Lipid Peroxidation in Rod Outer Segments

Transcription

1 Investigative Ophthalmology & Visual Science, Vol. 33, No. 7, June 1992 Copyright Association for Research in Vision and Ophthalmology Lipid Peroxidation in Rod Outer Segments Role of Hydroxyl Rodicol and Lipid Hydroperoxides Monica A. De La Poz* ond Robert E. Anderson Lipid peroxidation of rod outer segment (ROS) membranes has been implicated in the pathogenesis of numerous ocular disease processes. The hydroxyl radical might be involved in initiating the reaction. An in vitro system was developed to study lipid peroxidation of the ROS and the role of the hydroxyl radical. Bovine ROS were suspended in various concentrations of ferrous sulfate, incubated for 10 min at 37 C, treated with diethylenetriamine pentaacetic acid to chelate the iron, and subjected to a thiobarbituric acid assay for malondialdehyde. A predictable increase in lipid peroxidation occurred in the presence of Fe +2. No effect was seen in the presence of Fe +3. Adding hydrogen peroxide, which would form the hydroxyl radical by reacting with Fe +2, had no effect at low concentrations. At higher concentrations, lipid peroxidation was inhibited, presumably from the oxidation of Fe +2 to Fe +3. Ethanol, a known hydroxyl radical scavenger, had no inhibitory effect in concentrations up to 0.50 mol/1. Conversely, cumene hydroperoxide and linoleic acid hydroperoxide, which form hydrophobic radicals, stimulated lipid peroxidation in the presence of Fe +2. These findings suggest that, under these experimental conditions, the hydroxyl radical is not an initiator of lipid peroxidation in ROS. They provide evidence that endogenous lipid radicals may initiate the reaction. Invest Ophthalmol Vis Sci 33: ,1992 Lipid peroxidation is a complex process that is detrimental to cell structure and function. In ocular tissue, it has been implicated in a wide range of disease processes, including cataractogenesis, 1 " 3 photic retinopathy, 4 " 9 retinopathy of prematurity, 1011 ocular siderosis, 1213 and ocular inflammation On a molecular level, the mechanism involves the abstraction of a hydrogen atom from a methylene carbon on a polyunsaturated fatty acid, and rearrangement of the double bonds to form an alkyl radical with a conjugated diene. 16 Attack by molecular oxygen produces a lipid peroxy radical, which can abstract a hydrogen atom from another methylene carbon on an adjacent molecule to form a lipid hydroperoxide. This self-perpetuating reaction continues until the lipid radicals are destroyed by free radical scavengers (such as vitamin From the Cullen Eye Institute, Baylor College of Medicine, Houston, Texas. ""Current address: Massachusetts Eye and Ear Infirmary, 243 Charles Street, Boston, MA Supported in part by grants from the National Eye Institute, Bethesda, Maryland (EY00871, 04149, and 02520); the Retina Research Foundation, Houston, Texas; the RP Foundation Fighting Blindness, Baltimore, Maryland; and Research to Prevent Blindness, Inc., New York, New York. Dr. Anderson is the recipient of a Senior Scientific Investigator Award from Research to Prevent Blindness, Inc. Submitted for publication: July 19,1991; accepted December 12, Reprint requests: Robert E. Anderson, PhD, MD, Cullen Eye Institute, Baylor College of Medicine, One Baylor Plaza, Houston, TX E or ascorbic acid) or by interacting with other free radicals. Breakdown of lipid hydroperoxides leads to the formation of various molecular species, including aldehydes, which are known to be toxic to cells. 17 Because rod outer segments (ROS) contain the highest level of long-chain polyunsaturated fatty acids of any membrane studied thus far, 18 they are particularly susceptible to peroxidative damage. The mechanism of initiation of ROS peroxidation is not understood completely. The hydroxyl radical, a highly reactive oxidant, has been proposed as the initiating species. 