Experimental Eye Research 130 (2015) 38e50

Contents lists available at ScienceDirect

Experimental Eye Research journal homepage: www.elsevier.com/locate/yexer

Chitosan oligosaccharides prevented retinal ischemia and reperfusion injury via reduced oxidative stress and inflammation in rats I-Mo Fang a, b, Chung-May Yang b, Chang-Hao Yang b, * a b

Department of Ophthalmology, Taipei City Hospital Zhongxiao Branch, Taipei, Taiwan Department of Ophthalmology, National Taiwan University Hospital, Taipei, Taiwan

a r t i c l e i n f o

a b s t r a c t

Article history: Received 17 March 2014 Received in revised form 24 November 2014 Accepted in revised form 1 December 2014 Available online 2 December 2014

The purpose of the present study was to investigate the protective effect and mechanism of chitosan oligonucleotides (COS) on retinal ischemia and reperfusion (I/R) injury. Rats pretreated with PBS, lowdose COS (5 mg/kg), or high-dose COS (10 mg/kg) were subjected to retinal ischemia by increasing their intraocular pressure to 130 mm Hg for 60 min. The protective effect of COS was evaluated by determining the electroretinograms (ERGs), morphology of the retina, and survival of retinal ganglion cells (RGCs). The oxidative damage was determined by imuunohistochemistry and ELISA, respectively. The expressions of inflammatory mediators (TNF-a, IL-1b, MCP-1, iNOS, ICAM-1) and apoptotic-related proteins (p53, Bax, Bcl-2) were quantified by PCR and Western blots. The detection of NF-kB p65 in the retina was performed by immunofluorescence. The protein levels of IkB and phosphorylated mitogen-activated protein kinases [MAPK; viz. extracellular signal-regulated protein kinases (ERK), c-Jun N-terminal kinases (JNK) and p38] and the NF-kB/DNA binding ability were assessed by Western blot analysis and EMSA. We found that pretreatment with COS, especially a high dosage, effectively ameliorated the I/R-induced reduction of the b-wave ratio in ERGs and the retinal thickness and the survival of RGCs at 24 h. COS decreased the expression of inflammatory mediators, p53 and Bax, increasing Bcl-2 expression and thereby reducing retinal oxidative damage and the number of apoptotic cells. More importantly, COS attenuated IkB degradation and p65 presence in the retina, thus decreasing NF-kB/DNA binding activity after I/R. In addition, COS decreased the phosphorylation levels of JNK and ERK but increased the phosphorylation level of p38. Pretreatment with p38 inhibitor (SB203580) abolished the protective effect of COS on retinal oxidative damage, as indicated by increased retinal 8-OHdG stains, and significantly increased the expression of inflammatory mediators (TNF-a, MCP-1, iNOS, ICAM1) in I/R-injured rats. In conclusion, COS prevented retinal I/R injury through its inhibition of oxidative stress and inflammation. These effects were achieved by blocking the activation of NF-kB, JNK, and ERK but promoting the activation of p38 activation. © 2014 Elsevier Ltd. All rights reserved.

Keywords: Chitosan oligosaccharides Ischemia-reperfusion injury NF-kB Mitogen-activated protein kinases (MAPK) Oxidative damage

1. Introduction Retinal ischemia/reperfusion (I/R) injury is critically involved in the pathogenesis of several major vision-threatening diseases, including diabetic retinopathy, hypertensive retinopathy, acute glaucoma, and retinal vascular occlusion (Tso and Jampol, 1982;

Abbreviations: COS, chitosan oligonucleotides; I/R, ischemia and reperfusion; ERG, electroretinogram; RGCs, retinal ganglion cells; TUNEL, TdT-dUTP terminal nick-end labeling; ROS, reactive oxygen species. * Corresponding author. Department of Ophthalmology, National Taiwan University Hospital, No.7, Chung-Shan South Road, Taipei, Taiwan. E-mail address: [email protected] (C.-H. Yang). http://dx.doi.org/10.1016/j.exer.2014.12.001 0014-4835/© 2014 Elsevier Ltd. All rights reserved.

Stefansson et al., 1992; Levine, 2001). These diseases are major causes of blindness worldwide and usually result in vision loss due to irreversible damage to the retinal neurons. Retinal damage induced by I/R injury is caused by the depletion of adenosine triphosphate during the ischemia status (Yokota et al., 2011) and by the generation of reactive oxygen species and proinflammatory mediators during the reperfusion status (Szabo et al., 1991; Laskowski et al., 2000). Chitosan oligosaccharides (COS), the hydrolyzed product of chitosan, is a mixture of oligomers of b-1,4-linked D-glucosamine residues and is abundant in the exoskeletons of crustaceans and the cell walls of fungi and insects (Pae et al., 2001). COS is known to have various biological activities, including antitumor,

I.-M. Fang et al. / Experimental Eye Research 130 (2015) 38e50

antimicrobial, anti-inflammation, anti-oxidative, and antiapoptotic effects (Pangestuti and Kim, 2010; Qin et al., 2002; Chen et al., 2006; Joodi et al., 2011). COS has good solubility in water and is easily absorbed in the intestine, which makes it an attractive ingredient in many healthy foods or dietary supplements. We have previously shown that COS effectively attenuated oxidative-stress related retinal degeneration in rats. However, whether COS is able to exert protective effects on an animal model of retinal I/R injury remains unknown. Transiently raising the intraocular pressure is a well-established animal model of retinal I/R injury. Interruption of the blood supply to the retina results in a wide variety of metabolic derangements, and the process of reperfusion itself is deleterious to injured retinal cells through the generation of free radicals and inflammatory cytokines. This model induces an extensive loss of retinal ganglion cells and the inner nuclear layer and an increase in apoptotic cells in the inner retina (Lam et al., 1999; Wu et al., 2004), both of which closely resemble the pathological changes observed in patients suffering from retinal ischemic insults, severe diabetic retinopathy, and acute glaucoma (Zheng et al., 2007). Because oxidative stress and inflammation are contributing factors in the pathogenesis of retinal I/R injury, we hypothesized that COS may display anti-inflammatory and anti-oxidative stress effects in protecting retinal cells from I/R injury. In view of the crucial role of NF-kB and mitogeneactivated protein kinase (MAPK) in regulating retinal inflammation and oxidative stress during retinal I/R injury, we evaluated the effects of COS on NF-kB activation and MAPK phosphorylation in a rat model of retinal I/R injury induced by transiently raising the intraocular pressure.

2. Material and methods 2.1. Reagents Chitosan oligosaccharide was purchased from SigmaeAldrich (St. Louis, MO, USA). The DNA fragmentation detection kit (TUNEL) was obtained from Calbiochem (La Jolla, CA, USA). Green Fluorescent Protein (GFP) antibody was purchased from BioVision (Mountain View, CA, USA). Mounting medium with 40 , 6-diamidino -2-phenylindole (DAPI) and phycoerythrin Streptavidin antibodies was obtained from Vector Laboratories (Burlingame, CA, USA). Antip65 antibodies were purchased from Rockland (Gilbertsville, PA, USA).

