MIRARY University This is to certify that the dissertation entitled "Oxygen Toxicity and Adaptation to Hyperoxia: Studies on the Retina of the Rainbow Trout (Salmo Gairdneri)“ presented by Paul E. Desrochers has been accepted towards fulfillment of the requirements for L degree in M Major professor Date 3 7 M5u;.....ur .‘ . - r‘ In . . . i 0.127" MSU LIBRARIES .—:_—I- RETURNING MATERIALS: Place in book drop to remove this checkout from your record. [lfl§§_will be charged if book is returned after the date stamped below. ‘___.-— ~ 1 OXYGEN TOXICITY AND ADAPTATION TO HYPEROXIA: STUDIES ON THE RETINA OF THE RAINBOW TROUT (SALMO GAIRDNERI) By Paul Emile Desrochers A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Physiology 1987 ABSTRACT OXYGEN TOXICITY AND ADAPTATION T0 HYPEROXIA: STUDIES ON THE RETINA OF THE RAINBOW TROUT (SALMO GAIRDNERI) BY Paul Emile Desrochers Molecular oxygen can accept electrons univalently' to form. toxic intermediates, i.e. superoxide (0;), hydrogen peroxide (H202) and the hydroxyl radical ('OH). The damaging effects of oxygen toxicity follow— ing hyperoxic exposure have been attributed to formation of these reac- tive oxygen metabolites. The retina of Salmo gairdneri is normally exposed to 0 tensions as high as 750 mmHg. This tissue has adapted to 2 these hyperoxic conditions and indeed is resistant to even greater oxy- gen tensions. There are several defense mechanisms to protect tissues against oxidative damage. One defense is to limit the concentrations of 1’ 2 and H202. enzymes (CuZnSOD and MnSOD) regulate the concentration of superoxide, the two puimary intermediates 0 The superoxide dismutase while catalase and glutathione peroxidase regulate the H202 concentra- tion. Since 'OH is believed to be formed by processes dependent on 0; and H202, regulation of these two metabolites limits 'OH formation. Based on the free radical theory of oxygen toxicity, tolerance to hyper- oxia is either due to a decreased propensity for generating reactive metabolites and/or the presence of efficient antioxidant defenses. The rate of oxidant formation was estimated in retinal homogenates under normoxic and hyperoxic conditions using cyanide-insensitive res- piration. Trout brain and liver tissues are sensitive to oxygen and were therefore used as internal controls for comparison to the oxygen Paul E. Desrochers resistant retina. The results indicate that CN--insensitve respiration by retinal homogenates was elevated during hyperoxic exposure. Similar results were found for the brain and liver homogenates, which indicated that oxygen resistance was not due to an inability to generate reactive oxygen metabolites. In addition, the antioxidant activities of the copper/zinc and manganese superoxide dismutases, catalase and gluta- thione/glutathione peroxidase enzymes were compared in these tissues. The results indicate that the efficiency of the antioxidant defenses is no greater in the retina than in the oxygen sensitive tissues. These results suggest that resistance to oxygen toxicity in the rainbow trout retina is not based on cellular properties controlling formation or removal of reactive oxygen metabolites but may be due to a predetermined genetic resistance of O -sensitive target sites in cells adapted to 2 chronic hyperoxic conditions. This dissertation is dedicated to the memory of Dr. Jack R. Hoffert, my mentor and friend. 11 ACKNOWLEDGMENTS I would like to thank the members of my guidance committee: Dr. Steve Heidemann for providing me with the opportunity to complete my dissertation work and for his support and guidance, Dr. Jacob Krier for his encouragement and willingness to help open doors which enabled me to continue this project, Dr. Bill Frantz for his availability and interest in research progress, Dr. Bob Roth for his scientific interest and keen review of this dissertation, and Dr. Steve Weiss for his willingness to contribute to my training, and for his critical insight and helpful suggestions. I would also like to thank Dr. P.0. Fromm for his support, encouragement and interest in my career development. A very special thanks is also extended to Esther Brenke for providing expert technical assistance, support, encouragement and friendship. Many thanks to the members and friends of the Physiology Department, both past and present, for their encouragement, support and assistance along the way. I am grateful for support from the National Institute of Health, Grant No. EY-00009, and funding through the Department of Physiology. Finally, I would like to thank my wife Denise, for her patience, perseverance and dedication, and Renee and Kevin for providing me with the opportunity to escape and play. iii TABLE OF CONTENTS LIST OF TABLES.................................................... LIST OF FIGURES................................................... INTRODUCTION...................................................... LITERATURE REVIEW................................................. Oxygen Toxicity............................................... Chemistry of Oxygen and its Reduction Products................ Cellular Sources and Reactivity of Various Oxygen Metabolites. Protection Against Oxygen Toxicity............................ Enzymatic Scavenging of Superoxide....................... Enzymatic Scavenging of Hydrogen Peroxide................ Secondary Antioxidant Defenses........................... A Potential Role for Oxygen Metabolites in Hyperoxic Injury... Resistance of Teleost Ocular Tissues to Oxygen Toxicity....... MATERIALS AND METHODS............................................. Experimental Animals.......................................... Tissue Preparation............................................ Cyanide Insensitive Oxygen Consumption Determinations......... Superoxide Dismutase Assay.................................... Cytochrome c Oxidase Assay.................................... Catalase Assay................................................ Glutathione Per0x1dase AssaYOOOOOOOOOOOOOOOOOOOOOOOOOO00...... Glutathione AssaYOOOOOOOOOOOOOOOOOOO...OOOOOOOOOOOOOOOOOOOOOOO iv Page vi vii l 4 l. 8 10 18 18 20 21 22 25 27 27 27 28 30 35 36 38 39 V RESULTSOO0..C0.0...0...C.II.0.0.0.0....OOOOOOOOOOOOOOOOOOOOO...0.0 Total and Cyanide Insensitive Oxygen Consumption by Trout RetinaOOOOO...COOOOOOOOOOOOOO00.0.00...0....00.0.. Primary Antioxidants in Trout Retina.......................... superOXide Dismutase ACCiVitYoooo0000000000000.000000.000 catalase ActiVitYOOOOO0..OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOI. Glutathione Peroxidase Activity and Glutathione Content.. DISCUSSION........................................................ Determination of Oxidative Flux in Rainbow Trout Retinal Homogenate......................................... Primary Antioxidant Defenses in the Rainbow Trout Retina...... A Rationale for the Adaptive Response of Tissues to Acute vs Chronic Hyperoxic Exposure..................... SUMMARY AND CONCLUSIONS........................................... APPENDIX A: Solutions............................................. APPENDIX B: Lowry Protein Determination........................... APPENDIX C: Superoxide Dismutase Assay............................ APPENDIX D: Cytochrome c Oxidase Assay............................ APPENDIX E: Catalase Assay........................................ APPENDIX F: Glutathione Peroxidase Assay.......................... 40 40 42 42 62 62 71 71 75 81 87 88 89 91 96 102 107 APPENDIX C: Assay for Estimation of Sulfhydryl Content............ 110 LIST OF REFERENCESOOO0.0000000000000000000000000000000000000000000 115 LIST OF TABLES TABLE IEEEE 1. Total and CN--insensitive rates of oxygen consumption...... 41 2. Assayed levels of superoxide dismutase activity............ 45 3. Superoxide dismutase activity based upon individual EDSO values.O..0..0....00...0.0I.OOIOOOOOOOOOOOOOOOOOOOO 52 4. Superoxide dismutase activity based upon potency ratios to standard preparations................................ 53 5. Assayed levels of cytochrome oxidase activity in selected tissues from the rainbow trout.......................... 57 6. Mitochondrial superoxide dismutase activity expressed per unit of cytochrome oxidase activity................. 57 7. Assayed levels of glutathione peroxidase activity and sulfhydryl content in selected tissues from the rainbow troutOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 66 vi LIST OF FIGURES FIGURE Page 1. The excitation and reduction products of molecular oxygen..... 10 2. Total and CN--insensitive respiration in rainbow trout tissue homogenates...OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 44 3. Percent inhibition of ferricytochrome c reduction as an assay of superoxide dismutase activity........................... 49 4. Percent inhibition of ferricytochrome c reduction vs. coded log dose for the determination of superoxide dismutase actiVitYOOOOOOOOOOOO0......0.0...0.000000000000000000000000 51 5. Superoxide dismutase activity of tissue homogenates from the rainbow trout.0.0.00......0..OOOOOOOOOOOOOOOOOOOOOOOOO. 55 6. Cytochrome oxidase activity of tissue homogenates from the rainbow troutOOOOOOOOOOO0.00.0000...OOOOOOOOOOOOOOOOOOOOOO. 59 7. Mitochondrial superoxide dismutase activity from the rainbow trout tissue homogenates expressed per unit cytochrome OXidaseact1v1ty...coo00000000000000...0.000000000000000... 61 8. Catalase activity determined in crude tissue homogenates from the rainbow trout...OOOOOOOOCOOOCOOOOOOOOOCOOOOOOOOOOO 64 9. Glutathione peroxidase activity measured in tissue homogenates from the rainbow troutOCOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 68 10. Total sulfhydryl and glutathione content of tissue homogenates from the rainbow troutOOOOOOOO0.0..00...OOOQOOOOOOOOOOOOIOO 7O vii INTRODUCTION The rainbow trout eye possesses a structure at the back of the retina, the choroidal rete mirabile, which, in conjunction with the choriocapillaris, has the ability to concentrate oxygen. The oxygen tension at the back of this tissue has been reported to be as high as 750 mmHg (Desrochers et a1., 1985). Exposure to oxygen tensions in the range of 425-760 mmHg has been shown to be toxic to the mammalian eye (Noell, 1955a, 1955b; Yanoff et a1., 1970). The teleost retina however, appears to have adapted to the hyperoxia generated at the back of the eye. In addition, the rainbow trout retina has consistently been shown to be resistant to oxygen tensions even greater than the high physiolog- ical concentration within the trout eye (Baeyens, Hoffert and From, 1973, 1974; Ubels, Hoffert and Fromm, 1977; Ubels and Hoffert, 1981). The oxygen molecule itself is not very reactive, however it can accept electrons in univalent steps to form three reactive intermediates. The successive addition of single electrons to oxygen will yield the super- oxide anion (0;), hydrogen peroxide (H202), and the hydroxyl radical ('0H) (Weiss, 1986). Each of these oxygen metabolites can be generated in aerobic cells and potentially can act as toxic oxidants. There are multiple sources for the formation of these reactive metabolites in vivo. Potential oxidant sources include components of the electron transport chain of the mitochondria, endoplasmic reticulum, and nuclear membranes. Various soluble enzymes and proteins are also capable of 1 2 reducing oxygen in one or two electron reduction steps (Freeman and Crapo, 1982; Weiss, 1986). All three of the reactive oxygen metabolites discussed have been implicated in oxidative injury to cells (Freeman and Crapo, 1982; Fridovich, 1983). Damage attributed to these reactive metabolites includes the inactivation of proteins via the oxidation of critical sulfhydryls, the oxidation of polyunsaturated fatty acids with- in membrane phospholipids and strand scission of nucleic acids (i.e., DNA and RNA). There are several defense mechanisms to protect tissues against oxi- dative damage. One key defense is to limit the concentrations of the I and H O . 'Because 0' ‘and H O are 2 2 2 2 2 2 thought to play a primary role in oxidative damage, either by acting intermediates formed, i.e., 0 directly or via the generation of even more reactive intermediates, defenses designed tn) control the concentrations of these two primary oxidants are of critical importance. The superoxide dismutase enzymes strictly regulate the concentration of superoxide generated within the cell. There are two forms of the superoxide dismutase (SOD) enzyme com- monly found in eukaryotic organisms. One type, a copper/zinc form of SOD, is found in the cytoplasm, whereas a manganese containing form is generally associated with mitochondria (Fridovich, 1986b). Two other antioxidant enzymes, catalase and glutathione peroxidase, provide impor- tant regulation of the H concentration (Chance, Sies and Boveris, 202 1979). It is believed that the 'OH radical is formed by processes which are dependent on 0; and H202 (Fridovich and Freeman, 1986). However, there are in) known enzymatic regulators for specifically reducing the hydroxyl radical concentration. Thus, limiting the concentration of the reactants (i.e. 0; and H202) limits formation of the product ('OH). 3 Currently, oxygen toxicity is thought to be due to the production of reactive oxygen metabolites at a rate that is in excess of the capacity of the cellular defense mechanisms to inactivate them (Jamieson et a1., 1986). Indeed, in the most extensively studied model, i.e., hyperoxic lung injury, there is evidence to indicate that production of the reac- tive oxygen metabolites is increased in the rat lung during hyperoxic exposure (Freeman and Crapo, 1981; Freeman, Topolosky and Crapo, 1982; Turrens, Freeman and Crapo, 1982; Turrens et al., 1982b), and synthesis of the primary antioxidants is induced (Kimball et a1., 1976). Given current views on the free radical theory of oxygen toxicity, two principal "strategies" that may account for the resistance of the trout eye to oxygen toxicity were considered: 1) a decreased propensity to generate reactive oxygen metabolites under hyperoxic conditions, and/or 2) the presence of a highly efficient antioxidant defense system within this tissue. Thus, in the present study, the rate of formation of the reactive metabolites by the trout retina was estimated in norm- oxic and hyperoxic conditions from a quantitative analysis of oxygen consumption in the presence of cyanide (Freeman and Crapo, 1981). In addition, the activities of the key antioxidant enzymes i.e., CuZn and Mn superoxide dismutases, catalase, and the glutathione/glutathione per- oxidase system were also measured in order to determine whether the trout retina had elevated levels of these primary antioxidants. Either of these factors or a combination of both, could provide a biochemical rationale for the resistance of the trout retina to oxygen toxicity. LITERATURE REVIEW Oxygen Toxicity Oxygen toxicity is a pathological condition which develops upon exposure to an elevated oxygen tension for an extended period of time. The severity of the condition is dependent upon the degree of hyperoxia and the duration of exposure. The gross manifestations of oxygen toxic- ity have been observed principally in the lung under hyperoxic condi- tions, a likely site since the lung is in direct contact with the high- est partial pressures of inspired oxygen. Most susceptible is the pul- monary vasculature, especially the capillary endothelial cells (Deneke and Fanburg, 1980; Crapo, 1986). The damaging effects of oxygen have also been demonstrated in a variety of other systems including the cardiovascular, renal, visual and central nervous systems (Haugaard, 1968; Balentine, 1982). Since the focus of this research is on the rainbow trout retina, this review will be limited to the toxic effects of oxygen on retinal tissues. Interest in the deleterious effects of oxygen on the eye began in the early 1950's with the realization that excessive oxygen administra- tion to premature infants caused retinal damage and eventual blindness, a condition which became known as retrolental fibroplasia (RLF) (Patz, Hoeck and De La Cruz, 1952; Patz, 1954). It was observed that this condition was due to an effect of oxygen on the immature retinal ves- sels. Specifically, there was an interruption. of 'retinal ‘vascular 4 5 development in hyperoxia, which was followed by prolific neovasculariza- tion on return to room air. In severe cases growth of the vasculature into the vitreous led to scar tissue formation and retinal detachment (Patz, 1984). These findings were confirmed experimentally in kittens, young rabbits and young mice (Ashton, Ward and Serpell, 1953; Gerschman et al., 1954b; Ashton, 1968). In the course of studying RLF several investigators observed that oxygen was also damaging to the choroidal vasculature and visual cells in eyes from several adult animal species. Yanoff, Miller and Waldhau- sen (1970) noted uveal and retinal detachments in 10 of 14 eyes removed from dogs exposed to 1 atmosphere of 100% oxygen for 48 to 50 hours (PaO 425-664 mmHg). They discovered that the initial damage was to the 2 choriocapillary endothelial cells, and that detachment was secondary to fluid leakage from the choroidal vasculature into the subretinal space. The visual cells of the adult rabbit were also found to be suscep- tible to the effects of hyperoxia (Noell, 19553, 1955b). In these studies adult rabbits were exposed to varying concentrations of oxygen (60-1001) at ambient pressure. Using the electroretinogram (ERG) as a measure of visual cell function, Noell noted a marked attenuation of the ERG, which occurred progressively earlier with increasing oxygen concen— trations. ‘Histological examination revealed selective degeneration of more than 70% of the photoreceptor cell population after 48 hours of 1002 02 exposure. Irreversible damage to the visual cells was observed in animals exposed to only 55-60% 0 at ambient pressure for a period of 2 just 7 days. It has also been reported that hyperbaric oxygen exposure (HBO) accelerates the onset of toxicity to the retina in vivo (Noell, 1962; 6 Bridges, 1966). Using the b-wave of the ERG as an index for quantifying the effects of oxygen, both investigators reported a greater reduction in amplitude of the rabbit ERG at higher pressures of 1007. oxygen. Noell (1962) again reported selective visual cell layer death after exposure of rabbits to 100% O at 3 atm for just 4 hours. The inner 2 layers were unaffected histologically, but the outer nuclear layer was reduced in thickness and number of nuclei, and deterioration of the photoreceptor outer and inner segments was observed. These findings were confirmed by Bresnick (1970), who studied electron micrographs of the adult rabbit retina after the animals were exposed to 100% 02 for 24-40 hours. The earliest ultrastructural changes observed immediately after 02 exposure were to the inner segments and nuclei of the photo- receptor cells. Swelling of various cellular organelle membranes was extensive with marked vesiculation followed by complete disorganization and destruction of the entire visual cell. Ubels, Hoffert and Fromm (1977) also used the ERG to investigate the toxic effects of HBO on the retina, however they performed their studies in vitro. Their design enabled them to discern whether the ERG changes were due to changes to the retinal vasculature or to the retinal cells themselves. They exposed the isolated eyecups from frogs and rats to 5 atm of pure oxygen and reported a marked attenuation of the b and c waves within 6 hours in the frog and severe attenuation of the a and b waves within 90-105 minutes in the rat, which led them to conclude that oxygen did indeed have a direct toxic effect on the retinal cells. Much of the early work on retinal oxygen toxicity was strictly descriptive with little understanding of the mechanism of toxicity. Several possibilities had been proposed based on the generalized 7 damaging effects that oxygen had (”1 cells from other tissues. These included: 1) the oxidation of sulfhydryl groups, 2) .alterations to several metabolic pathways and 3) the oxidation of polyunsaturated fatty acids (Bridges, 1966). To date, the same possibilities for mechanism of damage remain under investigation (Freeman and Crapo, 1982; Weiss, 1986). A good part of the earlier work aimed at determining the mechanism of oxygen toxicity was performed by Haugaard and associates. Decreased metabolism of various tissue slices from the rat had been demonstrated after exposure to 'HBO (Stadie, Riggs and Haugaard, 1945). Shortly thereafter, Haugaard (1946) was able to establish a link between inact- ivation of enzymes containing essential sulfhydryl groups and oxygen exposure, concluding that inactivation occurred through sulfhydryl oxi- dations. It was later reported that inhibition of glycolysis in rat heart homogenates occurred through reversible oxidation of sulfhydryl groups in glyceraldehyde-3-phosphate dehydrogenase after exposure to 1 atm oxygen with resultant effects on metabolism (Horn and Haugaard, 1966). Oxygen may also exert toxic effects by oxidizing nonprotein sulfhydryls essential in maintaining enzyme sulfhydryls in their reduced state and in preventing the accumulation of mixed disulfides which are able to exert deleterious effects on cell metabolism (Haugaard, 1968). More specifically, hyperoxia has also been reported to interrupt normal cellular metabolism in retinas from several species. Hyperbaric oxygen exposure is responsible for decreased oxidative metabolism in dog retinas (Baeyens, Hoffert and Fromm, 1973), and for inhibition of lac- tate dehydrogenase activity in frog and dog retinas (Baeyens, Hoffert and Fromm, 1974). The effect of hyperbaric oxygenation on Na+-K+ ATPase 8 activity was also studied (Ubels and Hoffert, 1981). This enzyme con- tains critical sulfhydryl groups and is essential for normal retinal function. A 507. decrease in enzyme activity after exposure of rat retinas to 5 atm of pure 0 for 4 hours was reported. In light of the 2 effects of oxygen on the ERG they proposed that the toxicity may be due to an inability to maintain transmembrane potentials. In addition to protein targets in the retina, the oxidation of poly- unsaturated fatty acids (PUFA) had also been proposed as a mechanism for oxygen toxicity. In fact, Yagi and associates have reported that lipo- peroxide levels are elevated in rabbit and chick embryo retinas after exposure to high concentrations of oxygen (Hiramitsu et a1., 1976; Yagi and Ohishi, 1977). This elevation of lipoperoxide was observed prior to changes in the electroretinograms. In summary, the gross manifestations of retinal oxygen toxicity appear to be due to: 1) an interruption of retinal vascular development in the immature retina, 2) oxidative inactivation of various enzymes, 3) oxidation of membrane-associated PUFA, 4) a decrease in cell function, as indicated by changes in enzyme activities and oxidative metabolism and 5) selective visual cell degeneration and cell death. Chemistgz.gfuggzggg_and its Reduction Products Molecular oxygen is a strong oxidant but exhibits limited reactivity because of its unusual electronic configuration. Oxygen in the ground state contains two unpaired electrons with parallel spins in its outer- most orbitals. This configuration presents a kinetic barrier to reac- tion, since a change in spin state is required in order for oxygen to oxidize a two electron donor. Spin inversion however, is unlikely since it is a rather slow process in comparison to the lifetime of the 9 collisional complex formed between oxygen and a two electron donor. In contrast, the spin constraints do not apply to reactions with single electron donors or atoms or molecules containing unpaired electrons (Fridovich, 1974; 1977; Hill, 1979; Weiss, 1986). Because of the spin restriction, oxygen reduction can generally be categorized into two types of reactions, enzyme-catalyzed multivalent reduction (e.g., mito- chondrial cytochrome oxidase) and enzyme-independent univalent reduc- tion. The univalent reduction of oxygen is favored and commonly occurs in biological systems (Fridovich, 1977; 1978). Molecular oxygen can accept a total of four electrons in either univalent or multivalent steps. If this reduction process were to take place in univalent steps, three reactive metabolites of oxygen would be generated (Figure 13 'Hill, 1979). Successive univalent reduction of dioxygen would result in the formation of the superoxide radical (0;), hydrogen peroxide (H202) and the hydroxyl radical ('OH) (Fridovich, 1977; Weiss, 1986). The superoxide radical may also be protonated to T 2 radical is a stronger form the perhydroxyl radical (H05), the conjugate acid of the O radical, with a pKa of 4.8 (Hill, 1979). The H05 oxidant than the 0; radical but at a physiological pH of 7.4 the super- oxide anion will predominate. A two electron addition to molecular oxygen results in the formation of the peroxide anion (OE-2), but at physiologic pH, it is rapidly protonated to form hydrogen peroxide (H202). Addition of the third electron to molecular oxygen results in formation of the hydroxyl radical ('OH). Finally, addition of the fourth electron results in the formation of the hydroxyl anion, which when protonated forms water. One other potentially damaging species of oxygen, singlet oxygen (102), is formed by the conversion of ground 10 state molecular oxygen to a more excited singlet oxygen state either through photochemical or thermal processes (Green and Hill, 1984). All of the above intermediate species of oxygen are oxidants capable of causing cellular damage. There is now evidence to indicate that these species are generated in vivo under a variety of pathophysiological conditions. hv 1 O2 ) 02 e- 07 < DKa = [‘08 > HO; fe- 032 <—E§§-Z-L£-> H0; < pKa = 11'8 > H202 19’ 'OH + OH- le‘ OH— Figure l The excitation and reduction products of molecular oxygen. Cellular Sources and Reactivity 2f Various Oxygen Metabolites There are numerous intracellular sources of the superoxide radical, including several soluble enzymes and proteins, bound proteins associa- ted with electron transport systems and small. molecules capable ‘of releasing superoxide through autoxidation reactions (Freeman and Crapo, 1982; Fridovich, 1983). Xanthine oxidase is the most widely studied superoxide-generating, soluble enzyme. ‘Under normal conditions 90% of its activity is as an NAD+-dependent dehydrogenase (type D form) which 11 1' 202 or 02 (McCord, 1985). Proteolytic modification of the enzyme during purifica- cannot transfer electrons to molecular oxygen to form H tion or during in vivo ischemia converts it from the dehydrogenase form to the superoxide generating oxidase form (type 0 form) (Roy and McCord, 1983). The relative proportion of 0' and H released from the active 2 202 site is dependent on the pH, oxygen concentration and substrate concen- tration (Fridovich, 1970; Nangano and Fridovich, 1985). Therefore under hyperoxic conditions an increased flux through the oxidase form of the enzyme would result in a greater production of the superoxide radical. 'I' 2 in their cycling reactions include aldehyde oxidase, dihydro-orotic Other soluble enzymes and proteins which produce 0 as an intermediate dehydrogenase, flavoprotein dehydrogenase, tryptophan dioxygenase, hemo- globin and myoglobin (for review see Fridovich, 1978; Freeman and Crapo, 1982). It has also been proposed that various components of the electron transport systems in mitochondria, endoplasmic reticulum and nuclear membranes are responsible for the formation of superoxide anions. Flohe et a1., (1977) have presented evidence suggesting that mitochondrial H202 is pmimarily, if not exclusively, produced from superoxide radi- cals. In an attempt to identify the site of formation, electron trans- port chain inhibitors were used on uncoupled, intact mitochondria and on submitochondrial particles. It was soon recognized that the ubiquinone- cytochrome b region, when fully reduced, readily reacted with oxygen 566 and was capable of producing H Their data as well as that of Nohl 202' and Hegner (1978), suggested that cytochrome b 566 was responsible for the formation of superoxide and that H production resulted from the 202 dismutase reaction. The ubiquinone-cytochrome b region has been 12 confirmed as a superoxide generating site by several investigators using heart and lung mitochondria from bovine, rat and porcine sources (Turrens and Boveris, 1980; Turrens et al., 1982b; Turrens, Freeman and Crapo, 1982a), however they have implicated ubisemiquinone as the aut- oxidizable molecule responsible for the 0; production within this region. Several investigators have also identified the enzyme NADH 1' 2 respiratory chain (Tyler, 1975; Turrens and Boveris, 1980; Turrens et dehydrogenase as a second 0 generating site along the mitochondrial al., 1982b; Turrens, Freeman and Crapo, 19823). Due to the rapid elec- tron transfer rates between intramolecular redox groups, they found that 1' 2 itself or at the iron-sulfur clusters associated with the enzyme. In it was difficult to ascertain whether 0 formation occurs at the enzyme either case, at least two 0; (generating regions have been identified along the mitochondrial electron transport chain when it is in a reduced state (Boveris and Cadenas, 1982). The electron transport system of the endoplasmic reticulum has also 1' 2 implicated in a number of hydroxylation reactions (Halliwell, 1977). In been identified as a source of 0 formation, and its generation has been intact rat liver microsomes, as well as with purified enzyme extracts, superoxide anions are generated at the level of both NADPH-cytochrome P reductase and cytochrome P (Bartoli et a1., 1977; Rosen, Finkel- 450 450 stein and Rauckman, 1982). Since the microsome is an artificially pro- duced organelle it is not known whether the 0; radical remains bound to the active site in the complex or if it is released into the free solu- tion in ‘vivo, however the results do indicate that the respective 1' 2 addition to the electron transport sites, there are also superoxide enzymes can interact with oxygen to yield 0 (Halliwell, 1977). In 13 generating enzymes associated with the microsomal and plasma membranes (Freeman and Crapo, 1982). The NADPH oxidase associated with the plasma membrane of phagocytes is also a well studied biologic source of super- oxides (Weiss and LoBuglio, 1982; Weiss, 1986). There are also several endogenous compounds which are capable of undergoing autoxidation reactions to produce superoxide radicals. Included among these are catecholamines, reduced cytochrome c, reduced ferrodoxins, hydroquinones, leukoflavins and thiols (for references and review see Fridovich, 1978; Freeman and Crapo, 1982; Fridovich, 1983). Whenever superoxide is generated in an aqueous system, hydrogen peroxide will also readily be formed through the dismutation reaction: 03 + o; + 21L:+ ------- > H202 + 02 (1) Over the biochemical pH range this reaction is quite rapid. The pre- viously mentioned sites of superoxide generation would therefore also be considered sites of H 0 generation. In fact, as stated, most if not 2 2 all of the H202 generated by mitochondria is due to the dismutation of O? In addition to the mitochondrial and microsomal sites, H202 is 2. generated in peroxisomes without the formation of the superoxide radical. There are several peroxisomal enzymes capable of generating H202 including; D-amino acid oxidase, fatty acyl-CoA oxidase and urate oxidase (Chance, Sies and Boveris, 1979). I and H 0 formation within the cell is evident, 2 2 2 but the reactivity of these two species within an aqueous environment The potential for O has been questioned (Sawyer and Valentine, 1981). Both 0; and H202 are unstable with respect to their dismutation products. Since the anions are stabilized by protons or cations, equilibrium reactions will be sensitive to pH, metal ions and solvent. The immediate surroundings at 14 the site of generation of these species will have a major impact on their stability and reactivity (Hill, 1979). Superoxide in aqueous solution can act as a moderately effective one electron reducing agent (e.g., reducing ferricytochrome c and various transition metal ion complexes) or as a selective oxidizing agent acting on reducing substrates with readily transferable hydrogen atoms (e.g., ascorbate, a-tocopherol, hydrazines and hydroxylamines) (Fee and Valen- tine, 1977; Halliwell, 1979; Hill, 1979; Sawyer and Valentine, 1981). It is 'not, 'however3 a ‘highly' reactive electron, transfer' oxidant of organic or inorganic substrates unless the peroxide anion formed can be stabilized. Furthermore, the reactivity of the superoxide radical with a given substrate is limited because of the rapidity with which it can dismutate to form oxygen and peroxide (Sawyer and Valentine, 1981). Although the view of superoxide as a dangerous species has been tempered because of its low reactivity, Fridovich (1986a), has cited several studies which have shown it to be directly responsible for various damaging effects, including the inactivation of catalase and glutathione peroxidase. The superoxide anion can also be protonated to form the perhydroxyl radical, a much stronger oxidant with a pKa of 4.8. Normally at neutral pH only about 1 percent will exist in this form (Weiss and LoBuglio, 'I' 2 surfaces would favor protonation, since the polyanionic surfaces attract 1982). However the generation of O in areas adjacent to membrane a high H+ concentration, (i.e., 3 pH units lower than the bulk solvent), and since protonation of the superoxide radical would occur as movement into the hydrophobic interior of membrane anion channels took place (Freeman and Ckapo, 1982; Fridovich, 1983). Therefore, to reiterate, 15 the environment (i.e., the pH and solvent) in which the superoxide radical is generated is extremely important in determining its reactivity. The H202 produced from the dismutation reaction is a very powerful oxidant, but reactions with organic substrates are slow. Reactions with transition metal ions and complexes however occur much more readily and may be responsible for the formation of a more reactive species of oxygen, i.e., the lwdroxyl radical (Fee and valentine, 1977; Chance, Sies and Boveris, 1979). Haber and Weiss had also proposed formation of the hydroxyl radical, as a result of direct interactions of the superoxide radicals with hydrogen peroxide (Willson, 1979). This reaction however has come under considerable debate (Fee and. Valentine, 1977; Cohen, 1977; Willson, 1979). There is no chemical evidence to showr that superoxide can directly reduce hydrogen peroxide in aqueous solutions in vitro (Fee and Valentine, 1977). However, there are numerous biological studies in which the product formed or effect observed was consistent with the formation and chemistry of the 'OH radical (Cohen, 1977). Pure 0; and H202 will not react together to form hydroxyl radical unless catalyzed by complexes of iron or some other transition metal (Fee and Valentine, 1977; Cohen, 1977). The following reactions have been proposed for ‘hydroxyl radical production in biological systems (Czapski, 1984): 1' 2 Fe (II) + H202 ----> Fe (III) + on" + 'OH (3) Fe (III) + o ----> Fe (II) + 02 (2) 02 + H202 ----> 02 + on + OH (4) 16 In these reactions, superoxide donates an electron to the ferric ion and in turn the ferrous ion then reduces H202 to form 'OH. The overall net reaction has been termed the iron catalyzed Haber-Weiss reaction. The availability of free or low molecular weight chelated forms of iron within cells may be limited or non-exsistent, so this reaction remains speculative in vivo. However, in a model microsomal system, it has recently been shown that superoxide is capable of releasing iron from ferritin, thereby potentially raising the low molecular weight iron pool available for 'OH formation (Thomas, Morehouse and Aust, 1985). The availability of iron or other transition metals to act in this reaction is an area of current investigation (for reviews see Halliwell and Gutteridge, 1984; Weiss, 1986). Hydroxyl radical formation can also occur in aqueous solution by the photolytic decomposition of H with ultraviolet radiation (Fee and 202 Valentine, 1977), a reaction of considerable importance in the eye. As the hydroxyl radical is a very powerful one electron oxidant, it is highly and indiscriminately reactive (Fee and Valentine, 1977). It will react with almost every type of molecule found in living cells, acting primarily through hydrogen atom abstractions, addition reactions (e.g. by adding to purine and pyrimidine ring structures of DNA), and electron transfer reactions (Willson, 1979; Halliwell and Gutteridge, 1984). This high reactivity however limits its lifetime as it will react with whatever biological molecule is in its vicinity, producing secondary radicals of variable reactivity (Halliwell and Gutteridge, 1984). It is believed that singlet oxygen formation by means of photody- namic action does occur in biological systems and would be of importance 17 in the eye (Fridovich, 1984). Photooxidation can occur by direct oxida- tions, whereby a molecule is photoexcited and an electron released which can subsequently react to form superoxide; or by sensitized oxidations in which excitation of a: photosensitizer molecule indirectly promotes oxidation of a second molecule (Foote, 1982). In sensitized oxidations, absorption of light promotes an electron to a higher energy orbital, which, after spin inversion, is termed a sensitizer triplet. Since the electrons in these orbitals are bound less strongly, they are readily removed by oxidizing agents. There are two basic types of photosensi- tized oxidation reactions (Foote, 1982). In type I reactions, the trip- let sensitizer reacts directly with a substrate and allows for transfer of a hydrogen atom or an electron, such as with hydrogen abstraction from polyunsaturated fatty acids. The newly formed radicals can then undergo further reaction with oxygen or an organic molecule. In type II reactions, the sensitizer triplet interacts with oxygen via energy transfer to yield singlet molecular oxygen, which can then further react with various