19 " 22 The generation of this species occurs through an iron-catalyzed Fenton reaction, which requires the presence of Fe In addition to initiating lipid peroxidation, the hydroxyl radical has been implicated in direct cellular damage. 24 " 26 The ability of the hydroxyl radical to initiate lipid peroxidation has been questioned by some investigators. 27 " 30 It was found that the reaction in vitro is not initiated by the hydroxyl radical, but rather, it requires the presence of both Fe +2 and Fe Using a modified version of an in vitro system to study ROS lipid peroxidation previously described, 31 we provide evidence suggesting that the hydroxyl radical is not an initiator of lipid peroxidation of ROS. We found support for a role for lipid radicals in the initiation process. Materials and Methods Bovine retinal tissue was used in the experimental protocol. The tissue was frozen after it was harvested 2091

2 2092 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / June 1992 Vol. 33 from cattle that had been dark adapted before they were killed. Purified ROS were isolated by discontinuous sucrose-gradient centrifugation as previously described. 8 Care was taken to saturate all solutions with argon to minimize lipid oxidation during the preparation and extraction procedures. The washed ROS pellet was resuspended in HEPES buffer, substituting glucose with pyruvate. 32 An aliquot was obtained for protein determination using the BCA assay. The reaction mixture consisted of ROS at a final concentration of 39 Mg/ml protein plus other agents as described subsequently in a total volume of 1 ml. The solution was incubated at 37 C for 10 min, and the incubation was stopped by adding the chelating agent diethylenetriamine pentaacetic acid (DTPA; Aldrich, Milwaukee, WI). A thiobarbituric acid assay for malondialdehyde was done. 16 The absorbance of the reaction mixture was measured at 535 nm on a Gilford 250 spectrophotometer (Gilford Instrument Labs, Inc., Oberlin, OH) against a blank that contained the reagents minus the ROS. The ROS subjected to the peroxidation scheme were treated in one of several ways. Incubation of ROS With Fe +2 or Fe +3 A 400-/xl aliquot of ROS was incubated with 100 /*1 of deionized water containing various concentrations of Fe +2 or Fe +3 at 37 C for 10 min. The incubation was stopped by adding 0.5 ml of DTPA at a final concentration of 2.27 mmol/1. Incubation of ROS With Hydrogen Peroxide or Ascorbic Acid Various concentrations of hydrogen peroxide or ascorbic acid in 50 fi\ of deionized water were incubated with 400-^1 aliquots of ROS. The samples were incubated for 10 min at 4 C. Then 50 ix\ of Fe +2 was added to the reaction mixtures at a final concentration of 80 JUM before incubation for 10 min at 37 C. The experiment was repeated to study the effect of Fe +2 and Fe +3. An aliquot of ROS was incubated with 1.0 mm hydrogen peroxide or 0.10 mmol/1 ascorbic acid at 4 C for 10 min. At the end of the incubation, various concentrations of Fe +2 or Fe +3 were added to the reaction mixtures, and the incubation continued for 10 min at 37 C. All of these incubations were stopped by adding 0.5 ml of DTPA at a final concentration of 2.27 mm. Incubation of ROS With Ethanol Aliquots of ROS were incubated for 10 min at 4 C with various concentrations of ethanol up to 0.54 mm. Then Fe +2 was added at a final concentration of 80 fim, and the solutions were incubated for 10 min at 37 C. The incubation was stopped by adding DTPA at a final concentration of 2.27 mm. Incubation of ROS With Cumene Hydroperoxide Aliquots of ROS in a volume of 55 n\ were incubated at 4 C for 10 min with 400 tx\ of HEPES buffer containing various concentrations of cumene hydroperoxide (isopropylbenzene hydroperoxide; Sigma, St. Louis, MO). Care was taken to saturate all solutions with argon and to do all experiments in dim light. Others 33 previously found that cumene hydroperoxide is soluble in HEPES buffer at the concentrations we used. At the end of the incubation period, 45 ix\ of water containing Fe +2 at a final concentration of 80 /xm was added to the reaction mixture, which then was incubated for 10 min at 37 C. The incubation was stopped by adding 500 /xl of DTPA at a final concentration of 2.27 mmol/1. Incubation of ROS With Linoleic Acid Hydroperoxide The synthesis of the hydroperoxide derivative of linoleic acid was based on a modified version of previously described procedures. 