system with the following thermocycles (18 cycles for β-actin and 30 cycles for TNF-α): 95°C for 30 seconds, 60°C for 1 minute, 72°C for 1 minute. To assess the amount of cDNA synthesized, we used β-actin as an internal standard. PCR products were subjected to agarose gel analysis and visualized by ethidium bromide staining. 

Cell Viability Assay

PC12 cells were seeded on collagen-coated 96-well plates at a density of 10⁴ cells per well in differentiation medium containing 100 ng/mL nerve growth factor. Cells were incubated with CSPG-DS at the indicated concentrations for 45 minutes and then washed with PBS and exposed for 15 minutes to 50 μM NOR-3 (a cell-permeable NO donor that provides rate-controlled release of NO; Calbiochem, San Diego, CA). The medium was washed away and replaced with DMEM for another 24 hours of incubation. Viable cells were then counted using the XTT viability kit (TOX-2; Sigma-Aldrich) according to the manufacturer’s instructions. 

Statistical Analysis

Student’s t-test was used in experiments in which only two groups were compared. For comparisons of more than two groups, we used Welch ANOVA when the variances differed significantly over the treatment groups but normal distribution of the data was not rejected. When deviation from equal variances was not significant, ANOVA was used for overall comparison of the means, followed by Dunnett testing for comparison of each of the means to the control values only at P = 0.05. 

Results of single representative experiments (rather than of pooled data) are presented because of the nonuniform nature of the studied populations, which included, for example, old rats and old rats that underwent thymectomy, thus inevitably resulting in some variance between the results of each independent experiment. Moreover, old rats underwent the experiment at ages ranging from 10 to 12 months, resulting in further variability. We used corresponding littermates as controls for each of the studied populations. In each experiment, a significant difference was observed between the treated population and its untreated control. 

Results

CSPG-DS Application in the Chronic Model of Elevated Intraocular Pressure

Lewis rats were subjected to chronic elevations in IOP according to a model in which IOP is elevated by repeated laser irradiation of the limbal and episcleral veins to prevent the aqueous humor from draining out of the eye. Previous studies suggest that cell death in this model starts on day 7 after the first laser session. ³⁰ We injected CSPG-DS according to one of three different intravenous regimens: a single injection given on the day of the insult, a single injection on day 7 after the insult, or a total of four injections given on days 7, 9, 11, and 14 after the first laser irradiation. Rats in the control group were injected with PBS. Doses used in vivo in the rats were chosen on the basis of our previous experience with the use of this disaccharide in mice ^{29 33} ; the dosage found to be efficient in a mouse (5 μg/mice) was converted for use in the rat (25 μg/rat). Repeated injection of CSPG-DS was the regimen found to be the most effective (Fig. 1) . With this regimen, the number of surviving RGCs (all values are mean ± SD per square millimeter within retinas located 800-1600 μm from the optic disk) in rats was 2068 ± 212 compared with 1430 ± 242 in PBS-injected controls (Fig. 1) . Welch ANOVA disclosed significant overall differences between the groups (P < 0.0001). Dunnett tests at P = 0.05 revealed significant differences between the repeated application regimen and the control. Given the statistical findings, we conducted all subsequent experiments according to the repeated-applications regimen. 

To determine whether the observed neuroprotective effect of the six-sulfated disaccharide CSPG-DS was specific, we analyzed the effect of another CSPG-derived disaccharide, a nonsulfated disaccharide, previously shown to have a modest effect relative to that of CSPG-DS. ²⁹ Elevated IOP was induced in rats, which were then treated with CSPG-DS, nonsulfated disaccharide, or PBS, according to the repeated-applications regimen (Fig. 2) . Photomicrographs illustrate the effect of CSPG-DS on RGC survival (Figs. 2A 2B) . Figure 2shows that CSPG-DS effectively protected RGCs from death (2203 ± 381 compared with 1152 ± 168 in the PBS-treated group; P = 0.0004; Welch ANOVA), whereas the numbers of surviving RGCs after treatment with the nonsulfated disaccharide did not differ significantly from the numbers that survived after treatment with PBS (Fig. 2C) . 

