Topical Administration of Peroxiredoxin-6 on the Cornea Suppresses Inflammation and Neovascularization Induced by Ultraviolet Radiation

Abstract

Purpose.: To investigate the effect of topical administration of peroxiredoxin-6 (PRDX6) on ultraviolet-induced corneal injury.

Methods.: Corneal transparency and neovascularization were observed with a slit-lamp microscope and hematoxylin and eosin staining. The oxidative damage was determined with a commercial malondialdehyde (MDA) kit. The expressions of PRDX6, polymorphonuclear neutrophil (PMN), vascular endothelial growth factor (VEGF), and pigment epithelium-derived factor (PEDF) were determined by immunohistochemistry and Western blot. The expressions of genes related with antioxidant defense systems and cell apoptosis were detected by RT-PCR.

Results.: The irradiated corneas appeared opaque and had high levels of MDA. Peripheral neovascularization and neutrophils appeared in the control and buffer-treated groups (with no treatment or PRDX6 diluent, respectively), whereas they were significantly suppressed in the PRDX6-treated group. The MDA content of the corneas in the PRDX6-treated group was significantly lower than that of the control and buffer-treated groups (P < 0.05). In the PRDX6-treated group the immunoreactivity of VEGF was lower, and that of PEDF was higher, than that in the control and buffer-treated groups. In addition, there were expression correlations between PRDX6 and PMN, VEGF, PEDF. The expressions of genes related with antioxidant defense systems and cell apoptosis were significant different between buffer- and PRDX6-treated groups (P < 0.05).

Conclusions.: The topically administered PRDX6 maintained the homeostasis of corneal cells, reduced inflammation, and suppressed neovascularization and apoptosis under ultraviolet irradiation.

Introduction

Low levels of reactive oxygen species (ROS) are necessary for some normal biological functions, such as intracellular pathways of signal transduction of the cell. However, high levels of ROS can drive oxidative stress and induce many diseases, such as keratitis and cataracts of the eye. ^1–3 To maintain the level of ROS, a balance between oxidant and antioxidant systems is necessary. Cells have considerable antioxidant defense systems, including superoxide dismutase (SOD), catalase (CAT), gluthathione peroxidase (GPx), and peroxiredoxins (PRDXs). It has been reported that PRDXs are more important than the others in controlling ROS levels in the lens, because CAT and GPx can be inactivated by oxidative attack and GPx cannot be transported from lens fiber cells to lens epithelial cells (LECs). ^4–6 The PRDX family contains six members that function in detoxifying ROS and providing cytoprotection from internal and external environmental stresses, as well as playing roles in cellular signaling by limiting ROS levels. ³ Of the six members, the expression of PRDX6 in the lens is higher than that of the others and the expression of PRDX6 in the cornea is higher than that of the lens, which suggests a potential role for PRDX6 in protection against oxidative stress in the cornea. ^3,7,8  

PRDX6-knockout mice show high susceptibility to oxidative stress, develop lens opacity following stress, and LECs from these mice harbor high levels of ROS. ^3,9,10 The results reported by Wang et al. ⁹ indicate that PRDX6 is a unique nonredundant antioxidant that functions independently of other peroxiredoxins and antioxidant proteins under conditions of excessive oxidative stress in vivo. ⁹ There is evidence that overexpression of PRDX6 reduces H₂O₂ generated in response to internal and external environmental stresses. ¹¹ The increased expression of PRDX6 is directly correlated with enhanced protection of cells against oxidative stress. ^12,13 By enhancing resistance to oxidative stress, PRDX6 may offer opportunities for intervention to slow progression of age-related cataractogenesis. ⁶ Many studies have demonstrated a critical function for PRDX6 in the generation and development of cataracts. However, there are no thorough investigations on the relationship of PRDX6 to corneal protection. 

Irradiation of eyes with ultraviolet-B (UV-B) caused serious enzymatic disturbances and inflammatory reaction in the cornea. ¹⁴ In the present study we investigate the effect of topical administration of PRDX6 on ultraviolet irradiation-induced corneal injury, and discuss the feasibility of keeping corneal homeostasis by improving the external environment by topically administering PRDX6. We found that topical application of PRDX6 improved corneal transparency, decreased edema, reduced inflammation, and suppressed neovascularization, accompanied by downregulation and upregulation of the expression of vascular endothelial growth factor (VEGF) and pigment epithelium-derived factor (PEDF), respectively. In addition, there were significantly different expressions of genes related to the antioxidative system and apoptosis. The results of the present study show that PRDX6 has the potential to be a useful adjunct therapy for keratitis induced by ultraviolet irradiation. 

Materials and Methods

Rat Corneal Ultraviolet Irradiation and Treatment

Wistar rats (6-weeks-old, male; Binzhou Medical University, Yantai, China), confirmed as being free of ocular disease, were used in the study. All treatments of animals were in accordance with the ARVO statement for the use of animals in ophthalmic and vision research. Rat corneal ultraviolet irradiation was conducted as previously reported, ¹⁵ with several modifications. In brief, the rats were anesthetized by intraperitoneal injection of 5% (w/v) chloral hydrate (0.7 mL/100 g body weight) and eyes (one eye per rat) were irradiated with 280-nm ultraviolet radiation 15 minutes after mydriasis by tropicamide (30 μL/eye; Yantai Yuhuangding Hospital, Yantai, China) on four occasions (once a day for 4 days), providing an accumulative illumination of 9 kJ/m². 

After corneal damage had been induced by the UV irradiation, the rats were randomly divided into three groups (40 rats/group). Rats were administered topically on the cornea four times a day with 15 μL PRDX6 once (a concentration of 2 mg/mL dissolved in 0.15 mol/L NaCl and 20 mmol/L PBS buffer, pH 7.2) in the PRDX-treated group, with 15 μL buffer (in which PRDX6 was dissolved) in the buffer-treated group and rats in the treatment-free control group were not given any treatment after irradiation. The treatments were administered for 14 consecutive days after induction of the injury. After different durations the animals were euthanized, and the corneas were dissected and stored at −80°C for malondialdehyde (MDA) analysis, RT-PCR, and Western blot detection or fixed for histologic processing. 