34 ' 35 Linoleic acid hydroperoxide was synthesized by incubating 100 ml of 1.0 mmol/1 linoleic acid (cis, cis-9,\2 octadecadienoic acid; Nu-Chek-Prep, Elysian, MN) in 0.2 mol/1 sodium borate buffer (ph 9.0) with 2 mg of lipoxidase type 1-B from soybean (linoleate: oxygen oxidoreductase, EC ; Sigma). The mixture was stirred at 21 C in the dark for approximately 12 hr. Residual precipitate was removed by centrifugation at 12,000 X g for 15 min. The clear aqueous reaction mixture then was mixed withfiveparts of chloroform-methanol (2:1, by volume); this resulted in the formation of a biphase. The lower chloroform layer was evaporated to dryness, and the lipid was taken up in 2 ml absolute ethanol. We assayed conjugated dienes by measuring the absorbance at 233 nm on a Gilford spectrophotometer against an ethanol blank. The ethanol solution then was applied to a silica gel HR plate (20 X 20 cm) and developed in hexane-diethyl ether-glacial acetic acid (50:50:1, by volume). The region on the plate containing the linoleic acid hydroperoxide was scraped into a vial and washed with a known volume of chloroform-methanol (2:1, by volume). After mixing with 0.2 parts of deionized water, the chloroform layer was evaporated to dryness and taken up in 2 ml of argon-saturated absolute ethanol. An absorption spectrum of the ethanol mixture verified a peak absorbance at 233 nm, indicating the presence of conjugated dienes.

3 No. 7 LIPID PEROXIDATION IN ROS / De La Poz and Anderson 2093 Various concentrations of linoleic acid hydroperoxide in a volume of 8 fa of absolute ethanol were incubated with 800-^1 aliquots of ROS. The specimens were incubated at 21 C for 45 min on a sonicator. At the end of this period, a 45-^1 aliquot of deionized water containing Fe +2 at a final concentration of 80 nm was added to the reaction mixture. The incubation was done for 10 min at 37 C and stopped by adding DTP A at a final concentration of 2.27 mmol/1. Results The effect of Fe +2 and Fe +3 on lipid peroxidation in bovine ROS is shown in Figure 1. In concentrations up to 0.40 mmol/1 Fe +2, there was a significant increase in lipid peroxidation. Conversely, Fe +3 did not stimulate peroxidation except at high concentrations. Thus, the initiation of lipid peroxidation in our isolated ROS preparation was dependent on iron in the reduced (Fe +2 ) state. The role of Fe +2 in initiating lipid peroxidation in ROS was shown also by experiments using ascorbic acid to reduce Fe +3 to Fe +2. When 0.10 mmol/1 ascorbic acid was included in the reaction mixture, there was a marked enhancement in lipid peroxidation in E in s O (0.a o Ferric (+3) Ferrous (+2) > i i min i i i inn Fig. 1. Effect of Fe 2+ and Fe 3+ on malondialdehyde production, measured by absorbance at 535 nm. ROS were incubated at 37 C for 10 min Ferric Ferrous Ferric + ASC Ferrous + ASC Fig. 2. Effect of ascorbic acid (ASC) on malondialdehyde production, measured by absorbance at 535 nm in the presence of Fe 2+ and Fe 3+. ROS were suspended for 10 min at 4 C in 0.10 mm ascorbic acid before incubation with either Fe 2+ or Fe 3+ for 10 min at 37 C. the presence of Fe +3, even at the lowest concentration (Fig. 2). By contrast, only a small increase in lipid peroxidation was seen at the lower Fe +2 concentrations. In the absence of Fe +2 and Fe +3, ascorbic acid alone in concentrations up to 1.0 mmol/1 in the reaction mixture had no effect on lipid peroxidation