Protein Modifications by 4-Hydroxynonenal and 4-Hydroxyhexenal in Light-Exposed Rat Retina

Abstract

purpose. 4-Hydroxynonenal (4-HNE) and 4-hydroxyhexenal (4-HHE) are reactive aldehydes derived from the nonenzymatic oxidation of n-6 and n-3 polyunsaturated fatty acids, respectively. Increasing evidence suggests that protein modifications by reactive aldehydes are involved in various diseases. The present study was undertaken to test whether protein modifications by 4-HNE and 4-HHE increase in retinal tissues after exposure of rats to damaging levels of light.

methods. Albino rats were exposed to 1 or 5 klux white fluorescent light for 3 hours and, at various times thereafter, the levels and localizations of aldehyde-modified proteins in retinas were assessed by densitometric analysis of semiquantitative Western dot blots and by immunohistochemistry, using 4-HNE- and 4-HHE-specific antibodies. In some rats, the protective antioxidant phenyl-N-tert-butylnitrone (PBN) was injected (50 mg/kg) before exposure to light. To assess retinal damage, outer nuclear layer (ONL) thickness was measured on hematoxylin-eosin (H&E)–stained sections, and apoptosis was semiquantitatively analyzed by TUNEL staining.

results. By dot blot analysis, 4-HNE- and 4-HHE-modified proteins were significantly increased in retina (both by 1.7-fold) and RPE fraction (1.5- and 1.8-fold, respectively) after 5-klux exposure. In retina, increases in 4-HNE- and 4-HHE-modified proteins were more prominent at 3 hours than at 24 hours or 48 hours after exposure to light. In rod outer segments, only 4-HHE-modified proteins increased significantly (1.4-fold). Retinal thinning, TUNEL staining in ONL, 4-HNE-, and 4-HHE protein modifications were all found in the same retinal regions. PBN treatment inhibited the light-induced increase of 4-HNE and 4-HHE modified proteins in retina and RPE fractions.

conclusions. Exposure to intense light increases 4-HNE and 4-HHE protein modifications in the retina, suggesting that free radical initiated, nonenzymatic reactions are involved in this process. These modifications may be early events that precede photoreceptor cell apoptosis.

Excessive light may enhance the progression and severity of age-related macular degeneration (ARMD) and some forms of retinitis pigmentosa. ^{1 2} The hazards of light from the operating microscope used in ophthalmic practice can cause photic maculopathy. ³ Exposure to light causes photoreceptor cell damage, and the apoptotic pathway is the main course of light-induced cell death. ⁴ Previous studies have clarified that exposure of the retina to intense light causes lipid peroxidation of retinal tissues, ^{5 6} and lipid peroxidation is propagated by free radicals, especially lipid radicals. ⁷ Retinal damage caused by exposure to light can be reduced by various types of antioxidants such as ascorbate, ⁸ dimethylthiourea, ⁶ N-acetylcysteine (NAC), ⁹ thioredoxin (TRX), ^{10 11} and PBN. ^{12 13} Thus, oxidative stress is likely to be involved in the pathogenesis of light-induced retinal damage. 

Current thinking suggests that free radicals formed during oxidative stress can directly attack critical biomolecules including polyunsaturated fatty acids (PUFAs) and initiate free radical chain reactions that result in lipid peroxidation in cellular membranes. This chain reaction acts as an amplifier for the generation of lipid radical species, causing PUFA degradation into a variety of oxidized products, including aldehydes. Some of these aldehydes have been shown to be extremely reactive and can damage intra- and extracellular molecules located some distance from the initial site of free radical attack, because aldehydes are relatively long-lived compared with free radicals. ^{14 15}  

Damaging aldehydes include 4-hydroxyalkenals, such as 4-hydroxynonenal (4-HNE) and 4-hydroxyhexenal (4-HHE). These are α,β-unsaturated aldehydes that are end products of lipid peroxidation of PUFAs. It is reported that 4-HNE is formed from n-6 PUFAs such as linoleic acid and arachidonic acid, ¹⁶ and 4-HHE is formed from n-3 PUFAs such as docosahexaenoic acid, eicosapentaenoic acid, and linolenic acid, ¹⁷ by several nonenzymatic steps. These highly reactive aldehydes can react readily with histidine, cysteine, or lysine residues of proteins, leading to formation of stable Michael adducts with a hemiacetal structure. ¹⁸ These aldehydes exhibit a variety of cytopathologic effects, such as inhibition of enzyme activity; inhibition of protein, RNA, and DNA synthesis; cell cycle arrest; and apoptosis. ¹⁵ Thus, it has been recognized that reactive aldehydes exert cytotoxicity due to their facile reactivity with proteins. ¹⁹ The use of specific antibodies recognizing the hemiacetal structure of Michael adducts ²⁰ enables its detection in tissues. 