Preparation of Recombinant Protein PRDX6

The PRDX6 gene was directly amplified by PCR from our human epididymal cDNA library ¹⁶ with the specific primers (forward [F]: 5′-tatccatatgcccggaggtctgcttc-3′; reverse [R]: 5′-ttactcgagaggctggggtgtgtagcg-3′), each containing NdeI and XhoI sites. The PCR products (full-length PRDX6) were digested with NdeI and XhoI, and inserted into pET-32b (+) vectors. The plasmid of pET-32b (+)/PRDX6 was transformed in Escherichia coli BL21 (DE3) strain according to the supplier's instructions. The transformed E. coli were grown to mid-log phase and the fusion protein expression was induced with 1 mmol/L isopropyl-1-thio-d-galactoside (IPTG) for 3 hours at 32°C. The cells were sonicated and the fractions analyzed on 15% (w/v) polyacrylamide gels and stained with Coomassie blue G-250 by standard procedures. ¹⁷ The histidine tag in the vector permitted the recombinant fusion protein to be purified by commercial Ni²⁺-chelating resin (Sepharose Fast Flow resin; GE Healthcare, Beijing, China) according to the manufacturer's instructions. After the protein concentration was measured ¹⁸ and endotoxins removed (Endotoxin Removal Kit; Jin si rui Biotechnology Company, Nanjing, China), the purified PRDX6 proteins were stored at −80°C. 

Histologic Examination by H&E Staining and Measurement of Corneal Epithelial Thickness

Rat corneas were immersed in 4% paraformaldehyde for 24 hours and dehydrated and embedded in paraffin by routine methods. Sections (4 μm) were then deparaffinized in toluene and rehydrated in decreasing alcohol concentrations. Staining with a commercial hematoxylin and eosin staining kit (Beyotime Institute of Biotechnology, Beijing, China) was according to the manufacturer's protocol. The results were examined by brightfield microscopy (DM LB2; Leica Microsystems, Nussloch, Germany). 

The full thicknesses of the total cornea, corneal epithelium, and stroma were measured according to the methods of Gul et al. ¹⁹ In brief, the thicknesses were calculated from 10 different points in sections from each specimen. The thickness of the total cornea, corneal epithelium, and stroma were statistically analyzed by using a commercial image analysis program (Leica Q Win Image Analyses System; Leica Microsystems Imaging Solutions, Cambridge, UK). 

Observation by Slit-Lamp Microscope and Measurement of Area of Corneal Neovascularization

At different times after irradiation, the rats were anesthetized with chloral hydrate and mydriasis was induced with tropicamide; their eyes were examined with a slip-lamp microscope (BQ900; Haag-Streit AG, Koeniz, Switzerland). A commercial image acquisition software was used (EyeCap V3, version 3.3.44; Haag-Streit). 

The area of cornea neovascularizaton was measured according to the methods of D'Amato et al., ²⁰ with several modifications. The vessel length (L) was measured with a reticule from the limbus and the number of clock hours (C) of limbus involved was also measured. A formula was used to determine the area of a circular band segment: C/12 × 3.1416 × [r ² − (r − L)²], where r = 3 mm, the measured radius of the rat cornea. 

Standardized Grading Scale for Corneal Opacity

The corneal opacity was evaluated according to the Dickey edema classification standard ²¹ on corneal opacity by the same corneal specialist. Grade 0: graft totally transparent; grade I: slight haziness; grade II: increased haze, but anterior chamber structures still clearly visible; grade III: advanced opacity with difficult-to-view anterior chamber structures; and grade IV: opaque cornea, without view of the anterior chamber. 

Immunohistochemistry

Rat corneas were immersed in 4% (w/v) paraformalehyde for 24 hours, dehydrated, and embedded in paraffin by routine methods. Sections (4 μm) were then deparaffinized in toluene and rehydrated in decreasing alcohol concentrations. Antigens were unmasked by incubating sections in boiling 0.01 mol/L sodium citrate (pH 6.0) in a microwave oven. Sections were brought back to room temperature and endogenous peroxidase activity was quenched with 3% (v/v) hydrogen peroxide for 10 minutes. Sections were blocked with Tris-buffered saline (TBS: 0.5 mol/L Tris-HCl, 0.15 mol/L NaCl, pH 7.6) containing 3% (w/v) BSA for 1 hour, and subsequently incubated with the relevant antibodies (anti-PRDX6, sc-55017; Santa Cruz Biotechnology, Inc., Santa Cruz, CA), diluted 1:500; anti-VEGF (sc-152; Santa Cruz Biotechnology), diluted 1:50; anti-PEDF (sc-16956; Santa Cruz Biotechnology), diluted 1:500; anti-PMN (20R-PR102; Fitzgerald Industries International, Acton, MA), diluted 1:500 in blocking solution overnight at 4°C. Finally, sections were washed with TBS containing 0.1% (v/v) Tween-20 (TBS-T) and incubated with a 1:500 dilution of horseradish peroxidase (HRP)–conjugated antibodies (Zhongshan Goldenbridge Biotechnology Co., Ltd., Beijing, China) for 1 hour at 37°C. The peroxidase activity was revealed by 3,3′-diaminobenzidene substrate. After being counterstained with hematoxylin, the sections were examined with brightfield microscopy (DM LB2; Leica Microsystems). 

The integrated optical density (IOD) expression of positive immunostaining was analyzed according to the methods of Dong et al. ²² Briefly, images from immunostained (PRDX6, PMN, VEGF, and PEDF) sections were processed with commercial image analysis software (Image-Pro Plus 6.0; Media Cybernetics, Silver Spring, MD). The unequal illumination (shading correction) and the measurement system were corrected with a reference slide. For each sample, different areas of eight sections were scored. The images of immunostained slides were converted to grayscale, and then the IOD was measured as a linear combination between the average gray intensity and the relative area occupied by positive cells. 