2.2. An animal model of retinal ischemia/reperfusion injury SpragueeDawley (SD) rats weighting 150e200 g were used in all subsequent experiments. All animal experiments in this study were carried out in strict accordance with the recommendations in the guide for the care and use of Laboratory Animals of the National Institutes of Health. The protocol was approved by the committee on the Ethics of Animal Experiments of the National Taiwan University. All surgeries were performed under sodium pentobarbital anesthesia, and all efforts were made to minimize suffering. The anterior chamber of the right eye was cannulated with a 27-gauge infusion needle connected to a bottle containing normal saline. The intraocular pressure was raised to 130 mmHg for 60 min by elevating the saline reservoir. Retinal ischemia and reperfusion were confirmed by the whitening of the fundus and restoration of the retinal blood flow. Sham-procedure right eyes were treated similarly but without the elevation of the bottle; thus, normal ocular tension was maintained.

39

2.3. Animal grouping and treatment SD rats were randomly divided into four groups: Group 1: intraperitoneal injection of phosphate-buffered saline (PBS) and then cannulation only without elevation of the bottle, serving as normotensive control (control). Group 2: intraperitoneal injection of PBS before inducing retinal ischemia/reperfusion injury (PBS-treated group). Group 3: intraperitoneal injection of a low dose of COS (5 mg/ kg) before inducing retinal ischemia/reperfusion injury (lowdose COS group). Group 4: intraperitoneal injection of a high dose of COS (10 mg/ kg) before inducing retinal ischemia/reperfusion injury (highdose COS group).

2.4. Electroretinogram (ERG) recordings The ERG was performed at 24 h and 7 days after retinal ischemia/reperfusion injury. The rats were dark-adapted for 1 h before performing the ERG. All manipulations were performed under dim red light illumination. After being anesthetized, the rats were place on a heating pad. A recording electrode was placed on the cornea after the application of 0.5% methyl cellulose. A reference electrode was attached to the shaved skin of the head, and a ground electrode was clipped the animal's tail. A single flash light (duration, 100 ms) 30 cm from the eye was used as the light stimulus. Responses were amplified with a gain setting of ±500 mV and filtered with low 0.3 Hz and high 500 Hz from an amplifier. The pattern of b-waves was recorded. The b-wave ratio was defined as the b-wave amplitude of the right eye/the b-wave amplitude of the left eye. The fold of the b-wave ratio represented the b-wave ratio at 24 h or day 7/the b-wave ratio at day 0. 2.5. Histological study and retinal thickness For all control and experimental animals, both eyes were collected at 24 h and at day 7 after treatment. The specimens were fixed with 4% paraformaldehyde in PBS. Sections (1 mm) were cut along the vertical meridian of each eye and passed through the optic nerve head for staining with hematoxylin and eosin (H&E). We measured different layer thicknesses to quantify the ischemic damage in the retina at day 7. The total retinal thicknesses (from the inner limiting membrane to the pigment epithelium), the outer nuclear layers (ONL), the inner nuclear layer (INL), and the inner plexiform layer (IPL) were measured. Alterations in the thickness of the retinal layers were measured for each eye in the same topographic region of the retina (1 mm from the optic nerve head) under 200 magnification. 2.6. Neu N stain in flat-mounted retinas and counting of NeuNpositive cells The density of retinal ganglion cells (RGCs) was evaluated by immunofluorescence staining with NeuN at 24 h after retinal I/R injury. Briefly, the retina was cryoprotected overnight in 30% sucrose, followed by three freezeethaw cycles and overnight incubation with monoclonal FITC-conjugated NeuN antibody (Millipore, Billerica, MA, USA). Finally, the retina was flat-mounted and viewed with a Leica TSL AOBS SP5 confocal microscope (Leica Microsystems, Exton, PA, USA). The number of Neu N-positive cells was counted in 4 selected retinal areas located at the same eccentricity (approximately 1.5 mm from the optic disk) in the four retinal quadrants. The cell number was quantified with image-analysis

40

I.-M. Fang et al. / Experimental Eye Research 130 (2015) 38e50

software (Image-Pro Plus, ver.6.0; Mediacybernetic, Atlanta, GA, USA). 2.7. Terminal deoxynucleotidyl transferase-mediated dUTPbiotinide end labeling (TUNEL)

determined using image-analysis software (Photoshop, ver.7.0; Adobe Systems, San Jose, CA, USA). The optical densities were calculated and standardized based on the density of the b-actin band. 2.10. Measurement of oxidative damage in the retina

Rats were euthanized, and frozen sections were prepared, as described above. TUNEL assays were performed according to the manufacturer's instructions. Sections were visualized on a fluorescent microscope (Nikon, Melville, NY, USA). 2.8. Semi-quantitative polymerase chain reaction (PCR) Total RNA in the retina was extracted with TRIzol reagent (Invitrogen-Life, Gaithersburg, MD, USA). One microgram of total RNA was annealed with 300 ng oligo (dT) (Promega, Madison, WI, USA) for 5 min at 65  C and reverse-transcribed into cDNA using 80 U Moloney murine leukemia virus reverse transcriptase (MMLVRT; Gibco-life, Grand Island, NY, USA) for 1 h at 37  C. The cDNA product was subjected to PCR with specific primers (Table 1). Amplification was performed in a thermocycler (MJ Research, Waltham, MA, USA) with a 1-min denaturation at 94  C and a 3-min extension at 72  C. The annealing temperature was between 62  C and 42  C, and the temperature was decreased at 1  C increments, followed by 21 cycles at 55  C. Finally, the temperature was elevated to 72  C for 10 min and then reduced to 4  C. We obtained a 10 mL sample of each PCR product to perform electrophoresis on 2% agarose gels containing ethidium bromide (SigmaeAldrich). The intensity was quantified using image analysis software (Photoshop, ver.7.0; Adobe Systems, San Jose, CA, USA), and the results were standardized against the intensity of rat b-actin, a housekeeping gene. 2.9. Western blot analysis For the rats sacrificed at 24 h after retinal reperfusion, the total protein was extracted from the retina by lysing the sample in RIPA buffer and protease inhibitors (Complete Mini; Roche Diagnostics Corp., Indianapolis, IN, USA)]. The extract and Laemmli buffer were mixed at a 1:1 ratio, and the mixture was boiled for 5 min. A 100 mg sample was separated on 10% SDS-polyacrylamide gels and then transferred to polyvinylidene difluoride membranes (Immobilon-P; Millipore, Billerica, MA, USA). The membranes were incubated with anti-iNOS, anti-ICAM-1, anti-TNF-a, anti-IL-1b, anti-MCP-1, antip53, anti-bax, anti-Bcl-2, anti-phosphorylated p38, antiphosphorylated ERK, anti-phosphorylated JNK, or anti-b-actin antibodies. Then, the membranes were incubated with horseradish peroxidase-conjugated secondary antibody and visualized by chemiluminescence (GE Healthcare). The density of blots was

