Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Free Radical Tissue Damages in the Anterior Segment of the Eye in Experimental Autoimmune Uveitis Sei-ichi Ishimoto, Guey-Shuang Wu, Seiji Hayashi, Jie Zhang, and Narsing A. Rao Purpose. To investigate increased free radical activity and the accumulation and localization of lipid peroxidation in the anterior segment of the eye with uveitis. Methods. Experimental autoimmune uveitis (EAU) was induced in rats using human S-antigen peptide. Conjugated dienes (CD) and keto-dienes (KD) were then extracted from the cornea, iris-ciliary body and lens of the EAU eyes. The quantity of CD and KD were determined by measuring ultraviolet absorption and estimating by means of a molar extinction coefficient. Frozen sections of EAU eyes were reacted with 3-hydroxy-2-naphthoic acid hydrazide (NAH), and NAH-carbonyl compounds were detected using a confocal laser scanning microscope. Statistical comparisons of CD and KD products between the EAU groups and controls were performed using the Student's t-test. Results. Compared to controls, CD and KD were significantly increased in the cornea and iris-ciliary body of EAU eyes. Lenses of EAU eyes showed a tendency to elevated levels of CD and KD. In EAU, primarily the anterior border layer and the posterior epithelium of the iris—and, to a lesser extent, the trabecular meshwork and corneal endothelium—revealed positive fluorescence staining for peroxidized carbonyl products. No staining was observed on the ciliary epithelium. Conclusions. Free radicals and lipid peroxidation products are generated in the anterior segment of the eye in EAU. Because the individual tissues in the anterior segment are composed of various levels of fatty acids and different concentrations of antioxidants, the extent of tissue damage from lipid peroxidation may represent a balance between the fatty acid composition and the antioxidant distribution in each of the tissues. Invest Ophthalmol Vis Sci. 1996; 37:630-636. V-Jxygen-free radicals, primarily the hydroxyl radical (•OH), hydrogen peroxide (H 2 O 2 ), and superoxide anion (O 2 - ~), are known to be toxic to tissue components. 1 In an inflammatory process, phagocytic cells, such as polymorphonuclear cells (PMNs) and macrophages, produce these oxygen-free radicals, which, in turn, induce a series of cytotoxic effects responsible for tissue damage. 2 In human uveitis, inflammation of ocular tissues often results in pathologic conditions such as retinal degeneration, secondary glaucoma, complicated cataract, iris atrophy, and corneal degeneration. It has From the Doheny Eye Institute, University of Southern California, School of Medicine, Los Angeles, California. Supported by National Institute of Health grant EY10212 Submitted for publication August 30, 1995; revised November 29, 1995; accepted December 1, 1995. Proprietary interest category: N. Reprint requests: Narsing A. Rao, Doheny Eye Institute, 1355 San Pablo Street, Los Angeles, CA 90033. 630 been demonstrated that lipid peroxidation of the retina subsequently induced retinal degeneration in experimental autoimmune uveitis (EAU).3 There have also been reports of the free radical-induced tissue damage in the anterior segment of the eye in a variety of pathogenic ocular conditions. Spector and Garner4 and Bhuyan and Bhuyan5 found that endogenous hydrogen peroxide and lipid peroxidation products were significantly increased in aqueous humor in senile cataract. Babizhaye and Bunin6 found that trabecular meshwork tissues in primary open angle glaucoma contained a high proportion of lipid peroxidation products. Artola et al7 demonstrated that injection of exogenous hydrogen peroxide into the anterior chamber of the rabbit induced lipid peroxidation in the iris epithelial cell membranes. These reports suggest that oxygen-free radical-mediated peroxidation may play a causative role in subsequent pathologic conditions of the anterior segment of the Investigative Ophthalmology & Visual Science, March 1996, Vol. 37, No. 4 Copyright © Association for Research in Vision and Ophthalmology Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933415/ on 