It has been reported that the formation of protein modifications by 4-HNE may be an indicator for assessing the effect of antioxidants against retinal light-induced damage. ^{9 21 22} However, a precise study regarding the protein modifications by 4-HNE and 4-HHE has not been reported in this model. In the present study, we assessed the formation of protein modifications by 4-HNE and 4-HHE in light-exposed rat retinas by using specific antibodies against 4-HNE and 4-HHE Michael adducts. Our results suggest a potential association between protein modification and retinal light-induced damage. 

Materials and Methods

Antibodies and Control Proteins

Mouse monoclonal anti-4-HNE- and anti-4-HHE-modified protein antibodies (anti-4-HNE and 4-HHE antibodies, respectively) were purchased from NOF Corporation (Tokyo, Japan). These antibodies recognize 4-HNE- and 4-HHE-histidine adducts, respectively (characterization of anti-4HHE antibody is indicated on the manufacturer’s Web site; http://www.jaica.com/biotech/). ²³ Ovalbumin modified with 4-HNE (4-HNE-OVA) and 4-HHE (4-HHE-OVA) was prepared according to the methods of Uchida et al., ²⁰ by reacting 5 mg/mL ovalbumin (Sigma-Aldrich, St. Louis, MO) with 10 mM 4-HNE (Alexis Biochemicals, San Diego, CA) or 10 mM 4-HHE (Alexis Biochemicals) at 37°C for 24 hours. 

Animal Care

All procedures were performed according to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and the University of Oklahoma Health Sciences Center (OUHSC) Guidelines for Animals in Research. All protocols were reviewed and approved by the Institutional Animal Care and Use Committees of the OUHSC and the Dean A. McGee Eye Institute. Sprague-Dawley (SDCD, Harlan Sera-Laboratory; Indianapolis, IN) rats were born and raised in our vivarium and kept under dim cyclic light (5 lux, 12 hours on–off, 7 AM–7 PM; central time). All animals used in this study were female and 5 to 7 weeks of age. 

Exposure to Light

All exposure to light began at 8 AM. The pupils were dilated with 1% cyclopentolate hydrochloride eye drops (Santen, Osaka, Japan) 30 minutes before exposure to light. Unanesthetized rats were exposed to 1 or 5 klux of diffuse, cool, white fluorescent light for 3 hours in wire-topped cages. After exposure to light, rats were kept in the dim cyclic light environment until euthanasia and enucleation at the indicated time-points. 

PBN Treatment

PBN, a potent free radical trapping agent, was synthesized and purified in our laboratory, to assure its purity. ¹³ In some rats, PBN (50 mg/kg of 20 mg/mL dissolved in saline) or saline alone was injected intraperitoneally 30 minutes before exposure to light, as described previously. ^{12 13}  

Preparation of Whole Retinal Samples for Western Dot Blot

Whole retinal samples were prepared as previously described, with slight modifications. ^{22 24} Briefly, after deep anesthesia was induced by the intramuscular injection of a mixture of ketamine (120 mg/kg) and xylazine (6 mg/kg), rats were perfused through the left cardiac ventricle with ice-cold PBS (pH 7.4) to wash out the blood, and the eyes were then enucleated. The cornea and the lens were removed, and the retina was separated from the eyecup. After cardiac perfusion with ice-cold PBS, the adhesion between the photoreceptor cell layers and retinal pigment epithelial (RPE) cell layers is weak, so they were easily separated. After the removal of the retina, the remaining eyecups were analyzed as an RPE cell fraction. Accordingly, this fraction also contained choroid and sclera. Samples were sonicated in radioimmunoprecipitation (RIPA) buffer (Upstate Biotechnology, Lake Placid, NY) containing a protease inhibitor cocktail (Upstate Biotechnology), 1 mM dithiothreitol (Bio-Rad, Hercules, CA), 2 mM diethylenetriaminepentaacetic acid (Sigma-Aldrich), and 100 μM butylated hydroxytoluene (Sigma-Aldrich) and centrifuged at 10,000g for 15 minutes at 4°C and the supernatants collected. 

Preparation of Rod Outer Segments for Western Dot Blot

Rod outer segment (ROS) fractions were prepared from retinas obtained as just described, by previously reported methods with slight modifications. ^{25 26} Briefly, the retina was homogenized in a Teflon glass homogenizer in 47% sucrose (1.174 g/mL) containing 10 mM Tris-HCl (pH 7.4), 70 mM NaCl, 2 mM MgCl₂, 0.1 mM EDTA, and proteinase inhibitors. The homogenate was placed at the bottom of a discontinuous sucrose gradient of 37% (1.137 g/mL) and 32% (1.118 g/mL) and centrifuged at 84,000g for 90 minutes at 4°C. The band containing ROS (1.118–1.137 interface) and the remainder of the retina (a band at the 1.137–1.174 interface pooled with the retinal pellet) were collected and diluted with buffer containing 10 mM Tris-HCl (pH 7.4), 100 mM NaCl, and 1 mM EDTA; centrifuged at 30,000g for 30 minutes at 4°C; and the pellets resuspended in RIPA buffer. The resuspended ROS pellet was designated as the ROS fraction, and the resuspended remainder of the retina was designated as residual fraction. 

