Involvement of Oxidative Stress on Corneal Epithelial Alterations in a Blink-Suppressed Dry Eye

Abstract

purpose. To investigate whether oxidative stress is involved in the etiology of the corneal disorder in blink-suppressed dry eye in a clinically relevant in vivo rat model.

methods. A series of treatments were performed under continuous exposure to low-humidity airflow. Rats were placed on a jogging board (JB) made of a plastic pipe for 7.5 h/d, and for 16.5 hours, they were placed in individual cages without a JB. This protocol was repeated for up to 30 days. Corneal surface alteration was evaluated by the score of punctate fluorescein staining. To assess oxidative stress status, the levels of damaged DNA, and the protein modification by reactive aldehydes in corneal epithelia were detected by immunohistochemistry, using 8-hydroxy-2-deoxyguanosine, 4-hydroxynonenal- and malondialdehyde-specific antibodies.

results. Significant increases in the fluorescein staining score were observed from days 1 to 30 compared with the initial value. The average score for the dry eye group was significantly increased compared with that for the nontreatment group at all time points throughout the experiment. Immunoreactivity of all oxidative stress markers increased in the dry eye treatment. Quantitative analysis of the positive-stained cells showed a significant increase in the number of positive cells after 10 and 30 days in the dry eye treatment group compared with the nontreatment group.

conclusions. These results suggest a relationship between the accumulation of oxidative stress and the etiology of corneal epithelial alterations in blink-suppressed dry eye.

Oxidative damage resulting from free radicals and/or H₂O₂ has been implicated in the pathogenesis of many chronic progressive diseases, such as cancer, inflammation, and neurodegenerative disorders. Oxidative damage has also been involved in several ocular diseases ¹ including age-related macular degeneration, ^{2 3} cataract ⁴ induced by aging, and overexposure to sunlight. The ocular surface epithelial cell layers, consisting of the conjunctiva and cornea, are the initial areas protecting the eye from pathogenic microbes, foreign invasion, and a dry environment. This tissue is exposed to atmospheric oxygen and sunlight including the ultraviolet range, ⁵ known as causative factors of oxidative stress in biological systems. Nevertheless, the potential role of oxidative damage in common human corneal diseases has not been thoroughly investigated. ⁶  

Dry eye is defined as a disturbance in tear film physiology that leads to various abnormal states of ocular surface cells. ^{7 8} Recent progress in understanding the pathophysiology of dry eye has demonstrated that the pathogenic mechanism is not limited to dysfunction of the lacrimal apparatus but also involves external factors such as a dehydrating environment and suppressed blink frequency due to visual tasking. ^{9 10 11 12} We have established a rat model of blink-suppressed dry eye using a novel procedure. ¹³ In our model, in place of tasking visual activity such as video display terminal (VDT) work or driving a car, we placed the rats on a jogging board (JB) to disturb postural equilibrium, requiring continuous visual fixation to promote postural stability. This dry eye model mimics the actions of office workers engaged in daily long-term VDT work and the symptom of so-called “office dry eye.” 

In this study, performed in our rat dry eye model, superficial punctate keratopathy (SPK) was accompanied by an increase in oxidative stress markers, the expression of antioxidant-related genes and reactive oxygen species (ROS) production in corneal epithelia. These results suggest a strong relationship between the accumulation of oxidative stress and the etiology of corneal surface disorder in blink-suppressed dry eye. Furthermore, we demonstrated the discordance of epithelial differentiation capacity using a 5-bromo-2-deoxyuridine (BrdU) incorporation assay. This indicates that oxidative stress activates cell regulatory molecules that chronically distort the regenerative capacity of the corneal epithelial cell layer under dry eye conditions. 