Western Blotting

Five rat corneas for each group were pooled and the total proteins were extracted by the methods of Lin et al. ²³ Total protein extracts, 40 μg for each sample, were mixed with loading buffer (50 mmol/L Tris, pH 6.8, 150 mmol/L NaCl, 2% [w/v] SDS, 20% [v/v] glycerol, 5% [v/v], mercaptoethanol, 0.002% [w/v], bromophenol blue) and boiled for 5 minutes. The proteins were then separated by 12% (w/v) SDS-PAGE and transferred to a polyvinylidene difluoride membrane. To block nonspecific binding, the membrane was incubated in blocking buffer (PBS with 5% [w/v] nonfat powdered milk) for 1 hour at room temperature. The membrane was first immunoblotted with the related antibodies (diluted 1:2000 for PRDX6; 1:200 for PMN, VEGF, PEDF; and 1:1000 for GAPDH [E021010‐01; EarthOx LLC, San Francisco, CA]) in blocking solution overnight at 4°C. After being washed with PBS-T (PBS with 0.05% [v/v] Tween-20), the membrane was incubated with a 1:3000 dilution of HRP-conjugated goat anti-rabbit IgG (for PRDX6, PMN, and VEGF), rabbit anti-goat IgG (for PEDF), and goat anti-mouse IgG (for GAPDH) for 1 hour at 37°C. After being washed with PBS-T, the immunoreactive protein on the membrane was visualized by enhanced chemiluminescence and was exposed to X-ray film (Bio-Rad, Hercules, CA). 

Commercial image analysis software (GeneTools, version 4.02; Syngene, Cambridge, UK) was used to analyze the IOD of positive immunostaining. Results of Western blot were expressed as the relative quantity of target proteins to that of GAPDH. 

MDA Detection

Frozen rat corneas were homogenized with a glass grinder. MDA was detected with the trace quantity detection kit according to the manufacturer's protocol (A003‐2, MDA Detection Kit; Nanjing Jiancheng Bioengineering Research Institute, Nanjing, China). Corneal MDA content was quantified by calculation from the formula: MDA content (nmol/mg prot) = [(sample absorption − sample blank absorption)/(standard substance absorption − standard substance blank absorption)] × 10 nmol/protein content of sample. 

RNA Isolation and Reverse Transcription Polymerase Chain Reaction

Five rat corneas for each group were pooled and the total RNA was extracted using a commercial total RNA reagent (RNAiso Plus, D9108A; TaKaRa Bio, Dalian, China) according to the manufacturer's instructions. Total RNA per sample (1 μg) was used as a template to generate the first strand of cDNA by using an efficient commercial kit (ReverTra Ace, MMLV Reverse Transcriptase RNase H-,TRT-101; Toyobo Co., Ltd., Osaka, Japan). Expressions of genes were detected by PCR with gene-specific primers (Table 1) in 20 μL reaction mixture containing 2 μL 10× PCR buffer (with MgCl₂), 2 μL dNTP mix (10 mmol/L), 1 μL of each primer (25 mmol/L), 1 μL Taq DNA polymerase (2.5 U/mL), 12 μL ddH₂O, and 1 μL cDNA template. The expression of GAPDH was used as the internal control. PCR was performed under the following conditions: 95°C for 10 minutes, followed by 28 cycles for GAPDH, PRDX2, PRDX5, PRDX6, GPx1, and SOD1, or 33 cycles for others at 95°C for 30 seconds, annealing (annealing temperatures of each gene are shown in Table 1) for 30 seconds, and 72°C for 30 seconds, with a final round of extension at 72°C for 10 minutes. All of the PCR-amplified products were analyzed by electrophoresis on a 1.5% (w/v) agarose gel; the IOD of each band was analyzed as before (GeneTools, version 4.02; Syngene). The results were expressed as the relative quantity of target genes/GAPDH. 

Table 1. 

 

Sequence of Primers

Gene Name	Primer (5′–3′)	Tm (°C)
GAPDH	F: AAGTTCAACGGCACAGTCA; R: CCACAGCTTTCCAGAGGG	53
NF-κB	F: GGCAGCACTCCTTATCAA; R: GGTGTCGTCCCATCGTAG	52
TNF-α	F: TCTCATTCCTGCTCGTGG; R: GGTATGAAATGGCAAATCG	50
EGF	F: ACCGAAGGTGGCTATGTC; R: GTGATGTCGTGCCTCTGC	53
PRDX1	F: ACCTGTAGCTCGACTCTG; R: ATCCTCCTTGTTTCTTGG	50
PRDX2	F: GAGGTGCTGGGAGTGTCT; R: GGTAGGTCGTTGACTGTGAT	53
PRDX3	F: GGGAAGGTTGCTCTGGTC; R: TTCTTTCTTGGCGTGTTG	51
PRDX4	F: ACCTAAGCAAAGCCAAGA; R: AACGCAGTGTCTCATCCA	49
PRDX5	F: TAATGATGCCTTCGTGACTG; R: GAGCTGGGTGGAGGAGAT	52
PRDX6	F: ATCCTCTACCCAGCCACC; R: CCACGCCACAATCTTTCT	52
CAT	F: TATTGCCGTCCGATTCTC; R: ATGCCCTGGTCAGTCTTG	51
SOD1	F: GCAGGGCGTCATTCACTT; R: AGACTCAGACCACATAGGGA	52
GPx1	F: CAGTCCACCGTGTATGCC; R: CCATTCACCTCGCACTTC	53
GRP78	F: TTCCTGCGTCGGTGTATT; R: TCGGCAGTTTCCTTCATT	49
p53	F: CTGAGTATCTGGACGACA; R: CAGGCACAAACACGAACC	52
Fas	F: CACGGACAGGAAACACTA; R: ACTTTCAGGACTTGGGAT	49
FasL	F: GGGTTAGGAATGTATCAA; R: AATGGTCAGCAACGGTAA	49
Bcl2	F: GATACTGGAGATGAAGACT; R: CCACCGAACTCAAAGAAGG	50
caspase 3	F: CTGGACTGCGGTATTGAG; R: GGAACATCGGATTTGATT	49
caspase 8	F: GTAAACTTTGGCGGACTG; R: AGCCTCTGAAATAGCACC	50
caspase 9	F: GCCTCATCATCAACAACG; R: CTGGTATGGGACAGCATCT	51