acceptors in solution. Two good singlet oxygen acceptor systems involve additions to olefins and dienes to form hydroperoxides and endoperoxides respectively (Foote, 1982). These types of oxidation reactions are very concentration dependent, since there is competition between substrate and oxygen for the sensitizer triplet. Low substrate concentration and high oxygen concentration would favor Type II reac- tions (Foote, Shook and Abakerli, 1984). In summary, there is sufficient evidence to show that formation of the various oxygen metabolites does occur in biological systems. In addition to the above mentioned reactive oxygen metabolites, there are numerous other species (e.g., ferryl, perferryl, alkoxy and peroxy l8 radicals) which may be generated as a result of secondary reactions (Hill, 1979; Willson, 1979; Halliwell and Gutteridge, 1984). The poten- tial for oxidative damage rests not only on those species with limited reactivity (0; and H O ) but also with the numerous secondary radicals 2 2 which may be formed. Protection Against Oxyggg_roxicity Enzymatic Scavenging 2£_Superoxide Biological interest in free radical chemistry grew tremendously with the discovery that the protein erythrocuprein was responsible for enzy- matically scavenging the superoxide radical (McCord and Fridovich, 1969). The accumulation of superoxide radicals was prevented by catal- ysis of the dismutation reaction to form H202 and 02, and the protein was therefore aptly renamed superoxide dismutase: 1' 1' + SOD 02 + 02 + 2H > H202 + 02 (5) The catalyzed reaction functions at a rate of 2 x 109 M-l.sec-1 at 25 C, which is 104 fold faster than the spontaneous dismutation reaction (Fridovich, 1983). There are three classes of superoxide dismutases common to respiring cells. All three catalyze the same reaction with comparable efficiency at rates which approach the diffusion limit (Fridovich, 1986b). The iron containing (FeSOD; 46 Rd) and manganese containing (MnSOD; 46 Kd) enzymes are characteristic of prokaryotes and are very similar in amino acid sequence. These enzymes are frequently dimeric containing one metal atom per subunit. The MnSOD form is also found in eukaryotes, primarily within the mitochondrial matrix, differing from prokaryotes only in that they are tetrameric. The class of SOD's containing 1 metal atom each of copper and zinc per subunit (CuZnSOD; 32 Rd) is usually 19 dimeric and commonly found in the cytosol of eukaryotes. At high molecular weight tetrameric form of this enzyme (135 Rd) has been found in human tissues and since it constitutes 90% of the SOD found in plasma it has been referred to as an extracellular SOD or ECSOD (Fridovich, 1983; Fridovich, 1986b). The three classes of SOD are distinguishable by their sensitivities to various reagents. The CuZnSOD form is inhibi- ted by 5 mM cyanide while the MnSOD and FeSOD forms are resistant to it. Hydrogen peroxide inactivates CuZnSOD and FeSOD but has no effect on MnSOD. Sodium dodecyl sulfate (2%) can be added to eliminate MnSOD activity without affecting CuZnSOD, and lastly alkaline pH (pH 10.0) can be used to suppress MnSOD activity without affecting CuZnSOD activity. A measure of activities at pH 7.8 and 10.0 would allow for the calcula- tion of both of these SOD forms (for references see Fridovich, 1986b).. The mechanism of action for all of the superoxide dismutases can be written: E - Men + 0; ----- > E - Men-l + 02 (6) E - Men-1 + o; ----- > E - Men - o;2 (7) + + n -2 H n - H E - Me - o2 ----- > E - Me + H02 -----> H202 (8) where E denotes the enzyme and Me the metal ion. In the CuZnSODs it is the copper that functions in electron transport during the catalytic cycle, while the zinc is thought to primarily play a structural role lending stability to the molecule. It has been proposed that the first superoxide anion at the active site reduces the metal ion and disrupts an imidizolate ligand bond. The released imidizolate is highly basic and would then give up that proton when a second superoxide anion enter- ed the active site and reoxidized the active metal ion to form HO2 as 20 product. Once released from the active site the H0; could then be protonated to form H in solution (Fridovich, 1986b). 202 Only a brief discussion of the enzyme has been presented here, since more extensive reviews are available in the literature (Fridovich, 1978, 1983, 1986b). Enzymatic Scavengiggfigf_flydrogen Peroxide As previously mentioned, hydrogen peroxide is slow to react, but it is capable of participating in the generation of the reactive hydroxyl radical. Accumulation of hydrogen peroxide is prevented by two related enzymes, the catalases and the peroxidases. Both catalyze the reduction of H202 to water. The enzymes differ in that catalases can use H202 as the source of electrons for the reduction process, whereas peroxidases use some other reductant. Catalases can also act as peroxidases using other electron donors when the concentration of H O is low. The res- 2 2 pective reactions are given below: Catalase H202 + H202 ---------- > 2H20 + 02 (9) Peroxidase H202 + H2R > 2H20 + R (10) The physiological reductant for peroxidases may be glutathione, ascor- bate or cytochrome c, and these would be continously renewed by NADH and NADPH (Fridovich, 1977). The peroxidase activityr of catalase also requires an electron donor but the compound used in vivo is not known (Fridovich, 1977). The catalase enzyme (240 Rd) is tetrameric with an Fe3+ - protopor- phyrin ring at its active site. It is found in varying concentrations in tissues and is reportedly highest in liver and kidney where it is localized in peroxisomes. In other tissues it may be found in microper- oxisomes or it may be distributed throughout the cytosol, e.g. 21 erythrocyte catalase (Chance, Sies and Boveris, 1979). The most common peroxidase is the selenium dependent form of gluta- thione peroxidase (GSHPx) which has four atoms of selenium per molecule in its active site. It is a 76 Rd enzyme composed of 4 subunits. It catalyzes the reaction of hydrogen peroxide with reduced glutathione (GSH) to form oxidized glutathione and water. Although it is specific for its hydrogen donor, GSH, it is non-Specific for the hydroperoxide and can react with a range of substrates from H to organic hydroper- 202 oxides. It is widely distributed with variable tissue concentrations, and is located in the cytosol and mitochondrial matrix but not in per- oxisomes (Chance, Sies and Boveris, 1979). The complimentary distribu- tion of these two enzymes apparently acts to effectively control the intracellular concentration of hydrogen peroxide. Secondary Antioxidant Defenses Although superoxide dismutases, catalase and glutathione peroxidase provide the first line of defense in the cell by preventing 0; and H202 from generating more reactive oxidants, other defense mechanisms have been described. Should the primary defense system become overwhelmed or if oxidant attack should occur at :1 site inaccessible to the primary antioxidants, then several other lines of defense against oxidative attack are available. The multilayered system of defense which has evolved to counter oxidative attack has recently been discussed (Fridovich and Freeman, 1986). As previousLy mentioned, glutathione peroxidase is capable of detoxifying lipid hydroperoxides (the byprod- ucts of lipid free radical reactions) and can therefore act as a secon- dary defense mechanism. Zhi addition to its role as substrate for the glutathione peroxidase enzyme, glutathione can also serve as a secondary 22 antioxidant. Glutathione and other thiol (SH) compounds exert a protec- tive action against oxidative damage by undergoing non-enzymatic hydro- gen abstraction to restore function to enzymes which have been oxidized. Glutathione may also serve as a nucleophile in conjugation reactions involving GSH transferase (Mannervik et a1., 1983). The thiol transfer- ases catalyze thiol interchange between glutathione and protein disul- fides or low molecular weight disulfide substrates to form oxidized glutathione: GSH + GS-S-R Thi°1 TranSferase > cssc + HS-R (11) The regeneration of GSH from mixed disulfides occurs by way of conver- sion to glutathione disulfide, which is then reduced by glutathione reductase utilizing NADPH (Mannervik et a1., 1983). Other secondary defenses include the antioxidants ascorbic acid and Vitamin E, which function as scavengers through hydrogen abstraction reactions (Nishikimi, 1975; Bielski, Richter and Chan, 1975; Tappel, 1972; Packer, Slater and Willson, 1979). There are a number of other mechanisms utilizing numerous enzymes, small molecules, metal ions, etc., which can act to prevent, limit and repair oxidative damage to cellular constituents. The primary antioxi- dants discussed above, represent the most commonly studied mechanisms and provide background for this study. éLPotential Role for Oxyggg_uetabolites'12_Hypgroxic Injury It has long been presumed that oxygen metabolite formation would be increased under hyperoxic conditions in vivo. Fridovich has demonstra- ted an increased rate of superoxide anion formation by the isolated enzyme xanthine oxidase ‘with increasing oxygen tensions (Fridovich, 1970; Nangano and Fridovich, 1985). However, it is difficult to obtain 23 measurements of the steady state levels of the reactive oxygen metabol- ites in biological systems because of their very short half lives. There is now sufficient indirect evidence to show that hyperoxic expos- ure can indeed increase the rate of formation of superoxide and hydrogen peroxide. Freeman and Crapo (1981), have used the method of CN--insensitive I and H 0 2 2 2 production in rat lung slices, since oxygen consumption in the presence respiration to obtain an approximate index of intracellular 0 of cyanide (which inhibits mitochondrial cytochrome oxidase), is repre- sentative of the one and two electron reduction of oxygen. They found that hyperoxic conditions will increase the proportion of CN--resistant respiration in the lung preparations relative to total oxygen consump- tion measured in the absence of cyanide. Cyanide insensitive respira- tion increased linearly as rat lung slices were incubated in various concentrations of oxygen from 5% to 95% 0 Isolated mitochondria were 2. also found to increase their rate of CN--insensitive respiration from 0 nmol 0 consumed/min/mg protein at 15% () to 1.34 nmol 0 consumed/ 2 2 2 min/mg nfltochondrial protein at 85% oxygen (Freeman and Crapo, 1981). Similar results were seen when lung homogenates were subjected to measurements in hyperoxia, with CN--insensitive respiration accounting for 7% of total respiration in air and 17% of total respiration when incubated in 80% oxygen (Freeman, Tapolosky and Crapo, 1982). Thus, it 'I' 2 H202 and implied an essential role for these metabolites in hyperoxic was concluded that hyperoxia enhanced the cellular production of 0 and lung damage (Freeman and Crapo, 1981; Freeman, Topolosky and Crapo, 1982). Superoxide production under hyperoxic conditions was also 24 investigated in submitochondrial particles that had been washed free of superoxide dismutase (Turrens et 31., 1982b). The cyanide insensitive respiration, epinephrine oxidation and cytochrome c reduction methods were used to quantitate superoxide production in lung submitochondrial particles. Superoxide production increased linearly with increasing oxygen tensions in porcine lung submitochondrial particles above an O2 tension of zero, and in rat lung submitochondrial particles above an oxygen concentration of 40%. Hydrogen peroxide release by nfltochondria and ndcrosomes has also been reported 11) be increased under hyperoxic conditions (Boveris and Chance, 1973; Chance, Sies and Boveris, 1979; Turrens, Freeman {and Crapo, 1982). A four fold increase in H202 formation was noted in pi- geon heart mitochondria and a 15-20 fold increase was seen in rat liver mitochondria when the p0 was raised to 19 atm of pure oxygen (Boveris 2 and Chance, 1973). Hyperoxia also caused a 10-fold increase in H202 release by porcine lung mitochondria when the 02 concentration was in- creased from room air to 100% oxygen and 3 linear increase in H202 for- mation by NAD(P)H supplemented microsomes was seen when oxygen concen- trations were increased above 21% (Turrens, Freeman and Crapo, 19823). Although these studies involved indirect, in vitro measurements, it is quite clear that hyperoxia enhanced the rate of formation of the partially reduced forms of oxygen. This evidence supports the current belief that hyperoxia induced oxygen toxicity is due to the production of reactive oxygen species at a rate in excess of the capacity of the cellular defense mechanisms to inactivate them and that the suscepti- bility of 3 tissue to oxygen toxicity would vary depending on the cells defense mechanisms (Jamieson et 31., 1986). 25 Resistance g£_Teleoat Ocular Tissues £2_Hyperoxic Injury In contrast to most tissues, the teleost retina appears to be resis- tant tn) hyperoxic injury; The choroidal rete mirabile at the back of the teleost eye operates in conjunction with the choriocapillaris to generate an oxygen tension which is far in excess of that found in arterial blood (Wittenberg and Wittenberg, 1962; Fairbanks, Hoffert and Fromm, 1969; 1974). Since the teleost retina is avascular, it is depen- dent on this high p0 to maintain normal visual function (Fonner, 2 Hoffert and Fromm, 1973). The oxygen tension profile across the back of the rainbow trout eye has been determined (Desrochers et 31., 1985). The tissue p0 at the choriocapillaris averaged 380 mmHg and 124 mmHg at 2 the retinal-vitreal interface. In fact, a retinal p0 as high as 750 2 mmHg was recorded in one fish. Oxygen tensions in this range have been shown to cause oxygen toxicity in several species (Yanoff et 31., 1970; Noell, 19553, 1955b). Further, the rainbow trout has repeatedly been shown to be resistant to oxygen toxicity. In a comparative study on the influence of hyperbaric oxygen on cellular respiration (Baeyens, Hoffert 3nd Fromm, 1973), dog retinal tissue respiration was significantly decreased after a 24 hr incubation at 2 atm oxygen, while trout retinal consumption was significantly increased. No change in respiration of frog retinal tissues was observed. Liver and brain tissue respiration was inhibited by hyperbaric oxygen in all three species after 24 hr exposures. The fact that oxygen exerted a toxic effect on trout brain tissue provided evidence that not all teleost neural tissue was equally insensitive to oxygen and that the resistance was unique to the retina. It was concluded that inhibition of respiration most likely resulted from inactivation of a critical enzyme of carbohydrate metabolism 26 (Baeyens, Hoffert and Fromm, 1973). A second comparative study by the same authors investigated the influence of oxygen on lactate dehydrogen- ase (LDH) activity (Baeyens, Hoffert and Fromm, 1974). LDH activity was determined in homogenates of retina, liver and brain from trout, frogs and dogs after exposure to hyperbaric oxygen (2 atm) for 24 hrs. All tissue preparations except trout retina exhibited lower LDH activity after hyperbaric oxygen exposure. Ubels, Hoffert and From (1977), provided additional evidence to support the hypothesis that the teleost retina was resistant to oxygen toxicity. Electroretinograms (ERG) were recorded in vitro using frog, rat and teleost retinas. Attenuation of the ERG after exposure to 5 atm oxygen was observed in frog retinas after a period of 6 hours and in rat retinas after just 90 - 105 minutes. Goldfish and trout ERG's however did not exhibit any significant changes attributable to oxygen exposure after the 6 hour period as normal function was observed in the goldfish retina for the duration of the exposure period. Ubels and Hoffert (1981) also demonstrated that trout retinal Na+-K+ ATPase was resistant to oxygen exposure at 15 atm pure oxygen, while Na+-K+ ATPase activity in the rat was inhibited by 66% at this pressure. It appears as though the trout retina is unique in that it is func- tionally dependent on the high p0 generated at the back of the eye and 2 is resistant to oxygen toxicity. The mechanism of this resistance is not yet understood, but in light of the numerous studies describing the toxic actions of oxygen, it must be hypothesized that the resistance is either due to an inability to generate reactive oxygen metabolites in hyperoxia and/or to an efficient antioxidant defense system within this tissue. MATERIALS AND METHODS Experimental Animals Rainbow trout (Salmo gairdneri) weighing 150-450 g were obtained from Balders Fish Farming Enterprise (Baldwin, MI). They were held in fiberglass tanks at 9:1 C, with a continous flow of dechlorinated tap water which was aerated by compressed air filtered through activated charcoal. The fish were fed a maintenance diet of commercial trout pellets and exposed to a photoperiod of 16 h light - 8 h dark. Tissue Preparation Trout were anesthetized with tricaine methane sulfonate (MS-222; Finquel, Ayerst Laboratories, Inc., NY) and secured in :ni inverted position in a holding trough. They were ventilated with water passed over their gills at a rate of 4.2 L/min. An incision was made along the mid-ventral line and the heart exposed. Sodium heparin (0.1 m1, 1000 units/ml; Upjohn Co., Kalamazoo, MI) was injected directly into the ventricle and allowed to circulate prior to perfusion in order to pre- vent clotting. The ventricle was then cut and 3 cannula (PE-50) insert- ed through to the bulbous arteriosus. The atria were cut and Cortland's saline solution (Wolf, 1963; Appendix A) was perfused through the tissues at a rate of 6 ml/min for approximately 8 minutes. Perfusion was stopped when the tissues, particularly the liver and gills, appeared to be free of blood. Because hemoglobin is known to interfere with many of the colorimetric assays used in this study, perfusion of the tissues 27 28 with saline was required prior to enzymatic analysis. The liver, eyes and brain were immediately excised, placed in a beaker containing Cortland's saline solution and kept on ice. Brain and liver tissues were trimmed, blotted dry, weighed and then diluted with the appropriate buffers used in the respective assays. The enucleated eyes were trimmed of the extraorbital muscles and weighed. An incision was made along the corneal limbus so that the iris, cornea and lens could be lifted from the eyecup with a forceps. The vitreous adheres to the retina in the trout, but an attempt was made to remove as much of the vitreous as possible without damaging the retina. The neural retina, pigmented epithelium, perfused choroid body and any remaining vitreous were then scooped from the eyecup, weighed and diluted with the appropriate solutions used in the respective assays. Cyanide Insensitive Oxygen Consumpgion Determinations To characterize the effect of oxygen concentration on the production of partially reduced oxygen species, oxygen uptake by tissue homogenates was measured in the absence and presence of 1.0 mM potassium cyanide, with and without the addition of 1.0 mM B-NADH (Grade II from yeast, Lot #93F-7255, Sigma Chemical Co., St. Louis, MO) (Freeman and Crapo, 1981; Freeman, Topolosky and Crapo, 1982). Oxygen consumption was determined polarographically using 3.0 ml waterjacketed glass chambers (Cole-Palmer Instrument Co., Chicago, Il) fitted with Clark type oxygen electrodes (Model 5331) attached to a YSI Model 53 oxygen monitor (Yellow Springs Instrument Co., Yellow Springs, 0H) and the change in p0 recorded on a 2 dual channel vertical—face recorder (Model 486, Cole-Palmer Instrument Co., Chicago, 11). Temperature was maintained at a constant 15:1 C using a Refrigerated Bath Circulator (Neslab Instruments, Inc., 29 Portsmouth, NH). A general description of the electrode and its prepa- ration is given by Green and Hill (1984). Electrode calibration and stability were determined by immersing the oxygen probes in 3 mls of constantly stirred KCL-KHZPOA incubation medium, pH 7.4 (Appendix A), equilibrated at 15 C and saturated with either room air (21% 02) or 100% oxygen. The probes are held by a Lucite plunger that fit snugly into the glass chambers. There is a small access groove along the length of the plunger, so in order to minimize the diffusion of oxygen out of the polarograph chamber during measurements made under 100% 02, the chambers were capped with an inverted plastic centrifuge tube which was continu- ously flushed with 100% 02. In order to obtain 3 sufficient amount of tissue for analysis, sam- ples from 2 perfused fish were pooled and used in this series of experi- ments. After weighing, the tissues were diluted with KCl-KH P0 incuba- 2 4 tion medium (pH 7.4) containing 0.1 mM Na -EDTA and 5 mM glucose. 