To rule out the possibility that the therapeutic effect observed after treatment with the disaccharide resulted from a CSPG-DS-induced decrease in IOP, we measured the IOP 7 days after the first laser session and again 7 days after the second session (on day 14). IOP values were not affected by disaccharide treatment (Fig. 2D) , suggesting that the effect of CSPG-DS on neuronal survival does not occur through an effect on IOP levels. 

CSPG-DS Application in Aged Lewis Rats with Chronically Elevated IOP

Aging is a crucial risk factor for the development and progression of glaucoma. ^{37 38 39 40} The increasing mean life expectancy worldwide has inevitably been accompanied by an increase in the numbers of persons with glaucoma ⁴¹ and other neurodegenerative diseases. We therefore compared RGC survival after inducing high IOP, with or without CSPG-DS treatment, in aged (10–12 months old) and adult (mean age, 2 months) Lewis rats. Significantly more RGCs survived after CSPG-DS treatment in aged and adult rats than after PBS treatment in controls (1924 ± 53 and 1313 ± 130, respectively, in adult rats, and 1010 ± 40 and 523 ± 105, respectively, in aged rats; P < 0.0001; Student’s t-test; Fig. 3 ). These results also indicate that CSPG-DS can protect against the effects of IOP elevation in aging populations. 

CSPG-DS Application in Chronic Elevation of IOP in Rats Deprived of T Cells

Among the factors that may contribute to the increased prevalence of glaucoma with age is decreased efficiency of the immune response. ⁴² To address this possibility, we examined the ability of CSPG-DS to protect RGCs in rats that underwent thymectomy at birth (resulting in lifelong deprivation of T cells). We compared the efficacy of such treatment to that in rats with elevated IOP and in which the T-cell population was normal. In the adult thymectomy control group, the number of RGCs that survived IOP elevation was 644 ± 157, whereas almost twice as many RGCs survived in the adult thymectomy rats treated with CSPG-DS (1131 ± 184; P < 0.0001; Student’s t-test; Fig. 4A ). 

T-cell deficiency is commonly observed in aged animals. We examined the effect of CSPG-DS on aged rats with chronically elevated IOP that underwent thymectomy. Treatment of these rats with CSPG-DS according to the dosage and repeated-applications regimen described in Figure 1resulted in RGC survival of 650 ± 99 compared with 317 ± 51 in PBS-treated aged controls that underwent thymectomy (P < 0.0001; Student’s t-test; Fig. 4B ). These results suggest that CSPG-DS can protect against neuronal loss after IOP elevation in aged and in immunocompromised rats. 

Topical Application of CSPG-DS Is Effective in Acute and Chronic Models of IOP Induction

To examine the therapeutic potential of CSPG-DS for rat RGCs when applied in the form of eye drops, we used both the acute transient model of elevated IOP and the model of chronic elevation. In the acute model, IOP insult causes more immediate loss of RGCs and the therapeutic window is narrower than in the chronic model (Ben Simon, unpublished observations, 2006). CSPG-DS was topically applied to normal adult rats according to a protocol designed to fit the time scale appropriate for each model (Fig. 5) . In the chronic model of IOP induction, topical administration of CSPG-DS according to the repeated-applications regimen described in Figure 1resulted in RGC survival of 1897 ± 60 (n = 6) compared with 1229 ± 130 in the PBS-treated control group (n = 4; P < 0.0001; Student’s t-test; Fig. 5A ). In the acute model, CSPG-DS was applied only once, immediately after IOP was induced. In this model, RGC survival was 1668 ± 64 (n = 6) compared with 1224 ± 62 (n = 5) in the PBS-treated group (P < 0.0001; Student’s t-test; Fig. 5B ). 