 F, forward primer; R, reverse primer; Tm, annealing temperature; NF-κB, nuclear factor-kappa B; TNF-α, tumour necrosis factor alpha; EGF, epidermal growth factor; GRP78, Bip/glucose-regulated protein 78; p53, tumor protein 53; Fas, TNF receptor superfamily, member 6; FasL, Fas ligand; Bcl2, B-cell lymphoma 2; caspase, cysteine-aspartic proteases.

Statistical Analysis

Summary data are reported as means ± SD. Group means were analyzed by using one-way ANOVA, where P < 0.05 was considered statistically significant. A commercial software package (SPSS 18.0; SPSS, Chicago, IL) was used to perform the correlation analysis. R > 0.25 and a value of P < 0.05 were considered to be statistically significant. 

Results

Observations with a Slit Lamp In Vivo

Before ultraviolet irradiation, the corneas were shiny and transparent. After irradiation, they appeared opaque and edematous, as indicated by the disappearance of the second light strip of the slit lamp. The corneal edema showed a slight decrease after being treated for 14 days in the buffer-treated group. There was a dramatic reduction of corneal edema and the corneas were almost completely hyaloid after 14 days of PRDX6 treatment (Fig. 1A). 

Figure 1. 

 

PRDX6 lessens corneal edema, prevents neovascularization, and decreases MDA content of irradiated corneas. (A) Appearance of eyes by slit-lamp microscopy. The normal corneas (normal) were glassy and transparent, whereas the irradiated corneas (1 day after irradiation) appeared opaque and edematous. After 4 days of treatment, peripheral neovascularization occurred in the buffer-treated group; and after 14 days, the neovascularization had spread all over the corneas in the buffer-treatment group, whereas there were only a few blood vessels in the limbus and none in the peripheral and central corneas of the PRDX6-treated group. There was dramatic reduction of corneal edema and the corneas were almost completely clear at 14 days in the PRDX6-treated group. (B) H&E staining of corneas. After 7 days of treatment, there was more neovascularization in the stroma layer of control and buffer-treated groups than that in the PRDX6-treated group and the difference became obvious at 14 days. Scale bar, 50 μm. (C) Detection of MDA content. After 7 days of treatment, the MDA of control group (without treatment after radiation) was about 2-fold that of the PRDX6-treated group, and the MDA content of the buffer-treated group was nearly 1.5-fold that in the PRDX6-treated group. After being treated for 14 days, there was still a significantly lower MDA content in the PRDX6-treated group than that of the control and buffer-treated groups. *Statistically significant difference between the control group and PRDX6-treated or buffer-treated groups of the same day (P < 0.05). †Statistically significant difference between PRDX6-treated and buffer-treated groups of the same day (P < 0.05). normal, corneas without irradiation or treatment; buffer, irradiated corneas were treated with the buffer in which PRDX6 was dissolved; PRDX6, irradiated corneas were treated with PRDX6 protein; control, irradiated corneas without any treatment; 1d, the day after irradiation; 4d, the fourth day of treatment after irradiation; 7d, the seventh day of treatment after irradiation; 14d, the fourteenth day of treatment after irradiation.

After being treated for 4 days, peripheral neovascularization toward the central cornea occurred in the buffer-treated group. After 14 days, neovascularization had spread to the corneas in the buffer-treated group but the PRDX6-treated group showed only a few blood vessels in the limbus and none in the peripheral or central cornea (Fig. 1A). 

Corneal Epithelial Thickness

The mean thicknesses of the epithelium, stroma, and total cornea increased after irradiation. At the seventh day after irradiation, the thicknesses of epithelium in the buffer- and PRDX6-treated groups were significantly lower than those of the control group (P < 0.05). At the fourteenth day after irradiation, the thicknesses of stroma and total cornea in PRDX6-treated group were significantly decreased below the control and buffer-treated group values (P < 0.05) (Fig. 1B, Table 2). 

Table 2. 

 

Thickness of Epithelium, Stroma, and Total Cornea

Group	Epithelium ± SD (μm)	Stroma ± SD (μm)	Total Cornea ± SD (μm)
Normal	21.51 ± 1.15	47.73 ± 2.28	68.18 ± 4.55
1d	31.36 ± 3.89	237.88 ± 9.46	284.85 ± 11.44
7d-control	35.0 ± 2.76	150.15 ± 12.12	173.27 ± 11.1
7d-buffer	22.73 ± 0.86*	143.03 ± 13.92	155.89 ± 9.26
7d-PRDX6	23.18 ± 0.46*	141.06 ± 16.65	156.82 ± 14.87
14d-control	28.27 ± 6.4	153.03 ± 11.44	172.21 ± 15.46
14d-buffer	31.18 ± 4.62	140.15 ± 3.47†	168.39 ± 12.38†
14d-PRDX6	24.17 ± 2.69	98.48 ± 9.46*†	125.0 ± 8.5*†

 Normal, cornea without irradiation, 7d-control, the seventh day of irradiated corneas without any treatment; 7d-buffer, the seventh day of irradiated corneas treated with buffer in which the PRDX6 protein was dissolved; 7d-PRDX6, the seventh day of irradiated corneas treated with PRDX6 protein; 14d-control, the fourteenth day of irradiated corneas without any treatment; 14d-buffer, the fourteenth day of irradiated corneas treated with buffer in which the PRDX6 protein was dissolved; 14d-PRDX6, the fourteenth day of irradiated corneas treated with PRDX6 protein.