Western Dot Blot for 4-HNE- and 4-HHE-Modified Proteins

After protein concentrations were determined by the DC protein assay kit (Bio-Rad), equal aliquots (5 μg) of retinal protein were applied to a 96-well dot blot apparatus (Bio-Rad) and then transferred to a polyvinylidene difluoride (PVDF) membrane (Millipore, Bedford, MA) by vacuum filtration. Equivalent sample loading was monitored by staining a membrane loaded in parallel with Coomassie brilliant blue R-250 (CBB). After blocking with 5% nonfat dry milk for 30 minutes at room temperature, the membrane was incubated with the anti-4-HNE (1:5000) or anti-4-HNE (1:5000) antibodies for 1 hour at room temperature, followed by incubation with peroxidase-linked anti-mouse IgG antibody (1:5000; GE Healthcare, Buckinghamshire, UK) for 1 hour at room temperature. Chemiluminescence was developed with extended-duration substrate (SuperSignal; Pierce, Rockford, IL) and detected with a digital imaging system (IS2000R; Eastman Kodak, New Haven, CT). Care was taken to ensure that the intensities of detected spots were within the linear range of the camera and that no pixels were saturated. Minimum detection limits of this Western dot blot method for 4-HNE- and 4-HHE-modifications are 1 ng in 4-HNE- and 4-HHE-OVA and 100 ng in retinal samples (data not shown). Intensities of dots were determined using Image J 1.32j software (available at http://rsb.info.nih.gov/ij/, developed by Wayne Rasband, National Institutes of Health, Bethesda, MD) and compared between the two groups by unpaired t-test or among three or five groups by one-way ANOVA followed by the Dunnett post hoc test. Either the right or left eye was analyzed in each rat. The coefficient of variance (CV) of this method was calculated as 12.7% ± 7.3% (mean ± SD) based on analysis of quadruplicate spots of six independent samples. 

Specificity Experiments for Antibodies

To test the specificities of anti-4-HNE and anti-4-HHE antibodies, 4-HNE-OVA (100 ng) and 4-HHE-OVA (100 ng) were transferred to polyvinylidene difluoride (PVDF) membrane by using the dot-blot vacuum manifold (described earlier) and detected with anti-4-HNE (1:5000) and anti-4-HHE (1:5000) antibodies and with the same antibodies (1:5000) preadsorbed at 37°C for 2 hours with at least a 400-fold excess of control proteins (4-HNE-OVA or 4-HHE-OVA for anti-4-HNE and anti-4-HHE, respectively). To test the effect of paraformaldehyde (PFA; used for tissue fixation) on protein modifications, OVA (100 ng) was preincubated with 4% PFA for 24 hours at room temperature, transferred to a PVDF membrane, and detected with anti-4-HNE- (1:5000) and anti-4-HHE- (1:5000) antibodies. 

Preparation of Retinal Tissue Sections

After CO₂-induced unconsciousness and cervical dislocation, both eyes were enucleated before (dim), just after (3 hours+0 hours), 3 hours (3 hours+3 hours), 24 hours (3 hours+24 hours) and 7 days (3 hours+7 days) after the 3 hours of exposure to intense light. Enucleated eyes were immersed in 4% PFA containing 20% isopropanol, 2% trichloroacetic acid, and 2% zinc chloride, for 24 hours at room temperature. After alcohol dehydration, the eyes were embedded in paraffin, and 4-μm-thick sagittal sections containing the whole retina including the optic disc were cut. 

Outer Nuclear Layer Thickness Measurement

The retinal sections obtained were stained with hematoxylin-eosin (H&E). For each section, digitized images of the entire retina were captured using a digital imaging system (Eclipse E800; Nikon, Tokyo, Japan) at 4× magnification with 1300 × 1030 pixels. To cover the entire retina, five images were obtained in each section. Outer nuclear layer (ONL) thickness was measured at 0.5, 1, 1.5, 2, 2.5, and 3.0 mm superior and inferior to the optic nerve head (ONH) and at the periphery 100 μm from the inferior and superior edge of the retina, using Image J 1.32j software. Both right and left eyes were analyzed in each rat and mean values were used for statistical analyses. The values among three groups (dim, 3 hours+24 hours, and 3 hours+7 days) were compared by 1-way ANOVA followed by the Dunnett post hoc test. 

TUNEL Staining

TUNEL was performed for 6 minutes with a kit (Apoptag Peroxidase In Situ Apoptosis Detection Kit; Chemicon, Temecula, CA) according to the kit manufacturer’s instruction, with 3′,3′-diaminobenzidine (Dako, Carpinteria, CA) used as a chromogen. For negative control experiments, the terminal deoxynucleotidyl transferase (TdT) enzyme was omitted. 