Materials and Methods

Animals

Eight-week-old female Sprague-Dawley rats (n= 8–12 in each experiment; Tokyo Laboratory Animal Science, Tokyo, Japan) were used. They were quarantined and acclimatized for 1 week before the experiments under standard conditions (SC) as follows: room temperature 23 ± 3°C, relative humidity of 60% ± 10%, alternating 12-hour light–dark cycle (8 AM to 8 PM), and water and food ad libitum. All procedures were performed according to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 

Blink-Suppressed Dry Eye Model

Rat dry eye was developed by placing rats on a JB in combination with dry condition stress (JBDC), according to previously published procedures. ¹³ In brief, the rats underwent a series of these dry-inducing procedures under dry condition stress, with a room temperature of 22 ± 3°C, relative humidity 25% ± 5%, and constant air flow at 2 to 4 m/s. Each rat remained in place on the JB for 7.5 h/d between 9 AM and 5 PM. The track of the JB was made of plastic piping 30 mm in diameter × 50 mm in length, suspended 60 cm above the bottom and 30 cm below the top frame by a wire. To prevent the rat from slipping from the JB, the track was covered with metallic mesh. In addition to being placed on the JB, the rats were exposed to constant air flow at 2 to 4 m/s aimed at the face, produced by an 18-cm diameter electric fan (Morita Denko, Osaka, Japan) fixed 25 cm horizontally from the JB track (Fig. 1) . Although the rats rarely left the JB, we checked on their condition at least once every hour during treatment. 

Corneal Fluorescein Staining

Changes in the corneal surface were determined by applying a fluorescein solution under a blue-free barrier filter. ¹⁴ Corneal staining was graded based on the area of staining according to previously described criteria. ¹³ The total area of punctate staining was denoted as grade 0 when there was no punctate staining, grade 0.5 when less than one sixteenth was stained, grade 1 when less than one eighth was stained, grade 2 when one fourth was stained, grade 3 when greater than one half was stained, and grade 4 when the entire area was stained. 

Immunohistochemistry for 8-Hydroxy-2-deoxyguanosine (8-OHdG), 4-Hydroxynonenal (4-HNE)-, and Malondialdehyde (MDA)-Modified Proteins

The rats were killed with an overdose of pentobarbital, and their eyeballs were removed and fixed in 4% paraformaldehyde and 0.25% glutaraldehyde in 0.1 M phosphate buffer (pH 7.4). The corneal specimens were dehydrated and embedded in paraffin. The cornea was cut at 4 μm along the equator, and immunohistochemical staining was performed. After they were blocked with normal horse serum (Vector Laboratories, Burlingame, CA), the sections were layered for diluted mouse anti-8-OHdG monoclonal antibody (Japan Institute for the Control of Aging [JaICA], Shizuoka, Japan) and anti-4-HNE- and anti-MDA-modified protein antibodies (JaICA). Antibody binding was detected with a horse anti-mouse IgG ABC kit (Vector Laboratories) and a black alkaline phosphatase substrate kit (Vector Laboratories) according to the manufacturer’s protocol. Antibody-treated sections were examined by light microscope, and the number of positive cells was counted in each section. 

Quantitative Real-Time RT-PCR Analysis

Total RNA from cornea epithelia was isolated by using an RNA extraction reagent (ISOGEN; Nippon Gene, Tokyo, Japan), according to the manufacturer’s instructions. Quantitative real-time RT-PCR analysis was then performed (Prism 7000 Sequence Detection System using TaqMan universal PCR master mix according to the manufacturer’s specifications; Applied Biosystems, Inc. [ABI], Foster City, CA). PCR primers were purchased in a kit (Assay-on-Demand Gene Expression Products; ABI). The specific ABI assay identification numbers for primer sets used in this study include: glutathione peroxidase 1 (GPX 1): Rn00577994_g1, catalase (Cat): Rn00560930_m1, superoxide dismutase 1 (SOD1): Rn00566938_m1, matrix metalloproteinase-9 (MMP-9): Rn00579162_m1, tumor necrosis factor (TNFα): Rn00562055_m1, Rodent GAPDH: 4308313. The gene-specific probes were labeled using reporter dye FAM, and the GAPDH internal control probe was labeled with different reporter dye VIC at the 5′ end. 

Thermal cycling conditions were as follows: 2 minutes at 50°C, 10 minutes at 95°C followed by 40 cycles of 95°C, for 15 seconds, and 60°C for 1 minute. Amplification data were analyzed by sequence detection software (Prism Sequence Detection Software version 1.0; ABI). The relative expression of RNA species was calculated using the comparative Ct (threshold cycle number) method. All data were controlled by GAPDH. The experiments were performed in triplicate to ensure the reproducibility of the results, and statistical analysis was performed as described later. 