Oral Administration of CSPG-DS in the Chronic Model of IOP

The observation that intravenous and topical administration of CSPG-DS had beneficial effects on RGC survival after IOP elevation led us to examine the therapeutic potential of this compound when administered orally. The choice of dosage was based on a previous finding in mice that a 10-fold increase in the effective dose of intravenously injected CSPG-DS was required to achieve a similar protective effect by oral administration (data not shown, 2006). The oral dosage used here was 100 or 250 μg/rat, administered according to the repeated-applications regimen used for intravenous application (days 7, 9, 11, and 14). As a control for examining the effectiveness of the oral treatment, we used rats treated by intravenous administration of 25 μg/rat. Oral administration of CSPG-DS resulted in significant protection. Indicated by RGC survival in the groups treated orally with either 100 μg/rat or 250 μg/rat (2119 ± 339 and 1804 ± 136, respectively, compared with 1375 ± 220 in the PBS-treated control group; Fig. 6 ), similar to the protection achieved with intravenous treatment (2101 ± 336). 

Weekly Oral Application of CSPG-DS Protects RGCs in the Chronic IOP Model

We performed an additional experiment designed to examine the effect of oral CSPG-DS administered weekly for a prolonged period (8 weeks). Rats subjected to chronic elevations in IOP were fed CSPG-DS (100 μg/rat) once a week for 8 weeks, starting 1 week after the first laser session. The resultant neuroprotection was significant: counting of surviving RGCs yielded 1943 ± 159 compared with 1093 ± 176 in the PBS-treated control group (P < 0.0001; Fig. 7A ). To validate our counting method, we counted four retinal fields at a magnification of either ×40 (Fig. 7A)or ×10 (Fig. 7B) . The two counts yielded relatively similar results in terms of the numbers of surviving RGCs/mm² (1093 ± 176 [PBS] and 1943 ± 159 [CSPG-DS] at ×40 magnification [Fig. 7A ] compared with 1192 ± 118 [PBS] and 2441 ± 219 [CSPG-DS] under ×10 magnification [Fig. 7B ]). 

CSPG-DS Induces Microglial Activation in Rats with Elevated IOP

The neuroprotective effect of T cells in the injured CNS ^{43 44} is not fully understood. A major factor in the contribution of activated T cells to CNS recovery is their ability to activate the local macrophages/microglia in the CNS in a way that causes them to express a neuroprotective phenotype characterized by MHC class II expression. ²⁰ Activation of microglia to adopt this neuroprotective phenotype can also be induced by CSPG-DS and is similarly accompanied by MHC class II expression. ²⁹ Therefore, we used MHC class II expression to evaluate CSPG-DS beneficial activation. To determine whether such activation of microglia occurs in eyes with elevated IOP, we used immunohistochemical staining to examine the eyes of Lewis rats that were subjected to chronic IOP elevation and were treated 7 days later with CSPG-DS (topical application) or with PBS (control). Rats were humanely killed and their eyes removed 10 days after chronic IOP was induced. Staining of slices for MHC class II showed that whereas in PBS-treated control rats only 33% of the slices were positive, stained cells were observed in 64% of the slices obtained from rats treated with CSPG-DS. Representative images are presented in Figures 8A and 8B . 

To verify that the observed microglial activation could be a direct effect of CSPG-DS, we examined purified cultures of microglia for CIITA, a non-DNA-binding coactivator that serves as the master control factor for MHC class II expression. ^{45 46} We compared mRNA coding for CIITA expression in CSPG-DS-activated microglia with that in three controls: (i) a positive control consisting of microglia activated by interferon (IFN)-γ, known to induce MHC class II expression ²⁰ ; (ii) a negative control consisting of microglia activated by lipopolysaccharide (LPS), known to activate microglia without inducing MHC class II expression; and (iii) untreated microglia. RT-PCR analysis disclosed a significant increase in CIITA in microglia activated by CSPG-DS (Fig. 8C) . 

The phenotype of activated microglia can be protective or destructive. A major characteristic of the destructive phenotype is secretion of TNF-α, which can cause neuronal death ²⁰ and is associated with the death of RGCs in glaucoma. Examination of the mRNA levels for TNF-α in microglia activated by CSPG-DS, IFN-γ, or LPS showed that microglia activated by CSPG-DS, in contrast to LPS-activated microglia, did not induce TNF-α expression (Fig. 8C) . 