Immunohistochemistry for 4-HNE- and 4-HHE-Modified Proteins

The sections were deparaffinized, and endogenous peroxidase activity was inactivated with 3% H₂O₂ for 5 minutes. After the reaction was blocked with a serum-free blocking reagent (Dako) for 30 minutes at room temperature, the sections were incubated with the anti-4-HNE (1:200) or anti-4-HHE (1:50) antibody diluted with antibody diluent (Dako) for 2 hours at 37°C and then with the peroxidase-linked anti-mouse IgG polymer (EnVision+ System; Dako) for 1 hour at 37°C. Signals were developed with 3′,3′-diaminobenzidine (Dako) as the chromogen for 2 minutes and 8 minutes for 4-HNE and 4-HHE, respectively. For negative control experiments, sections were incubated with primary antibodies that had been preincubated at 37°C for 2 hours with at least a 600-fold excess of control proteins (4-HNE-OVA for anti-4-HNE and 4-HHE-OVA for anti-4-HHE). 

Semiquantification of Immunohistochemistry and TUNEL Staining

The staining intensity of immunohistochemistry for 4-HNE- and 4-HHE-modified proteins and TUNEL was semiquantitatively analyzed in the outer nuclear layer (ONL; Fig. 1 ). After staining, digitized color images covering the entire retina (five images, 1300 × 1030 pixels each) were obtained with a digital imaging system (Eclipse E800; Nikon; Fig. 1A ). The entire retina was divided into eight locations as 0 to 1 mm, 1 to 2 mm, and 2 to 3 mm inferior (Inf-1, -2, and -3, respectively) or superior (Sup-1, -2, and -3, respectively) from the ONH, and 0 to 0.5 mm from the inferior (Inf-peri) or superior (Sup-peri) peripheral edge of the retina (Fig. 1A) . To adjust the background intensity among the five images from one retinal slice, we opened the images in grayscale (Photoshop, ver. 7; Adobe Systems, San Jose, CA), and an imported small black square (reference of highest intensity) was superimposed on the images. Black and white signals were inverted, and the staining intensities of the vitreous cavity (for lowest reference) and the imported reference for highest intensity (now white box) were set at levels of 0 and 255, respectively (Fig 1B) . After these conversions, the intensity of immunostaining for 4-HNE (4-HNE index)- and 4-HHE (4-HHE index)-modified proteins and TUNEL (TUNEL index) were calculated as  

\[\mathrm{Staining\ index\ (\%)}{=}{\Sigma}_{X}_{{>}\mathrm{threshold}}{[}(X{-}\mathrm{threshold}){\times}\mathrm{area}(\mathrm{pixels}){]}/\mathrm{total\ area}(\mathrm{pixels}){\times}100,\]

where X is the staining intensity indicated by a number between 0 and 255. 

The threshold value was determined at the Inf-peri region, based on our observations that the changes of immunostaining and TUNEL before and after exposure to light were minimal in this area. Because the staining intensity in the ONL showed a Gaussian-like distribution (Fig. 1C) , the thresholds were determined as intensity values at the upper 2.5% (=1.96 SD) for the 4-HNE and 4-HHE indices and 0.5% (=2.576 SD) for the TUNEL index. After determination of thresholds, the area stained above threshold (percentage) was determined in the other seven locations, and a representative result is presented in Figure 1D . Either the right or left eye was analyzed in each rat, and the values between the two groups (dim and 3 hours+24 hours) were compared by unpaired t-test. 

Results

Specificities of Anti-4-HNE- and 4-HHE Antibodies

Initially, we tested the specificities of anti-4-HNE and 4-HHE antibodies by Western dot-blot analysis (Fig. 2) . The anti-4HNE antibody recognized 4-HNE-OVA, but not 4-HHE-OVA, and the anti-4-HHE antibody recognized 4-HHE-OVA, but not 4-HNE-OVA (Fig. 2A , lanes 1 and 2). Anti-4-HNE immunoreactivity was completely blocked by preincubation with 4-HNE-OVA, but not by 4-HHE-OVA (Fig. 2B , top row), and the anti-4-HHE immunoreactivity was completely blocked by preincubation with 4-HHE-OVA, but not by 4-HNE-OVA (Fig. 2B , bottom row). Thus, both antibodies specifically recognize the appropriate aldehyde-modified proteins with minimal cross-reactivity. To test whether the aldehyde-based fixative (4% PFA) used in subsequent immunohistochemical experiments might cause artifactual aldehyde modification of proteins, we assessed 4-HNE and 4-HHE immunoreactivities of OVA treated with 4% PFA. As indicated in Figure 2A(lane 4), no detectable immunoreactivity was observed with either antibody, demonstrating that the aldehyde-based fixation does not result in 4-HNE and 4-HHE protein modifications. CBB staining is shown as a control for protein loading. 