*  Statistically significant difference between control and PRDX6- or buffer-treated groups of the same day (P < 0.05).

†  Statistically significant difference between PRDX6- and buffer-treated groups of the same day (P < 0.05).

MDA Content

One day after irradiation, the MDA content in the rat corneas increased approximately 10-fold more than that of the nontreated corneas. The MDA content decreased gradually thereafter, and the reduction was most obvious in the PRDX6-treated group. After 7 days of treatment, the MDA content of the control group was twice that of the PRDX6-treated group. The MDA content of the buffer-treated group was nearly 1.5-fold more than that in the PRDX6-treated group. At 14 days, there was still a significant reduction of MDA content in the PRDX6-treated group compared with that of the control and buffer-treated groups (Fig. 1C). 

Area of Corneal Neovascularization

The area of corneal neovascularization increased in both buffer- and PRDX6-treated groups after irradiation, although the increase in the PRDX6-treated group was low. At the fourth, seventh, and fourteenth days after irradiation, the area of corneal neovascularization in the PRDX6-treated group was significantly smaller than that of the buffer-treated group (P < 0.05) (Table 3). 

Table 3. 

 

Area of Corneal Neovascularization Units

Group	4d Mean ± SD (mm2)	7d Mean ± SD (mm2)	14d Mean ± SD (mm2)
Buffer-treated	6.13 ± 1.46	10.08 ± 2.26	17.02 ± 2.67
PRDX6-treated	2.63 ± 0.83*	4.6 ± 1.38*	6.71 ± 2.09*

 Buffer-treated, irradiated corneas were treated with the buffer in which PRDX6 was dissolved; PRDX6-treated, irradiated corneas were treated with PRDX6 protein.

*  Statistically significant difference between PRDX6- and buffer-treated groups of the same day (P < 0.05).

Corneal Opacity

Ten rats per group were detected to evaluate the corneal opacity. On the first day after irradiation, the corneal opacity was grade IV. At the fourth day, the corneal opacity of the buffer-treated group was two in grade II and eight in grade III. The corneal opacity of the PRDX6-treated group was five in both grade II and grade I. At the seventh day, the corneal opacity of the buffer-treated was six in grade II and four in grade III. The corneal opacity of the PRDX6-treated group was two in grade II, five in grade I, and three in grade 0. At the fourteenth day, the corneal opacity of the buffer-treated was seven in grade II and three in grade III. The corneal opacity of the PRDX6-treated group was two in grade I and eight in grade 0. 

Immunoreactivity of PRDX6

PRDX6 was expressed in the epithelial cells of the normal cornea. After ultraviolet irradiation, the immunoreactivity of PRDX6 in the epithelial cells increased and staining appeared in the stroma layer. Seven days after irradiation, the immunoreactivity of PRDX6 in the epithelial cells and the stromal layer was increased in the control and buffer-treated groups. By 14 days after treatment, PRDX6 staining also appeared in the epithelial cells and the stroma layer, but was slightly lower in the control and buffer-treated groups. It is noteworthy that the immunoreactivity of endogenous PRDX6 was tenuous when the corneas were topically administered with exogenous PRDX6 for 7 and 14 days (Fig. 2A). IOD showed that the immunoreactivity of epithelial and stromal PRDX6 in the PRDX6-treated group was significantly lower than that of the control and buffer-treated groups at both 7 and 14 days (P < 0.05) (Fig. 2C). 

Figure 2. 

 

Expression of PRDX6 and PMN in normal and irradiated rat corneas. (A) Expression of PRDX6. (B) Expression of PMN. PRDX6 and PMN existed in the epithelium of the normal corneas. The expression of PRDX6 and PMN in the epithelium increased and the staining appeared in the stroma of irradiated corneas. After being treated for 7 days, the expression of PRDX6 in the epithelial cells and the stroma increased in the control and buffer-treated groups. After 14 days treatment, the staining also appeared in the epithelium and the stromal layer and was slightly lower in the control and buffer groups. The expressions of endogenous PRDX6 and PMN were tenuous when the corneas were topically administered with exogenous PRDX6 protein for 7 and 14 days. normal, corneas without radiation or treatment; PRDX6, irradiated corneas were treated with PRDX6; 7d, the seventh day of treatment after radiation. Scale bar, 50 μm. (C) The IOD analysis showed that the immunoreactivity of epithelial and stromal PRDX6 in the PRDX6-treated group was significantly lower than that of the control and buffer-treated groups at both 7 and 14 days. (D) The IOD analysis showed that the immunoreactivity of epithelial and stromal PMN in the PRDX6-treated group was significantly lower than that of the control and buffer-treated groups at both 7 and 14 days. *Statistically significant difference between control group and PRDX6-treated or buffer-treated groups of the same day (P < 0.05). †Statistically significant difference between PRDX6-treated and buffer-treated groups of the same day (P < 0.05).

Immunoreactivity of Polymorphonuclear Neutrophil

Polymorphonuclear neutrophil (PMN) appeared in the epithelial cells of the normal cornea. The staining was deeper in the posterior epithelial cells than that in the anterior epithelial cells. One day after ultraviolet irradiation, PMN in the posterior epithelial cells increased and the cells appeared in the stroma. After 7 days of treatment, PMN within the epithelium and the stroma increased in the control and buffer-treated groups, and after 14 days the staining in the two groups decreased slightly. It is noteworthy that the appearance of PMN was tenuous when the corneas were topically administered with PRDX6 (Fig. 2B). IOD analysis showed that the immunoreactivity of epithelial and stromal PMN in the PRDX6-treated group was significantly lower than that of the control and buffer-treated groups at both 7 and 14 days (P < 0.05) (Fig. 2D). 