Table 1 Primer sequences for TNF-a, IL-1b, MCP-1, iNOS, ICAM-1 and b-actin for semiquantitative PCR. Genes

Primer sequences

Product size (bp)

TNF-a

50 -ATGATCCGAGATGTGGAACTGGCA-30 50 -GCTCCTCTGCTTGGTGGTTTGCTA-30 50 -TGTGATGAAAGACGGCACAC-30 50 -CTTCTTCTTTGGGTATTGTTTGG-50 50 -CTGGGCCTGTTGTTCACAGTTGC-30 50 -CTACAGAAGTGCTTGAGGTGGTTG-3 50 -TATCTGCAGACACATACTTTACGC-30 50 TCCTGGAACCACTCGTACTTG-30 50 -CCTGTTTCCTGCCTCTGAAG-30 50 -CCTGGGGAAGTACTGTTCA-30 50 -CTGGAGAAGAGCTATGAGCTG-30 50 -AATCTCCTTCTGATCCTGTC-30

295

IL-1b MCP-1 iNOS ICAM-1

b-actin

255 436 344 830 246

The levels of 8-OHdG and MDA in the retina were measured _ using MDA and 8-OHdG ELISA Kits (Cayman, Ann Arbor, MI, USA; Cosmo Bio., Tokyo, Japan). The procedure was carried out as described in the protocol provided by the manufacturer. The absorbance was measured at 450 nm using a microplate reader (Bio-Rad, Richmond, CA, USA). 2.11. Immunohistochemistry Immunohistochemistry was carried out by simultaneously blocking and permeabilizing sections with 0.2% Triton in PBS containing 5% goat serum for 1 h at room temperature, incubating with 8OHdG, acrolien and NF-kB p65 primary antibodies diluted in blocking solution overnight at 4  C, and incubating with the appropriate fluorescent secondary antibodies (all diluted 1:1000) in blocking solution for 3 h at room temperature. Nuclei were counterstained with DAPI. 2.12. Nuclear protein extract and electrophoretic mobility shift assay of NF-kB (EMSA) Rats were sacrificed at 24 h, and the retinas were harvested and minced in 0.5 mL of ice-cold buffer A. The suspension was centrifuged at 5000 g at 4  C for 10 min. The sediment was suspended in 200 ml of buffer B. The suspension was incubated on ice for 30 min. The sample was centrifuged at 12,000 g at 4  C for 30 min. The supernatant containing the nuclear proteins was collected. The EMSA was performed with an NF-kB DNA-binding protein-detection system (Pierce Biotechnology, Rockford, IL, USA). A 10 mg nuclear protein aliquot was incubated in binding buffer with a biotinlabeled NF-kB consensus oligonucleotide probe (50 -AGTTGAGGGGACTTTCCCAGGC-30 ) for 30 min. The specificity of the DNA/protein binding was determined by adding a 100-fold molar excess of unlabeled NF-kB oligonucleotide for competitive binding 10 min before adding the biotin-labeled probe. 2.13. Statistical analysis The results are expressed as the means ± SD. Statistical analysis for multiple comparisons was determined using an analysis of variance (ANOVA) followed by Bonferroni analysis. A p value of 0.05 or less was considered significant. All data were analyzed using the SPSS 10.0 for Windows statistical package. 3. Results 3.1. Protective effects of COS on retinal function after I/R injury The relative b-wave ratio was approximately equal to 1 in the normal controls. For the I/R-injured rats, whether pretreated with PBS, low-dose COS, or high-dose COS, the ratios were all significantly decreased, at both 24 h and 7 days after injury (p < 0.05 in all paired comparisons; n ¼ 10). Notably, in rats pretreated with highdose COS, the relative b-wave ratio was higher than that in the PBStreated and the low-dose COS groups, both at 24 h (p ¼ 0.007 and p ¼ 0.018; n ¼ 10) and 7 days (p ¼ 0.003 and p ¼ 0.011; n ¼ 10) after the I/R injury (Fig. 1).

I.-M. Fang et al. / Experimental Eye Research 130 (2015) 38e50

41

Fig. 1. Protective effects of chitosan oligosaccharides (COS) on the b-wave ratio in electroretinograms (ERGs) at 24 h and on day 7 after ischemia/reperfusion (I/R) injury. (A). Individual typical ERG records of the four groups at 24 h after injury. (B) The relative b-wave ratio was significantly decreased in the PBS-treated, low-dose COS, and high-dose COS groups compared with the normal group, at both 24 h and 7 days after I/R injury. The data are expressed as the means ± SD. (*P < 0.05 compared with the PBS-treated group; #P < 0.05 compared with normal; & P < 0.05 by two-way ANOVA with post hoc Bonferroni test; n ¼ 10 for each group).

3.2. Effects of COS on retinal histology after I/R injury Retinal I/R injury resulted in a marked thinning of the whole retina, compared with the normal retina at day 7. There was clear preservation of the structure in the COS group, especially in the high-dose COS group, compared with the PBS-treated group (Fig. 2A). Seven days post-I/R injury, a significant decrease in the total retinal thickness, IPL, INL and ONL was evident in the eyes of PBS-treated groups (P < 0.05 versus control in all comparisons, n ¼ 5). Treatment with low-dose and high-dose COS maintained the total thickness, IPL, INL and ONL after retinal I/R injury. The total retinal thickness, IPL, INL and ONL were significantly higher in the low-dose and high-dose COS groups than in the PBS-treated groups (p < 0.05 in all comparisons, n ¼ 5) (Fig. 2B).

3.3. Effects of COS on the density of retinal ganglion cells after I/R injury Retinal I/R injury induced a diffuse loss of retinal neurons in the GCL. Low-dose or high-dose COS treatment led to the significantly higher survival of Neu N-positive neurons in the GCL when compared with PBS-treated groups at 24 h (p < 0.05; n ¼ 3). The effects of decreased neuron loss in the ganglion cell layer were more noticeable in the high-dose COS treatment group than in the low-dose group (p < 0.05; n ¼ 3) (Fig. 2C and D).