05/04/2017 Free Radical Damage in Uveitis 631 eye in uveitis. Oxygen radical-derived tissue damage in the anterior segment has not been studied in uveitis. In the current study, we investigated the generation of lipid peroxidation products and their localization in the anterior segment, including the cornea, the trabecular meshwork, the iris-ciliary body, and the lens, in EAU. solved in 1 ml ethanol (anhydrous "photrex" grade; J.T. Baker Chemical, Phillisburg, NJ) for ultraviolet measurement. Absorptions for CD and KD were measured at 233 nm and 280 nm (Shimadzu [Kyoto, Japan] spectrophotometer, UV-160), respectively. The quantity of CD and KD products was estimated using molar extinction coefficients of 25,200 and 20,000, respectively. MATERIALS AND METHODS Detection of Peroxidized Carbonyl Products Peroxidized carbonyl products were detected histochemically by the method previously described.10 Briefly, eyes were embedded in optimum cutting temperature compound (Miles Laboratory, Naperville, IL) and immediately snap frozen in liquid nitrogen. Ten-micrometer frozen sections were prepared, fixed in 5% trichloracetic acid, and reacted with 3-hydroxy2-naphthoic acid hydrazide (NAH) catalyzed by a trace of p-toluenesulfionic acid. The fluorescent products also were coupled with Fast Blue B (FBB) for color visualization. The fluorescent NAH-carbonyl compounds were detected by Zeiss confocal laser scanning microscope (Carl Zeiss, Thornwood, NY) using an excitation at 488 nm and an emission at 520 nm. Induction of Experimental Autoimmune Uveitis Human S-antigen peptide, DTNLASSTIIKEGIDKLG, was synthesized on 4-hydroxymethylphenoxymethylresin using an automated peptide synthesizer (model 430A; Applied Biosystems, Foster City, CA) and was desalted on a Sephadex G-10 column (Sigma, St. Louis, MO).8 The purity was assessed by a reversephase high-pressure liquid chromatography (Bio-Rad RP304 column; Bio-Rad, Richmond, CA). Twenty-eight female Lewis rats, each weighing 150 to 175 g, were obtained from Charles River Laboratory (Wilmington, MA). Fourteen rats were immunized with 100 fig human S-antigen peptide in an 1:1 emulsion with complete Freund's adjuvant containing 4 mg/ml Mycobacterium tuberculosis strain H37RA (Difco Statistical Analysis Laboratory, Detroit, MI) in a total of 0.2 ml into one footpad. At the time of immunization, animals also received an intravenous injection in the tail vein with 1 fig pertussis toxin (List Biological Laboratory, Campbell, CA) in 0.3 ml sterile saline. Another 14 nonimmunized female rats were used for control. All animals were treated in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Light Microscopic Study At the peak of inflammation, between 14 and 16 days after immunization, the eyes were enucleated under anesthesia and fixed in formalin for paraffin embedding. Seven-micrometer sections were stained with hematoxylin and eosin. Measurement of Conjugated Dienes and KetoDienes The cornea, iris-ciliary body, and lens were dissected from the eyes. Lipids were extracted by the method previously reported.9 Briefly, each determination used either the cornea or the lens from one eye or the irisciliary body complex from two eyes. The cornea and iris-ciliary body tissues were homogenized and extracted in 1 ml chloroformrmethanol (2:1), and the pooled extracts were washed with 0.2 ml water. Lens tissues underwent the same procedure with 2 ml solvent and washing with 0.4 ml water. Solvents were evaporated under nitrogen, and the residue was dis- Statistical comparisons of CD and KD products between the EAU groups and controls were performed using the Student's Mest. The null hypothesis was rejected for P values < 0.05. RESULTS All 14 Lewis rats sensitized with human S-antigen peptide developed uveitis, characterized by conjunctival hyperemia, corneal edema, and anterior chamber cells and flare. Some of these animals developed hypopyon and hyphema. Clinically, the ocular inflammation reached its peak between 14 and 16 days after immunization. Controls showed none of these clinical signs. Light Microscopic Findings In the eyes of EAU group, massive inflammatory cells, mostly PMNs and