Immunoreactivity of VEGF

VEGF was expressed at low levels in the epithelium of the normal cornea. One day after irradiation there was a dramatic increase in immunoreactivity of VEGF in both the epithelium and stroma. After being treated for 7 days, the immunoreactivity of VEGF in the PRDX6-treated group was lower than that in the control and buffer-treated groups, and the staining almost recovered to the level of the normal cornea. After 14 days of treatment, the immunoreactivity of VEGF in the control, PRDX6-, and buffer-treated groups increased, but a lower staining after PRDX6 treatment than that in the control and buffer-treated groups was still present (Fig. 3A). IOD analysis showed that the immunoreactivity of epithelial and stromal VEGF in the PRDX6-treated group was significantly lower than that of the control and buffer-treated groups at both 7 and 14 days (P < 0.05) (Fig. 3C). 

Figure 3. 

 

Expression of VEGF and PEDF in normal and irradiated rat corneas. (A) Expression of VEGF. VEGF was slightly expressed in the epithelial cells of the normal corneas and there was a dramatic increased expression of VEGF in the posterior epithelial cells and stroma of irradiated corneas (1 day after irradiation). After treatment for 7 days, the expression of VEGF in the PRDX6-treated group was lower than that of the control and buffer-treated groups and the staining of VEGF in the PRDX6-treated group had almost recovered to the level of the normal cornea. After being treated for 14 days, the expressions of VEGF in the control, buffer-, and PRDX6-treated groups were all increased. The lower staining in the PRDX6-treated group than that in the control and buffer-treated groups was evident. (B) Expression of PEDF. PEDF was moderately expressed in the epithelial cells of normal corneas. The staining was decreased in the epithelium of irradiated cornea, whereas the staining in the stroma was detected (1 day postirradiation). At day 7 and 14 after treatment, the expression of PEDF was maintained low in the control and buffer-treated groups, although the staining was increased in the PRDX6-treated group. normal, corneas without irradiation or treatment; PRDX6, irradiated corneas were treated with PRDX6 protein; 7d, the seventh day of treatment after irradiation. Scale bar, 50 μm. (C) The IOD analysis showed that the immunoreactivity of epithelial and stromal VEGF in the PRDX6-treated group was significantly lower than that of the control and buffer-treated groups at both 7 and 14 days. (D) The IOD analysis showed that the immunoreactivity of epithelial PEDF in the PRDX6-treated group was significantly higher than that of the control and buffer-treated groups at 7 and 14 days (P < 0.05), and the immunoreactivity of stromal PEDF in the PRDX6-treated group was significantly higher than that of the control group at day 7. *Statistically significant difference between control group and PRDX6- or buffer-treated groups of the same day (P < 0.05). †Statistically significant difference between PRDX6- and buffer-treated groups of the same day (P < 0.05).

Immunoreactivity of PEDF

PEDF was moderately expressed in the epithelial cells of nonirradiated corneas. One day after irradiation the staining was decreased in the epithelium and staining in the stroma was detected. After 7 and 14 days of treatment, the immunoreactivity of PEDF appeared at a low level in the control and buffer groups, although the staining increased significantly in the PRDX6-treated group (Fig. 3B). IOD analysis showed that the immunoreactivity of epithelial PEDF in the PRDX6-treated group was significantly higher than that of the control and buffer-treated groups at 7 and 14 days (P < 0.05). The immunoreactivity of stromal PEDF in the PRDX6-treated group was significantly higher than that of the control group at day 7 (P < 0.05) (Fig. 3D), which was consistent with the results of Western blot (Fig. 4). 

Figure 4. 

 

Expression of PRDX6, PMN, VEGF, and PEDF detected by Western blot. (A) Western blots showed that the expression of PRDX6, PMN, and VEGF in the PRDX6-treated group was lower than that of the buffer-treated group at both the seventh and fourteenth days, whereas the expression of PEDF was the reverse. PMN and VEGF were expressed at a low level in the unirradiated corneas, whereas the expressions were dramatically increased after irradiation (1d). The expression of PEDF was sharply decreased after irradiation (1d) below that of the unirradiated cornea. (B) Densitometry of protein expression compared with GAPDH showed significant differences between the buffer- and PRDX6-treated groups in endogenous PRDX6, PMN, VEGF, and PEDF at days 7 and 14 (P < 0.05). normal, cornea without irradiation; 7d, the seventh day of treatment after irradiation; PRDX6, irradiated corneas were treated with PRDX6 protein. *Significant difference between PRDX6- and buffer-treated groups (P < 0.05). Results of one representative experiment of two others.

Correlation Analysis

The expression of PRDX6 relative to PMN and to VEGF was positively correlated, R = 0.788 and 0.508, respectively. There was a negative correlation between PRDX6 and PEDF, R = 0.504 (Table 4). 

Table 4. 

 

Correlation Analysis

	PRDX6	PMN R (Sig.)	VEGF R (Sig.)	PEDF R (Sig.)
PRDX6	1	0.788 (0.000)	0.508 (0.000)	−0.504 (0.000)
PMN		1	0.439 (0.002)	−0.369 (0.010)
VEGF			1	−0.550 (0.000)
PEDF				1

R  , correlation coefficient; Sig, significant difference; −, negative correlation.

Different Expressions of Genes

Compared with that in the buffer-treated group, the expression of PRDX4, PRDX5, CAT, and GPx1 was increased in the PRDX6-treated group by the seventh day (P < 0.05). By the fourteenth day, the expression of PRDX2, PRDX3, PRDX4, and CAT in the PRDX6-treated group was lower than that of the buffer-treated group, whereas the expression of SOD1 and GPx1 was higher than that of the buffer-treated group (P < 0.05). PRDX1-6, SOD1, CAT, and GPx1 were all expressed in normal corneas, but the expression of SOD1 was at low. Compared with that in controls, the expression of PRDX1, SOD1, and GPx1 was increased after 1 day of irradiation, whereas the expression of PRDX3 was dramatically decreased (Fig. 5). 

Figure 5. 