3.4. Effects of COS on retinal apoptosis after I/R injury Increased cell apoptosis in each retinal layer was noted in the PBS group compared with the control group by in situ TUNEL staining. In the IR-injured rats pretreated with COS, especially in

42

I.-M. Fang et al. / Experimental Eye Research 130 (2015) 38e50

the high dosage group, the number of TUNEL-positive cells in GCL, INL and ONL was markedly reduced compared with rats pretreated with PBS (p < 0.05, low-dose vs. PBS-treated; p < 0.001, high-dose vs. PBS-treated; n ¼ 3) (Fig. 3A, B). The level of p53 was significantly higher in the PBS-treated group compared with the control group (p < 0.001; n ¼ 3). The expression level of p53 was significantly lower in the COS group compared with the PBS-treated group (p < 0.01; n ¼ 3). These effects were more remarkable in the high-dose COS treatment group than in the low-dose COS treatment group (p < 0.05; n ¼ 3) (Fig. 3C). Measurement of Bcl-2 and Bax by Western blotting showed that IR-injured rats pretreated with PBS showed decreased Bcl-2 and increased Bax expression, which resulted in a decreased Bcl-2/Bax ratio (p < 0.001, PBS-treated vs. normal; n ¼ 3). However, pretreatment with low-dose or high-dose COS clearly helped increase the Bcl-2/Bax ratio compared with the PBS-treated group (p < 0.05, in all paired comparisons; n ¼ 3) (Fig. 3D, E). 3.5. Effects of COS on retinal oxidative damage after I/R injury Quantitative assessment of oxidative lipid and DNA damage were tested using and malondialdehyde (MDA) and 8hydroxydeoxyguanosine (8-OHdG) ELISA kits, respectively. The levels of 8-OHdG and MDA were significantly higher in the PBStreated group compared with the control group (p < 0.05, in all paired comparisons; n ¼ 3). Pretreatment with COS resulted in reduced MDA and 8OHdG levels compared with the PBS-treated group, in a dose-dependent manner (p < 0.05, low-dose vs. PBStreated group; p < 0.01, high-dose vs. PBS-treated group; n ¼ 3) (Fig. 4A and B). Acrolein and 8-OHdG are biomarkers of oxidative damage to lipids and DNA. Retinal I/R injury induced increase in immunhistochemical staining for 8-OHdG in GCL, INL and ONL. Treatment with high-dose COS effectively reduced the immunohisto-chemical staining for 8-OHdG in retinas. Similarity, retinal I/R injury caused a striking increase in immunohistochemical staining for acrolein throughout the retina. Treatment with high-dose COS markely reduced the immunohistochemical staining for acrolein in retinas (Fig. 4C and D). 3.6. Effect of COS on the mRNA expression of inflammatory mediators after I/R injury At 24 h, the mRNA expression levels of TNF-a, IL-1b, MCP-1, iNOS, and ICAM-1 were significantly higher in the IR-injured rats pretreated with PBS compared with normal rats (p < 0.05 in all paired comparisons; n ¼ 8). In the IR-injured rats pretreated with high-dose COS, the expression of TNF-a, IL-1b, MCP-1, iNOS, and ICAM-1 was significantly lower than those in the PBS-treated group (p < 0.05, high-dose vs. PBS-treated group; n ¼ 8). In addition, the levels of TNF-a, MCP-1 and ICAM-1 mRNA were more reduced in the high-dose COS group than in the low-dose COS group (p ¼ 0.05 in all paired comparisons; n ¼ 8). At day 7 after I/R injury, the mRNA expression of IL-1b, MCP-1, iNOS, and ICAM-1 was significantly lower in the PBS-treated group, compared with low- or high-dose COS group (p < 0.05, Fig. 2. Effects of COS on retinal histology and the density of retinal ganglion cells at 7 days after I/R injury. (A) Representative H&E-stained retinal sections from control or PBS-treated or high- or low-dose COS groups. (B) Seven days post-I/R injury, a significant decrease in the total retinal thickness (from the inner limiting membrane to the pigment epithelium), the inner plexiform layer (IPL), the inner nuclear layer (INL) and the outer nuclear layers (ONL) was found in the eyes of PBS-treated group. The total retinal thickness, IPL, INL and ONL were significantly higher in the low-dose and high-dose COS groups than in the PBS-treated group. (C) Representative confocal

images of Neu N-labeled neurons (green) in the superior, inferior, temporal, and nasal parts of the flat-mounted retinas. (D) The percentage of surviving neurons in the GCL was significantly higher in the COS group than in the PBS-treated group. The data are expressed as the means ± SD. (*p < 0.05 compared with the PBS-treated group; #p < 0.05 compared with normal; & p < 0.05 by one-way ANOVA with post hoc Bonferroni test; n ¼ 5 for each group). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

I.-M. Fang et al. / Experimental Eye Research 130 (2015) 38e50

43

Fig. 3. Effects of COS on cell apoptosis and apoptosis-related proteins at 24 h after I/R injury. (A) TUNEL-positive nuclei in each layer of retinal tissue were noted in the PBS group, whereas only sparse TUNEL-positive cells in GCL of retinas were noted in the COS group. (B) In the COS-pretreated groups, especially in the high-dose group, the number of TUNELpositive cells in GCL, INL and ONL was markedly reduced compared with the PBS group. The levels of (C) p53, (D) Bcl-2 and Bax in the retina were evaluated with western blot analysis. (E) Effects of COS on the Bcl-2/Bax ratio. Treatment with low- or high-dose COS caused a significant increase in the Bcl-2/Bax ratio at 24 h. The data are expressed as the means ± SD. (*p < 0.05 compared with the PBS-treated group; #p < 0.05 compared with normal; & p < 0.05 by one-way ANOVA with post hoc Bonferroni test; n ¼ 3 for each group).

low-dose vs. PBS-treated group; p < 0.01, high-dose vs. PBS-treated group; n ¼ 8) (Fig. 5).

3.7. Effect of COS on the protein expression of inflammatory mediators after I/R injury At 24 h after I/R injury, the protein expression levels of TNF-a, IL1b, MCP-1, iNOS and ICAM-1 were significantly higher in the PBS-

treated group compared with the control group (p < 0.05 in all paired comparisons; n ¼ 8). The expression of TNF-a, IL-1b, MCP-1, iNOS and ICAM-1 proteins was significantly lower in the high-dose COS groups than it was in the PBS-treated group (p < 0.05, low-dose vs. PBS-treated group; p < 0.01, high-dose vs. PBS-treated group; n ¼ 8). At day 7 after I/R injury, the protein expression of IL-1b, iNOS, and ICAM-1 was significantly lower in the PBS-treated group,

44

I.-M. Fang et al. / Experimental Eye Research 130 (2015) 38e50

Fig. 4. Effects of COS on oxidative damage to retinas 24 h after I/R injury. The levels of (A) 8-hydroxy-2'-deoxyguanosine (8-OHdG) and (B) malondialdehyde (MDA) in the retina, which represented the oxidative damage to the DNA and lipids, were examined using ELISA. The low- or high-dose COS groups showed significantly reduced levels of 8-OHdG and MDA compared with the PBS-treated group. Immunohistochemical stainings for (C) 8-OHdG and (D) acrolein, biomarkers of oxidative damage to DNA and lipids, were evaluated at 24 h post-I/R injury. Bars show the means ± SD. (*p < 0.05 compared with the PBS-treated group; #p < 0.05 compared with normal; & p < 0.05 by one-way ANOVA with post hoc Bonferroni test; n ¼ 5 for each group).