mononuclear cells, have infiltrated the iris, ciliary body, and trabecular meshwork (Fig. 1). A large number of PMNs also were evident in the anterior and posterior chamber, and proteinaceous material and fibrin debris were present in the anterior chamber. Some PMNs were adherent to the corneal endothelium and surface of the lens capsule (Fig. 2). The retina and choroid also were infiltrated with PMNs and mononuclear cells. Conjugated Dienes and Keto-Dienes Products Results for CD and KD, expressed as mean ± SD in nmoles, are shown in Table 1. A statistically significant Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933415/ on 05/04/2017 632 Investigative Ophthalmology & Visual Science, March 1996, Vol. 37, No. 4 FIGURE l. Anterior segments of the experimental autoimmune uveitis eye. Massive inflammatory cells, mostly polymorphonuclear cells (PMNs) and mononuclear cells, infiltrate the iris and ciliary body. Note the disorganized trabecular meshwork and die presence of PMNs and mononuclear cells in die trabecular meshwork. AC = anterior chamber; TM = trabecular meshwork. Hematoxylin and eosin, X100. increase of CD was found in the experimental eyes compared to control eyes in the cornea (P < 0.001), the iris-ciliary body (P < 0.001), and the lens (P = 0.005). A significant increase of KD was found in experimental animals in the cornea (P = 0.004) and the iris-ciliary body (P = 0.01). Compared to the control group, KD products within the lens tended to increase in the experimental group, but the increase was not statistically significant. tion products and by the detection of peroxidized carbonyi products. For the confirmation and quantification of in vivo lipid peroxidation, we used two parameters in this study, CD and KD, because any one parameter may not be sufficient to provide an accurate measure of complex lipid peroxidation processes.11 Conjugated dienes represent the primary formation of monomeric fatty acid hydroperoxides, as well as any secondary dimeric and polymeric hydroperoxides that retain the diene conjugation in the molecule. Ketodienes reflect the fraction of hydroperoxide rearranged to ketodiene structure. These CD and KD are parameters as consequential products of free radical-initiated lipid peroxidation, indicating oxygen-dependent degradative processes of polyunsaturated fatty acids (PUFAs). In the current study, the cornea and iris-ciliary body showed a significant increase of CD and KD,and the lens showed slightly elevated levels of lipid peroxidation products in EAU (Table 1). Although all cellular constituents, including DNA strands, enzymes, iron channels, structural proteins, and membrane lipids are potential targets for these reactive free radicals, cell membranes and structural lipoproteins are particularly susceptible to lipid peroxidation because of their abundant PUFAs.12"14 Therefore, the elevation of lipid peroxidation products within the cornea and iris-ciliary body in EAU indicates the presence of destructive processes of these tissues mediated by oxygen-free radicals. In contrast, the accumulation of lipid peroxidation products in lens tissue is less in acute inflammations such as EAU, because the lens fiber membranes contain fewer than 1% PUFAs.15 Be- Localization of Peroxidized Carbonyi Products Positive fluorescence stainings were observed in the experimental group, primarily in the iris and, to a lesser extent, in the trabecular meshwork (Figs. 3, 4). In the iris, fluorescence staining appeared as small clumps spread throughout the entire surface of the anterior border layer and the posterior epithelium of the iris. No staining was observed on the ciliary epithelium. In the cornea, granular fluorescence stainings were present along the endothelium (Fig. 4). These results were supported by visualization of bluish-purple carbonyl-NAH-FBB products. The control group was negative for fluorescence and NAHFBB products. DISCUSSION We demonstrated biochemical and histochemical evidence of oxygen-free radical-mediated tissue damage in the anterior segment of the eye in EAU. The presence and localization of free radical tissue damage were confirmed by the measurement