CSPG-DS Protects PC-12 Cells from Death Induced by Nitric Oxide

Studies by our group have indicated that CSPG-DS can also directly exert a supportive effect on neurons, endowing them with better ability to withstand stressful conditions. Cells of the pheochromocytoma-derived cell line PC12, when cultured in the presence of nerve growth factor, differentiate to resemble sympathetic neurons morphologically and functionally. ^{47 48} This cell line thus serves as an excellent in vitro model for studying chemical disruption of processes associated with neuronal differentiation, synthesis, storage, and release of neurotransmitters and the effects of neurotoxic factors. ⁴⁹ In a previous study, we used PC12 cells as an in vitro neuronal model to study the protective effect of CSPG-DS under conditions of glutamate toxicity. ²⁹ Nitric oxide toxicity is considered a cause of neuronal death in glaucoma and in relevant animal models of IOP induction. ^{50 51} To determine whether CSPG-DS can also protect neurons from death induced by an excess of NO, we used a classical model of toxicity induction in PC12 cells. Survival of PC12 cells in the presence of NO increased with increasing dosages of added CSPG-DS (1–50 μg/mL). Overall comparison of the means was significant (P = 0.005). A Dunnett test showed that at P = 0.05, only the 50-μg/mL dose had a significantly higher mean than the control (Fig. 9) . 

Discussion

Results of this study showed that CSPG-DS was highly effective in protecting against RGC loss in models of acute and chronic high IOP. Neuroprotective efficacy was evident not only in adult rats but also in aged and in immunocompromised rats. Our data suggest that the beneficial effect of CSPG-DS is mediated, at least in part, by local control of the microglial response and by a direct effect of CSPG-DS on neuronal survival. 

The natural source of CSPG-DS is the extracellular molecule CSPG, which increases in abundance after various CNS insults. ^{28 52} Enzymes that naturally degrade CSPG in vivo have not yet been identified (suggested candidates are hyaluronidase and matrix metalloproteinases); nevertheless, patterns of CSPG expression after CNS injury are characterized by a significant increase in the presence of this molecule within hours of injury and a gradual decline thereafter. ⁵³  

Recent findings reported by our group have suggested that CSPG-DS can beneficially affect neural and immune cells at the site of damage. ²⁹ In vitro, CSPG-DS protects neurons from the aftermath of insult (axonal collapse, glutamate toxicity, mechanical injury, and others) through activation of neuroprotective signaling pathways such as PYK2 and PKC. In vivo, CSPG-DS has a significantly protective effect on retinal neurons experiencing the toxic effects of glutamate or aggregated β-amyloid. In addition, CSPG-DS induces the activation of microglia, causing them to adopt a neuroprotective phenotype associated with the direct activation of ERK1/2 and PYK2, ²⁹ two major signaling cascades in microglial activation pathways. 

These findings prompted us to suggest that CSPG-DS might be protective in animal models of neurodegenerative conditions. We tested this hypothesis in rats with elevated IOP. This model simulates open-angle glaucoma, in which the observed loss of RGCs results not only from elevated IOP ³² but also from neurodegeneration evoked by processes such as increased NO levels, observed after IOP elevation and considered a major factor contributing to RGC loss. ⁵⁴ Neurons affected by factors that operate after IOP induction are likely to be amenable to neuroprotective therapy. ⁵⁵ In the present study we demonstrated that, consistent with our previous findings, ²⁹ adding CSPG-DS to the culture medium of RGCs subjected to excessively high levels of NO results in significantly reduced cell death. 