 

Expression of antioxidative genes in normal and irradiated rat cornea. (A) RT-PCR agarose gel. (B) Densitometry of RNA expression compared with GAPDH. Compared with the buffer-treated group, the expression of PRDX4, PRDX5, CAT, and GPx1 was higher in the PRDX6-treated group at the seventh day (P < 0.05). At the fourteenth day, the expression of PRDX2, PRDX3, PRDX4, and CAT in the PRDX6-treated group was lower than that of the buffer-treated group, whereas the expression of SOD1 and GPx1 was higher than that of the buffer-treated group (P < 0.05). In the normal cornea, PRDX1–6, SOD1, CAT, and GPx1 were all expressed, and the expression of SOD1 was at a low level. Compared with that of the normal cornea, the expression of PRDX1, SOD1, and GPx1 was increased after irradiation one day (1d), whereas the expression of PRDX3 was dramatically decreased. normal, cornea without irradiation; 7d, the seventh day of treatment after irradiation; PRDX6, irradiated corneas were treated with PRDX6 protein; M, DL2000 DNA marker; NC, PCR negative control that lack the cDNA template in the reaction mixture. The results are expressed as the relative quantity of target gene/GAPDH. *Statistically significant difference between normal group and PRDX6- or buffer-treated groups of the same day (P < 0.05). †Statistically significant difference between PRDX6- and buffer-treated groups of the same day (P < 0.05). Results of one representative experiment of two others.

On day 7, the expression of FasL and Fas in the PRDX6-treated group was higher than that of the buffer-treated group (P < 0.05). On the fourteenth day, the expression of p53, Fas, caspase 8, caspase 3, and caspase 9 in the PRDX6-treated group was lower than that in the buffer-treated group, whereas the expression of Bcl2 was higher than that in the buffer-treated group (P < 0.05) (Fig. 6). 

Figure 6. 

 

Expression of P53, FasL, Fas, caspase 8, Bcl 2, caspase 3, and caspase 9 in normal and irradiated rat corneas. (A) RT-PCR agarose gel. (B) Densitometry of RNA expression compared with GAPDH. At the seventh day, the expression of FasL and Fas in the PRDX6-treated group was higher than that of the buffer-treated group (P < 0.05). At the fourteenth day, the expression of p53, Fas, caspase 8, caspase 3, and caspase 9 in the PRDX6-treated group was lower than that of the buffer-treated group, whereas the expression of Bcl 2 was higher than that of the buffer-treated group (P < 0.05). normal, cornea without irradiation; 7d, the seventh day of treatment after irradiation; PRDX6, irradiated corneas were treated with PRDX6 protein. The results were expressed as the relative quantity of target genes/GAPDH. *Statistically significant difference between normal and PRDX6- or buffer-treated groups of the same day (P < 0.05). †Statistically significant difference between PRDX6- and buffer-treated groups of the same day (P < 0.05). Results of one representative experiment of two others.

There was no expression of TNF-α detected in the unirradiated and irradiated corneas. On the fourteenth day, the expression of EGF and NF-κB in the PRDX6-treated group was lower than that of the buffer-treated group (P < 0.05). The expression of GRP78 was decreased in the PRDX6-treated group below that of the buffer-treated group by both 7 and 14 days (P < 0.05) (Fig. 7). 

Figure 7. 

 

Expression of TNF-α, EGF, NF-κb, MMP9, and GRP78 in normal and irradiated rat cornea. (A) RT-PCR agarose gel. (B) Densitometry of RNA expression compared with GAPDH. There was no expression of TNF-α detected in the corneas of either unirradiated or irradiated groups. At the fourteenth day, the expression of EGF and NF-κB in the PRDX6-treated group was lower than that of the buffer-treated group (P < 0.05). The expression of GRP78 was low in the PRDX6-treated group compared with that of the buffer-treated group at both the seventh and fourteenth days (P < 0.05). normal, cornea without irradiation; 7d, the seventh day of treatment after irradiation; PRDX6, irradiated corneas were treated with PRDX6 protein. The results were expressed as the relative quantity of target gene/GAPDH. *Statistically significant difference between normal group and PRDX6- or buffer-treated groups of the same day (P < 0.05). †Statistically significant difference between PRDX6- and buffer-treated groups of the same day (P < 0.05). Results of one representative experiment of two others.

Discussion

As a protective molecule, PRDX6 controls the reactive oxygen species (ROS) level of cells. The critical role of intracellular PRDX6 in protecting LECs has been confirmed. ^3,9,10 The result of our previous study showed that exogenous PRDX6 could protect spermatozoa from oxidative damage induced by H₂O₂. ²⁴ The application of PRDX6 protein onto an incised wound and burned tracheal epithelium leads to enhanced healing, smaller cicatrix, and soon to complete recovery, ^25,26 but the possible mechanism remained to be investigated. In the present study, we evaluated the effects and possible mechanism of exogenous PRDX6 in an in vivo model involving ultraviolet irradiation of rat corneas, which appeared opaque, edematous, inflammatory, and developed neovascularization. We found that topically administered PRDX6 reduced corneal opacity and edema and diminished inflammation and neovascularization. 

Neutrophils provide the first line of immunological defense, in part through production of ROS. ²⁷ However, the neutrophils may also release into the injured tissue oxidative, hydrolytic, and pore-forming molecules that can damage host cells. Therefore, an exaggerated or constant influx and the presence of neutrophils are detrimental. ^28,29 To illustrate the relationship between topical administered PRDX6 and inflammation further, we detected PMN antigen, which indicates the presence of neutrophils. The migration and accumulation of neutrophils in the stroma of irradiated corneas were demonstrated, but these processes were efficiently depressed when the corneas were topically administered exogenous PRDX6. The results of our study indicated that exogenous PRDX6 reduces the migration and accumulation of neutrophils in the irradiated corneas. 