2 Homogenates of these tissues were made using a Corning Ten Broeck tissue grinder and kept on ice. Two aliquots (2.8 ml) of the homogenates were removed and placed in separate chambers so that simultaneous recordings of total respiration and CN--resistant respiration could be performed. Initial measurements were made in the absence of NADH at 21% 02. Homog- enates were allowed to equilibrate to temperature and become saturated with room air. A volume of 100 ul of 30 mM KCN was added to the chamber used to measure CN--resistant respiration. Incubation medium was then added to the respective chambers to bring the volume to 3.0 mls. After determination at 21% oxygen the plungers were lifted above the level of the homogenates and 100% oxygen was flushed in the airspace between homogenate and plunger for a period of at least 15 minutes. After 30 saturation with 100% oxygen the electrodes were returned to the homogen- ates and 02 uptake was recorded at this elevated oxygen tension. The chambers were then replaced and the procedure repeated with fresh aliquots of homogenate to which 100 ul of 30 mM NADH was added. Total respiration and CN—-insensitive respiration rates were determined using the oxygen solubility coefficient at 15 C of 0.034 ml 02/m1 fluid. Dissolved oxygen was determined using the following equation: 0.034 x total volume x F02 x (DBP/760) x 1000 = ul 02 (12) where: F02 is the fraction of oxygen and DBP is the dry barometric pres- sure. Respiration rates were expressed as nmol O consumed/min/mg 2 protein (Umbreit, Burris zuui Stauffer, 1964). Protein determinations were made using the Lowry method (Oyama and Eagle, 1956; Appendix B). A Paired-t analysis was used to determine whether the addition of 1.0 mM NADH had a significant effect on baseline rates of oxygen con- sumption. The same test was applied to determine the effect of hyper- oxic exposure on total and CN--insensitive rates of oxygen consumption. Differences between tissue homogenates for the same set of conditions were determined using a one-way analysis of variance and the Tukey's w-procedure for multiple comparisons testing (Steel and Torrie, 1960). Superoxide Dismutase Assay Superoxide dismutase (SOD) was assayed on the basis of its ability to inhibit the reduction of ferricytochrome c by superoxide anion gener- ated by the xanthine-xanthine oxidase system: xanthine/xanthine oxidase -----> 0; (13) 0; + cyt c Fe(III) -----> 02 + cyt c Fe(II) (14) t r + SOD 02 + 02 + 2H > H202 + 02 (15) (McCord and Fridovich, 1969). The rate of reduction of ferricytochrome 31 c was followed at 550 nm using a recording spectrophotometer (Beckman Model DB-G Grating Spectrophotometer, Beckman Instruments, Inc., Fullerton, CA). The xanthine oxidase concentration was adjusted to pro- duce a rate of cytochrome c reduction of 0.025 absorbance units/min in the assay. One unit of SOD activity is defined as that amount of enzyme which inhibits the rate of cytochrome c reduction by 50%, under the specified conditions. Since there is a small amount of ferricytochrome c reduction which cannot be inhibited with SOD, the true uninhibited rate of reduction must be determined from the difference between the reduction rate in the absence of any SOD and the rate in the presence of excess amounts of SOD (e.g., 20 BESOD units in the reaction system). The degree of inhibition at each dose can likewise be determined from the difference between the rate of reduction observed and the rate in the presence of excess amounts of SOD and then expressed as percent inhibition of the true uninhibited rate of ferricytochrome c reduction. Bovine erythrocyte superoxide dismutase (BESOD), E.C. No. 1.15.1.1, was used as the purified SOD standard (Sigma Chemical Co., St. Louis, MO (Lot #16C-8030; 2900 U/mg protein; 2880 U/mg solid; assayed as per McCord and Fridovich, 1969). Preparation of the standard is described in Appendix C. Since measurements were made in crude tissue homogenates, several controls were run in an effort to detect endogenous interfering substan- ces. Crude homOgenates may contain substances which are capable of interfering with assay reactions. For example, any substance in the crude homogenate which can react with reduced cytochrome c may alter apparent SOD activity. One of the most common problems is the presence of contaminating cytochrome oxidase or peroxidases. Either of these 32 enzymes can. re-oxidize ferrocytochrome <2 and artefactually increase "apparent" SOD activity of the homogenate. These enzymes, if present, would regenerate ferricytochrome c and thus add to the 'apparent' SOD activity of the homogenate. However, sodium azide (10-5 M), an effec- tive inhibitor of the cytochrome oxidases and peroxidases (Flohe and Otting, 1984), did not alter the rate of reduction of ferricytochrome c in any of the homogenates. Thus, the oxidases and peroxidases were not present in sufficient concentrations to interfere with the SOD assay. In addition, a parallel-line analysis of variance was applied to the data in order to reveal the presence of interfering substances (Eldred and Hoffert, 1981). The applicability of these statistics is dependent on a linear response so that the slopes of the dose response curves can be compared to an SOD standard. For each tissue sample, a dose response curve was generated whereby percent inhibition was determined over at least 4 concentrations with duplicates at each concentration. This res- ponse was linear below 50-60% inhibition for the bovine erythrocytic SOD standard and for the tissue homogenates. However the slope of the individual tissue response curves was dependent upon the concentration of superoxide dismutase within the tissue. Tissue samples which were more dilute than the standard would have lower slopes and lie to the right of the standard, while samples more concentrated than the standard were shifted to the left of the standard with higher slopes. All of the lines would theoretically pass through the origin. In order to test for endogenous substances which might interfere with the response curve, this dilution effect must be eliminated without affecting the linear response of the curves. The standard and tissue homogenate dose res- ponse curves should differ at any given level of inhibition by a potency 33 ratio which would be independent of the level of response. This was accomplished by converting the response to log of percent inhibition and plotting the responses against a log dose (Goldstein, 1965). Since, for the most part, successive doses were increased by doubling, the values could be coded. Ascending doubled doses were coded by assigning inte~ gers values to them, beginning with the value 1 at the lowest dose (25 ul). Coded log units therefore differed by log 2. The procedure for decoding the Coded Log Dose Scale is given in Appendix C. This trans- formation eliminated the dilution effects, so that the response of the tissue homogenates could now be statistically compared to the response of the standard. If no endogenous interfering substances are present, then the dose response curve for the tissue homogenate should be paral- lel to the dose response curve for the BESOD standard. If this were not the case, the relative potency computed from converging lines would differ with the level of response, and the homogenate could not be assigned 3 single potency. Neither the EDSO's or the potency ratios were affected by the transformation. Tissue superoxide dismutase activity was measured in homogenates which had been stored frozen for less than one week. Tissues were initially diluted (retina, 300 mg/ml; brain, 250 mg/ml; liver, 80 mg/ml) with an ice cold phosphate/EDTA solution (pH 6.5) (Appendix A) and homogenized on ice with a Corning Ten Broeck tissue grinder. The homog- enates were frozen in :1 dry ice/alcohol bath, stored frozen and used within one week. After thawing the homogenates for assay, they were appropriately diluted with sodium phosphate buffered saline (pH 7.8). The dilute homogenates and BESOD standard (2.9 ml distilled H20/vial) were then sonicated over ice for four consecutive 15 second periods at 5 34 second intervals using a Sonifier Cell Disruptor at a power setting of 20 watts (Mbdel W-185-C, Heat Systems Co., Melville, L.I., NY, fitted with a Micro Tip, Branson Instruments, Inc., Danbury, CT). To disrupt cell membranes further, tissue homogenates were subjected to three cycles of freezing and thawing in a dry ice-ethanol bath and a 50 C water bath, respectively. The homogenates were then centrifuged at 13,200g for 30 min at 4 C (Sorvall, Inc., Model RC2-B, SM-24 rotor, Newtown, CT). The supernatants were then removed and recentrifuged at the above settings for an additional 20 minutes. The final supernatants were then removed and kept on ice until used in the assay. Cytosolic SOD (CuZnSOD form) was distinguished from mitochondrial SOD (MnSOD form) by the farmer's sensitivity to 2 mM cyanide in the assay medium (Flohe and Otting, 1984). :A detailed description of the procedure and assay design is given in Appendix C. The variation about the dose response lines was estimated using the ANOVA approach to regression analysis as described by Neter and Wasser- man (1974). All lines were initially tested to ensure linear regression functions and significant positive slopes (i.e., F-tests). Comparison of the standard and sample regression lines followed from the analysis of variance. The error variances of the lines were tested for homogen- eity in order that the comparison for parallelism be appropriate. For a valid assay, the crude homogenates should differ from the standards in dilution only and should be parallel. One unit of SOD activity was defined as that amount of standard or tissue homogenate which inhibited the true rate of ferricytochrome c reduction by 50%. The volume of standard added at the 50% inhibition point was termed the ED50(S). The concentrations of the standard (SOD 35 units/ul) could then be obtained from the reciprocal of the ED The 50' ED50 for the tissue homogenates was also determined at 50% inhibition (EDSO(T)), and the potency of the tissue homogenate was expressed as the ratio of ED50(S)/EDSO(T). The potency ratio can be used to express tissue homogenate SOD activity in terms of activity of the commercially obtained BESOD standard. The 95% confidence limits about the EDSO's and the potency ratios were also determined (Goldstein, 1965). To facilitate comparisons between samples, the amount of wet tissue weight at the tissue ED was determined as well as the tissue protein 50 concentration using the method of Lowry (Oyama and Eagle, 1956). The SOD activity could then be expressed as units/mg wet tissue weight or units/mg tissue protein. :A one way analysis of variance was used to detect statistical differences in tissue SOD activities. Cytochrome £_Oxidase Assay Cytochrome c oxidase catalyzes the electron transfer from ferrocyto- chrome c to dioxygen in the respiratory chain of the mitochondria. Since the MnSOD form is contained within the mitochrondria, a measure of cytochrome c oxidase (COX) activity within these organelles would pro- vide a means of comparison for this form of SOD between tissues. Cyto- chrome c oxidase activity was determined using a slight modification of the method described by Rafael (1983) where COX is assayed by following the rate of oxygen uptake polarographically by means of an oxygen elec- trode. Oxygen concentration in the assay system was measured with a Clark-type oxygen electrode. Cytochrome c was maintained in a reduced state with the addition of ascorbate, and N,N,N',N'-tetramethyl-p- phenylenediamine dihydrochloride (TMPD) was added to mediate electron transfer. The blank rates of autoxidation due to ascorbate and TMPD 36 alone and in the presence of ferrocytochrome c were determined and sub- tracted from the measured oxygen uptake. The uptake per unit time of oxygen dissolved in the assay solution at 25 C was used as the measure of COX catalytic activity. The electrode and experimental setup was the same as that used in the cyanide insensitive oxygen consumption determinations. Cytochrome c oxidase and SOD activities were determined in the same tissue homogen- ates, so aliquots of the homogenates prepared for SOD activity were used. The homogenates were diluted 1+ 1 with a 1% Lubrol solution (type WX from Sigma, prepared in the phosphate/EDTA buffer) and kept at 0-4 G. Cytochrome c oxidase activity was calculated in terms of the rate of oxygen consumed (nmol 02/min/mg protein) and also in terms of the standard unit, the katal (kat), and expressed as nkat/mg protein. A detailed description of the assay design and determination of activity is given in Appendix D. A one way analysis of variance was used to determine statistical differences in tissue activity and Tukey's w- procedure used for a multiple comparisons test. Catalase Assay The catalatic activity of catalase was assayed in crude tissue ex- tracts using 3 rapid spectrophotometric variation of the technique for determining catalase activity by measuring H by titration with per- 202 manganate (Cohen, Dembiec and Marcus, 1970). When used on crude tissue homogenates the authors stress the importance of incorporating ethanol and Triton X-100 to increase the observable catalase levels. Ethanol acts by decomposing Complex II, an inactive complex of catalase with that forms spontaneously in crude homogenates, and Triton X-100 H202 acts either by releasing catalase from its particulate form in the 37 peroxisomes or by allowing free access of the substrate. A detailed description of the assay is given in Appendix E. Liver samples were placed in cold tissue homogenizers, diluted with 10 volumes of ice cold isotonic buffer, and homogenized while in an ice water bath. Pboled retina (6 eyes) and brain samples from three fish were treated similarly but diluted with only 3 volumes of isotonic buffer. Homogenates were centrifuged for 10 min at 700g (Sorvall, Inc., Model RCZ-B, SM—24 rotor, Newtown, CT) to remove cellular debris. A 3 ml aliquot of each supernatant was removed and ethanol added to a final concentration of 0.17 M (i.e. 10 ul EtOH/ml of supernatant). Samples were then incubated for 30 min in an ice water bath (0-2 C) to decompose Complex II. After incubation, 10% Triton X-100 was added to a final concentration of 1.0%. Aliquots of the tissue samples were then diluted with ice cold isotonic buffer so that they would fall along the linear portion of the standard curve. The catalase standards were prepared from crystalline bovine liver catalase (Sigma Chemical Co., St. Louis, MO, #C-40, lot# 53F-7070; 11,000 units/mg protein). The stock standard solution was prepared with ice cold distilled water to a concentration of 0.20 mg/ml. Various dilutions were prepared from this stock catalase solution, such that a range of 5 ug to 0.5 ug of catalase were added to the reaction vessels. All reactions were carried out in an ice water bath (0-2 C). The enzyme catalyzed decomposition of H202 was followed. The reagents were added sequentially at fixed intervals and allowed to react for exactly 3 minutes. The remaining H 0 was then reacted with a standard excess of 2 2 KMnO4 and the absorbance of the residual KMnOA was measured at 480 nm using a Gilford single beam spectrophotometer with optics from a Beckman 38 DU Spectrophotometer (Beckman Instruments, Inc., Fullerton, CA). The cuvette chamber of the spectrophotometer was maintained at 2-4 C with a refrigerated water bath to prevent the turbidity which developed when measurements were made at room temperature. Catalase activity was expressed in terms of Bergmeyer units (units/g tissue), International units (units/g tissue) and Kat.f. (Katalasefahig- keit) units (Kat.f. min'-l.mg--l protein). See Appendix E for calcula— tions and expression of activity. Glutathione Peroxidase Assay Glutathione peroxidase catalyzes the reduction of H and organic 202 hydroperoxides using glutathione as the substrate. The oxidized form of glutathione (GSSG) can be cycled back to the reduced form (GSH) with the addition of exogenous glutathione reductase and NADPH, thereby maintain- ing substrate concentration for the enzyme. Glutathione peroxidase activity is therefore measured indirectly by following the decrease in absorbance of the reaction mixture at 340 nm as NADPH is converted to NADP (Paglia and Valentine, 1967). Cytosolic fractions of the tissue homogenates were prepared for the determination of selenium-dependent glutathione peroxidase activity (Tappel, Chaudiere and Tappel, 1982). Tissues were diluted 1:5 (wzv) (livers 1:7) with cold 0.25 M sucrose and homogenized in an ice water bath using a glass Corning Ten Broeck homogenizer. Homogenates were centrifuged at 4 C for 15 min at 13,000g, the supernatant pipetted off and then recentrifuged for 30 min at 28,000g (Sorvall Inc., Model RCZ-B, SM-24 rotor, Newtown, CT). The final supernatants were removed for assay of selenium dependent peroxidase activity using 2.2 M H as the 202 substrate. A detailed description of the assay is given in Appendix F. 39 Activity was determined from the molar extinction coefficient for NADPH at 340 nm (6,220 M-l.cm_1; Mattenheimer, 1971) and expressed as mUnits/mg protein or nmoles.minm1.mgul protein. Protein determinations were made using the dye binding method of Bradford (1976). A one way ANOVA was used to determine statistical differences in tissue activity and Tukey's wbprocedure used for a