Also demonstrated in this study was a direct effect of CSPG-DS on the activation of microglia to adopt a neuroprotective phenotype. Ever since it was first proposed that glaucoma is a neurodegenerative disease characterized by a self-perpetuating process of degeneration, neuroprotective approaches have become a focus of therapeutic endeavor. ^{55 56 57} Special attention has been devoted to therapeutic modalities that neutralize mediators of toxicity ^{2 58 59 60} or that compensate for deficiencies in growth and survival factors. ^{61 62} A study by our group suggested that immune responses mediated by activated T cells stimulate naturally occurring mechanisms of neuroprotection, which in turn are associated mainly with local activation of the microglia, causing them to adopt a neuroprotective phenotype. ²⁰ Results of the present study suggest an additional pathway for microglial modulation that does not require the recruitment of T cells or of a systemic immune response. Our previous findings indicated that microglia activated by CSPG-DS can protect neurons from the consequences of mechanical injury in the organotypic hippocampal slice culture model. ²⁹ In addition, microglial activation by CSPG-DS is associated with MHC class II expression, recently shown to characterize various modes of microglial activation (e.g., by IFN-γ or IL-4). ²⁰ MHC class II-expressing microglia were reported to be neuroprotective in various models. ^{20 52 63 64 65} To further characterize the observed effect of CSPG-DS on microglia, we assayed the mRNA that encodes CIITA in microglia activated by CSPG-DS. MHC class II expression is known to be regulated mainly at the level of transcription, and most signals that activate or repress its expression under physiological or pathologic situations converge on one or more of three promoters that drive transcription of the MHC2TA gene (which encodes CIITA expression). CIITA, a non-DNA-binding coactivator that serves as the master control factor for the expression of MHC class II, is the most important factor dictating the expression of this protein. ^{45 46} RT-PCR analysis disclosed an increase in CIITA levels in microglia activated by CSPG-DS. 

The adverse effect of microglial activation is commonly associated with the secretion of TNF-α, which can lead to cell death and which reportedly contributes to neuronal damage in the elevated IOP model of glaucoma. ⁶⁶ We demonstrated here that the activation of microglia with CSPG-DS, in contrast to other modes of activation, does not induce TNF-α secretion. A previous study by our group demonstrated similar support of neuronal survival and renewal by microglia triggered by IL-4-activated microglia. ²⁰  

A specific receptor for CSPG-DS has not yet been identified. It is possible that CSPG-DS interacts with CSPG receptors as an agonist or as an antagonist. Several receptors were suggested as candidates for interaction with different members of the CSPG family, such as CD44 with aggrecan ⁶⁷ and TLR2 and TLR4 with biglycan. ⁶⁸ It is unlikely, however, that CSPG-DS acts as a CSPG antagonist or operates through the modulation of CSPG-related enzymes because its effects in vitro are detectable in the absence of CSPG. The occurrence of such a mechanism in vivo, however, cannot be ruled out. The significant difference in activity levels observed between the sulfated and the nonsulfated CSPG-derived disaccharides suggests that the effects of CSPG-DS are highly specific. 

The lack of difference in IOP levels observed here between CSPG-DS-treated and PBS-treated rats indicates that whatever the cellular mechanisms underlying the CSPG-DS-mediated effect, they cannot be attributed to a reduction in IOP. 

Treatment with CSPG-DS yielded a neuroprotective effect in adult rats and in immunocompromised aged rats. Because the immune system deteriorates with age, it seems reasonable to expect that older persons will be less well equipped than younger persons to withstand injurious conditions in the CNS in general. This is likely to be a major factor contributing to the increased prevalence of neurodegenerative diseases, including glaucoma, in elderly populations. The observed beneficial effect of a prolonged period of weekly treatment on RGC survival substantiates the potential of CSPG-DS treatment for chronic conditions. The neuroprotective effect of CSPG-DS on RGC survival demonstrated in this study, as in other studies that have tested neuroprotective compounds, is based on counting a specific area of the retina ^{11 32 69} (800–1600 μm from the optic disk). This retinal region shows a uniform RGC distribution and can reliably be used for comparison between treatments. However, it should be noted that RGCs are not uniformly distributed throughout the retina ^{11 34 35} ; thus, analysis of the entire retina ^{11 34 35} might provide additional insight into the potential effect of CSPG-DS. 