in ROS. These studies provide additional evidence that Fe +2, but not Fe +3, is responsible for initiating lipid peroxidation in our system. Hydrogen peroxide oxidizes Fe +2 to Fe +3. In the process, it generates one molecule each of hydroxyl radical and ion. The effect of the hydroxyl radical in initiating lipid peroxidation in ROS was tested by incubating ROS with 1 mmol/1 hydrogen peroxide and various concentrations of Fe +2 (Fig. 3). Contrary to our expectations, lipid peroxidation was not enhanced by the addition of hydrogen peroxide. At low concentrations of Fe +2, lipid peroxidation was reduced to the same level observed for Fe +3. Only at the highest concentration of Fe +2 could the effect of hydrogen peroxide be overcome completely. Hydrogen peroxide had no effect in the presence of Fe +3, except perhaps to reduce peroxidation slightly. Hydrogen peroxide inhibits lipid peroxidation of ROS induced by Fe +2 in a concentration-dependent manner (Fig. 4). In the presence of 80 /im Fe +2, the inhibitory effect

4 2094 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / June 1992 Vol Q O.a < effectiveness of 80 um Fe +2 in inducing lipid peroxidation of ROS in a dose-dependent manner (Fig. 5). This substance alone had no effect on lipid peroxidation in ROS. Similarly, the presence of linoleic acid hydroperoxide in the reaction mixture caused a dosedependent increase in lipid peroxidation of ROS, induced by 80 fxm Fe +2 (Fig. 5). Used alone, this substance did not stimulate peroxidation in ROS. Discussion It has been shown previously that constant illumination 7 " 9 and intravitreal injection of ferrous sulfate 12 produce a retinal degeneration in frogs and rats that is specific for photoreceptor cells, with early biochemical changes in membranes including a loss of docosahexaenoic acid (22:6n-3) and an increase in lipid hydroperoxides. These results suggest that lipid peroxidation may play a role in certain types of retinal degeneration. Earlier in vitro studies 31 found that incubation of ROS membranes with 20 /xm ferrous iron in the presence of 2 mmol/1 ascorbic acid resulted in predictable lipid peroxidation, an increase in lipid hydroperoxides, and a decrease in 22:6n-3. Fig. 3. Effect of hydrogen peroxide (HP) on malondialdehyde production, measured by absorbance at 535 nm in the presence of Fe 2+ and Fe 3+. ROS were suspended for 10 min at 4 C in 1.0 mm hydrogen peroxide before incubation with either Fe 2+ or Fe 3+ for 10 min at 37 C..40 was most evident with hydrogen peroxide concentrations greater than 100 fxm. In the absence of Fe +2 and Fe +3 5 hydrogen peroxide had no effect in concentrations up to 100 mmol/1. These studies suggest that hydroxyl radical is not involved in the initiation of lipid peroxidation in our bovine ROS preparation and provide additional evidence that iron in the reduced form (Fe +2 ) initiates peroxidation in membranes. Ethanol is an effective scavenger of hydroxyl radicals. To test the role of hydroxyl radical in lipid peroxidation in another way, isolated bovine ROS were incubated with Fe +2 and various concentrations of ethanol. In concentrations up to 0.54 mmol/1, ethanol had no significant effect on lipid peroxidation of ROS induced by 80 um Fe +2 (Fig. 5), suggesting that the peroxidation was not initiated by the hydroxyl radical. Cumene hydroperoxide and linoleic acid hydroperoxide are organic molecules and thus are more hydrophobic than hydrogen peroxide. When incubated with Fe +2, they form free radicals that are more stable than the hydroxyl radical. 36 The presence of cumene hydroperoxide in the reaction mixture increased the 0.00 i i mm i Him 10 10* n 10 v 10 Fig. 4. Effect of hydrogen peroxide (HP) on malondialdehyde production measure by absorbance at 535 nm in the presence of 80 MM Fe 2+. ROS were suspended for 10 min at 4 C in various concentrations of hydrogen peroxide before incubation with 80 um Fe 2+.