Measurements of Corneal ROS Production

Intracellular generation of ROS during the dry eye treatment was measured using 2′,7′-dichlorofluorescein diacetate (DCFH-DA; Invitrogen-Molecular Probes, Inc., Eugene, OR). DCFH-DA is a nonfluorescent, membrane-permeable compound which on oxidation becomes fluorescent and membrane impermeable. Whole corneal epithelium was scraped with an ophthalmic surgical blade and placed in a 96-well plate containing 200 μL of Krebs-Ringer bicarbonate buffer. The cells were incubated in the dark with 20 μg/mL DCFH for 30 minute at 37°C. The plates were read at an excitation of 480 nm and emission of 530 nm (Cytofluor 4000; PerSeptive Biosystems, Inc., Framingham, MA). 

Changes in Corneal Epithelial Differentiation

Proliferation Assay.

Rats received BrdU (200 mg/kg) intraperitoneally 3 hours before euthanatization by the administration of an overdose of pentobarbital. Immediately after euthanatization, their eyeballs were removed and fixed in 10% neutral-buffered formalin, and the corneal specimens were dehydrated and embedded in paraffin. The cornea was cut at 4 μm along the equator, and BrdU was detected with an anti-BrdU staining kit (Zymed Laboratories Inc., South San Francisco, CA), according to the manufacturer’s protocol. The percentage of BrdU-positive cells in the total basal cells was calculated. 

Migration Assay.

Sixteen hours or 112 hours after BrdU injection, rats were euthanatized, and cross sections of the cornea were prepared according to the method just described. At each point, the percentage of total BrdU-positive cells in the basal cell layer was calculated, and the reduction rate over 4 days was compared with nontreatment and JBDC. 

Detection of Apoptosis

Apoptosis was evaluated by TUNEL assay (Apoptosis In Situ Detection Kit; Wako Pure Chemical, Osaka, Japan). TUNEL-positive cells were detected according to the manufacturer’s protocol. The corneal section was prepared by the method just described, and the number of positive cells was counted in each cornea. 

Statistical Analysis

For the fluorescein staining score, data were analyzed by the Mann-Whitney test between two groups. Other data were analyzed by Student’s t-test between two groups. Differences were accepted as being statistically significant for P < 0.05. 

Results

Corneal Fluorescein Staining

In our previous study, we compared corneal epithelial fluorescein staining among three groups: rats in dry conditions for the entire time of their daily ride on the JB (JBDC), those placed in dry conditions for the entire time without the JB, and standard conditions with daily JB sessions for 10 days. We found that JBDC was necessary to produce a sustained increase in corneal fluorescein staining. ¹³ In this study, we examined the change in the fluorescein staining score after 30 days of JBDC. Significant increases in the corneal staining score were observed from days 1 to 30 compared with the initial value. The average score for the JBDC group was significantly increased compared with that for the nontreatment group at all time points throughout the experiment (Fig. 2) . 

Effect of Dry Eye Treatment on 8-OHdG, Protein Modifications by MDA and 4-HNE in Rat Corneal Epithelia

To confirm the oxidative stress status in corneal epithelia during JBDC, we assessed the formation of oxidative stress–specific markers on days 10 and 30. Representative patterns of the immunohistochemical localization of each oxidative stress marker in the corneal epithelia on day 10 are shown in Figure 3A . All three markers show nuclear and/or perinuclear localization on surface and wing cells. Increased immunoreactivity was observed in JBDC. In the immunoreactive pattern, positive staining is shown on nuclear and/or perinuclear localization of 8-OHdG and 4-HNE, consistent with a previous report. ^{15 16} Quantitative analysis of the positive-stained cells showed a significant increase in the number of positive cells after 10 and 30 days in the JBDC group compared with the nontreatment group (Fig. 3B) . 