compared with high-dose COS group (p < 0.05, high-dose vs. PBStreated group; n ¼ 8) (Fig. 6). 3.8. The inhibitory effect of COS on NF-kB activation after I/R injury Increased staining of the NF-kB p65 subunit in each layer of the retina was noted in the PBS-treated group. Treatment with low-

and high-dose COS prominently reduced the expression of p65 in the retina (Fig. 7A). The levels of IkВ were significantly reduced in the PBS-treated group. Treatment with COS significantly increased the expression of IkВ, especially in the high-dose COS group (Fig. 7B). The increased activity of NF-kB/DNA binding in retina after I/R injury was inhibited markedly by treatment with COS. This inhibitory

I.-M. Fang et al. / Experimental Eye Research 130 (2015) 38e50

45

Fig. 5. Effects of COS on the mRNA expression of inflammatory mediators at 24 h and 7 days. The mRNA expression levels of (A) TNF-a, (B) IL-1b, (C) MCP-1, (D) iNOS, and (E) ICAM1 were evaluated by using PCR. At 24 h, the mRNA expression levels of TNF-a, IL-1b, MCP-1, iNOS, and ICAM-1 were significantly higher in the IR-injured rats pretreated with PBS compared with normal rats. In the IR-injured rats pretreated with high-dose COS, the expression of TNF-a, IL-1b, MCP-1, iNOS, and ICAM-1 was significantly lower than those in the PBS-treated group. At day 7 after I/R injury, the mRNA expression of IL-1b, MCP-1, iNOS, and ICAM-1 was significantly lower in the PBS-treated group, compared with low- or highdose COS group. The data are expressed as the means ± SD. (*p < 0.05 compared with the PBS-treated group; #p < 0.05 compared with normal; & p < 0.05 by one-way ANOVA with post hoc Bonferroni test; n ¼ 8 for each group).

effect was especially prominent in the high-dose COS group. Adding a 100-fold molar excess of unlabeled NF-kB probe completely inhibited the binding of the labeled probe to the NF-kB/DNA complex (Fig. 7C).

3.9. Effects of COS on mitogen-activated protein kinases phosphorylation after I/R injury The expression of phospho-JNK and phospho-ERK was significantly higher in the PBS-treated group compared with normal rats at 24 h after I/R injury (p < 0.05 in all paired comparisons; n ¼ 3). Pretreatment with low- or high-dose COS significantly inhibited JNK and ERK phosphorylation compared with pretreatment with PBS (p < 0.05, low-dose vs. PBS-treated group; p < 0.01, high-dose vs. PBS-treated group; n ¼ 3) (Fig. 7D, E). In contrast, the phosphop38 levels were significantly decreased in the PBS-treated group at 24 h after I/R injury compared with the normal group (p < 0.01; n ¼ 3). Pretreatment with COS significantly up-regulated p38 phosphorylation in a dose-dependent manner compared with the

PBS-treated group p < 0.05, low-dose vs. PBS-treated group; (p < 0.001, high-dose vs. PBS-treated group; n ¼ 3) (Fig. 7F).

3.10. p38 inhibitor abolished the protective effect of COS on retinal oxidative stress and the expression of inflammatory mediators after I/R injury To further investigate the role of p38 MAP kinase in the retinal protective effect of COS, we intravitreously injected p38 inhibitor SB203580 or PBS 30 min before intraperitoneal injection of highdose of COS and induction of retinal I/R injury. Immunohistochemistry showed treatment with high-dose COS markedly reduced retinal staining for 8OHdG, however, pretreatment with p38 inhibitor SB203580 partially abolished the inhibitory ability of COS on retinal oxidative damage as indicated by the presence of 8OHdG stains in the retinas of rats with retinal I/R injury (Fig. 8A). Western blot analysis showed the expression of TNF-a, MCP-1, iNOS, and ICAM-1 protein was significantly decreased in the I/R injured rats treated with high-dose COS. Similarity, pretreatment

46

I.-M. Fang et al. / Experimental Eye Research 130 (2015) 38e50

Fig. 6. Evaluation of the protein expression of inflammatory mediators at 24 h and 7 days after I/R injury. The protein levels of (A) TNF-a, (B) IL-1b, (C) MCP-1, (D) iNOS and (E) ICAM-1 were evaluated by using western blot analysis. At 24 h after I/R injury, the protein expression levels of TNF-a, IL-1b, MCP-1, iNOS and ICAM-1 were significantly higher in the PBS-treated group compared with the control group. The expression of TNF-a, IL-1b, MCP-1, iNOS and ICAM-1 proteins was significantly lower in the high-dose COS groups than it was in the PBS-treated group. At day 7 after I/R injury, the protein expression of IL-1b, iNOS, and ICAM-1 was significantly lower in the PBS-treated group, compared with with highdose COS group. The data are expressed as the means ± SD. (*p < 0.05 compared with the PBS-treated group; #p < 0.05 compared with normal; & p < 0.05 by one-way ANOVA with post hoc Bonferroni test; n ¼ 8 for each group).

with p38 inhibitor SB203580 significantly increased the expression of TNF-a, MCP-1, iNOS, and ICAM-1 protein in the I/R injured rats treated with high-dose COS, than without pretreatment with SB203580 (p < 0.05 in all paired comparisons; n ¼ 3) (Fig. 8BeE). 4. Discussion In this study, we demonstrated that pretreatment with COS effectively protected retinal cells from I/R injury in a dosedependent manner, as indicated by the reduction of b-wave decrement in the ERGs and the preservation of retinal morphology and neurons in the ganglion cell layer. In addition, pretreatment with COS attenuated retinal oxidative damage, inhibited the expression of inflammatory mediators (IL-1 b, TNF-a, MCP-1, iNOS and ICAM-1) and pro-apoptotic proteins, p53 and Bax, and increased the expression of anti-apoptotic protein Bcl-2. Importantly, the administration of COS inhibited the activation of NF-kB and down-regulated the levels of phospho-JNK and phospho-ERK but up-regulated the level of phospho-p38, which may have

contributed to the anti-inflammatory, anti-apoptotic, and antioxidative effects observed after retinal I/R injury. It has been reported that multiple inflammatory mediators, such as TNF-a, IL-1b, MCP-1, iNOS, and ICAM-1, are up-regulated and implicated in the pathogenesis of retinal I/R injury (Osborne et al., 2004). TNF-a was proposed as a pivotal cytokine mediating neuron death in retinal I/R injury (Hangai et al., 1996; Barone et al., 1997; Tezel and Wax, 2000; Fontaine et al., 2002). IL-1b mediated the degeneration of inner retinal elements after I/R injury (Yoneda et al., 2001). MCP-1 is an important chemokine in attracting monocytes/macrophages to target tissues (Jo et al., 2003). Nitric oxide (NO) is generated by iNOS-induced delayed neuronal cell death after ischemia (Samdani et al., 1997). It had been reported that the activation of NF-kB is associated with an increased expression of these inflammatory mediators after I/R injury (Dvoriantchikova et al., 2009a, 2009b; Gustavsson et al., 2008; Biermann et al., 2010). In this study, we clearly demonstrated that COS could inhibit NF-kB activation and decrease the expression of IL-1b, TNF-a, MCP-1, iNOS and ICAM-1. Our findings are consistent