5 No. 7 LIPID PEROXIDATION IN ROS / De La Paz and Anderson Log Concentration (nm) Fig. 5. Effect of ethanol (ETOH), cumene hydroperoxide (CHP), and linoleic acid hydroperoxide (LHP) on ratio of absorbance of the. specimen at 535 nm to absorbance of untreated control at 535 nm. ROS were suspended in various concentrations of ETOH, CHP, or LHP for 10 min at 4 C and then incubated with 80 MM Fe 2+ for 10 min at 37 C. Malondialdehyde production was measured as absorbance at 535 nm. Control specimen consisted of ROS alone incubated with 80 nm of Fe 2+ for 10 min at 37 C. Lipid peroxidation in ROS is an incompletely understood process. It is known that the reaction involves the formation of free radicals, but factors involved in the initiation of the reaction are not clear. We used an in vitro system to study lipid peroxidation, and we gained a greater understanding of factors influencing the reaction. Incubating ROS with Fe +2 (but not Fe +3 ) led to a predictable increase in lipid peroxidation. Thus, the classic Fenton-type reaction occurred in the ROS. The relatively small increase in peroxidation in the presence of Fe +3 was probably caused by small amounts of Fe +2 that contaminated the reaction mixture. The absolute requirement for Fe +2 was shown in the experiment where ROS incubated with Fe +3 were not oxidized unless ascorbic acid was present. Numerous studies have implicated the hydroxyl radical in the initiation of lipid peroxidation. 19 " 22 The evidence is overwhelming in certain systems. However, there are reports that, in some experimental systems, the hydroxyl radical is not involved in the initiation step. 27 " 30 The latter appears to be the case for ROS, based on the following evidence: 1. Hydroxyl radicals generated by incubating hydrogen peroxide with Fe +2 in the presence of ROS did not enhance lipid peroxidation. At higher concentrations, hydrogen peroxide inhibited lipid peroxidation, probably by promoting the oxidation of Fe +2 to Fe +3. As expected, this inhibition can be overcome with relatively high concentrations offe Ethanol, a known scavenger of hydroxyl radicals, 37 did not inhibit lipid peroxidation induced by Fe +2. Previous studies 31 found that mannitol, another scavenger of hydroxyl radicals, also had no effect on lipid peroxidation in ROS. The hydroxyl radical is a highly reactive and shortlived species that reacts with most organic compounds at nearly diffusion-controlled rates. 38 For it to have an effect on lipid peroxidation, it must have access to reactive methylene hydrogen atoms. In the ROS, these would be located on long fatty acyl chains in the interior of the bilayer. The short-lived nature of the hydroxyl radical makes it unlikely that it could migrate from the site of generation to the hydrophobic membrane interior where peroxidation must be initiated. Conversely, cumene hydroperoxide and linoleic acid hydroperoxide enhance Fe +2 -induced lipid peroxidation of ROS. We postulate that these endogenous hydroperoxides form more stable free-radical species in comparison with the hydroxyl radical. There is resultant facilitated entry of these species into the interior of the bilayer to initiate peroxidation of long fatty acyl chains. We do not know how Fe +2 initiates lipid peroxidation in the absence of a source of radicals. It is possible that fatty acid hydroperoxides were formed during the preparation of the ROS, even though we took precautions to protect the membranes from oxidation. Alternatively, studies have shown that the retina can generate lipid hydroperoxides through enzymatic oxidation of endogenous polyunsaturated fatty acids. Others found that 22:6n-3 can be oxidized by retinal enzymes to form docosanoids. 