Changes in Antioxidant-Related Gene Expression in Dry Eye

The expressions of antioxidant and proinflammatory genes were assessed on days 1, 5, 10, 20 and 30. Superoxide dismutase (which reduces O₂· to H₂O₂), catalase, and glutathione peroxidase (which reduces H₂O₂ to H₂O) are the abundant antioxidants in mammalian cells and play a major role in the defense against oxidative stress. Cytologic Cu- and Zn-containing SOD (SOD 1) is the major isoform of SOD, which is relatively abundant in human cornea. ¹⁷ The relative expressions of all three genes, SOD1, Cat, and GPX 1, were at the same level as the nontreatment group up to day 10. On days 20 and 30, an approximately twofold increase with nontreatment was observed in all genes examined. Statistical significance appeared at day 30 compared with levels in the nontreatment group (Fig. 4) . 

Changes in Proinflammatory-Related Gene Expression

Increased levels of proinflammatory cytokine TNF-α and MMP-9 have been reported in tear fluid in patients with dry eye and experimentally induced murine dry eye. ^{18 19} In MMP-9, the relative expression level was the same level as in the nontreatment group on days 1 and 5. On day 10, an almost three-fold increase was observed in the nontreatment group, and this elevation was sustained up to day 30. Significantly increased expression was observed on day 20 compared with the nontreatment group (P < 0.005). In TNF-α, the relative expression was the same as in the nontreatment group up to day 20. On day 30, there was an approximately twofold increase in the expression level in the nontreatment group, although the difference was not significant (Fig. 5) . 

Measurements of ROS Production from Corneal Epithelia

Changes in ROS production in corneal epithelia were assessed on day 10. A significant increase in fluorescein intensity was observed in the JBDC corneas compared with that in nontreated corneas (Fig. 6) . 

Changes in Differentiation Capacity in Corneal Epithelia

To investigate the relationship between increased oxidative stress and the homeostasis of corneal epithelia, we assessed changes in differentiation capacity on day 10. The mature corneal epithelial cell layer is maintained by a balance between three important mechanisms: cell proliferation, central and vertical migration, and exfoliation into tear fluid. ²⁰ BrdU is incorporated into nuclear DNA as a substitute for thymidine during the S-phase of DNA replication. In the normal cornea, proliferation exclusively occurs in basal cells. In the proliferation assay, the appearance of BrdU-positive cells was localized in the basal cell layer in the nontreated and JBDC groups. In the JBDC group, fewer BrdU-positive cells appeared compared with the number in the nontreatment group, and a significant reduction of positive cells was defined by quantitative analysis (Figs. 7A 7D) . Once BrdU is incorporated into a cell, the label remains detectable in the nucleus for an extended period, even if the cells proceed to mitosis. In the migration assay, positive cells appeared not only in basal cell but also in wing and surface cells, which suggests that the upward movement of basal cells to the surface proceeded during the 4 days after BrdU incorporation. In the JBDC group, the number of BrdU-positive cells remaining in the basal layer was higher than that in the nontreatment group (Fig. 7B) . Quantitative analysis of the residual BrdU-positive cells in the basal layer showed that the declining ratio was significantly lower than in the nontreatment group (Fig. 7E) . Terminally differentiated cells move up toward the corneal surface and undergo apoptotic exfoliation. ²¹ TUNEL-positive cells were localized in the surface cells of the nontreated and JBDC corneal epithelia. In the JBDC group, dramatic increases in TUNEL-positive cells were observed compared with the number in the nontreatment group, and significant increases in positive cells were defined by quantitative analysis (Figs. 7C 7F) . 