I.-M. Fang et al. / Experimental Eye Research 130 (2015) 38e50

47

Fig. 7. Effects of COS on the activation of NF-kB and MAPK at 24 h after I/R injury. (A) Immunohistochemical study of the expression of the NF-kB p65 subunit in retinas. (B) Evaluation of IkB in each group using Western blot analysis. The Y scale represents the ratio of IkB blot density to the b-actin blot density. (C) The NF-kB/DNA binding activity in the retina was measured using EMSA. The protein densities of (D) phospho-JNK, (E) phospho-ERK and (F) phospho-p38 were measured using Western blot analysis, and the ratio to total levels was calculated. Data are expressed as the means ± SD. (*P < 0.05 compared with the PBS-treated group; #P < 0.05 compared with normal; & P < 0.05 by one-way ANOVA with post hoc Bonferroni test; n ¼ 3 for each group).

48

I.-M. Fang et al. / Experimental Eye Research 130 (2015) 38e50

Fig. 8. p38 inhibitor abolished the protective effect of COS on retinal oxidative stress and the expression of inflammatory mediators in I/R-injured rats. p38 inhibitor SB203580 or PBS were injected intravitreously 30 min prior to high-dose of COS treatment in I/R-injured rats. (A) Immunohistochemistry showed pretreatment with SB203580 partially abolished the inhibitory ability of COS on retinal oxidative damage as indicated by the presence of 8OHdG stains in the retinas of rats with retinal I/R injury. Pretreatment with SB203580 significantly increased the expression of (B) TNF-a, (C) MCP-1, (D) iNOS, and (E) ICAM-1 protein than without pretreatment with SB203580 in the I/R injured rats treated with high-dose COS. Data are expressed as the means ± SD. (*P < 0.05 compared with the PBS-treated group; #P < 0.05 compared with normal; & P < 0.05 by one-way ANOVA with post hoc Bonferroni test; n ¼ 3 for each group).

I.-M. Fang et al. / Experimental Eye Research 130 (2015) 38e50

with those of previous studies, which demonstrated that COS exhibits anti-inflammatory activities through the inhibition of NF-kB activation and, thus, the suppression of NF-kB-dependent genes (Khodagholi et al., 2010; Wei et al., 2012). NF-kB is a redox-sensitive transcription factor that plays a critical role in neuronal cell death in retinal I/R injury. In this study, we demonstrated that pretreatment with COS inhibited I/R injuryinduced NF-kB activation by blocking IkB degradation and p65 subunit translocation. ROS is an inducer of NF-kB (Schreck et al., 1992; Aggarwal, 2000). COS has been reported to have antioxidative effects with ROS scavenging ability. Je et al. (2004) have shown that COS is able to quench various radicals by the action of nitrogen on the C-2 position of COS. Xie et al. (2001) reported that the scavenging mechanisms of chitosans are related to their hydrogen donating ability to free radicals to form stable molecules. In this study, we showed that pretreatment with COS diminished oxidative damage in retinas after I/R injury. This finding indicated that the COS inhibition of NF-kB activation in our study may be partly mediated by reducing the amount of ROS in the retina. p53 is a gene that regulates cell cycle progression and modulates apoptosis and/or DNA repair in cellular responses to stress (Ryan et al., 2001; Bates and Vousden, 1999). p53 may mediate apoptosis through intrinsic pathways that are associated with the Bcl-2 family of proteins. The Bcl-2 family of proteins consists of anti-apoptotic (Bcl-2 and Bcl-xL) and pro-apoptotic members (Bax and Bak). The balance between the antiapoptotic and proapoptotic members of the Bcl-2 family may determine the susceptibililty of a cell to apoptosis (Ju et al., 2008). Increasing evidence has indicated that activated NF-kB could promote cell apoptosis by up-regulating p53 expression and interrupting the balance between Bax and Bcl-2 (Qin et al., 1999; Liang et al., 2007). In this study, we demonstrated that COS effectively inhibited NF-kB activation and suppressed p53 expression. In addition, COS significantly up-regulated the expression of anti-apoptotic protein Bcl-2 and suppressed the expression of the pro-apoptotic protein Bax. These findings, along with the decrease in TUNEL-positive cells in the GCL and INL of the retina in COS groups 24 h after I/R injury, suggest that COS might decrease apoptosis in the retina by suppressing p53 and reducing the Bax-Bcl-2 ratio via the inhibition of NF-kB. In addition, in the present study, the rats were subjected to retinal ischemia by increasing their intraocular pressure to 130 mmHg and were found that the nuclei in GCL, INL, and ONL were TUNEL-positive in retinas. This finding was different from several previous reports, who demonstrated that TUNEL-positive cells were located in GCL and INL in the retinal I/R-injured model that was induced by increasing the intraocular pressure to 110 mmHg (Qi et al., 2013; Zhang et al., 2013). It is possible that different induction pressure may account for this discrepancy. Our finding is consistent with previous report by Sun et al., who showed obvious TUNEL-positive nuclei in the INL, RGC layers, and ONL in retinas after 60 min of increasing intraocular pressure to 130 mmHg (Sun et al., 2010). The MAPK family contains many important signaling molecules that are involved in the production of pro-inflammatory cytokines and the modulation of NF-kB (Tang et al., 2001). The three main members of the MAPK family are p38, ERK-1/2, and JNK. ERK is generally involved in cell proliferation, and JNK and p38 provide mediated responses to cellular stress (Chen et al., 2001; Tian et al., 2000). In our study, we demonstrated that phospho-JNK and phospho-ERK are up-regulated, whereas phospho-p38 is downregulated after retinal I/R injury. In addition, treatment with COS significantly reduced the phosphorylation of JNK and ERK but increased p38 phosphorylation after retinal I/R injury. In addition, treatment with p38 inhibitor SB203580 prior to COS administration abolished the protective effect of COS on retinal oxidative stress and the expression of inflammatory mediators after I/R injury. p38