Effect of Exposure to Light on Protein Modifications by 4-HNE and 4-HHE in Rat Retina

We next tested whether short-term exposure to intense light would result in increased levels of protein modifications by 4-HNE and 4-HHE in whole retinal samples. At 3 hours+3 hours, levels of 4-HNE- and 4-HHE-modified proteins were significantly increased in the retina by 5-klux light (both 1.7-fold) and in RPE fractions by exposure to 1-klux (1.5- and 1.8-fold, respectively) or 5-klux (1.5- and 1.8-fold, respectively) light compared with the nonexposed (dim) control (P < 0.01 for all comparisons; Fig. 3 ). Exposure to 5-klux light for 3 hours increased the levels of 4-HNE- and 4-HHE-modified proteins in retina and RPE fractions immediately after the exposure to light and they remained elevated for up to 48 hours (Fig. 4) . The increases of 4-HNE-modified protein levels reached statistical significance at 3 hours+0 hours (1.9-fold, P < 0.01), 3 hours+3 hours (1.9-fold, P < 0.01), and 3 hours+24 hours (1.7-fold, P < 0.05) in the retina, and at 3 hours+0 hours (1.6-fold, P < 0.05), and 3 hours+3 hours (1.6-fold, P < 0.05) in the RPE fraction, compared with the dim control (Fig. 4B) . The increases of 4-HHE-modified protein levels reached statistical significance at 3 hours+0 hours (1.5-fold, P < 0.05) and 3 hours+3 hours (1.6-fold, P < 0.01) in retina compared with the dim control (Fig. 4B) . 

Light-Induced Protein Modifications by 4-HNE and 4-HHE in ROS

The light-induced increase in protein modifications by 4-HNE and 4-HHE in ROS and residual retinal fractions was determined by Western dot-blot analysis. In residual retinal samples, protein modifications by both 4-HNE (1.6-fold, P < 0.05) and 4-HHE (1.5-fold, P < 0.01) were significantly increased after exposure to light compared with dim control (Figs. 5A , lanes 3 and 4; 5B ). In ROS fractions, protein modification by 4-HHE (1.4-fold, P < 0.05) increased significantly after exposure to light, but 4-HNE (1.2-fold) did not, when compared with the dim control (Figs. 5A , lanes 1 and 2; 5B ). 

Effect of PBN on Light-Induced Protein Modifications by 4-HNE and 4-HHE

The effect of the antioxidant PBN on light-induced protein modifications by 4-HNE and 4-HHE was tested. At 24 hours after exposure to light, pretreatment with PBN strongly inhibited the expression of TUNEL-positive photoreceptor cells in the ONL (Fig. 6C) , consistent with previous reports. ^{12 13} Western dot blots indicated that the levels of 4-HNE- and 4-HHE-modified proteins in retina (0.7-fold, P < 0.05; and 0.7-fold, P < 0.05, respectively) and RPE fractions (0.4-fold, P < 0.01; and 0.4-fold, P < 0.01, respectively) were significantly lower in PBN-treated than in saline-treated rats at 3 hours+3 hours (Figs. 6B 6C) . 

H&E, TUNEL, and Immunohistochemical Staining for 4-HNE- and 4HHE-Protein Modifications

We assessed retinal damage and localization of proteins modified by 4-HNE and 4-HHE, using morphologic methods. By H&E staining, we observed a remarkable reduction in ONL thickness at 3 hours+7 days (Fig. 7A) . Most nuclei in the ONL showed TUNEL-positive staining at 3 hours+24 hours, which was not observable at 3 hours+0 hours or 3 hours+3 hours (Fig. 7B) . As a control, no signal was observed on the section when the TdT enzyme was omitted (Fig. 7B , TdT(−)). Labeling of sections with antibody against 4-HNE-modified proteins showed only weak immunoreactivity throughout the retina before exposure to light (Fig. 7C , Dim). In contrast, the intensity of 4-HNE immunoreactivity increased at 3 hours+0 hours (Fig. 7C)in the ganglion cell layer (GCL), inner plexiform layer (IPL), outer plexiform layer (OPL), rod inner segments (RIS), and RPE layer (Fig. 7C , top, inset). At 3 hours+3 hours (Fig. 7C)and 3 hours+24 hours (Fig. 7C) , staining intensities in the OPL and RIS decreased compared with that at 3 hours+0 hours, and intense nuclear staining appeared in the ONL. Immunoreactivity of 4-HNE-modified proteins was almost completely blocked by preincubation of anti-4-HNE antibody with 4-HNE-OVA (Fig. 7C , 4-HNE-OVA(+)). When retinal sections were labeled with antibody against 4-HHE-modified proteins, only weak staining was observed throughout the retina before exposure to light (Fig. 7D , Dim). However, at 3 hours+0 hours (Fig. 7D) , nuclear and/or perinuclear staining was observable in the ONL (bottom, inset) and RPE (top, inset). At 3 hours+3 hours (Fig. 7D)and 3 hours+24 hours (Fig. 7D) , nuclear staining was apparent in more cells in the ONL. This staining was almost completely blocked by incubation of anti-4-HHE antibody with 4-HHE-OVA (Fig 7D , 4-HHE-OVA(+)). It is important to note that increased immunoreactivity to both 4-HNE and 4-HHE was observable at time points preceding detectable TUNEL labeling. 