The effects of CSPG-DS were demonstrated here using different (acute and chronic) models of IOP elevation, various modes of CSPG-DS application (systemic, topical, and oral), different treatment periods, and recipients with different characteristics (normal, immune deficient, and aged rats). These findings suggested that all the treatment routes converge into a common mechanism of controlling the cellular and molecular environment of the RGCs. The efficacy of CSPG-DS under all these conditions appears to augur well for the therapeutic use of this compound. However, further studies are needed to establish the most suitable regimen and the most effective mode of application under different pathologic situations. 

The significant neuroprotection of RGCs obtained by the exogenous application of CSPG-DS to aged rats in this study further highlights the potential of this compound for the treatment of glaucoma in aged populations. Taken together, our results suggest that CSPG-DS is a highly promising candidate for development as a treatment modality for neurodegenerative disorders in elderly and immunodeficient populations. 

²  These authors contributed equally to the work presented here and should therefore be regarded as equivalent authors.

⁴  Deceased 2004.

 Supported by grants from Gail and Richard Siegal (MS) and the Horowitz Foundation. MS holds the Maurice and Ilse Katz Professorial Chair in Neuroimmunology.

 Submitted for publication September 13, 2005; revised February 12, March 29, May 22, June 26, and August 2, 2006; accepted December 26, 2006.

 Disclosure: S. Bakalash, P; A. Rolls, P; O. Lider, P; M. Schwartz, P

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Repeated intravenous injections of CSPG-DS protect retinal ganglion cells from the consequences of chronically high IOP. Chronically high IOP was induced in Lewis rats. CSPG-DS (25 μg/rat) was injected intravenously according to three different regimens: a single injection on the day of the first laser session (day 0), a single injection on the day of the second laser session (day 7), and repeated injections on days 7, 9, 11, and 14. Rats in the control group were injected with PBS according to the repeated-applications regimen. Whole-mounted retinas were excised 3 weeks after the first laser session and 24 hours after the injection of dye into the optic nerve. Values (mean ± SD) are from 1 of 3 representative experiments. Overall differences between the groups were significant (P < 0.0001; Welch ANOVA). Mean values obtained with the repeated-applications regimen differed significantly from control values (Dunnett test; P = 0.05).

Figure 2.

 

CSPG-DS promotes RGC survival in Lewis rats subjected to chronic elevation in IOP but does not affect the pressure itself. Chronically high IOP was induced in Lewis rats, which were then treated with CSPG-DS, a CSPG-derived nonsulfated disaccharide, or PBS. Dosage and repeated-applications regimen were as in Figure 1 . (A) A montage (magnification, ×40) of representative retinas from each of the three treatment groups: (i) CSPG-DS; (ii) nonsulfated disaccharide; (iii) PBS. (B) High magnification (×400) of a representative field from each of the three treatment groups. (C) Surviving RGCs in each retina were counted, and differences among the three treatment groups were significant (P = 0.0004; ANOVA). Mean RGC count of CSPG-DS-treated rats was significantly higher and that of rats treated with nonsulfated disaccharide was nonsignificantly higher than that of PBS-treated rats. (D) IOP values (mean ± SD) 7 and 14 days after the first laser session.

Figure 3.

 

CSPG-DS promotes RGC survival in aged Lewis rats with chronically elevated IOP. Chronically high IOP was induced in adult (2-month-old) and aged (10- to 12-month-old) rats. In each group, one subgroup was treated with CSPG-DS according to the dosage and the repeated-applications regimen described in Figure 1 , and the other was treated with PBS according to the same regimen. Surviving RGCs in each retina were counted. Values (mean ± SD) are from one of two experiments in which similar results were obtained (***P < 0.0001; Student’s t-test).

Figure 4.

 

CSPG-DS protects RGCs from chronically elevated intraocular pressure in T-cell-deprived rats, regardless of age. (A) Adult rats (2 months old) that underwent thymectomy were subjected to chronic elevations in IOP. Seven days later they were injected with CSPG-DS or PBS according to the repeated-applications regimen described in Figure 1 . Results shown here are from 1 of 2 repeats. (B) Aged rats (10 to 12 months old) that underwent thymectomy were subjected to chronic elevations in IOP and were injected with CSPG-DS or PBS according to the same regimen. Retinas were labeled, excised, and analyzed for viable RGCs 21 days after the first laser session. Values (mean ± SD per mm²) are from 1 of 2 experiments in which similar results were obtained (***P < 0.0001; Student’s t-test).