multiple comparisons test. Glutathione Assay This procedure is a spectrophotometric method for the determination of sulfhydryl content in the protein-bound (PB-SH) and nonprotein-bound (NP-SH) fractions of various tissues. The method is based on the reduc- tion of 5,5'-dithiobis-(2-nitrobenzoic acid), by SH groups to form 1 mole of 2-nitro-S-mercaptobenzoic acid per mole of SH. The nitromercap- tobenzoic acid anion has an intense yellow color which absorbs at 412 nm and is used to measure SH groups (Sedlak and Lindsay, 1968). Tissues were homogenized in approximately 20 volumes of 0.02 M EDTA :hi 3 Ten Broeck homogenizer in an ice water bath; (i.e. tissues were diluted in order to obtain absorbance readings which fell between 0.1 and 0.8 absorbance units when 0.5 ml and 2.5 ml of homogenate were used for the estimation of total and non—protein bound sulfhydryls respec- tively). The homogenates were kept in an ice bath until used. A de- tailed description of the assay is given in Appendix G. Sulfhydryl concentrations were determined from the molar extinction coefficient of 13,600 M-1.cm-l at 412 nm and expressed on a per mg pro— tein basis. Protein concentrations were determined using the method of Lowry (Oyama and Eagle, 1956). A one way analysis of variance was used to determine statistical differences in tissue activity' and Tukey's w—procedure used for a multiple comparisons test. RESULTS Total and Cyanide Insensitive Oxygen Consumption by Trout Retina Oxygen uptake by tissue homogenates was measured in the absence and presence of 1.0 mM potassium cyanide in order to establish the total rate of O2 consumption as well as the cyanide resistant rate of oxygen utilization. The latter uptake rate was used as an estimate of the rate of formation of partially reduced species of oxygen (i.e., 0; and H202 production). Measurements were made in room air (21% 02) and in 100% oxygen and are summarized in Table 1. Measurements were made in homog- enates from the hyperoxic resistant retina and compared to values obtained from the oxygen sensitive brain and liver homogenates. Total oxygen uptake at 15 C in room air was 0.58 nmol 02/min/mg protein in the retina, and 1.36 and 2.03 nmol 02/min/mg protein in liver and brain res- pectively. cyanide insensitive respiration accounted for 21%, 26% and 7% of total respiration in the retina, liver and brain respectively (Table 1). There was a significant increase in total oxygen consumption by the retinal homogenate after it was incubated in 100% oxygen, however no change in total consumption was observed for the other two tissue homogenates. Cyanide insensitive consumption however rose dramatically in all 3 tissue homogenates after incubation in 100% oxygen, increasing to values of 0.46, 0.62 and 0.55 nmol 02/min/mg protein for retina, liver and brain respectively. Total and CN--insensitive oxygen consumption was also determined 40 41 13 w.>mxse ecm Amo.o v av <>oz< has one mafia: maceuwvccu .umwu mcomfiumasou mfiafiu~sa you munvmooua seem any new uo>fiH coca poaoa >Hucmuamwcwfim I A .ummu mcowfiumanu waeaufine new ouskuoun I3 m.>mxse use Amo.c v av <>oz< was one means mcofiufivccu meow msu new sauna coca posed uaucmoawwcwum I n .Amo.o v ev maesaeee u confine mega: has Econ 5 venomous cor—3 sump coHuaesmcoo m>fiuuwamwu Eoum acouwmwwv hHucmqucwwm In. .Amo.o v av mfimzamcw u confine mean: mn¢z 25 o._ no cofiuavwm unozufis wumcoono; meow 80pm ucmuwMMfiv maucmowmwcwfim I « .mwu m\uoua ms c.nm H 0.0m u um>fia "may w\uoun we n.o H 5.05 n cfimum “mfiu m\uoua ms o.m H ~.e~ n mcfiuom m n a “cm H m oa.ow wz.ou eo.ou -.cu oz.ou No.0“ sewage Neoa eeeueeeeeoo a. oe.o p~e.o a. we.o emm.o a .aee.o ee.c No e>auaeeeeeHuuzo _m.iH me.ou ~m.cu em.cH -.ow ~_.ou eemsxo noel am.~ e~._ a. oe.~ _o.~ .nme.o n Ne.o eoaeeeeeeeo No fleece 33.0“ ec.ow mo.cu ec.cw eo.ou eo.cu ea< seem coweeEeeeeo _m.o em.o a. m~.o ge_.o e3.o -.o No e>euaeeeeeHuuzu mo.~H mw.ou em.ow om.ou e_.ow e~.ou efi< seem ~3.~ em.~ mm.~ mo.~ .nee.o mm.o eeaueeeeeoo No sauce :32 :32 :92 :32 £22 :92 no“: oneness sees oneness see: unease: mcuumx .Acfiououa wE\cwE\voEom=oo Imufiammu oumcomoeo: osmmfiu usouu soncamm N o Heeev =efiuwmcomcal 20 use HouOH .# wHQMH 42 after the addition of 1.0 mM NADH as substrate for the mitochondrial electron transport chain. The addition of NADH did not significantly alter total or CN—-insensitive consumption in either 21% oxygen or 100% oxygen for the retinal homogenate. However the addition of 1.0 mM NADH to both brain and liver homogenates significantly increased both total and CN--insensitive rates of consumption after incubation in either 21% or 100% oxygen. The lack of a statistically significant increase in total consumption at 100% oxygen for the liver homogenate was probably due to variability within this data set. In the presence of 1.0 mM NADH, CN-—insensitive consumption of oxygen in room air and 100% oxygen accounted for 20% and 66% of total respiration in retinal homogenates, 11% and 30% in brain homogenates, and 26% and 39% in liver homogenates, respectively. Therefore CN--resistant respiration accounted for a significantly greater fraction of total consumption after hyperoxic exposure, and these CN--resistant rates were 2-4 fold greater than those measured in room air. Although the retinal rate of CN--insensitive consumption in the presence of substrate was less than that measured in brain and liver, the increase in CN--insensitive consumption from room air to 100% oxygen, was not significantly different among the three tissue homogenates (Table 1; Figure 2). Primagy Antioxidants 12 Trout Retina Superoxide Dismutase Activity Retinal tissue exhibited significant SOD activity which was elimina- ted by boiling the samples for 15 minutes at 100 C. The calculated means for total superoxide dismutase activity in the retina of 3 rainbow trout specimens are reported in Table 2 and are expressed as tissue units/mg protein, tissue units/mg tissue (wet weight) and BESOD units/mg 43 .mfimzfimcm uIvouwmm m magma .Amo.o v av paw Eoou ca ucmE Imusmmme Eon“ ucmpommac zHucmu«uficwwm mouocwv I « .Aam H xv Camacho wE\:HE\emE:mcoo NC ~05: mm emmmwunxw ma cofiuouwammm .20M :5 o.~ mo mocmwwua ecu ea ewusmmwe we: seeumufiowou w>fiu Ifimcomcfil zo .:aHH use :wmun .HmcHqu .mwumcwmoEo: osmmwu usouu soncwmu CH :Owumufiomwu w>wufimcwmcfil 20 can Hmuoe .N ouswfim 44 N shaman mw>~4 z~~p~mzumz~uuzu oz< 4wq oo._ xocouom eumccmum nemmmuHmuOu mcfiumm "mum vumecmum 20mmm ozu ou uuocwmp «a xuw>fiuom wcwwmouaxw u0m mofiumu zocmuoo wi%w .H: nmmnamqvo om huH>Huom some: .334em use 3: .5?va am Samoa.“ H33 23% “a: momufimv am Pym commm "one mood; am cannons: .ouo Icmwoeox onmmuu no cofiuofiom eumocmum acmmm mo Afiov oeofio> m> cofiuuovou no such mo cowuwnwccw ucouuoa uo mcofimmouwou noocfia mm vmuuoHo who moommwu uoouu sopC#mu Eouu mumo o>fiumucwmouoou mo mw>u=o wmcoommu omoa .hufi>fiuom ommussmfie oofixouoasm mo zmmmm so we cowuoswmu o oEoucoouxowuuwm mo cowuwnwscfi ucoopom .m ousmwm 49 AJDV wzaqo> mum 0mm mNN DON mun om" mm" Don wk on mm pl - P F p . _ .3 smm .3 mom .3 com as as :5 omom omom omom _ _ _ . _ . _ . . Ohm \— _ b P h n n - com a. 00m J m4mzbu>~hu< wmaa mo.“ u zocouom oumocmumomommmufimuou mewuox Om .H: now u ANAV om oomczamw>wq "A: mofi u Aime om sea>aeee fleece assume “He eoN n Amv em eeeeeeem CCmmm ”mum A0 xficcwoo< wowv uCHvouwv noumm vaprouwe mooam> zocouoq one oQO .oooom wmumcmmoeo; mommfiu ho cofiusfiom oumecmum nemmm mo Aasv w55~o> mxu mucmmmuoou wmoo omega “meow on oovou w> sewage—HE; ucmunmo on mo meowmmouwou passed no owuuoHaImu mum m manage Eouw mum: .>u«>«uom owmu=Emfio ovfixoumasw mo cofiumcfisumuov can now omov on coooo .m> cowuosvou o oEoHSUOuonuuom mo cofluwnfi:Cfi acouumm .e ououfim 51 m4 a: 8m4 “NI—V Dmaw wmoo Do..— ownou e was: 04 m.m o.m m.N O.N m4 0." md 0.0 n p - b - p - - _ a: ecu Amv Ohm am now am come: a: so“ mm>~4 flame omnm . _ . _ . _ _ _ _ . . . om a zonbnmazzu N no... >._. a > H hu< wmufi>fiuu< m~.¢ Aw.wm~ m._-v m.~m~ Amnm.e ano.¢v o-.q moms: I pm>HA o~.- Ao.~wm w.cm_v N.@~N Aaaq.e «mw.mV m~_.e «o.n_ Ao.mmm o.qw~v m.mo~ Afimm.¢ www.mV omo.c >ua>auu< now nm.- Am.mcm ~.~m~v m.o- Anam.q qmm.mv mm~.q Hmuoe I um>wq No.m Ao.wc~ c.mq~v e.om~ Aqmn.m www.mV oeo.m m<.m A~.ow~ n.5wgv ~.~NN Anwe.q eco.mV ccfi.¢ >u~>fiuu< m¢.~ A~.fiuo< cow cw.¢ Aw.mm~ ~.-~V ~.~N~ Aomq.m mc~.mV w-.m Hmuofi I :Hmum mm.m Am.q- m.mw~v N.~cN Acc~.¢ flew.mV o~o.< mo.¢ Aw.¢mu o.co~V o.ma~ A~o~.q amm.mV mmm.m huw>wuo< c~.m Am.mw~ m.~¢~V n.wm~ Awnw.m eoq.mV coo.m Down: I mcaumm q¢.m AN.NNN m.ow_v o.cc~ Anw~.¢ on.mV Nqo.e em.m AN.~mm m.m-v n.om~ Am~N.¢ mon.mv ona.m hufi>fiuo< now on.n Am.mmm o.m-V N.mN~ Acm~.¢ ~m~.¢v ~n~.e Hmuoe I mcwumm >ua>fiuo< com Amuwswa mucoufiwcoo Nmov omnm vooouea Amu«E~H wocoefimcou Nmav Omnm ooooo oumcoonom osmofia .ummmm u oeouzuouxufiuuom way an oocfimmw mm cfiououe me\muwc= mm oommwuoxo mg mufi>uuo< cowom .32.: mocmvfiucou Nme o>fiuooammuomufiosu :33 33:33:55 Hov mufics venous: vow vooou cw mm mm ewuuooou who mmsam> .mm:Hm> om Hmoeq>wecfi coo: momma >uw>fiuo< wmcu35mwa ovqxopoonm .m ofinme 53 Table 4. Superoxide dismutase activity based upon potency ratios to standard preparations. Activity is expressed as Units/mg protein as defined by the commercially prepared standard. Homogenate:Standard Potency Ratio (95% Confidence Limits) SOD Activity Homogenate BESOD units/mg protein SOD Activity 1.05 (0.97 1.13) 8.74 1.17 (1.05 1.29) 6.34 Retina - MnSOD 1.32 (1.07 1.63) 4.17 Activity 1.06 (0.98 1.15) 5.25 1.19 (1.08 1.32) 4.05 Brain - Total 2.54 (1.42 4.57) 12.31 SOD Activity 0.91 (0.85 0.98) 7.31 0.82 (0.73 0.94) 3.38 Brain - MnSOD 0.62 (0.44 0.87) 1.20 Activity 0.93 (0.86 1.02) 3.21 1.62 (1.31 1.99) 4.88 Liver - Total 0.93 (0.84 1.02) 10.52 SOD Activity 0.99 (0.93 1.06) 16.89 1.10 (0.96 1.27) 13.51 Liver - MnSOD 1.04 (0.82 1.31) 4.30 Activity 0.75 (0.63 0.89) 4.26 1.07 (1.00 1.14) 4.20 54 .Amo.o v av umou mcomfiumaeou maawuane you ounewuouols m.>oxoe ocm mucmfium> mo mam>~mcm >m3 wco m wcfim: mosmmwu oau nonuo “Xx. some ucoquMHo xaucmu«mwcwfiw mouocoo I a .Am u 2 ”cm H xv samuoua as \muwco cow mm ommmouoxm ma one omuuooou we xufi>wuom come: one Hmuou :uom .uoouu soccfimu ozu Eouw mmumcomoEo; wswmwu mo >uw>fiuum ommussmfio ocfixouoaom .m opsmfim 55 mw>n4 z~Z>HB< 8m .28» I D >h~>nh0< wmoxsv ocm oucmwpm> mo mammamcm >93 oco m udwm: mmsmmwu oBu amnuo wcu Boom ucwuommwo hauchHWficmHm mouocoo I_« .Am n 2 new + xv cfiououa mE\umxc mm oommmuaxo ma muw>auu< .usouu soncamu mzu Eoum moumcouosoz wnmmwu mo >uu>fiuom mmmefixo meouzuouao .o wuowfim 59 mw>HJ o muanm zH._. H > H ._.U< mm<0 H x0 w201100h>0 N I 31.08:! SIN/10‘1” 60 .Hmc.c v av uwou meowHumoEou oHoHuHoe p00 museo00uolz m.%oxpe can museum; mo me>Hmcm .33 9.0 m mch: mmsmmHu o3u posuo ecu Eouu ucwuommHo HHucmonHcmHm mouocoo I « .ommvon oeopfioouxu umemuHcs new we vommopoxo mH zuH>Huu< .quZuom omogxo mEou500u>o uHc: poo oommmuoxm moumcomoEo; mommHu usouu aoachu osu Eouw qu>Huo< mmmu26mHn oconuwaom Hmwuvcocuoqu .n wustm 61 m muswam mw>H.._ zH H ._.U< mm<0 H x0 w201100h>0\ >._. H > H ._.U< 000...: BSVO I X0 BNOMHSOIAIJ ROW-VS]. I NO 62 Catalase Activity Catalase was measured in crude tissue homogenates after the incor— poration of ethanol and Triton X-100. No catalase activity was detected in either retinal or brain homogenates from the rainbow trout. For liver, three dilutions of the homogenates were run in each experiment, their activities were calculated and then averaged. Trout liver tissue was very high in catalase activity. The mean liver activity when expressed in Bergmeyer units at 2 C ‘was 1.42 1 0.46 units/mg liver (X 1_S.D.) which corresponded to 14.53 1 4.66 International units (units/mg tissue). Catalase activity when expressed as Kat.f. units is useful for comparisons of crude tissue homogenates. Liver activity was 4 1_0.93 x 10-4 min-'l.mg--l protein (Figure 8). measured at 4.36 x 10- The incorporation of Triton X-100 alone into) the liver' homogenates accounted for a 37% increase in activity while the incorporation of both ethanol and Triton X-100 into the homogenate accounted for a 350% increase in liver catalase activity. Since a standard curve was run with each assay it was possible to estimate the equivalent amount of pure catalase to that which was measured in the homogenate. An estimate of 54.9 1 19.6 ug of catalase was determined per mg protein in the liver homogenates. Glutathione Peroxidase Activity and Glutathione Content Selenium-dependent glutathione peroxidase activity was measured in- directly by following the rate of regeneration of GSH in the presence of glutathione reductase and NADPH using H as substrate. The change in 202 absorbance at 340 nm was followed over time as NADPH was converted to NADP and the nmol of NADPH consumed/min/ml was determined from the 1 extinction coefficient for NADPH at 25 C (6,220 M- .cm-l). Protein 63 .Hm u 2 "cm H xv cHououo mE\cHE \wuH65 .m.umx we wwmmwuoxo mH qu>Huo< .uoouu sonchu ecu Eouw mwumcwwoso; wommHu ovauu cH vmcHEumuoo qu>Huom ommHmumo .m ousmHm 64 mw>H.._ zH._. H > H k0< wmoxoe vcm mocmHum> Ho mHm%Hmcm Hos oco m wchs moommHu osu umcuo msu Eoum ucouonHo HHucmuHchmHm wouocoo I « .Am n 2 "mm H xv cHouOMQ wE\muHCDE mm oommmuaxo mH zuH>Huu< .usouu socchu any Eouw moum IcouoEo; oommHu cH rmuomoos qu>Huom wmmvonuoo oconumuoHo .m ousuHm 68 mm>H4 2HhH>HhU< wm<0Hx0mmm w20HIhHuu< .usouu soachu mcu Eoum mmum Icwmoeo: wommHu Ho ucmucoo oconucusHm ecm quomszsm Hmuoe .3 spam; 7O mw>H.._ zH00>Iu400 1.5.0... I D hzmhzou 4>m0>zunsm mammHh 00 .0 .00 .0 ..0H .0 .mH .0 Hews N13108:! SIN/HS WON" DISCUSSION According to the free radical theory of oxygen toxicity, hyperoxic tissue damage is caused by the intracellular formation of activated oxygen metabolites, i.e., superoxide (0;), hydrogen peroxide (H202) and hydroxyl radical ('OH), at a rate that overwhelms the cellular defenses (Jamieson et 31., 1986). Based on this theory, tolerance to oxygen toxicity is thought to be due to either a decreased propensity of 02- adapted cells to generate the reactive oxygen metabolites and/or the presence of a highly efficient antioxidant defense system to limit the concentrations of the primary oxidants, i.e., the superoxide dismutase, catalase and glutathione peroxidase enzymes. In the present study these two oxidative defense "strategies" were examined in an attempt to dis- cern the factors responsible for the resistance In) oxygen. toxicity observed in the rainbow trout retina. Determination 2£_Oxidative Flux 12_Rainbow Trout Retinal Homogenate Trout retina exposed to hyperoxic conditions do not exhibit any evidence of oxidative damage. In contrast, a variety of other tissues are rapidly damaged under identical conditions (Nishiki et 31., 1976; Freeman and Crapo, 1981; Turrens, Freeman and Crapo, 1982). This damage is thought to be mediated by an increased generation of reactive oxygen metabolites in the presence of supraphysiologic p0 Indeed, in all 2. tissues studied to date hyperoxia causes an increase in CN--insensitive 02 consumption. The resistance of the trout retina to hyperoxia may be 71 72 due to a decreased propensity or ability of the retinal cells to gener- ate oxidative metabolites under these conditions. This was examined in the present study using retinal homogenates. Brain and liver tissues from the rainbow trout previously have been shown to be sensitive to oxygen (Baeyens, Hoffert and Fromm, 1973; 1974), and therefore were used in this study as internal controls for comparisons with the oxygen resistant retina. The results of the oxygen consumption measurements indicate that there was an enhanced rate of CN--resistant respiration by the rainbow trout retina during hyperoxic exposure. These results were similar to measurements made in the brain and liver tissues and demon- strated that CN--resistant respiration is increased comparably in all three tissues during hyperoxia. Several studies have shown that CN-- resistant respiration is a reliable indicator of the effect of p02 on the production of partially reduced species of oxygen (Hassan and Fridovich, 1977; Freeman and Crapo, 1981; Freeman, Topolosky and Crapo, 1982; Turrens et 31., 1982b; Turrens, Freeman and Crapo, 19823). The method however, does not provide an absolute measure of cellular 0' and 2 7 generation is 2 very dependent on the conditions at the time of measurement. For exam- H202 production (Freeman and Crapo, 1981). The rate of 0 ple an overestimation of the rate of oxidant formation may be caused by a number of factors. Superoxide production by mitochondria is enhanced when there is a decrease in ADP stores and the components of the respir- atory chain are reduced (Boveris and Cadenas, 1982). This situation arises with the addition of cyanide to inhibit cytochrome oxidase and subsequently reduce the mitochondrial respiratory carriers. An estimate of the formation of partially reduced oxygen metabolites would therefore be greater in the presence of cyanide than when measured in situ 73 (Freeman and Crapo, 1982). In addition, only oxygen reduced to water or to an oxygen species which can covalently react with cellular constitu- ents will be measured polarographically. Oxygen may be consumed in the oxidation of substrates which may or may not generate or release reac- tive intermediates, thereby overestimating their production (Freeman and Crapo, 1981). This method may also underestimate oxidant formation. Cyanide is an effective inhibitor of one electron reduction at the ubiquinone-cytochrome b region of the respiratory chain (Boveris and Cadenas, 1982). Since this is one of the two major sites of superoxide generation, inhibition would result in an underestimation of 0; forma- tion. Although 1.0 mM CN- will partially inhibit the CuZnSOD and cata- lase in the homogenate, the spontaneous dismutation of 0;, and dismuta- tion by MnSOD could all result in the regeneration of oxygen and an overall underestimation of the true rate of production of the partially reduced oxygen metabolites: o; + o; + 2H+ —59§99-> H202 + 02 (16) (Freeman and Crapo, 1981). Thus, there are numerous pathways through which CN--insensitive 02 consumption can: be influenced, nonetheless, the measurement currently provides the best index of the rate of formation of partially reduced forms of oxygen. The addition of NADH as substrate for the mitochondrial electron transport chain did not have a significant effect on the rate of total 02 consumption in retinal homogenates, probably because this tissue is heavily dependent on aerobic glycolysis to meet metabolic needs. How- ever the addition of NADH to brain and liver homogenates significantly increased the rate of 02 consumption in these tissues. These results 74 suggested that the endogenous pool of reducing equivalents may have been rate limiting in the latter two homogenates and might partly explain why total consumption was not increased on transition to hyperoxia in the absence of NADH (Table l). The addition of NADH to brain and liver homogenates in the presence of cyanide caused an increase in consumption and was most likely due to an increase in the reduction of the cyto- chrome carriers. Hyperoxic exposure significantly increased the partial reduction of oxygen in all three tissues. More importantly, the change in cyanide-insensitive oxygen consumption from room air to hyperoxia was the same in all three tissues and provided evidence that the retina was indeed capable of enhancing the rate of reduction of the partially reduced oxygen metabolites. Thus, the results demonstrate that all three tissues repond similarly and that retinal tissues do not protect themselves against oxygen toxicity by a decreased propensity or ability to generate reactive oxidants in hyperoxia. Finally, these results are consistent with those described in other tissues tolerant to hyperoxia. The hypothesis that tolerance to hyper- oxic exposure may be due to a decreased production of the reactive oxygen metabolites has been similarly advanced in other models of hyper- oxic damage. Rats exposed to 100% oxygen generally will die within 72 hours of exposure. However, they are tolerant to 1 atm pure oxygen if they are first exposed to 85% O for a period of 7 days before subse- 2 quent challenge with 100% oxygen (Crapo and Tierney, 1974). It had been presumed that hyperoxic exposure resulted in an enhanced rate of 0; and H202 production, but this had not been formally tested until the studies by Freeman and Crapo (1981). They reported that CN--resistant respira- tion was in fact, greater in rats pre-exposed to 85% oxygen than in 75 controls. Their findings led them to conclude that the increased tolerance of the adapted rats was not due to a decrease in intracellular 02/H202 production (Freeman and Crapo, 1981; Crapo and Tierney, 1974). Indeed, no study to date has successfully demonstrated that any tissue is able to regulate its sensitivity to hyperoxia by decreasing the generation of oxygen metabolites intracellularly (Fridovich and Freeman, 1986). Primary Antioxidant Defenses 12_the Rainbow Trout Retina If oxidants are generated within the tissue then the primary antiox- idants designed to control 0; and H202 ical role in the resistance of the rainbow trout retina to oxygen toxic— might be expected to play a crit- ity. The antioxidant activities of the CuZn and Mn superoxide dismu- tases, catalase, glutathione peroxidase and glutathione content of the rainbow trout retina were determined. Brain and liver tissues from the same species were also examined in an effort to compare the efficiency of the antioxidant defense system of the retina to the efficiency of the defense systems within these oxygen sensitive tissues. 