Quantification of ONL Thickness, TUNEL, 4-HNE, and 4-HHE Staining

We further analyzed the H&E, TUNEL, and immunohistochemical staining by semiquantitative methods. ONL thickness measured on H&E stained sections was significantly decreased at 1.5 mm superior to the ONH at 3 hours+24 hours, and 0.5 to 2.0 mm superior and inferior to the ONH at 3 hours+7 days compared with the dim control (Fig. 8A) . At 3 hours+24 hours, TUNEL indices (calculated as described in the Materials and Methods section) were significantly higher at Inf-1 (15.9%) and -2 (8.9%) and Sup-1 (20.0%) and -2 (16.0%) compared with the dim control (2.1%, 2.4%, 1.1%, and 1.1%, respectively; Fig. 8B ). These results indicate that retinal damage caused by exposure to light was severe across the retina near the ONH. Finally, we determined staining intensity indices (as described in the Materials and Methods section) for 4-HNE- and 4-HHE-modified proteins in the ONL, since staining for both was prominent in the ONL (Figs. 7C 7D) . At 3 hours+24 hours, the 4-HNE indices were significantly increased at Inf-1 (43.3%) and -2 (43.4%) and Sup-1 (45.4%) and -2 (53.5%) compared with the dim control (5.4%, 9.8%, 2.9%, and 5.7%, respectively; Fig. 8C ). In addition, at the same time point, the 4-HHE indices were significantly higher at Inf-1 (27.5%) and -2 (21.6%) and Sup-1 (26.5%) and -2 (28.2%) compared to dim control (6.1%, 3.9%, 7.3%, and 4.6%, respectively; Fig. 8D ). Thus, the increase of protein modifications by exposure to light was prominent across the retina near to ONH. More important, the retinal regions with the highest 4-HNE and 4-HHE indices were the same regions with elevated TUNEL staining and decreased ONL thickness. 

Discussion

Protein modification by reactive aldehydes is thought to occur in many tissue proteins under conditions of oxidative stress. ²⁰ As the retina may be particularly sensitive to oxidative stress, ^{5 7} we used a semiquantitative Western dot-blot technique ²⁷ to measure the total levels of reactive aldehyde-modified proteins in the retinal tissues. Protein modifications by 4-HNE and 4-HHE in the retina were significantly increased after exposure to 5-klux light, but the increases were not significant after the exposure to 1-klux light (Fig. 3) . Significant increases in protein modifications by both aldehydes were observed immediately and 3 hours after exposure to light in both retina and RPE fractions (Fig. 4) . These results indicate that exposure to intense light increases protein modifications in retinal tissues by reactive lipid aldehydes. Although the antibodies we used are reported to be highly specific to histidine Michael adducts, ²³ we can speculate that other protein modifications such as cysteine and lysine Michael adducts also occur in light-exposed retina, because the formation of all these protein modifications are thought to be generated by nonenzymatic oxidation pathways. ^{16 17} The animals used in this study were all female and relatively young (5–7 weeks of age). The effect of age and gender on protein modifications by reactive aldehydes is a subject of interest but remains to be tested. 

It is hypothesized that aldehydes have the potential to cause damage, not only at the initial site of oxidative insult, but also at distant locations because of their relatively long life compared with reactive free radicals. ^{14 15} Sustained increases in both modifications in the retina were observed at least up to 48 hours, although these levels did not remain statistically significant over the entire time range (Fig. 4) . Increased staining intensities for 4-HNE were observed in the RIS and OPL immediately after exposure to light, whereas nuclear staining in the ONL became predominant at 3 hours after exposure and later (Fig. 7C) . There is abundant evidence suggesting that extensive aldehyde modification of proteins inhibits the proteasomal system. ²⁸ In fact, recent evidence suggests that the chymotrypsin-like activity of the 20S proteasome is inhibited by 4-HNE modification. ²⁹ Given the sustained increase in aldehyde-modified proteins in light-exposed retinas, we can speculate that the inhibition of degradative machinery may be responsible for the accumulation of these proteins, but this remains to be tested. Accordingly, our results support the previously proposed theory regarding cellular damage caused by reactive aldehydes generated by nonenzymatic oxidation of n-3 and n-6 PUFAs. ^{14 15}  

In ROS, protein modifications by 4-HHE increased significantly after exposure to light, but 4-HNE did not, suggesting that ROS proteins are modified by different oxidized lipid species than proteins in other retinal locations. In ROS, the content of n-3 PUFAs is much higher than n-6 PUFAs (e.g., four times higher in albino rats ³⁰ and seven times higher in pigmented mice ³¹ ). Therefore, the increase in 4-HHE-modified proteins in ROS could result from the increased levels of n-3 PUFAs, the precursors of 4-HHE, in this photoreceptor organelle. It should be noted that immunohistochemistry did not detect the increase in 4-HHE-modified proteins in ROS throughout the time course analyzed (Fig. 7D) . We speculate that the epitopes for 4-HNE and 4-HHE antibodies may be masked or lost during sample fixation and/or subsequent preparation such as dehydration and deparaffination. 