Discussion

ROS not only has harmful functions as a byproduct of cellular metabolism but also has an important role in cell signaling and regulation, and the balance between the formation and detoxification of ROS is tightly controlled by a homeostatic mechanism. ²² Oxidative stress results from an imbalance between the formation of ROS by a metabolic reaction and antioxidant defense mechanisms. ²³ In this study, significant increases in the oxidative stress markers 8-OHdG, HNE, and MDA; the expression of antioxidant-related genes; and increased ROS production were observed in corneal epithelia during JBDC. These results suggest that ROS levels exceeded the antioxidant capacity in the corneal epithelia in our dry eye model. We previously reported in the same model that chronic SPK and the reduced barrier function of corneal epithelia is accompanied by thinning and abnormal arrangement in the cell layer, and poorly developed microvilli in surface cells. ¹³ In this study, we further found that the discordance of epithelial differentiation capacity—namely, decreased proliferation and upward migration and increased apoptotic cells—were induced in this model (Fig. 7) . The longer maintenance of high levels of 8-OHdG is explained by the exhaustion and/or disturbance of the DNA repair system. ^{24 25} HNE and MDA, endogenous products of lipid peroxidation, are known to react with deoxynucleoside to produce a variety of adducts and damage to DNA. ^{26 27 28} In response to DNA damage induced by oxidative stress, tumor suppressor protein p53 is stabilized and mediates the genes involved in growth control, DNA repair, and apoptosis. ^{24 29} p53 also regulates p21^WAF1/Cip1, an important regulator of the cell cycle that acts by binding and inhibiting several cyclin-dependent kinase/cyclin complexes. ^{25 30 31} In addition, HNE is known to induce cell cycle arrest, apoptosis, and diminished DNA synthesis independent of DNA damage. ^{32 33 34} We can hypothesize that, in dry eye conditions, chronic exposure to oxidative stress activates cell regulatory molecules that chronically distort the regenerative capacity of the corneal epithelial cell layer implicated in the appearance of the corneal surface disorder. 

To protect against the potentially damaging effects of ROS, cells possess several antioxidant enzymes such as superoxide dismutase, catalase, and glutathione peroxidase. ³⁵ In this study, significant increases in oxidative stress markers in the corneal epithelia were sustained during 10 to 30 days’ treatment (Fig. 3) . In contrast, a significant increase in antioxidant-associated gene expression was observed after 30 days’ treatment, although the increase rates of each gene were progressive during treatment (Fig. 4) . Many studies have shown that adaptive modulation of the antioxidant defense enzyme occurs in response to the increase in oxidative stress status from stress exposure. ^{36 37 38} Our observation can explain that, although JBDC chronically increased oxidative stress in corneal epithelial cells, the level was insufficient for the induction of acute adaptation. 

MMPs are enzymes involved in the breakdown of extracellular matrix and are suggested to play a role in inflammatory cell trafficking and inflammation through the degradation of type IV collagen. ³⁹ The concentration of MMP-9 and inflammatory cytokines in tear fluid was found to be significantly increased in eyes with severe corneal epithelial disease or sterile corneal ulcers. ^{18 40} We demonstrated that a dramatic and persistent increase in the gene expression of MMP-9 was observed during JBDC (Fig. 5) . This result indicates that MMP-9 plays a role in the pathogenesis of SPK and reduced corneal epithelial barrier function in our model. A recent study demonstrated that ROS stimulates MMP-9 secretion in the human fetal membrane. ⁴¹ It is also reported that the ROS scavenger N-acetyl-l-cysteine inhibits MMP-9 secretion in atherosclerotic lesions. ⁴² Along with increased MMP-9, the activation of mitogen-activated protein kinase (MAPK) signaling pathways has been shown on the ocular surface of experimental dry eye induced by pharmacologic blockade of tear secretion. ¹⁹ MAPKs can be activated by a wide variety of different stimuli, but in general, extracellular signal-regulated kinases are activated by mitogenic stimuli, whereas two other types of MAPKs, JNK and p38, are more responsive to environmental stress. ⁴³ In mammals, it has been reported that stress-activated kinase members c-Jun N-terminal kinase and p38 MAPK activated by ROS are generated by various stress conditions. ^{44 45 46 47} Taken together, our findings raise the possibility that oxidative stress is one of the initiators activating the signal transduction pathways implicated in the breakdown of the integrity and inflammatory response of the ocular surface in dry eye. 