49

MAPKs exert different cellular functions depending on the stimulus and timing of the activation. Several experimental studies have shown significantly increased the phospho-p38, phospho-JNK, and phospho-ERK levels after retinal ischemia, and the blockade of these MAPKs provides significant protection from ischemic damages (Ishizuka et al., 2013; Roth et al., 2003). However, Dreixler et al. demonstrated that retinal ischemic preconditioning resulted in the activation of p38 MAPK, which contributed to the neuroprotective effects in retinal I/R injury (Dreixler et al., 2009). Jiang et al. (2012) found that the activation of p38 MAPK could activate the antiapoptosis genes Bcl-2 and Bcl-XL and promoted the protection of retinal cells from ischemic injury. In our study, COS was given before the induction of I/R injury, similar to the ischemic preconditioning described by Dreixler et al. We found that the expression of anti-apoptotic proteins Bcl-2 and phosphorylated p38 were increased simultaneously, indicating that the COSinduced activation of p38 might exert neuroprotective effects through the activation of Bcl-2. Further studies will be necessary to further elucidate the roles of p38 in COS-mediated protection after retinal I/R injury. 5. Conclusions In this study, we have demonstrated that COS can prevent retinal I/R injury through its antioxidant and anti-inflammation activities. These effects are achieved by blocking the activation of NF-kB, JNK, and ERK but promoting the activation of p38. Our data indicate that COS has potential as a prophylactic agent for retinal I/ R-related diseases. References Aggarwal, B.B., 2000. Apoptosis and nuclear factor-kappa B: a tale of association and dissociation. Biochem. Pharmacol. 60, 1033e1039. Barone, F.C., Arvin, B., White, R.F., Miller, A., Webb, C.L., Willette, R.N., Lysko, P.G., Feuerstein, G.Z., 1997. Tumor necrosis factor-alpha. A mediator of focal ischemic brain injury. Stroke 28, 1233e1244. Bates, S., Vousden, K.H., 1999. Mechanisms of p53-mediated apoptosis. Cell. Mol. Life Sci. 55, 28e37. Biermann, J., Lagreze, W.A., Dimitriu, C., Stoykow, C., Goebel, U., 2010. Preconditioning with inhalative carbon monoxide protects rat retinal ganglion cells from ischemia/reperfusion injury. Investig. Ophthalmol. Vis. Sci. 51, 3784e3791. Chen, J., Fujii, K., Zhang, L., Roberts, T., Fu, H., 2001. Raf-1 promotes cell survival by antagonizing apoptosis signal-regulating kinase 1 through a MEK-ERK independent mechanism. Proc. Natl. Acad. Sci. U. S. A. 98, 7783e7788. Chen, Q., Liu, S.Q., Du, Y.M., Peng, H., Sun, L.P., 2006. Carboxymethyl -chitosan protects rabbit chondrocytes from interleukin-1beta-induced apoptosis. Eur. J. Pharmacol. 541, 1e8. Dreixler, J.C., Barone, F.C., Shaikh, A.R., Du, E., Roth, S., 2009. Mitogen -activated protein kinase p38a and retinal ischemic preconditioning. Exp. Eye Res. 89, 782e790. Dvoriantchikova, G., Barakat, D., Brambilla, R., Agudelo, C., Hernandez, E., Bethea, J.R., Shestopalov, V.I., Ivanov, D., 2009a. Inactivation of astroglial NFkappa B promotes survival of retinal neurons following ischemic injury. Eur. J. Neurosci. 30, 175e185. Dvoriantchikova, G., Agudelo, C., Hernandez, E., Shestopalov, V.I., Ivanov, D., 2009b. Phosphatidylserine-containing liposomes promote maximal survival of retinal neurons after ischemic injury. J. Cereb. Blood Flow Metab. 29, 1755e1759. Fontaine, V., Mohand-Said, S., Hanoteau, N., Fuchs, C., Pfizenmaier, K., Eisel, U., 2002. Neurodegenerative and neuroprotective effects of tumor Necrosis factor (TNF) in retinal ischemia: opposite roles of TNF receptor 1 and TNF receptor 2. J. Neurosci. 22, RC216. Gustavsson, C., Agardh, C.D., Hagert, P., Agardh, E., 2008. Inflammatory markers in nondiabetic and diabetic rat retinas exposed to ischemia followed by reperfusion. Retina 28, 645e652. Hangai, M., Yoshimura, N., Honda, Y., 1996. Increased cytokine gene expression in rat retina following transient ischemia. Ophthalmic Res. 28, 248e254. Ishizuka, F., Shimazawa, M., Umigai, N., Ogishima, H., Nakamura, S., Tsuruma, K., Hara, H., 2013. Crocetin, a carotenoid derivative, inhibits retinal ischemic damage in mice. Eur. J. Pharmacol. 703, 1e10. Je, J.Y., Park, P.J., Kim, S.K., 2004. Free radical scavenging properties of hetero- chitooligosaccharides using an ESR spectroscopy. Food Chem. Toxicol. 42, 381e387. Jiang, S.Y., Zou, Y.Y., Wang, J.T., 2012. p38 mitogen-activated protein kinase-induced nuclear factor kappa-light-chain-enhancer of activated B cell activity is required