PBN treatment before exposure to light effectively inhibited modification of proteins by 4-HNE- and 4-HHE in both retina and RPE fractions (Figs. 6B 6C) . PBN is a well-known synthetic free-radical–trapping agent that has been shown to protect rat retinas from light-induced damage, ^{12 13} suggesting that aldehyde-producing free radical reactions were inhibited by the antioxidant activity of PBN. It has recently been reported that protein modifications by 4-HNE are enhanced by photo-oxidative stress and inhibited by the antioxidants and/or antioxidative enzymes such as a precursor of glutathione, ⁹ an inducer of TRX, ²² and the adenovirus-mediated expression of catalase. ²¹ Thus, the relevance of 4-HNE-protein modifications as an indicator of oxidative stress in the retina has been proposed, and our results further support the usefulness of the detection of 4-HNE-protein modifications as an oxidative stress marker. In addition, detection of the 4-HHE-protein modification may be an equally useful oxidative stress marker for assessing the effect of antioxidants in retinal oxidative stress. 

At 24 hours after exposure to light, we observed a dramatic increase in TUNEL-positive cells in the ONL, which was followed at 7 days by a remarkable loss of ONL nuclei (Figs. 7A 7B) . Immunohistochemical staining for aldehyde-modified proteins was observed in the ONL immediately and 3 hours after exposure to light (Figs. 7C 7D) . At 24 hours after exposure, the 4-HNE and 4-HHE indices were significantly increased on both sides of the retina near the ONH (Figs. 8C 8D) , corresponding well with the locations that showed increased TUNEL staining at 24 hours (Fig. 8C)and severely decreased ONL thickness at 7 days after exposure (Fig. 8A) . Thus, the protein modifications by these aldehydes are an early event that precedes apoptosis and subsequent photoreceptor cell loss. The apoptosis-inducing effects of 4-HNE and 4-HHE have been well-studied ^{15 19} and have recently been reported in cultured RPE cells. ³² In our study, inhibition of 4-HNE- and 4-HHE-protein modification by PBN correlated with the reduction in TUNEL staining in ONL (Fig. 6A) . Collectively, our results suggest the possible association of protein modifications by these aldehydes and the initiation of photoreceptor cell apoptosis after exposure to light. 

Several recent studies suggest the relationship between abnormal protein oxidation–modification and retinal disease, including detection of cross-linked species of tissue metalloproteinase inhibitor 3 and vitronectin and docosahexaenoic acid-derived carboxyethyl pyrrole protein adducts in drusen from patients with ARMD ^{33 34} ; increased protein nitration associated with light-induced photoreceptor cell apoptosis in rats ³⁵ ; and increased protein carbonylation associated with photocytotoxicity of lipofuscin-loaded human RPE cells ²⁷ ; and with light-induced photoreceptor cell apoptosis in mice. ¹¹ In addition, increased 4-HNE-protein modifications in cone inner segments have been reported in a transgenic pig model of retinitis pigmentosa ³⁶ as well as in H₂O₂- and 4-HNE-treated RPE cells in vitro. ³² Thus, the protein modifications by biologically active molecules may be a critical process in oxidative stress-induced retinal cell apoptosis. 

In summary, we showed that exposure to intense light increased protein modifications by 4-HNE and 4-HHE in retinal tissues, and radical reactions were thought to be involved in the initiation of these modifications. The present results provide further evidence that protein modifications by 4-HNE and 4-HHE are early events that precede photoreceptor cell apoptosis induced by exposure to intense light. 

 Supported by National Eye Institute Grants EY04149, EY00871, and EY12190; the National Center for Research Resources Grant RR17703; Oklahoma Center for the Advancement of Science and Technology-Oklahoma Applied Research Support (OCAST-OARS); Research to Prevent Blindness, Inc.; and the Foundation Fighting Blindness. Masaki Tanito is a recipient of a Research Fellowship from the Japanese Society for the Promotion of Science (JSPS) for Young Scientists.

 Submitted for publication May 27, 2005; revised July 5, 2005; accepted August 24, 2005.

 Disclosure: M. Tanito, None; M.H. Elliott, None; Y. Kotake, None; R.E. Anderson, 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.

 

Methods for semiquantification of immunohistochemistry and TUNEL staining. (A) Composite of five retinal images of immunohistochemistry for 4-HNE-protein modifications. The entire retina was divided into eight locations (Inf-1, -2, -3, and -peri; Sup-1, -2, -3, and -peri) for calculation of staining indices. (B) Representative converted image used for staining index measurement. (C) Histogram of staining intensity in ONL at Inf-peri. In this example, the threshold for staining intensity was calculated as 83. (D) Histogram of ONL staining intensity at Inf-3. In this example, the area above threshold (staining index) was calculated as 34%. X is the staining intensity indicated by a number between 0 and 255.

Figure 2.

 

Specificity of antibodies against aldehyde-modified proteins indicated by Western dot-blot analysis. (A) 4-HNE-OVA, 4-HHE-OVA, and OVA incubated with PBS or 4% PFA were detected with anti-4-HNE or 4-HHE antibody. CBB staining indicates equal sample loading (bottom row). (B) 4-HNE-OVA and 4-HHE-OVA were detected with anti-4HNE and 4-HHE antibodies, respectively, pretreated with or without 4-HNE-OVA or 4-HHE-OVA.