The mechanism whereby oxidative stress was upregulated in the corneal epithelia in our dry eye model is not clear from this experiment. Tear fluid contains various antioxidants to protect the ocular surface from radical insults such as ascorbic acid, lactoferrin, uric acid, and cysteine. ^{48 49 50} In addition to suppressed blink frequency, decreased tear production, clearance, and stability were observed in our model (Nakamura S et al. IOVS 2006;47:E-Abstract 5589). ¹³ It can be speculated that, because of an imbalance in the tear film status, a synergistic effect between prolonged exposure to atmospheric oxygen and insufficient supplementation of antioxidant agents may have induced the overexpression of ROS production on the ocular surface. 

In conclusion, we have shown, for the first time to our knowledge, increases in oxidative stress markers, changes in antioxidant-related gene expression, and discordance in differentiation capacity in corneal epithelia in dry eye conditions. Our data suggest a strong relationship between the accumulation of oxidative stress and the etiology of corneal epithelial alterations in blink-suppressed dry eye. Further study is needed to elucidate the full mechanisms involved and the relative contribution of oxidative stress in patients with this type of dry eye. The management of oxidative stress may provide a new approach for the prevention and therapeutic treatment for this syndrome. 

 Submitted for publication August 30, 2006; revised November 10 and December 7, 2006; accepted January 29, 2007.

 Disclosure: S. Nakamura, Ophtecs Corporation (E); M. Shibuya, Ophtecs Corporation (E); H. Nakashima, Ophtecs Corporation (E); R. Hisamura, Ophtecs Corporation (E); N. Masuda, Ophtecs Corporation (E); T. Imagawa, None; M. Uehara, None; K. Tsubot a, None

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

 

Figure 1.

 

Photograph of dry eye treatment. Electric fan (1), track of jogging board (2). Experiments were performed in a dry, air-conditioned facility, with room temperature of 22 ± 3°C, relative humidity of 25% ± 5%, and constant air flow at 2 to 4 m/s. Each rat remained in place on the JB for 7.5 h/d. After 4 hours on the JB, the rats were returned to their cages for 30 minutes for food and water and again placed on the JB for 3.5 hours. For the remaining 16 hours, they were individually placed in cages with water and food ad libitum.

Figure 2.

 

Changes in the fluorescein score during dry eye treatment. (A) Typical fluorescein staining pattern during 30 days of treatment. (B) Average score in the dry eye group was significantly increased compared with nontreatment groups at all time points. Data represent the mean ± SE of results in 16 eyes. *P < 0.01 versus nontreatment group.

Figure 3.

 

Representative immunohistochemistry and the number of cells positive for oxidative stress markers in the corneal epithelia in dry eye. Representative images on day 10. (A) Quantitative analysis of positive 8-OHdG (left), MDA (center), and 4-HNE (right) cells. (B) Arrowhead: positive cells. Data represent the mean ± SE of 16 corneas. **P < 0.05 versus the nontreatment group.

Figure 4.

 

Changes in antioxidant-related gene expression in corneal epithelia. Data represent the relative expression rate compared with the nontreated group rats undergoing the standard protocol for the same days as the dry eye group. Data represent the mean ± SE of three measurements from 9 to 12 corneas. SOD1 (A), Cat (B), and GPX 1 (C). **P < 0.01, *P < 0.05 versus the nontreatment group.

Figure 5.

 

Changes in proinflammatory-related gene expression. Data represent the relative expression rate compared with the nontreated group rats undergoing the standard protocol for the same duration as the dry eye group. Data represent the mean ± SE of three measurements from 9 to 12 corneas. **P < 0.005 versus the nontreatment group.

Figure 6.

 

Changes in ROS production in corneal epithelia after 10 days of JBDC. ROS was determined by the DCF fluorescence level. Data represent the mean ± SE of 10 corneas. *P < 0.05 versus the nontreatment group.

Figure 7.

 

Changes in corneal epithelial differentiation. Changes in proliferation capacity. (A, D) Changes in upward movement of basal cells to the surface in corneal epithelia. (B, E) Reduction rate of BrdU-labeled cells in basal cells over 4 days were compared between the nontreatment and dry eye groups. Initial: 16 hours after BrdU injection. TUNEL positive cells. (C, F) Data represent the mean ± SE of 16 corneas. *P < 0.05 versus the nontreatment group. Scale bar: (A–C) 100 μm.