1' 2 cells (Fridovich, 1986b). Indeed, all mammalian cells are known to Superoxide dismutase is the key enzyme for regulating 0 in aerobic contain at least two forms of intracellular SOD; a cytoplasmic (CuZn) form and a mitochondrial (MnSOD) form. There are numerous assay methods for the measurement of the super- oxide dismutases, each of which possesses certain advantages and disad- vantages. In this study, SOD activity was measured by following the SOD inhibitable rate of ferricytochrome c reduction (McCord and Fridovich, 1969). Fridovich (1986b) has listed some of the potential problems that may be encountered with the xanthine oxidase/cytochrome c assay, 76 including: 1) the presence of agents capable of directly reducing cyto- chrome c, which if present, would increase 'apparent' SOD activity, 2) the presence of potential inhibitors of xanthine oxidase and/or 3) the presence of oxidases and peroxidases which could reoxidize ferrocyto- chrome c. The latter two situations would underestimate SOD activity. Each potential pitfall was directly tested for in this study. Cyto- chrome c was tested for contamination by SOD, but was determined to be free of SOD. The effect of oxidases and peroxidases on reduction by crude homogenates was tested for and did not appear to be present in any of the homogenates. In addition, the rate of reduction of cytochrome c was measured after the addition of boiled homogenate and no significant change in the rate was observed for any of the homogenates. This pro- vided an indication that it was unlikely that small autoxidizable mole- cules in the homogenates had an effect on cytochrome c reduction. As an overall check, an indirect statistical method was used to detect the presence of any endogenous interfering substances in the homogenates (Eldred and Hoffert, 1981). Although most investigators have determined SOD activity by simply diluting and readjusting tissue extracts until the 50% inhibition level is attained, it may not reflect true SOD activity. In most assays it is assumed that the unknown preparation differs from the standard only by a dilution factor. This should be tested in enzyme assays which rely on substrates that are not specific to the indicator. The statistical approach incorporates :1 parallel line analysis of variance to compare tissue response lines to those of 3 purified SOD standard (Goldstein, 1965). All tissue dose response lines tested in this study were paral- lel to their respective standards, an indication that there were no 77 endogenous interfering substances which were contributing to or altering the 'apparent' SOD activity. It was determined in this study that the trout retina, like all other tissues, contains significant SOD activity. It is difficult how- ever to compare the values reported here to others reported in the literature since the specific activity of SOD is expressed in a number of ways, e.g., per mg protein, per mg DNA, per whole tissue, or per number of cells. Thus, internal controls were selected, i.e., tissues from the rainbow trout which were not normally exposed to hyperoxic con- ditions and had been shown to be sensitive to oxygen toxicity. From measurements of SOD-inhibitable ferricytochrome c reduction it was shown that the retina did not contain more total SOD activity than that mea- sured in the oxygen sensitive trout tissues. Despite the fact that total SOD activity was not highest in the retina, most evidence suggests that oxidative flux is greatest at the mitochondrial level and that mitochondrial SOD (MnSOD) plays 3 disproportionate role in oxidative defense (Stevens and Autor, 1977; Crapo et 31., 1980; Forman and Fisher, 1981; Freeman et 31., 1986). Thus, manganous SOD activity was measured in the retina and compared to activity in the brain and liver. Because the number of mitochondria differs between the tissues, MnSOD activity expressed per mg protein does not reflect the mitochondrial content of MnSOD. Therefore activity was also expressed per unit cytochrome oxidase. Cytochrome oxidase has previously been used as a mitochondrial marker enzyme (Freeman and Crapo, 1981; Freeman et 31., 1986). When MnSOD activity was expressed per unit cytochrome oxidase, there was no indication that MnSOD/mitochondria was greater in the retina relative to the other tissues. Thus, total SOD content as well as MnSOD content of 78 the oxygen resistant trout retina is no different than that measured in the oxygen sensitive brain and liver tissues from this species. Taken 1' together, the data indicate that regulation of 02 is not solely respon- sible for the resistance of the retina to hyperoxia. With sufficient superoxide dismutase to handle the superoxide flux the resistance to oxygen toxicity in the trout retina might be due to a more efficient defense mechanism to handle the byproduct of the dismuta- tion reaction, i.e., H202 (see equation 1). The two primary antioxidant enzymes which scavenge H202 are catalase and the glutathione/gluthathi- one peroxidase redox system. Catalase decomposes H into H 0 and 0 . 202 2 2 During the catalytic cycle, the active heme site undergoes successive divalent oxidation and reduction reactions with H 0 Catalase is pre- 2 2' sent in a variety of aerobic cells although its concentration can vary significantly among tissues. The enzyme may be found in the cytosol but most often it is localized within peroxisomes or microperoxisomes (Chance, Sies and Boveris, 1979). However, catalase activity could not be detected in retina or brain homogenates from the rainbow trout. Indeed, neither the presence of Triton X-100 to release catalase from peroxisomes nor the addition of ethanol to decompose the inactive form of catalase unmasked catalase activity. These findings, however, are consistent with the fact that mammalian retina and brain tissue are known to exhibit only low catalase activity (Bhuyan and Bhuyan, 1977; Cohen, Dembiac and Marcus, 1970). No further attempt was made to quan- titate catalase levels in these tissues, since liver activity was within the normal expected range (Cohen et 31., 1970). Although there is no evidence that catalase plays an important role in retinal defense, H O 2 2 concentration may still be limited by 79 glutathione peroxidase. Glutathione peroxidase can reduce 11202 by catalyzing the reduction of H to H 0 as reduced glutathione is 202 2 oxidized to form its corresponding disulfide. Glutathione peroxidase is distributed throughout the cytosol, and indeed it reacts more readily with H202 at low substrate concentrations than does catalase, suggesting that it may play a more important role in cellular decomposition of H202 (Fridovich and Freeman, 1986). For this reason, selenium dependent glutathione peroxidase activity was also determined. Selenium dependent glutathione peroxidase activity was detected in all three tissues but retinal glutathione peroxidase was not greater than that found in the other tissues. Thus, both catalase and glutathione peroxidase activities in the trout retina were not significantly greater than that measured in the other two oxidant sensitive tissues and therefore do not appear to play a significant role in the protection of the trout retina against oxidative damage. 13m: glutathione S-transferases have also reportedly served as antioxidants in rat lung as selenium independent glutathione peroxidases (Jenkinson et al., 1983), however, it has been reported that no GSH peroxidase activity of this kind could be detected in trout tissues by differential assay of GSH peroxidase with different substrates (Bell, Cowey and Youngson, 1984). Total sulfhydryl and non-protein sulfhydryl content was also mea- sured in the trout retina and compared to brain and liver. The concen— tration of soluble protein and non-protein sulfhydryl groups were deter- mined with Ellman's reagent (Sedlack and Lindsay, 1968). Reduced gluta- thione reportedly accounts for over 90% of the non-protein sulfhydryls in. the retina. (Winkler' and Giblin, 1983), therefore ‘values of 'non- protein sulfhydryls obtained with Ellman's reagent are reported as 80 glutathione. In addition to its role as substrate for the glutathione peroxidase enzyme, glutathione as well as other thiol (SH) compounds may provide protection against oxidative injury. These substrates act to repair oxidative damage by undergoing non-enzymatic hydrogen abstraction to restore function to oxidized proteins (Mannervik et 31., 1983). How- ever, no differences in either total sulfhydryl content or glutathione content were measured in any of the tissues studied. Thus, glutathione as substrate for glutathione peroxidaee or as a secondary antioxidant does not appear to contribute significantly to the oxidative resistance observed in the trout retina. Although it is difficult to make comparisons to values reported in the literature for the previously stated reasons, some relative compari- sons could be drawn, i.e., superoxide dismutase and glutathione peroxi- dase activities were highest in the liver and very low in brain tissues from the carp (Mazeaud, Maral and Michelson, 1979). High SOD and cata- lase activities have also been reported in the liver of mackerel (Aksnes and Njaa, 1981) and in livers of several freshwater fish species (Wdzieczak et 31., 1982; Desrochers and Hoffert, 1983). The measure of glutathione peroxidase activity in the trout liver in the present study compares favorably with some published values in rat and trout (Lawrence and Burk, 1976; Smith and Shrift, 1979), but it is considerably higher than values obtained from rainbow trout liver by Tappel et 31., (1982). The glutathione content of the trout retina is also comparable to the level determined in the rat retina (Winkler and Giblin, 1983). Taken together the results of the present study indicate that the primary antioxidants do not appear to play a significant role in the protection of the trout retina against oxygen toxicity. 81 éLRationale for the Adaptive Response 2£_Tissues ‘52 Acute !g_Chronic Hyperoxic Exposure In all models of acute hyperoxia, one or all of the key antioxidant enzymes (i.e., CuZn and Mn superoxide dismutase, catalase and glutathi- one peroxidase) have been induced in a relatively short period of time (Jamieson et 31., 1986). The effect of acute hyperoxic exposure on tissue antioxidant activity has primarily been studied in the lungs since, in vivo, pulmonary epithelial and lung capillary endothelial cells would be exposed to the same concentration of oxygen as that which is inspired. Due to the saturation of hemoglobin in air at 1 atm and the steep p0 gradients along the vascular supply and within the tissue, 2 most tissues are not exposed to high oxygen tensions, even in hyperoxic situations (Crapo and Tierney, 1974; Chance, Sies and Boveris, 1979). Crapo and Tierney (1974) first investigated the induction of antioxi- dants as a possible mechanism for tolerance to oxygen toxicity. They noted that SOD activity changed in parallel to oxygen tolerance and remained elevated for the duration of the tolerant state. Several investigators have reported the induction of a variety of antioxidants in lung tissue after hyperoxic exposure in rats (Kimball et 31., 1976; Crapo, Sjostrom and Drew, 1978; Crapo et 31., 1980; Jenkinson et 31., 1983). Significant increases in the primary antioxidants CuZn and Mn superoxide dismutase, catalase and glutathione peroxidase (GSHPx), as well as reduced glutathione and total non-protein sulfhydryl content have all been shown to occur in lung tissue after 5-14 days of exposure to 85-90% oxygen. It is thought that induction of these antioxidant de- fense systems is responsible for the partial protection of the tissues observed with subsequent hyperoxic challenge. Although a causal 82 relationship between induction of the enzymes and hyperoxic exposure is not yet clearly established, there is sufficient evidence to indicate that a number of the antioxidant enzymes do increase their activities after exposure to 85-90% oxygen for at least 5 days. The results presented here appear to contradict the free radical theory of hyperoxic damage, i.e. current theories on the biochemical basis for resistance to hyperoxia all point to a central role for the primary antioxidants. .A critical difference appears tui'be related to the type of oxygen exposure to which the cells are exposed, i.e., acute vs. chronic hyperoxic exposure. Tissues exposed to normoxic conditions (i.e., 20% oxygen) will undergo a rapid induction of one or more of their antioxidant defense enzymes when exposed to elevated 02 tensions (Crapo and Tierney, 1974; Kimball et 31., 1976; Forman and Fisher, 1981; Freeman et 31., 1986). However, the increased level of antioxidant defense will only yield a partial protection against oxidative damage when subsequently challenged with hyperoxia, i.e., tolerance to acute hyperoxic challenge is limited and oxidative damage is still observed (Crapo et 31., 1980; Freeman et 31., 1986). In contrast, the rainbow trout retina is 3 unique tissue that is completely resistant to hyper- oxic damage, yet does not rely on the primary antioxidant defenses. In contrast to all other tissues acutely exposed to hyperoxia, the rainbow trout retina is adapted to chronic hyperoxic exposure. Thus, the resis— tance of a particular tissue to oxidative injury may be dependent upon the means by which adaptation was attained, i.e. acute induction of antioxidant defenses vs an inherent genetic adaptation to hyperoxia. If so, one would predict that other tissues chronically adapted to hyper- oxia would exhibit similar resistance despite non-induced levels of the 83 primary antioxidants. Interestingly, recent evidence obtained in the teleost swimbladder and in a HeLa cell line adapted to 80% oxygen, support the contention that a genetic resistance to chronic hyperoxic exposure is afforded by unique, yet undefined, mechanisms. Superoxide dismutase, catalase and glutathione peroxidase have each been measured in the fish swimbladder, a tissue chronically exposed to extremely high oxygen tensions (Matkovics et 31., 1977; Mazeaud, Maral and Michelson, 1979; Morris and Albright, 1981; 1984). Interestingly, catalase and peroxidase activities in the swimbladders of the carp and toadfish were not significantly greater than the other tissues studied. In addition, no correlation was seen between average percent oxygen contained in the swimbladder and catalase or glutathione peroxidase activities from six marine species studied (Morris and Albright, 1984). Superoxide dismu- tase activity was also not significantly greater in the swimbladder of the carp when compared to other tissues (Matkovics et 31., 1977; Mazeaud et 31., 1979). However, swimbladder SOD was observed to be greater than gill, kidney and heart tissues in the toadfish (Morris and Albright, 1981). Morris and Albright (1981), postulated that the low SOD activity measured in the carp may be due to the lower oxygen tension in the physostomatous swimbladder of that species. However, in their study (Morris and Albright, 1981), SOD activity in the liver, 3 tissue which had been shown to have the highest superoxide dismutase activity of the tissues studied, was not reported (Matkovics et 31., 1977; Mazeaud et 31., 1979). In addition, their measure of SOD activity in the gas gland of the toadfish swimbladder was not corrected for’ contamination 'by blood, which they reported to have the highest superoxide dismutase activity (Morris and Albright, 1981). In a later study these authors do 84 report a positive correlation between average percent oxygen measured in the swimbladder and superoxide dismutase activity for six marine fish (Morris and Albright, 1984). However, activity was not measured in any of the other tissues and SOD activity did not change within any one species of fish. as swimbladder' oxygen tension.