50

I.-M. Fang et al. / Experimental Eye Research 130 (2015) 38e50

for neuroprotection in retinal ischemia/reperfusion injury. Mol. Vis. 18, 2096e2106. Jo, N., Wu, G.S., Rao, N.A., 2003. Up-regulation of chemokine expression in the retinal vasculature in ischemia-reperfusion injury. Investig. Ophthalmol. Vis. Sci. 44, 4054e4060. Joodi, G., Ansari, N., Khodagholi, F., 2011. Chitooligosaccharide-mediated neuroprotection is associated with modulation of Hsps expression and reduction of MAPK phosphorylation. Int. J. Biol. Macromol. 48, 726e735. Ju, W.K., Lindsey, J.D., Angert, M., Patel, A., Weinreb, R.N., 2008. Glutamate receptor activation triggers OPA1 release and induces apoptotic cell death in ischemic rat retina. Mol. Vis. 14, 2629e2638. Khodagholi, F., Eftekharzadeh, B., Maghsoudi, N., Rezaei, P.F., 2010. Chitosan prevents oxidative stress-induced amyloid beta formation and cytotoxicity in NT2 neurons: involvement of transcription factors Nrf2 and NF-kappa B. Mol. Cell. Biochem. 337, 39e51. Lam, T.T., Abler, A.S., Tao, M.O.M., 1999. Apoptosis and caspase after ischemiareperfusion injury in rat retina. Investig. Ophthalmol. Vis. Sci. 40, 967e975. Laskowski, I., Pratschke, J., Wilhelm, M.J., Gasser, M., Tilney, N.L., 2000. Molecular and cellular events associated with ischemia/reperfusion injury. Ann. Transpl. 5, 29e35. Levine, L.A., 2001. Models of neural injury. J. Glaucoma 10, S19eS21. Liang, Z.Q., Li, Y.L., Zhao, X.L., Han, R., Wang, X.X., Wang, Y., Chase, T.N., Bennett, M.C., Qin, Z.H., 2007. NF-kappaB contributes to 6-hydroxydopamine -induced apoptosis of nigral dopaminergic neurons through p53. Brain Res. 1145, 190e203. Osborne, N.N., Casson, R.J., Wood, J.P., Chidlow, G., Graham, M., Melena, J., 2004. Retinal ischemia: mechanisms of damage and potential therapeutic strategies. Prog. Retin. Eye Res. 23, 91e147. Pae, H.O., Seo, W.G., Kim, N.Y., Oh, G.S., Kim, G.E., Kim, Y.H., Kwak, H.J., Yun, Y.G., Jun, C.D., Chung, H.T., 2001. Induction of granulocytic differentiation in acute promyelocytic leukemia cells (HL-60) by water-soluble chitosan oligomer. Leuk. Res. 25, 339e346. Pangestuti, R., Kim, S.K., 2010. Neuroprotective properties of chitosan and its derivatives. Mar. Drugs 8, 2117e2128. Qi, Y., Chen, L., Zhang, L., Liu, W.B., Chen, X.Y., Yang, X.G., 2013. Crocin prevents retinal ischaemia/reperfusion injury-induced apoptosis in retinal ganglion cells through the PI3K/AKT signalling pathway. Exp. Eye Res. 107, 44e45. Qin, Z.H., Chen, R.W., Wang, Y., Nakai, M., Chuang, D.M., Chase, T.N., 1999. Nuclear factor kappaB nuclear translocation upregulates c-Myc and p53 expression during NMDA receptor-mediated apoptosis in rat striatum. J. Neurosci. 19, 4023e4033. Qin, C., Du, Y., Xiao, L., Li, Z., Gao, X., 2002. Enzymic preparation of water-soluble chitosan and their antitumor activity. Int. J. Biol. Macromol. 31, 1111e1117. Roth, S., Shaikh, A.R., Hennelly, M.M., Li, Q., Bindokas, V., Graham, C.E., 2003. Mitogen-activated protein kinases and retinal ischemia. Investig. Ophthalmol. Vis. Sci. 44, 5383e5395. Ryan, K.M., Phillips, A.C., Vousden, K.H., 2001. Regulation and function of the p53 tumor suppressor protein. Curr. Opin. Cell. Biol. 13, 33e337.

Samdani, A.F., Dawson, T.M., Dawson, V.L., 1997. Nitric oxide synthase in models of focal ischemia. Sroke 28, 1283e1288. Schreck, R., Albermann, K., Baeuerle, P.A., 1992. Nuclear factor kappa B: an oxidative stress-responsive transcription factor of eukaryotic cells. Free Radic. Res. Commun. 17, 221e237. Stefansson, E., Machemer, R., de Juan Jr., E., McCuen 2nd, B.W., Peterson, J., 1992. Retinal oxygenation and laser treatment in patients with diabetic retinopathy. Am. J. Ophthalmol. 113, 36e38. Sun, M.H., Pang, J.H., Chen, S.L., Han, W.H., Ho, T.C., Chen, K.J., Kao, L.Y., Lin, K.K., Tsao, Y.P., 2010. Retinal protection from acute glaucoma-induced ischemiareperfusion injury through pharmacologic induction of heme oxygenase-1. Investig. Ophthalmol. Vis. Sci. 51, 4798e4808. , C., Braquet, P., 1991. Ischemia and Szabo, M.E., Droy-Lefaix, M.T., Doly, M., Carre reperfusion-induced histologic changes in the rat retina. Demonstration of a free radical-mediated mechanism. Investig. Ophthalmol. Vis. Sci. 32, 1471e1478. Tang, G., Minemoto, Y., Dibling, B., Purcell, N.H., Li, Z., Karin, M., Lin, A., 2001. Inhibition of JNK activation through NF-kappaB target genes. Nature 414, 313e317. Tezel, G., Wax, M.B., 2000. Increased production of tumor necrosis factor-alpha by glial cells exposed to simulated ischemia or elevated hydrostatic pressure induces apoptosis in cocultured retinal ganglion cells. J. Neurosci. 20, 8693e8700. http://www.iovs.org/external-ref?access_num¼10.1038/35104568&link_ type¼DOI. Tian, D., Litvak, V., Lev, S., 2000. Cerebral ischemia and seizures induce tyrosine phosphorylation of PYK2 in neurons and microglial cells. J. Neurosci. 20, 6478e6487. Tso, M.O., Jampol, L.M., 1982. Pathophysiology of hypertensive retinopathy. Ophthalmology 89, 1132e1145. Wei, P., Ma, P., Xu, Q.S., Bai, Q.H., Gu, J.G., Xi, H., Du, Y.G., Yu, C., 2012. Chitosan oligosaccharides suppress production of nitric oxide in lipopoly -saccharideinduced N9 murine microglial cells in vitro. Glycoconj. J. 29, 285e295. Wu, W.C., Lai, C.C., Chen, S.L., Sun, M.H., Xiao, X., Chen, T.L., Tsai, R.J., Kuo, S.W., Tsao, Y.P., 2004. GDNF gene therapy attenuates retinal ischemic injuries in rats. Mol. Vis. 10, 93e102. Xie, W., Xu, P., Liu, Q., 2001. Antioxidant activity of water-soluble chitosan derivatives. Bioorg. Med. Chem. Lett. 11, 1699e1701. Yokota, H., Narayanan, S.P., Zhang, W., Liu, H., Rojas, M., Xu, Z., Lemtalsi, T., Nagaoka, T., Yoshida, A., Brooks, S.E., Caldwell, R.W., Caldwell, R.B., 2011. Neuroprotection from retinal ischemia/reperfusion injury by NOX2 NADPH oxidase deletion. Investig. Ophthalmol. Vis. Sci. 52, 8123e8131. Yoneda, S., Tanihara, H., Kido, N., Honda, Y., Goto, W., Hara, H., Miyawaki, N., 2001. Interleukin-1b mediates ischemic injury in the rat retina. Exp. Eye Res. 73, 661e667. Zheng, L., Gong, B., Hatala, D.A., Kern, T.S., 2007. Retinal ischemia and reperfusion causes capillary degeneration: similarities to diabetes. Investig. Ophthalmol. Vis. Sci. 48, 361e367. Zhang, X.Y., Xiao, Y.Q., Zhang, Y., Ye, W., 2013. Protective effect of pioglitazone on retinal ischemia/reperfusion injury in rats. Investig. Ophthalmol. Vis. Sci. 54, 3912e3921.