of lipid peroxida- 2. Anterior segments of die experimental autoimmune uveitis eye. Numerous cellular infiltrates are evident in the anterior and posterior chamber. Proteinaceous material and fibrin debris are present in the anterior chamber. Some polymorphonuclear cells adhere to die corneal endothelium and die anterior lens capsule. C = cornea; IR = iris; L = lens. Hematoxylin and eosin, X80. FIGURE Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933415/ on 05/04/2017 633 Free Radical Damage in Uveitis TABLE l. Conjugated Diene and Keto-Diene Production in Experimental Autoimmune Uveitis KD CD Tissue Mean ± SD (nmol) P Value Mean ± SD (nmol) P Value Cornea EAU Control Iris-ciliary body EAU Control Lens EAU Control 2.32 ± 0.29 1.28 ± 0.21 <0.001 2.91 ± 0.45 0.98 ± 0.25 0.004 1.42 ± 0.32 0.71 ± 0.20 <0.001 0.52 ± 0.19 0.27 ± 0.12 0.01 3.08 ± 0.61 2.32 ± 0.33 0.005 0.88 ± 0.35 0.65 ± 0.17 CD = conjugated diene; KD = keto-diene; SD = standard deviation; n = number of determinations; EAU = experimental autoimmune uveitis. cause of this unique proportion of PUFAs, oxidative damage to the lens may require a longer exposure to oxygen-free radicals, as is the case in chronic uveitis.16 Histochemically, peroxidized carbonyl products are located primarily on the anterior border layer and posterior epithelium of the iris and, to a lesser extent, on the trabecular meshwork, and the corneal endothelium in EAU. Numerous PMNs were seen in the iris and the trabecular meshwork. Inflammatory cell infiltration also was seen in the anterior and posterior chamber surrounding the lens capsule as well as adherent to the corneal endothelium. Such PMNs are capable of inducing oxidation of PUFAs of these tissues. Generally, PUFAs of cell membranes consist primarily of linoleic acid (18:2), linolenic acid (18:3), arachidonic acid (20:4), and docosahexaenoic acid (22:6), all of which are potential targets for lipid peroxidation, their susceptibility increasing as the number of double bonds increases.14 Table 2 shows PUFA composition reported for ocular tissues. In the anterior segment, the proportion of PUFAs ranges from 34% for the iris to less than 1% for the lens. 1517 " 19 It is evident that the extent of pathologic peroxidation in the anterior segment in EAU appears to correlate with PUFA content of these tissues. The photoreceptor membranes contain a higher proportion of PUFAs (more than 50% of 22:6) than other organs in the body.20'21 Therefore, the retinal outer photoreceptor layer, by virtue of its structure, is extremely susceptible to peroxidation. 22 ' 23 In fact, peroxidized carbonyl products were found localized primarily within the outer photoreceptor layer in the retina in EAU.24 In contrast to the iris epithelium, no oxidized carbonyl products were detected on the ciliary epithelium. Under normal and inflammatory conditions, the eye is well protected from the damaging effects of oxygen and its metabolites by the control of antioxidant enzymes and other antioxidant agents, which are widely distributed in the eye.25 29 Particularly, antioxidant enzymes, such as superoxide dismutase, glutathione peroxide, and catalase, are found in abundance, predominantly in the ciliary epithelium, corneal epithelium and endothelium, and retinal pigment epithelium.25"28 Therefore, the ciliary epithelium and corneal endothelium seem to be well protected against oxygen metabolites in comparison to the iris and trabeculum. In addition, the ciliary body secretes aqueous humor that contains antioxidants, including ascorbate, as normal constituents.4'29 Therefore, the aqueous humor, which is closer to the ciliary body by its aqueous dynamics, contains a higher concentration of antioxidants and, thus, a lower accumulation of oxygen metabolites. Consequently, the exposure of oxygen-free radicals to the ciliary epithelium may be lower. It is supposed that damage to individual tissue correlates with a balance between the exposure to oxygen and its metabolites and the protection of the tissues afforded by the antioxidants contained in each. Treatment for ocular inflammatory disease has been based on suppressing a presumed immune response