*was increased over' a period of 4 days. They concluded that SOD may play a role in the adap- tation of swimbladders to high levels of oxygen, but since it was not induced with increasing oxygen concentrations it appears that defense mechanisms other than the induction of the primary antioxidants play a key role in protecting the swimbladder against the rapid increases in oxygen concentration that occur during inflation (Morris and Albright, 1984). Additional evidence that primary antioxidants do not play a role in cells genetically adapted to chronic hyperoxia is demonstrated by recent studies with an oxygen adapted HeLa cell line. Joenje et 31., (1985), successfully selected a substrain of HeLa cells capable of withstanding increasing oxygen tensions. Indeed, they report that the hyperoxia- adapted HeLa cell line proliferates in 80% oxygen yet does not exhibit any of the hyperoxia induced increases in superoxide dismutase, catalase or glutathione peroxidase enzyme changes normally associated with acute hyperoxic exposure (Joenje et 31., 1985). In their study, comparisons were drawn between the HeLa cell line adapted to 80% oxygen and the parent cell line maintained in 20% oxygen. Under hyperoxic conditions, -adapted cells exhibited similar CN--resistant 20% 0 -adapted and 80% 0 2 2 respiration rates, suggesting that the increased 02 tolerance was not due to a decreased cellular production of activated oxygen species in hyperoxia (Joenje et 31., 1985). The cellular levels of CuZnSOD, MnSOD, 85 catalase, glutathione peroxidase and glucose-6-phosphate dehydrogenase activities were normal or slightly below normal in the 80% O ~3dapted 2 cells, leading the authors to conclude that these enzymes were of no significance for the increased 0 tolerance observed (Joenje et 31., 2 1985). They concluded that the increased tolerance was due to a geneti- cally determined increased resistance of oxygen sensitive cellular targets. In summary, the biochemical basis for insensitivity to hyperoxic exposure appears to be dependent upon the oxygen tension to which the cells are normally exposed. In cells exposed to a normoxic environment and challenged with hyperoxia, the threat of oxidative damage causes an induction of the primary antioxidants, which then enables the cells to tolerate subsequent hyperoxic challenge. However, in cells adapted to chronic hyperoxia, the resistance to oxidative damage is independent of the primary antioxidant defense system. Cells adapted to chronic hyper- oxia do not exhibit a decreased ability to generate reactive oxygen metabolites or enhanced antioxidant activities. Nonetheless, these cell types are completely resistant to oxidative injury. The biochemical basis for this resistance has not been determined but may be due to a genetic modification of key oxygen sensitive target sites. In the present study, it may be postulated that the lipid and/or protein composition of the rainbow trout retina has been altered such that oxidation of the lipid and/or protein molecules by oxidants is unlikely. Indeed, single amino acid substitutions in proteins or (enzymes can drastically decrease their sensitivity to oxidative damage (Janoff, George-Nascimento and Rosenberg, 1986). Recent evidence also suggests that hyperoxia may interfere with DNA synthesis and specifically damage 86 enzymes important in nucleic acid metabolism. Oxidative damage may occur at the level of thymidine kinase, which, when inactivated would impair the phosphorylation of thymidine to form nucleotides which are incorporated into the DNA molecule (Junod et 31., 1985; Junod, Petersen and Jornot, 1987). Taken together, the genetic resistance to oxidative injury of trout retinal cells, may be related to an inherent resistance of key oxygen sensitive target sites. The identification and analysis of key 0 -sensitive regulatory enzymes like thymidine kinase in the 2 teleost retinal cells, may provide specific insights into the basis of the resistance of the rainbow trout retina to hyperoxia. SUMMARY AND CONCLUSIONS Although the rainbow trout retina is continually' exposed to :ui elevated oxygen tension, it exhibits a resistance to oxygen toxicity. The mechanism for this resistance was investigated in the present study. Using the method of CN--insensitive 0 respiration for an estimate of 2 oxidant formation, it was determined that the rainbow trout retina can generate reactive oxygen species and that formation of these metabolites is not decreased under hyperoxic conditions in this tissue. In addi- tion, the primary antioxidants, superoxide dismutase, catalase and glutathione/glutathione peroxidase do not exhibit greater antioxidant activity in the trout retina when compared to the oxygen sensitive liver or brain tissues. These results are consistent with recent studies which report no correlation 'between resistance: to chronic. oxidative stress and an induction of the antioxidant defense systems. Clearly the mechanism for resistance to oxygen toxicity remains to be determined in this model, but may be related to 3 predetermined genetic resistance of Oz-sensitive target sites in cells which function in chronic hyperoxic conditions. Thus, the free radical theory of ”antioxidant induction to protect against hyperoxic exposure" requires modification, and should exclude those specialized tissues in which chronic exposure to hyperoxic conditions is the norm. 87 APPENDICES Appendix A Solutions Cortland's Saline Solution (pH 7.6) mmol/L AgjL NaCl 124.06 7.249 CaC12.2H20 1.56 0.229 KCl 5.10 0.380 NaH2P04.H20 2.97 0.410 NaHCO3 11.97 1.000 MgSOa.7HZO 0.93 0.229 Glucose 5.55 1.000 Oxygen Consumption Media (pH 7.4) mmol/L g/L KCl 105.0 7.828 KHZPO4 20.0 2.722 NaZ-EDTA 0.1 0.037 Glucose 5.0 0.901 Phosphate/EDTA 50mmol/L (pH 6.5) mmol/L g/L KHZPO4 --- 4.533 KZHP04 --- 20908 NaZ-EDTA 1.0 0.372 Adjust pH with 2.0 M KOH 88 Appendix B Lowry Protein Determination Principle Tyrosine and tryptophan in proteins react with Folin's phenol reagent to give a blue color which is read spectrophotometrically. Reference Oyama, V.I. and H. Eagle., Proc. Soc. Expt'l. Biol. Med., 91:305-307, 1956. Reagents A. Lowry A 1. Sodium carbonate (anhydrous) 60.0 g 2. Sodium hydroxide 12.0 g 3. Sodium potassium tartrate 0.6 g 4. Distilled water to make: 3000.0 ml B. Lowry B 1. Copper sulfate solution (CuSOA.SH20) 0.5 g% C. Lowry C 1. Lowry A 50.0 parts 2. Lowry B 1.0 part D. Phenol reagent according to Folin & Ciocalteu (2.0 N) 1. Phenol reagent (Sigma Chemical Co.) 1.0 part 2. Distilled water 1.0 part E. Human albumin protein standard 8.0 g% (Lot # PRS-426, Dade Reagents, Inc., Miami, Fl) 1. Dilute 3 ml vial with 300 ml distilled water to give 800 ug/ml. 89 90 Procedure 1. 2. 3. 4. 5. 6. 7. Concentrations of protein standards used for determination of standard curve: 0, 20, 40, 60, 80, 120 and 150 ug/ml. Volume of protein solution (standard or unknown) diluted to 1 ml with distilled water. Add 5 ml of Lowry C to protein solution. Incubate 20 min at room temperature. Add 0.5 ml phenol reagent and vortex rapidily. Incubate 30 min at room temperature, mix occassionally. Read at 660 nm on spectrophotometer. Appendix C Superoxide Dismutase Assay Principle Superoxide dismutase (SOD) is assayed on the basis of its ability to inhibit the rate of reduction of ferricytochrome c by superoxide anions generated by the xanthine - xanthine oxidase system. Manganese SOD is distinguished from the Cu,Zn SOD form by the latter's sensitivity to 2.0 mM potassium cyanide. Reference McCord, J.M. and I. Fridovich., J. Biol. Chem., 244:6049-6055, 1969. Reagents A. Stock Dispenser Solution 1. Xanthine 13.1 mg 2. Naz-EDTA 55.8 mg 3. NaZHPO4 9.680 g 4. KH2P04 0.928 g 5. Distilled water to make: 500 ml B. Stock Dispenser Solution with 6mM KCN for measurement of MnSOD activity 1. KCN 39.1 mg 2. Stock Dispenser Solution (A. above) to make: 100 ml C. Stock Ferricytochrome c Solution, 0.06 mM (Type III from Horse Heart, MW 12,384; Sigma Chemical Co., St. Louis, Mo.); (R. DiGuilio, personal communication) 1. This protein is contaminated with H 0 and reduced Cytochrome 2 c which absorbs at 550 nm, therefore the concentration must be adjusted spectrophotometrically. The extinction 91 2. 3. 92 coefficient (EM) for reduced-oxidized cytochrome c is 21,000 M-1.cm-1 at 550 nm. .Any spectrophotometric measurements should be performed on stock diluted 1:10. About 10-30% of commercial cytochrome c (cyt c) is in the reduced form, the remainder is oxidized. Therefore to make 250 ml of 0.06 mM cyt c (ox), a greater amount of the dry protein must be dissolved. Assume only 70% is oxidized, so for 250 ml of 0.06 mM cyt c (ox), one must dissolve 265.4 mg instead of 185.8 mg of dry cyt c. To calculate the exact ratio of cyt c (red) to cyt c (ox) and to adjust cyt c (ox) to 0.06 mM, proceed as follows: a) Dilute 265.4 mg/250 ml dHZO 1:10 and measure A550. Add a few grains sodium. dithionite, remeasure the A550 and subtract the "minus dithionite" A550. A550 (+DT) = 0.1615 = reduced + oxidized A (-DT) = 0.0250 = reduced 550 0.1365 0.1365 x 10 (dil. factor) = 1.365, since EM = 21,000, 1.365/21.0 = 0.065 mM cyt c (ox). b) To make the 250 ml cyt c stock solution 0.06 mM dilute it appropriately (i.e. C1Vl = C2V2)° If the A550 change of cyt c stock 1 dithionite is less than 1.26, add more dry cyt c and repeat concentration determination. c) Recheck the concentration of the dilute cytochrome c stock as in 33) above and then divide the stock into 12- 15 ml splits. FTeeze the cytochrome c in a dry ice - ethanol bath and store at 0 C. C. Stock Xanthine Oxidase Solution (Make fresh stock prior to run- ning the dose response curve for each tissue.) 1. 38 ul xanthine oxidase (from cow milk, Lot #10641720-84, 20 Units/ml; Boehringer' Mannheim, Indianapolis, IN). The concentration should be adjusted to yield an initial uninhib- ited reaction rate of 0.025 to 0.030 OD units/min. Since there is a small amount of ferricytochrome c reduction which cannot be inhibited with SOD, this initial rate must be .fifi .. '1'. I ‘3‘ ‘y’» I," 93 determined from the difference between the rate of reduction in the absence of any SOD and the rate in the presence of maximal amounts of SOD (e.g., 20 BESOD units in the reaction system). 2. Distilled H O to make 50 ml. 2 3. Stock should be kept on ice during assay. D. Stock Standard SOD Solution 1. One preprepared vial of lyophilized standard* 2. Distilled water 2.9 ml *Standard is prepared in advance as follows: BESOD (#8-8254; Sigma Chemical Co.) 10.1 mg Distilled H20 to make: 500 ml This solution is divided into 5 m1 aliquots which are frozen, lyOphilized, and stored dessicated 3t 0 C until used. 3. Working solution for dose response curve was made by diluting 0.5 m1 of stock to 10 mls with dis- tilled water (5 units/ml). E. Stock Tissue Homogenate Solution 1. See description of tissue homogenate preparation in text. Procedure 1. Prepare a reference cuvette as follows and place in the reference well of the thermostatically controlled (25 C) spectrophotometer: Appropriate Dispenser Stock Solution 1.0 m1 Cytochrome c Stock Solution 0.5 ml Extract or Standard appropriate dosage Distilled H20 to bring volume to 3.0 ml 2. Sample cuvette for determination is prepared as follows: Appropriate Dispenser Stock Solution 1.0 ml Cytochrome c Stock 0.5 m1 Extract or Standard appropriate dosage Distilled H 0 to bring volume to 2.0 ml 2 Final 94 Place sample cuvette into the sample well of the spectrophoto- meter, turn on the recorder chart drive (chart speed 1"/min) and turn on the spectrophotometer. Start the reaction by injecting 1.0 m1 of xanthine oxidase stock solution into the sample cuvette and mix with glass rod. Record the change in absorbance per minute at 550 nm. (Eldred, 1979). Assay Reaction Mixture Composition Assay 10 uM ferricytochrome c 50 uM xanthine 50 mM phosphate buffer, pH 7.8 at 25 C 0.1 mM EDTA 2.0 mM KCN (when appropriate) sample dosage Design 1. 6. Determine the uninhibited rate of ferricytochrome c reduction by measuring the basal rate in the absence of SOD in triplicate. Determine the rate of ferricytochrome c reduction in the presence of excess SOD (20 BESOD units) in duplicate. The difference in these rates is a measure of the true initial rate of reduction and should be on the order of 0.025-0.030 0D units /min. Generate 3 dose response curve using at least four doses between 10 and 60% inhibition of the uninhibited rate of reduction, for the BESOD standard, running duplicates at each dose. Generate a tissue dose response curve to measure total SOD activ- ity as was done for the standard. Repeat Steps 1 & 4 this time using the Stock Dispenser Solution containing 6.0 mM KCN in order to measure Mn-SOD activity. Use the average rate of reduction obtained from Step 2 to obtain the true initial rate of reduction. Make fresh Xanthine Oxidase Stock Solution and repeat steps 1-5 for the next tissue. 95 Decoding 12_the Coded Log Dose Scale 1. For the present discussion, let A be the lowest dose used and coded 1 on the arbitrary coded log dose scale. 2. If the successive doses are doublings, triplings, or quadru- plings, let B equal 2, 3, or 4 respectively and the coded doses will differ by log B units. 3. Any point on the arbitrary log scale can be decoded by subtract- ing the value 1, then multiplying by log B and adding log A. Expression 2: SOD Activity The unit of SOD activity may be defined as the amount of enzyme in- hibiting the rate of reduction of ferricytochrome c by 50% (i.e., ED50 1 95% confidence limits), or the activity may be expressed in terms of potency ratios with associated 95% confidence limits in relation to standards of known and verifiable activity. Appendix D Cytochrome c Oxidase Assay Principle Cytochrome c oxidase (COX) catalyzes the electron transfer from ferrocytochrome c to dioxygen in the respiratory chain of mitochondria. The catalytic activity of cytochrome c oxidase is determined by follow- ing the rate of oxygen uptake polarographically by means of an oxygen electrode. Cytochrome c is kept reduced by ascorbate, with TMPD media- ting the electron transport. The blank rate of oxygen consumption, due to the low autoxidation rates of ascorbate and TMPD alone and in the presence of cytochrome c, is subtracted from the measured oxygen uptake. Reference Rafael, J., In: Methods p£_Enzymatic Analysis, Third Edition, edited by H.U. Bergmeyer. New York: Academic Press, 1983, p. 266-273. Reagents A. Phosphate/EDTA solution (phosphate buffer, 50 mmol/L, pH 6.5; EDTA 1 mmol/L) 1. KHZPO4 2.266 g 2. KZHPOQ 1.454 g 3. NaZ-EDTA 0.186 g 4. Distilled water 450 ml 5. Adjust pH to 6.5 with 2.0 M KOH 6. Adjust volume with distilled water to make: 500 ml B. Ascorbate (0.1 M) 1. L-Ascorbic acid (Sigma Chemical Co.) 0.176 g 2. Phosphate/EDTA solution 9 ml 96 3. 4. 5. 97 Adjust pH to 6.0 with 2 M KOH (no higher, turns yellow) Adjust volume with Phosphate/EDTA (A) to make: 10 ml Freeze 400 ul aliquots in dry ice - ethanol bath and store frozen until ready for use. C. N,N,N',N'-Tetramethyl-p-phenylenediamine, (50 mM) 1. 2. 3. 4. TMPD dihydrochloride (Eastman Kodak) 0.118 g Place in cold flask on ice Add ice cold distilled water to make: 10 ml Rapidly freeze 100 ul aliquots in dry ice-ethanol bath and store frozen and protected from light. If solution turns darker than a very faint purple, discard and make fresh solution. D. Reduced cytochrome c (0.7 mM; Type III from Horse Heart, MW 1. 12,384; Sigma Chemical Co.) See below for purification procedure E. Tris/Acetate Solution (50 mM; pH 7.5) 1. Trizma base (Sigma Chemical Co.) 6.055 g 2. Distilled water 900 ml 3. Adjust pH to 7.5 using glacial acetic acid 4. Adjust volume with distilled water to make: 1.0 L F. Tissue Homogenate preparation see text Purification p£_§ytochrome E Only 10-30% of the commercially available cytochrome c is in the reduced form, the remainder is in the oxidized form. Therefore the cytochrome c must be reduced with sodium dithionite and then desalted before the correct concentration can be adjusted spectrophotometrically. A. Preparation of column 1. 2. 3. Clean and dry 3 7 mm ID x 300 mm column. Prepare bottom plug using an 18 guage needle by cutting the barrel end at the needle edge and shaving it to fit inner hole of an invertible rubber stopper. Wrap parafilm around the joint. The opening should be level with the stopper bottom, and the stopper should fit snugly into the tube. Attach PE 160 tubing to needle end and clamp. 4. 98 Pack glass wool around the bottom plug and make sure no air bubbles are trapped in the parafilm or glass wool by tamping with a steel rod. Packing and testing column 1. 2. 3. Prepare sephadex G-25 (coarse) by weighing out 4 grams and adding 40 mls of the buffer to be used. Heat at 100 C for 2-3 hrs and cool overnight at 10 C. After cooling, remove air bubbles from sephadex by evacuating with vaccum filter flask for 10-15 minutes with intermittent swirling. Attach column. to support and check that 11: is perfectly vertical. Add buffer (4-6 mls) to column and let 2-3 mls drain out then close outlet. Gently add gel slurry to column with pasteur pipette and allow to settle. If gel begins to pack before completely poured, mix top layer. Fill column to bed height and allow it to pack. Gel should pack as one continous column. Run 100-150 ml of buffer through column to stabilize column, then mark bed level on outside. To test column for leakage and to obtain separation volumes run 1 test separation of Blue Dextran 2000 and Bromphenol blue (Cooper, 1977). Running cytochrome c sample 1. 2. 3. Run 100-150 ml of Tris/acetate solution through column to prepare column for cytochrome c desalting. Dilute cytochrome c sample with distilled water and add a few grains of sodium dithionite to fully reduce the cytochrome c (e.g., 0.1305 g of cytochrome c diluted with 0.3 ml water). Store concentrated sample on ice. Allow the Tris/acetate solution to drain until it just reaches the bed surface. Place the 0.3 ml cytochrome c solution unto the column with a drawn out pasteur pipette and allow it to drain just below the bed surface. D. Procedure 8. 99 Layer 1.0 ml of TTis/acetate solution on top of column and allow it to run into the column so that all of the sample is into the column. Add Tris/acetate solution to top of column, connect reservoir containing Tris/acetate solution and elute at 1 ml/min. Cytochrome c will chromatograph in three layers, the initial front will be comprised of the polymer form and should be discarded. The next layer, the darkest band, is the purified cytochrome c and should be collected and saved on ice. The last layer contains the de-aminated cytochrome c and also should be discarded. The column should be washed with buffer and preserved with 0.02% sodium azide. Determination of Cytochrome c concentration 1. 2. The chromatographed cytochrome c should be diluted to 3 ml and then its concentration determined spectrophotometrically. Reduced cytochrome c absorbs at 550 nm with an extinction coefficient of 21,000 M-1.cm-1. The spectrophotometric measurements should be performed on stock diluted 1:10. The final concentration should be adjusted to 0.7 mM and then 400 ul aliquots should be frozen in a dry ice - ethanol bath and kept frozen until used. Pipette the following reagents successively into the waterjacketed (25 C) electrode chamber; the smaller volumes ((0.2 m1) are injected into the chambers through the access groove using a 2" needle attached to a Hamilton syringe: l. 2. 3. 4. 5. Phosphate/EDTA solution 2.400 ml Wait for 3-5 min for final equilibration with air. Mixing should be stopped during the following addi- tions to prevent the introduction of air bubbles. Ascorbate solution 0.200 ml TMPD solution 0.040 ml Record oxygen consumption for 2-3 min until the assay blank remains constant. a an 6. 7. 8. 