Figure 5.

 

Topical application of CSPG-DS protects RGCs from the consequences of chronic or acute elevation in IOP. Lewis rats were subjected to chronic (A) or acute (B) elevation in IOP, after which five drops of CSPG-DS were applied to the cornea at 5-minute intervals for 25 minutes. (A) In the chronic model, CSPG-DS was applied according to the repeated-applications regimen described in Figure 1 . Retinas were excised and RGCs were counted 21 days after the first laser session. (B) In the acute model, CSPG-DS was applied immediately after the IOP elevation, and surviving RGCs were counted after 14 days. Values presented in each experiment (mean ± SD) are from 1 of 2 experiments in which similar results were obtained (***P < 0.0001; Student’s t-test).

Figure 6.

 

Oral application of CSPG-DS protects retinal ganglion cells from the consequences of chronic elevation in IOP. Chronically high IOP was induced in Lewis rats. CSPG-DS (25 μg/rat) was administered orally (100 or 250 μg/rat) or injected intravenously (as a positive control), according to the repeated-applications regimen described in Figure 1 . Rats in another control group were injected with PBS according to the same regimen. Dye was injected into the optic nerve, and 24 hours later (3 weeks after the first increase in IOP) whole-mounted retinas were excised. Values (mean ± SD) from two repeated experiments were pooled (12 rats/group). ANOVA F test was significant (P < 0.0001), and the results for each of the three treatment groups differed significantly from the results for the PBS-treated group (Dunnett test; P = 0.05).

Figure 7.

 

Weekly oral treatment with CSPG-DS protects RGCs from the consequences of chronic elevation in IOP. Chronically high IOP was induced in Lewis rats. CSPG-DS (100 μg/rat) was administered orally once a week for 7 weeks to the rats in the treatment group. Rats in the control group were treated with PBS according to the same regimen. Dye was injected into the optic nerve (7 weeks after the first increase in IOP), and 24 hours later retinas were excised and whole-mounted. (A) RGCs were counted manually (magnification, ×40). (B) RGCs were calibrated automatically (magnification, ×10). (C) Representative retinal fields from groups treated weekly with PBS or CSPG-DS (magnification, ×10). Values are mean ± SD; (n = 6 rats per group; ***P < 0.0001; Student’s t-test).

Figure 8.

 

CSPG-DS induces microglial activation in Lewis rats subjected to chronic elevation of IOP. Chronically high IOP was induced in Lewis rats. CSPG-DS or PBS (control) was applied topically (as eye drops) 7 days after the first laser session. Three days later the eyes were excised and sectioned, and their retinas were stained for MHC class II. (A, B) Representative photomicrographs of MHC class II-labeled retinas injected with PBS and CSPG-DS, respectively (arrows: representative MHC class II-labeled cells). (C) In vitro expression of CTIIA mRNA and TNF-α by microglia activated with CSPG-DS, measured by RT-PCR. mRNA from IFN-γ-treated and LPS-treated cells was used as a positive and a negative control, respectively. β-actin was used as a housekeeping gene.

Figure 9.

 

CSPG-DS promotes neural survival after NO toxicity in PC12 cell cultures. Death of cultured PC12 cells was induced by a toxic dose of NO (50 μm). Cells were treated with CSPG-DS or were left untreated. Numbers of surviving cells were determined by XTT assay and expressed as the percentage increase (mean ± SD) in cell viability relative to the untreated control. Representative data from 1 of 3 experiments are shown. Overall differences between the groups were significant (P < 0.005; Welch ANOVA). Mean value obtained with the 50-μg/mL dose differed significantly from that of the control (Dunnett test; P = 0.05).

The authors thank Hilary Voet for assistance with the statistical analysis and Shirley R. Smith and Shelly Schwarzbaum for editing the manuscript.