and reducing its inflammatory tissue damage. Although corticosteroids and immunosuppressive drugs such as cyclosporine have been effective in treating patients with uveitis, serious side effects have limited their use.30 3I Hence, a search for new therapeutic approaches for uveitis is fostered. The possible therapeutic advantage of antioxidant treatment for inflammatory disease has been focused. Bhuyan et al5 showed that one of the antioxidant agents, vitamin E, could prevent the peroxidative degeneration of lens lipid in vitro. Wu et al32 also demonstrated an efficient systemic antioxidant therapy for free radical damage of the retina in EAU. Alio et al33 reported the beneficial effects of topical antioxidant therapy on the evolution of infiltration in experimental keratitis. Therefore, it is reasonable to speculate that antioxidant Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933415/ on 05/04/2017 634 Investigative Ophthalmology & Visual Science, March 1996, Vol. 37, No, 4 FIGURE 3. {tap). False color, contocal imaging of fluorescence aistrmution in iNAn-reaciea anterior segments of the experimental autoimmune uveitis eye. Positive fluorescence stainings primarily locate in the iris {arrowhead) and the trabeculum {arroiv). The false colors refer to increasing fluorescence values, from blue (lowest), green, yellow, and red to pale pink (highest). AC = anterior chamber; IR = iris; TM = trabecular meshwork. Magnification, X500. FIGURE 4. {bottom). False color, confocal imaging of fluorescence distribution in NAH-reacted the anterior segments of the experimental autoimmune uveitis eye. Positive fluorescence stainings appear as the lesions spread throughout the anterior border layer of the iris. Note granular fluorescence stainings along the corneal endothelium {arrow). False colors indicate the same fluorescence values in Figure 3. C = cornea; AC = anterior chamber; IR = iris. Magnification, X500. Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933415/ on 05/04/2017 Free Radical Damage in Uveitis TABLE 2. 635 Polyunsaturated Fatty Acid Composition in Ocular Tissues Proportion ofPUFAs (%) Tissue 18:2 18:3 Cornea Epithelium Stroma Endothelium Trabeculum Iris-ciliary body Lens Retina — 1 2 4 12 — — — — — 5 — 20:4 22:5 22:6 2 — 9 11 12 17 — 4 2 — — 2 — 1 — — 2 Others Total Species 2 Rabbit Rabbit Rabbit Bovine Rabbit Rabbit Bovine 16 13 18 34 Cl 61 1 — — 51 Reference Number 19 19 19 18 17 15 20 PUFAs = polyunsaturated fatty acids. drugs may be useful for the treatment of ocular inflammation in humans. Probably in concert with conventional anti-inflammatory drugs, antioxidant therapy would be able to reduce the currently irreversible tissue damage—such as iris atrophy, secondary glaucoma, corneal degeneration, and complicated cataract— that occurs with persistent anterior uveitis. 9. 10. Key Words anterior segment, experimental autoimmune uveoretinitis, free radicals, lipid peroxidation, uveitis 11. Acknowledgments The authors thank David Stanforth for his expert technical assistance. 12. 13. References 1. Blake DR, Allen RE, Lunec J. Free radicals in biological systems: A review orientated to inflammatory processes. BrMedBull. 1987; 43:371-385. 2. Rossi F, Bellative P, Berton G, Grzeskowiak M, Papini E. Mechanism of production of toxic oxygen radicals by granulocytes and macrophages and their function in the inflammatory process. Pathol Res Pract. 1985; 180:136-142. 3. Goto H, Wu GS, Chen F, Kristeva M, Sevanian A, Rao NA. Lipid peroxidation in experimental uveitis: Sequential studies. CurrEyeRes. 1992; 6:489-499. 4. Spector A, Garner WH. Hydrogen peroxide and human cataract. Exp Eye Res. 1981;33:673-681. 5. Bhuyan KC, Bhuyan DK. Molecular mechanism of cataractogenesis: III: Toxic metabolites of oxygen as initiators of lipid peroxidation and cataract. Curr Eye Res. 1984;3:67-81. 6. Babizhayev MA, Bunin AY. Lipid peroxidation in open-angle glaucoma. Ada Ophthalmol. 1989; 67:371377. 7. Artola A, Alio JL, Bellot JL, Ruiz JM. Lipid peroxidation in the iris and its protection by means of viscoelastic substances (sodium hyaluronate and hydroxypropylmethylcellulose). Ophthalmic Res. 1993; 25:172176. 