39 Hydroperoxides are intermediates in both the cyclooxygenase and lipoxygenase reactions. Thus, there may be a small endogenous pool of lipid hydroperoxides in the retina, which, under certain conditions, can serve as a source of lipid free radicals and promote peroxidation of the long-chain polyunsaturated fatty acids that are abundant in the ROS membranes. Key words: rod outer segment, lipid peroxidation, free radical reaction, hydroxyl radical, cumene hydroperoxide, lipid hydroperoxide

6 2096 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / June 1992 Vol. 33 Acknowledgments The authors thank Maureen B. Maude for technical assistance and advice and Richard A. Alvarez for help in producing the figures. References 1. Borkman RF and Lerman S: Evidence for free radical mechanisms in aging and ultraviolet-irradiated ocular lenses. Exp Eye Res 25:303, Pirie A, Rees JR, and Holmberg NJ: Diquat cataract: Formation of the free radical and its reaction with constituents of the eye. Exp Eye Res 9:204, Varma SD: Superoxide and lens of the eye: A new theory of cataractogenesis. International Journal of Quantitative Chemistry 20:479, Hiramitsu T, Majima Y, Hasegawa Y, and Hirata K: Role of lipid peroxide in the induction of retinopathy by x-irradiation. Nippon Ganka Gakkai Zasshi 78:819, Kagan VE, Kuliev 1Y, Spirichev VB, Shvedova AA, and Kozlov YP: Accumulation of lipid peroxidation products and depression of retinal electrical activity in vitamin E-deficient rats exposed to high-intensity light. Bulletin of Experimental Biology and Medicine 91:144, Noell WK, Walker V, Kang B, and Berman S: Retinal damage by light in rats. Invest Ophthalmol 5:450, Wiegand RD, Giusto NM, and Anderson RE: Lipid changes in albino rat rod outer segments following constant illumination. In Problems of Normal and Genetically Abnormal Retinas, Clayton RM, Haywood J, Reading HW, and Wright A, editors. London, Academic, 1982, pp Wiegand RD, Giusto NM, Rapp LM, and Anderson RE: Evidence for rod outer segment lipid peroxidation following illumination of the rat retina. Invest Ophthalmol Vis Sci 24:1433, Anderson RE, Rapp LM, and Wiegand RD: Lipid peroxidation and retinal degeneration. Curr Eye Res 3:223, Hittner HM, Godio LB, Rudolph AJ, et al: Retrolental fibroplasia: Efficacy of vitamin E in a double-blind clinical study of preterm infants. N Engl J Med 305:1365, Kretzer FL, Hittner HM, Johnson AT, Mehta RS, and Godio LB: Vitamin E and retrolentalfibroplasia:ultrastructural support of clinical efficacy. Ann N Y Acad Sci 393:145, Rapp LM, Wiegand RD, and Anderson RE: Ferrous iron-mediated retinal degeneration: Role of rod outer segment lipid peroxidation. In Problems of Normal and Genetically Abnormal Retinas, Clayton RM, Haywood J, Reading HW, and Wright A, editors. London, Academic, 1982, pp Hiramitsu T, Hasegawa Y, Hirata K, Nishigaki I, and Yagi K: Lipoperoxide formation in the retina in ocular siderosis. Experientia 32:1324, Burke JM: Vitreal superoxide and superoxide dismutase after hemorrhagic injury: The role of invasive cells. Invest Ophthalmol Vis Sci 20:435, Sery TW and Petrill R: Superoxide anion radical as an indirect mediator in ocular inflammatory disease. Curr Eye Res 3:243, Buege JA and Aust SD: Microsomal lipid peroxidation. Methods Enzymol 30:302, Tappet