It is well established that corneal neovascularization is tightly regulated by a dynamic, natural equilibrium between local proangiogenic factors (such as VEGF) and antiangiogenic molecules (such as PEDF). ^28–32 Exposure to H₂O₂ increases VEGF expression in endothelial cells. ³³ PEDF is a potential neuroprotective agent in H₂O₂-induced retinal neuron death, which can be increased in expression and protect cells from injury induced by ROS. ³⁴ In the present study, the inhibited expression of VEGF and promoted expression of PEDF were detected after the topical administration of PRDX6. In addition, the positive correlation between endogenous PRDX6, PMN, and VEGF, and the negative correlation between PRDX6 and PEDF were demonstrated. 

To analyze the possible mechanism of topically administered PRDX6 on the ultraviolet-injured cornea, we monitored oxidative stress of corneas in different groups. MDA, the final product of lipid oxidation, was the marker of oxidative stress. A high level of intracellular ROS can result in dysfunction of cellular organelles, including the endoplasmic reticulum (ER), which could cause cell apoptosis. ^35,36 Bip/glucose-regulated protein 78 (GRP78), one of the ER's stress-related chaperone proteins, is one of the ER-stress markers. The results of the present study showed lower MDA content and decreased expression of GRP78 in the PRDX6-treated group. In addition, Schreck et al. ³⁷ showed that ROS can activate the transcription factor nuclear factor-kappa B (NF-κB), which can induce cell damage. A low expression of the NF-κB gene in the PRDX6-treated group was also detected in the present study, consistent with our results of MDA detection. The results of our study indicate that topically administered PRDX6 kept the homeostasis of the corneal cell, which might be the reason for the improvement in corneal transparency, decrease in edema, and diminution of inflammation and neovascularization. 

It has been reported that there are at least three pathways related with apoptosis induced by ultraviolet light. The first is cell apoptosis dependent on DNA damage. DNA is the major target of UV irradiation. DNA damage increases the expression of p53, which could induce the cell apoptosis. ^38,39 The second is cell apoptosis mediated by a death receptor. There are six members of the death receptor family, of which Fas has been thoroughly investigated. Fas can be activated by FasL or other stimulating antibodies, which attract proteins such as pro-caspase 8, which can be hydrolyzed and activated to induce apoptosis. ^39–41 The third is apoptosis mediated by mitochondrial damage. Mitochondria, as the production center of ATP in the cell, play important roles in the apoptosis. The expression of Bcl2 is decreased after ultraviolet irradiation, which can promote Smac/DIABLO protein release into the cytoplasm from mitochondria. The binding of Smac/DIABLO and apoptosis-inhibited protein (IAP), can activate caspase 9 and caspase 3, which cause apoptosis. ^42–44 We found that (1) the expression of p53 increased after irradiation on the fourteenth day in the buffer-treated group, and the expression was inhibited by the topical administration of PRDX6; (2) the expression of FasL and Fas were changed, and the expression of caspase 8 was increased without administered PRDX6; (3) the expression of Bcl2 was lower in the buffer-treated group than that in the PRDX6-treated group, which induced higher expression of caspase 3 and 9 in the buffer-treated group. This is consistent with the result of Sakamoto and Nakamura, ³⁶ who showed ultraviolet-B irradiation stimulated caspase 3 activity in human corneal epithelial cells. ⁴⁵ According to the present results, we presume that topically administered PRDX6 inhibits cell damage by affecting the three classical apoptotic pathways. 

Lack of PRDX6 in LECs had been shown to cause the mRNA and protein expression of PRDX1 and 5 to decrease and PRDX2, 3, and 4 to increase, whereas the expression of CAT and GPx1 decreased. However, the increased expression of PRDX2, 3, and 4 was not able to attenuate the changes in the PRDX6-deficient LECs, and PRDX6 played a pivotal role in blocking ER stress. ⁴⁶ Wang et al. ⁹ and Manevich and Fisher ⁴⁷ also reported that the role of PRDX6 in mice was not redundant. PRDX6 has both GSH peroxidase and PLA2 activities, ⁴⁷ and maintains Ca²⁺ homeostasis, ^48,49 which makes PRDX6 different from other antioxidants. The results of the present study show that the expression of PRDX4 and 5, CAT, and GPx1 increased in the PRDX6-treated group at the seventh day after irradiation compared with that of the buffer-treated group. The expression of PRDX2, 3, and 4, CAT decreased and SOD1 and GPx1 increased in the PRDX6-treated group at the fourteenth day after irradiation compared with that of the buffer-treated group. Accordingly, we presume that topically administered PRDX6 kept the homeostasis in rat corneal cell by regulating the expression of genes of antioxidant defense systems, and the expression of these genes was for different durations. 

We also detected the expressions of TNF-α and EGF in the present study. There was no positive expression of TNF-α in normal and irradiated corneas, which indicated the corneal damage induced by ultraviolet light was not related to the pathway related with TNF-α. There was expression of EGF in the normal cornea, which is consistent with the result of Han et al. ³² By the fourteenth day the expression of EGF in the buffer-treated group was significantly higher than that of the PRDX6-treated group, which might be the reason for the greater neovascularization in the buffer-treated group. However, the relationship between EGF and neovascularization remains to be investigated. 

In conclusion, the results of our study indicate that topical administration of PRDX6 maintains the homeostasis of corneal cells, diminishes inflammation, and suppresses neovascularization and apoptosis under ultraviolet irradiation. 

Acknowledgments

The authors thank Jun Mei Hao, PhD (Yantai Yuhuangding Hospital, China), for help in pathology analysis, and Colleen A. McMullen (Department of Physiology, University of Kentucky) and Trevor G. Cooper for language editing. 

Footnotes

 Supported by China National Natural Science Funds Grant 31071262 and Yantai Science and Technology Program Grant 2009107.

Footnotes

³  These authors contributed equally to the work presented here and should therefore be regarded as equivalent authors.

Footnotes

 Disclosure: H. Shi, None; H.J. Yu, None; H.Y. Wang, None; W.T. Wang, None; S.H. Jin, None; P. Zhu, None; S.J. Li, None; C.T. Rong, None; J.Y. Li, None