8. Donoso LA, Merryman CF, Sery TW, et al. S-antigen: 14. 15. 16. 17. 18. 19. 20. 21. Characterization of pathogenic epitope which mediates experimental autoimmune uveitis and pinealitis in Lewis rats. CurrEyeRes. 1987;6:1151-1159. Folch J, Lees M, Sloan-Stanley GH. A simple method for the isolation and purification of total lipides from animal tissues. / Biol Chem. 1957; 226:497-509. Pompella A, Comporti M. The use of 3-hydroxy-2naphthoic acid hydrazide and fast blue B for the histochemical detection of lipid peroxidation in animal tissues: A microphotometric study. Histochemistry. 1991;95:255-262. Halliwell B, Gutteridge JMC. Free Radicals in Biology and Medicine. Oxford: Clarendon Press; 1989:139170. Rice-Evans CA, Diplock AT. Current status of antioxidant therapy. Free Radic Biol Med. 1993; 15:77-96. Floyd RA, Carney JM. Free radical damage to protein and DNA: Mechanisms involved and relevant observations on brain undergoing oxidative stress. Ann Neurol. 1992;32:S22-S27. Cheeseman KH. Mechanisms and effects of lipid peroxidation. Mol Aspects Med. 1993; 14:191-197. Zelenka PS. Lens lipids. Curr Eye Res. 1984;3:13371359. Yugay MT, Pereira PC, Mota MC, Cunha-Vaz JG. Possible role of free radical oxidation in postuveal cataract pathogenesis. ARVO Abstracts. Invest Ophthalmol Vis Sri. 1995;36:S605. Zhang Y, Yousufzai SYK, Abdel-Latif AA. Comparative studies on fatty acid composition and phospholipases A2 and C activities in rabbit and bovine iris-ciliary body. Exp Eye Res. 1993;56:151-155. Nawar WW, Kim SK, Yates-Berg D, Anderson PJ. Fatty acid composition of total lipid from TM of calf and cow. ARVO Abstracts. Invest Ophthalmol Vis Sri. 1990; 31:357. Bazan HEP, Bazan NG. Composition of phospholipids and free fatty acids and incorporation of labeled arachidonic acid in rabbit cornea: Comparison of epithelium, stroma and endothelium. Curr Eye Res. 1984;3:1313-1319. Stone WL, Farnsworth CC, Dratz EA. A reinvestigation of the fatty acid content of bovine, rat and frog retinal rod outer segments. Exp Eye Res. 1979;28:387-397. Wheeler TG, Benolken RM, Anderson RE. Visual Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933415/ on 05/04/2017 636 22. 23. 24. 25. 26. 27. Investigative Ophthalmology & Visual Science, March 1996, Vol. 37, No. 4 membranes: Specificity of fatty acid precursors for the electrical response to illumination. Science. 1975; 188: 1312-1314. Rao NA. Role of oxygen-free radicals in retinal damage associated with experimental uveitis. Trans Am Ophthalmol Soc. 1990;88:797-850. Anderson RE, Rapp LM, Wiegand RD. Lipid peroxidation and retinal degeneration. Curr Eye Res. 1984; 3: 223-227. Wu GS, Stanforth DA, Rao NA. Localization of EAUmediated oxidative damage in retina by fluorescent derivatization of cellular carbonyls. ARVO Abstracts. Invest Ophthalmol Vis Sri. 1994; 35:1517. Armstrong D, Santangelo G, Connole E. The distribution of peroxide regulating enzymes in the canine eye. Curr Eye Res. 1981; 1:225-242. Rao NA, Thaete LG, Delmage JM, Sevanian A. Superoxide dismutase in ocular structures. Invest Ophthalmol Vis Sri. 1985;26:1778-1781. Atalla L, Fernandez MA, Rao NA. Immunohistochemi- 28. 29. 30. 31. cal localization of catalase in occular tissue. Curr Eye Res. 1987;6:1181-1187. Attala LR, Sevanian A, Rao NA. Immunohistochemical localization of glutathione peroxidase in ocular tissue. Curr Eye Res. 1988; 7:1023-1027. Bode AM, Green E, Yavarow CR, et al. Ascorbic acid regeneration by bovine iris-ciliary body. Curr Eye Res. 1993; 12:593-601. Whitcup SM, Nussenblatt RB. Treatment of autoimmune uveitis. AnnNY Acad Sri. 1993;696:307-318. Masuda K, Nakajima A. A double-masked study of cyclosporine in Behcet's disease. In: Schindler R, ed. Cyclosporine Treatment in Behcet's Disease. Berlin: Springer-Verlag; 1985:162-164. 32. Wu GS, Walker J, Rao NA. Effect of Deferoxamine on retinal lipid peroxidation in experimental uveitis. Invest Ophthalmol Vis Sri. 1993; 34:3084-3089. 33. Alio JL, Artola A, Serra A, Ayala MJ, Mulet ME. Effect of topical antioxidant therapy on experimental infectious keratitis. Cornea. 1995; 14:175-179. Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933415/ on 05/04/2017