A: Lipid peroxidation and fluorescent molecular damage to membranes. In Pathobiology of Cell Membranes, Trump BF and Arstila AJ, editors. New York, Academic, 1975, pp Anderson RE and Andrews LD: Biochemistry of photoreceptor membranes in vertebrates and invertebrates. In Visual Cells in Evolution, Westfall J, editor. New York, Raven, 1982, pp Gutteridge JMC, Richmond R, and Halliwell B: Inhibition of the iron-catalysed formation of hydroxyl radicals from superoxide and of lipid peroxidation by desferrioxamine. Biochem J 184:469, King MM, Lai EK, and McCay PB: Singlet oxygen production associated with enzyme-catalyzed lipid peroxidation in liver microsomes. J Biol Chem 250:6496, Fong K-L, McCay PB, Poyer JL, Keele BB, and Misra H: Evidence that peroxidation of lysosomal membranes is initiated by hydroxyl free radicals produced during flavin enzyme activity. J Biol Chem 248:7792, McCay PB, Pfeifer PM, and Stipe WH: Vitamin E protection of membrane lipids during electron transport functions. Ann N Y Acad Sci 203:62, Imlay JA, Chin SM, and Linn S: Toxic DNA damage to hydrogen peroxide through the Fenton reaction in vivo and in vitro. Science 240:640, Zweier JL, Kuppusamy P, and Lutty GA: Measurement of endothelial cell free radical generation: Evidence for a central mechanism of free radical injury in postischemic tissues. Proc Natl Acad Sci U S A 85:4046, Rubanyi GM: Vascular effects of oxygen-derived free radicals. Free Radic Biol Med 4:107, Lam S, Tso MOM, and Gurne DH: Amelioration of retinal photic injury in albino rats by dimethylthiourea. Arch Ophthalmol 108:1751, Minotti G and Aust SD: The requirement for iron (III) in the initiation of lipid peroxidation by iron (II) and hydrogen peroxide. J Biol Chem 262:1098, Morehouse LA, Tien M, Bucher JR, and Aust SD: The effect of hydrogen peroxide on the initiation of microsomal lipid peroxidation. Biochem Pharmacol 32:123, Svingen BA, Buege JA, O'Neal FO, and Aust SD: The mechanism of NADPH-dependent lipid peroxidation. J Biol Chem 254:5892, Aust SD and Svingen BA: The role of iron in enzymatic lipid peroxidation. In Free Radicals in Biology, Vol 5, Pryor WA, editor. New York, Academic, 1982, pp Anderson RE, Maude MB, and Nielsen JC: Effect of lipid peroxidation on rhodopsin regeneration. Curr Eye Res 4:65, Pu GA-W and Anderson RE: Ethanolamine accumulation by photoreceptor cells of the rabbit retina. J Neurochem 42:185, Naash MI and Anderson RE: Characterization of glutathione peroxidase in frog retina. Curr Eye Res 3:1299, Yamamoto A, Fujii Y, Yasumoto K, and Mitsuda H: Product specificity of rice germ lipoxygenase. Lipids 15:1, Gardner HW: Isolation of a pure isomer of linoleic acid hydroperoxide. Lipids 10:248, Pryor WA: The role of free radical reactions in biological systems. In Free Radicals in Biology, Vol 1, Pryor, WA editor. New York, Academic, 1976, pp Morehouse K and Mason RP: The transition metal-mediated formation of the hydroxyl free radical during the reduction of molecular oxygen by ferredoxin-ferredoxin: NADP + oxidoreductase. J Biol Chem 263:1204, Anbar M and Neta P: A compilation of specific bimolecular rate constants for the reactions of hydrated electrons, hydrogen atoms and hydroxyl radicals with inorganic and organic compounds in aqueous solution. International Journal of Applied Radiation and Isotopes 18:493, Bazan NG, Birkle DL, and Reddy TS: Docosahexaenoic acid (22:6n3) is metabolized to lipoxygenase reaction products in the retina. Biochem Biophys Res Commun 125:741, 1984.