Figure 3.

 

Western dot blot for 4-HNE and 4-HHE protein modifications in retinal samples exposed to 1- or 5-klux light. The eyes were enucleated before (dim) and at 3 hours+3 hours. (A) Representative Western dot blots of retina and RPE fractions (blots of two eyes from two rats in each group). (B) Densitometric analysis of protein dots. The mean ± SD densities are shown (n = 6 in each group). **P < 0.01 compared with dim control by one-way ANOVA followed by the Dunnett post hoc test.

Figure 4.

 

Western dot blot for 4-HNE- and 4-HHE-protein modifications in retinal samples exposed to 5-klux light. The eyes were enucleated before (dim) and at 3 hours+0 hours, 3 hours+3 hours, 3 hours+24 hours, and 3 hours+48 hours. (A) Representative Western dot blots of neural retina and RPE fractions (blots from two eyes of two rats in each group). (B) Densitometric analysis of dots. The mean ± SD densities are shown (n = 6 in each group). *P < 0.05 and **P < 0.01, respectively, compared with dim control by one-way ANOVA followed by the Dunnett post hoc test.

Figure 5.

 

Western dot blot for 4-HNE and 4-HHE protein modifications in ROS and residual retinal fractions. The eyes were enucleated before (dim) and at 3 hours+3 hours of exposure to 5-klux light. (A) Representative Western dot blots of ROS and residual retinal fractions. (B) Densitometric analysis of dots. The mean ± SD densities are shown (n = 5 in each group). *P < 0.05 and **P < 0.01, respectively, between dim and 3 hours+3 hours by unpaired t-test. NS, not significant.

Figure 6.

 

TUNEL staining and Western dot blots for 4-HNE- and 4-HHE-protein modifications in whole retinal samples from PBN- or saline-treated animals. (A) Representative TUNEL staining in saline- and PBN-treated animals. The eyes were enucleated at 3 hours+3 hours of exposure to 5-klux light. (B) Representative Western dot blots of retina and RPE proteins from saline- and PBN-treated animals (blots of two eyes from two rats in each group). The eyes were enucleated before (dim) and at 3 hours+3 hours of exposure to 5-klux light. (C) Densitometric analysis of dots. The mean ± SD densities are shown (n = 6 in each group). *P < 0.05 and **P < 0.01, respectively, between saline- and PBN-treated rats, by unpaired t-test.

Figure 7.

 

H&E, TUNEL, and immunohistochemical staining for 4-HNE and 4-HHE protein modifications. Eyes were enucleated before (dim) and at 3 hours+0 hours, 3 hours+3 hours, 3 hours+24 hours, and 3 hours+7 days of exposure to 5-klux light. Representative images at 1 mm superior to the ONH are shown (representative of five rats analyzed in each group). (A) H&E staining. Severe loss of ONL was observable at 3 hours+7 days. (B) TUNEL. Nuclear staining was observed in the ONL at 3 hours+24 hours. TdT(−), no TdT enzyme on the same section of 3 hours+24 hours. (C) Immunohistochemistry for 4-HNE-modified proteins. Nuclear staining was observed in the RPE layer at 3 hours+0 hours (top inset) and later, and in ONL at 3 hours+3 hours (bottom inset) and later. The 4-HNE-OVA(+), blocking experiment was performed on a 3 hours+24 hour section. (D) Immunohistochemistry for 4-HHE-modified proteins. Nuclear–perinuclear staining was observed in the RPE layer (top inset) and in the ONL (bottom inset) at 3 hours+0 hours. Nuclear staining in the ONL became more dramatic at 3 hours+3 hours and 3 hours+24 hours. The 4-HHE-OVA(+), blocking experiment was done on a 3 hours+24 hours section.

Figure 8.

 

Quantification of ONL thickness, TUNEL, 4-HNE, and 4-HHE indices. (A) ONL thickness before (dim), and at 3 hours+24 hours and 3 hours+7 days of exposure to 5-klux light. The mean ± SD thickness is shown (n = 5; in each group). *P < 0.05 and **P < 0.01, respectively, when comparing 3 hours+7 days with the dim control; ##P < 0.01 compared between 3 hours+24 hours and dim by one-way ANOVA followed by the Dunnett post hoc test. TUNEL (B), 4-HNE (C) and 4-HHE (D) indices before (dim) and at 3 hours+24 hours of exposure to 5-klux light. The mean ± SD results are shown (n = 5 in each group) *P < 0.05 and **P < 0.01, when comparing 3 hours+24 hours and dim control, by unpaired t-test.

The authors are grateful to Mark Dittmar (Dean A. McGee Eye Institute) for maintaining the animal colonies used in this study and to Louisa J. Williams and Linda S. Boone (Dean A. McGee Eye Institute) for excellent retinal section preparation.