9. Final 100 Cytochrome c solution 0.120 ml Record oxygen consumption for 2-3 min until the assay blank remains constant. Homogenized sample containing 1% Lubrol (1:1) 0.040 ml Oxygen consumption is monitored on the recorder. Assay Reaction Mixture Composition 50 mM Phosphate buffer, pH 6.5 1 mM EDTA 7 mM Ascorbic Acid 0.7 mM TMPD 30 uM Cytochrome c sample dosage Assay Design 1. Set the electrical zero point on the recorder and then adjust the recorder to 100% with air saturated Phosphate/EDTA solution (25 C), allow 3-5 minutes of mixing for saturation. Add Phosphate/EDTA solution, ascorbate, and TMPD, record the rate of consumption and determine the autoxidation rate for the latter two reagents. Add the tissue sample using a syringe microburet attached to PE-50 tubing, and record oxygen consumption. Inject the cytochrome c solution and record the rate of consumption. Repeat steps 1-4, only this time reversing steps 3 & 4, so that the uptake rate due to the addition of cytochrome c alone can be determined. Total recorder range should equal 1 and total assay blanks should not exceed 0.02-0.03/min. Repeat steps 1-5 so that duplicate measurements are obtained for each tissue sample. A new chamber should be used after each determination. Traces of cytochrome c are removed by washing the chambers with 0.1 N HCl and rinsing thoroughly with distilled water. 101 Calculation pf Activity 1. Determine the assay blank rates of consumption due to the ascor- bate + TMPD and due to the cytochrome c, average them for the duplicate measurements, and subtract the total blank rate from the measured uptake rate to obtain the true consumption per unit time. The catalytic activity of the sample is determined using the following equation: 02 consumption = 259;§_!.x —E-umoles oxygen consumed/L/second where: 240 = (equivalent to 240 umol 0 dissolved/L at 25 C X = oxygen consumption recorded (corrected for assay blank) V = assay volume in ml v = volume of sample added in ml t = time for oxygen consumption in seconds Since cytochrome oxidase catalyzes the transfer of one mole of electrons per 1/4 mole of oxygen, then 4 umoles of electrons are transfered per umole of oxygen. The katal is equivalent to the transfer of 1 mole of electrons per second. Therefore the rate of consumption can be converted to ukat/L by multiplying by 4. Appendix E Catalase Assay Principle The catalatic activity of catalase is assayed in crude tissue extracts using a rapid spectrophotometric variation of the technique for determining catalase activity by measuring H O 2 2 by titration with permanganate. Reference Cohen, 0., D. Dembiec and J. Marcus., Anal. Biochem., 34:30-38, 1970. Reagents I. Isotonic Buffered Solution A. Stock Solution 1. N3Cl 180.0 g 2. NazHPO4 27.3 g 3. NaH2P04.H20 4.86 g 4. Distilled water 1.9 L 5. Adjust pH to 7.40 6. Adjust volume with distilled water to make: 2.0 L B. Working Solution 1. Isotonic Buffer - Stock Solution 90.0 ml 2. Distilled water to make: 1.0 L II. Ethanol 1. Ethanol (100%) 10.0 ml III. 10% Triton X-100 (Sigma Chemical Co.) 1. Triton X-100 2.5 ml 2. Add isotonic buffer to make: 25 ml 102 IV. VI. VII. VIII. IX. 103 3. Dissolve in solution by sonifying. Monobasic Potassium Phosphate (1.0 M) 1. KHZPO4 2. Distilled water to make: Dibasic Potassium Phosphate (1.0 M) l. KZHPO4 2. Distilled water to make: Phosphate Buffer A. Stock Solution (1.0 M) 1. KB P04 (1.0 M) 2 2. K HPO4 (1.0 M) B. Worki:g_Solution (0.01 M) 1. Phosphate Buffer - Stock Solution 2. Distilled water to make: 3. Adjust pH to 7.0 Hydrogen Peroxide Solution (6 mM) 1. 30% H202 2. Phosphate Buffer to make: Sulfuric Acid (6 N) 1. Concentrated H2804 2. Distilled water to make: Potassium permanganate solution (0.01 N) 1. KMnO4 2. Distilled water to make: Catalase solution A. Stock Solution 1. Catalase* 2. Distilled water to make: B. Working Solution 13.609 100 17.418 100 9.0 10.0 2.0 200 0.06 100 168.0 1.0 0.316 1.0 2.0 10.0 1. Stock Catalase, at least 4 dilutions between 0.05-0.5 2. Distilled water to make: 10.0 * Crystalline beef-liver catalase (Sigma Chemical Co., St. Louis, Mo., #C-40). 8 ml ml ml ml ml ml ml ml ml ml ml ml 104 Assay Procedure The reagents are added sequentially at fixed intervals and allowed to react for exactly 3 minutes. The remaining H 0 is then reacted with 2 2 a standard excess of KMnOa and the absorbance of the residual KMnO is measured at 480 nm. All reactions are carried out in an ice-water 83th (0-2 0). 1. Place duplicate 0.5 ml aliquots of the cold catalase samples (either standards or diluted tissue samples) into cold 20-40 ml test tubes. 2. Include duplicate blanks consisting of 0.5 ml distilled water. 3. For tissues exhibiting weak catalase activity (e.g., brain) it is necessary to include a different blank (i.e., by simply reversing the order of addition of the acid and the peroxide) to correct for substances that can reduce H202 or KMnOa. 4. Initiate the enzymic reactions sequentially at fixed intervals by adding 5.0 ml of cold 6 mM H and vortex thoroughly. 0 2 2 5. After exactly 3 minutes, stop the reactions sequentially at the same fixed intervals by rapidly adding 1.0 ml 6N H280A and vortexing. Spectrophotometry The H202 is measured by reacting it with a standard excess of KMnO and then measuring the residual KMnO 4 spectrophotometrically. The temperature of the cell within the spelctrophotometer is maintained at 2-4 C. 1. Prepare duplicate spectrophotometric standards by adding 7.0 ml of 0.01 N KMnO to a mixture of 5.5 ml of 0.01 M phosphate 4 buffer and 1.0 ml 6 N H SO 2. Combine the duplicate 2spehctrophotometric standards and read the absorbance at 480 nm using a single beam spectrophoto- meter. 3. For convience in. performing later calculations, adjust the digital readout to read 1.000 absorbance unit for the standard. 4. The remaining enzyme-reaction samples and the blanks are taken one at a time. To each, rapidly add 7.0 m1 KMnO reagent, mix 4 105 thoroughly, and read at 480 nm within 30-60 seconds. To avoid precipitation of MnO2 in the cell, rinse with distilled water between samples. Calculations As described, the decomposition of H 0 by catalase follows first 2 2 order kinetics (See Equation 1). k = log (so/$3) x 2.3/t (1) where: k = first order reaction rate constant, The and 3 time interval over which reaction is allowed to occur (3 minutes), S o 3 substrate concentration at time zero, substrate concentration at 3 minutes. values SO and S are expressed in absorbance units. 1. 3 To obtain 50’ subtract the average absorbance of the reaction system blanks from the absorbance of the spectrophotometric standard. To obtain S3, subtract the absorbance of the reaction samples from the spectrophotometric standard. Results can now be expressed in terms of the first order reaction rate constant (k). Alternatively, a value k may be expressed in terms of units of activity as described by one of the following three methods. (Luck, 1965). a) A catalase unit as defined by Bergmeyer is the amount of enzyme which liberates half the peroxide oxygen from 3 H202 solution in 100 seconds at 25 C. Since the subtrate concentration is reduced by one half for a given half life then: 1 tau == MILEZ = 1n 2 = 0.693 [sec k A unit of enzyme activity is therefore related to the 1 (2) 7" W half-life time (tau) of a first order reaction: 100 _ 100 = k observed [sec] (3) tau observed — 0.6937k 0.00693 1 unit = '1'!- b) c) d) 106 An international unit of catalase is defined as the umoles of substrate converted per mg protein per min at 25 C. To convert to I.U. the H202 concentration must be defined. Since, according to Bergmeyer, 1 unit is defined for t = 100 seconds and therefore for a first order reaction k = 0.00693 sec-l, it follows that by substituting this value for k and S0 = 30 umoles H O /5 ml in equation 1, that approximately 2 2 10.2 umoles H202 are decomposed in 60 seconds. (The value 30 is from an initial H202 concentration of 10 mM or 30 umoles H202/5 ml). The value 10.2 is obtained from the difference between 30 and the value obtained from equation 1 with the above substitutions. Therefore 10.2 International Units are equal to 1 unit according to Bergmeyer or I.U. = 10.2 x units according to Bergmeyer. Activity can also be expressed in terms of Kat.f. units (Katalasefahigkeit) in order to facilitate comparisons of crude tissue homogenate activities: Kat.f. = 1/2.3 x k/W (4) where 1: is the first order reaction rate constant [min-l], the value 2.3 is for logarithmic conversion and W is the amount of protein in a reaction volume of 50.0 ml. The value of W is equal to (50.0/5.5) x the amount of protein in the 0.5 ml homogenate volume for the method described here. Protein concentrations are determined using the method of Lowry (Oyama and Eagle, 1956), so that activity' can be expressed on a per mg protein basis. Appendix F Glutathione Peroxidase Assay Principle Glutathione peroxidase catalyzes the oxidation of glutathione by H202 or an organic hydroperoxide. The oxidized form of glutathione (GSSG) can be converted back to the reduced form (GSH) through the addi- tion of exogenous glutathione reductase and NADPH, thereby maintaining substrate concentration for the enzyme. Glutathione peroxidase activity can therefore be measured indirectly by following the decrease in absor- bance of the reaction mixture at 340 nm as NADPH is converted to NADP. Reference Paglia, D.E. and WRN. Valentine., J. Lab. Clin. Med., 70:158-169, 1969. Reagents A. Potassium Phosphate buffer/EDTA (0.05 M phosphate; 5 mM EDTA) 1. KZHPO4 0.531 g 2. KHZPO4 0.265 g 3. EDTA-Na2 0.186 g 4. Distilled water 90 ml 5. Adjust pH to 7.0 6. Adjust volume with distilled water to make: 100 ml B. Reduced Nicotinamide Adenine Dinucleotide Phosphate (8.4 mM) 1. NADPH 0.016 g 2. Distilled water to make: 2 ml C. Glutathione Reductase (Type VII; 75 units/mg solid; Lot #22F-8080; Sigma Chemical Co., St. Louis, MO) 1. Glutathione reductase 1.3 mg 107 IV. V. VI. VII. 108 2. Distilled water to make: 1.0 ml Sodium Azide (1.125 M) 1. NaN3 1.829 g 2. Distilled water to make: 25 m1 Glutathione (0.15 M) 1. GSH 1.153 g 2. Distilled water to make: 25 m1 Hydrogen Peroxide (0.0022 M) 1. 30% H202 (w/v) 25 ul 2. Distilled water to make: 100 ml Sucrose (0.25 M) 1. Sucrose 8.558 g 2. Distilled water to make: 100 ml Assay Procedure 1. 4. The following reagents are mixed in a 3 ml cuvette or precombined in large fresh pools and kept on ice, so as to expedite sample measurements. a. Phosphate buffer/EDTA 2.58 ml b. B-NADPH 0.10 ml c. Glutathione reductase 0.01 ml d. Sodium azide 0.01 ml e. Reduced glutathione 0.10 ml The above mixture is pipetted into the which controlled spectrophotometer and allowed to equlibrate to 25 C reference and sample cuvettes, are then placed into the thermostatically for 2-3 minutes. The tissue sample (100 ul) is then added to the sample cuvette, while an equivalent volume of distilled water is added to the reference cuvette and mixed. Sample size is reduced to 50 ul to more accurately measure tissues which exhibit a greater degree of activity (i.e. liver). The difference in volume is made up with phosphate buffer. The reaction is initiated with the addition of 0.1 ml of the H 0 2 2 solution to both the reference and sample cuvettes. 109 Spectrophotometry Enzyme activity is measured indirectly by following the decrease in absorbance of the reaction mixture at 340 nm as NADPH is converted to NADP at 25 C. A Beckman DB-G Spectrophotometer equipped with a linear recorder is used to obtain a continuous recording of the absorbance changes. True enzymatic activity is measured through the use of a dual beam spectrophotometer which subtracts out the nonenzymatic oxidation of GSH which occurs simultaneously in the reference cuvette. Calculations Glutathione peroxidase activity was determined from the molar extinction coefficient for NADPH at 340 nm, which is 6220 l.mol.-1.cm-l (Mattenheimer, 1971). Protein determinations were made using the method of Bradford (1976) so that activity could be expressed as mUnits/mg protein or nmoles.min-l.mg protein-l. The general formula for the calculation of the international enzyme unit is: 6 A 10 1 TV _ -1 -1 -1 E x d X X '—E X 'SV' - nmoles.ml .min or mU.ml where° = change in absorbance A E = molar extinction coefficient d = diameter of the cuvette in cm TV = total assay volume SV = sample volume t = reaction time in minutes mU/ml milliunits/ml or nmoles/ml/min and 106 converts moles/liter to umoles/ml Appendix G Assay for Estimation of Sulfhydryl Content Principle This procedure is a simple spectrophotometric method for the routine determination of the sulfhydryl content in the protein-bound (PB-SH) and nonprotein-bound (NP-SH) fractions of various tissues. The method is based on the reduction of 5,5'-dithiobis-(2-nitrobenzoic acid) by SH groups to form 1 mole of 2-nitro-S-mercaptobenzoic acid per mole of SH. The nitromercaptobenzoic acid anion has an intense yellow color which absorbs at 412 nm and can be used to measure SH groups. Reference Sedlak, J. and R.H. Lindsay. Anal. Biochem., 25:192-205, 1968. Reagents A. 5,5'-dithiobis-(2-nitrobenzoic acid) solution (0.01 M) 1. DTNB 0.099 g 2. Methanol (absolute) 25 ml B. EDTA-Na2 solution (0.2 M) 1. EDTA-Na2 37.224 g 2. Distilled water to make: 500 ml C. Tris buffer (0.2 M) pH 8.2 1. Trizma base 13.369 g 2. Trizma-H01 14.157 g 3. EDTA-Na2 (0.2 M EDTA) 100 ml 4. Distilled water 850 ml 5. Adjust pH with 1N HCl to 8.2 6. Adjust volume with distilled water to make: 1.0 L D. Tris buffer (0.4 M; pH 8.9) 110 V. VI. VII. Assay 1. 2. 3. 4. 5. 6. 111 Trizma base Trizma-HCl EDTA-Na2 (0.2 M) Distilled water Adjust pH with 1 N HCl to 8.9 Adjust volume with distilled water to make: EDTA-Na solution (0.02 M) 2 l o EDTA-N82 2. Distilled water to make: Methanol-Absolute 1. 100% Methanol 50% Trichloroacetic acid 1. Trichloroacetic acid (100%) 2. Distilled water to make: Procedure Total sulfhydryl content is determined by reacting 0.5 m1 of the tissue homogenates buffered in 0.2M Tris at pH 8.2 with Ellman's reagent (5,5'-dithiobis-(2-nitrobenzoic acid)) in the presence of absolute methanol. A. All buffers and solutions should be gassed 2-3 minutes with a 10.633 1.928 25 200 250 7.445 1.0 100 25 50 g g ml ml ml ml ml ml vigorous stream of nitrogen prior to their use in the assay. Make up the following reaction mixtures in an ice water bath 1. Tissue homogenate (Run in duplicate) a. Place 0.5 ml of tissue homogenate in 15.0 ml tubes. b. Add 1.5 ml of 0.2 M Tris buffer (pH 8.2). c. Add 0.1 ml of 0.01 M DTNB solution and mix. d. Bring up to 10.0 ml with 7.9 ml of methanol. 2. Sample blank 3. Place 0.5 ml of tissue homogenate in 15.0 ml tube. be Add 105 m1 Of 002 M Tris bUffer (pH 8.2). c. Bring up to 10.0 ml with 8.0 ml of methanol. 3. Reagent blank (Run in duplicate) test test 1 VJ “a: 2.4 \fkji .‘. II. C. F. 112 a. Place 2.0 ml of 0.2 M Tris buffer (pH 8.2) in 15.0 ml test tubes. b. Add 0.1 ml of 0.01 M DTNB solution and mix. c. Bring up to 10.0 ml with 8.0 ml of methanol. Stopper the tubes and let stand 30 minutes with occassional shaking. Filter twice through the same folded filter paper such that a clear filtrate is produced. An alternative method would be to allow color to develop over 15 minutes then centrifuge at 3000g at room temperature for 15 min. The duplicate reagent blanks are mixed and used to set the spectrophotometer to full scale (100% Transmittance). The absorbance of the filtrates is read at 412 nm in a Beckman DB-G dual beam spectrophotometer using 1 cm quartz cuvettes with the respective sample blanks in the reference cuvette. Non-protein bound sulfhydryl content is determined using 2.5 ml of the tissue homogenate and reacting it with 50% trichloroacetic acid in order to precipitate out the proteins. Color is developed within 10-15 minutes and the mixture is filtered in order to obtain a clear filtrate. A. B. C. All solutions should be gassed with 100% N2 for 2-3 minutes. Make up the following reaction mixture in an ice water bath (o-2c). 1. Tissue homogenates (Run in duplicate). a. Place 2.5 ml of tissue homogenates in 15.0 ml test tubes. b. Add 2.0 ml of distilled water. c. Add 0.5 ml of 50% trichloroacetic acid. d. Shake tubes 10-15 minutes. e. Filter twice through the same folded filter paper such that a clear filtrate is produced. A 2 ml aliquot of the filtrate buffered with 0.4 M Tris at pH 8.9 is then reacted with Ellman's reagent and the absorbance measured against a reagent blank at 412 nm within 5 minutes. Make up the following reaction mixtures for spectroscopy. 113 1. Tissue homogenates a. Place 2.0 ml of the clear filtrate (from part B above) into 15.0 ml test tubes. b. Add 4.0 ml of 0.4 M Tris buffer (pH 8.9) and mix. c. Add 0.1 ml of 0.01M DTNB, mix and read absorbance within 5 minutes. 2. Reagent blank (Run in duplicate). a. Place 6.0 ml of 0.4 M Tris buffer (pH 8.9) into 15.0 ml test tubes. b. Add 0.1 ml of 0.01M DTNB, mix and read absorbance within 5 minutes. D. The duplicate reagent blanks are mixed and used to set the spectrophotometer to full scale (100% Transmittance). E. The absorbance of the tissue homogenates is read at 412 nm in a Beckman DB-G dual beam spectrophotometer using 1 cm quartz cuvettes with the reagent blank in the reference cuvette. III. Determination of Protein-Bound Sulfhydryls. A. Protein-bound sulfhydryl content is determined as the difference between the total and nonprotein-bound sulfhydryl groups. Calculation p£_Sulfhydryl Content 1. The total sulfhydryl and non-protein sulfhydryl content of the tissues is calculated from the experimentally determined molar extinction coefficient of 13,100 l.mol"l.cm-1 at 412 nm. The general formula for the calculation is: A 10 TV _ -1 X X SV' ~ umoles.ml where: A absorbance E - molar extinction coefficient d = diameter of the cuvette in cm TV 8 total assay volume SV a sample volume and 106 converts moles/liter to umoles/ml 2. 114 The protein bound sulfhydryl content is calculated by subtracting out the non-protein bound sulfhydryls from the total sulfhydryl content of the tissues. The Lowry method (Oyama and Eagle, 1956), was used to determine the protein concentration so that activity could be expressed on a per mg protein basis. LI ST 01" REFERENCES LIST OF REFERENCES Aksnes, A. and L.R. Njaa. Catalase, glutathione peroxidase and super- oxide dismutase in different fish species. Comp. Biochem. Physiol. 69B: 893-896, 1981. Ashton, N. Some aspects of the comparative pathology of oxygen toxicity in the retina. Brit. J. Ophthal. 52:505-531, 1968. Ashton, N., B. Ward and G. Serpell. Role of oxygen in the genesis of retrolental fibroplasia. Brit. J. Ophthal. 37:513-519, 1953. Baeyens, D.A., J.R. Hoffert and P.O. Fromm. A comparative study of oxygen toxicity in the retina, brain and liver of the teleost, amphibian and mammal. Comp. Biochem. Physiol. 45A:925-932, 1973. Baeyens, D.A., J.R. Hoffert and P.O. Fromm. A comparative study of the influence of oxygen on lactate dehydrogenase. Comp. Biochem. Physiol. 478:1-7, 1974. Balentine, J.D. Pathology p£_0xygen Toxicity. New York: Academic Press, 1982. Bartoli, G.M., T. Galeotti, G. Palombini, G. Parisi and A. Azzi. Different contribution of rat liver microsomal pigments in the formation of superoxide anions and hydrogen peroxide during development. Arch. Biochem. Biophys. 184:276-281, 1977. Bell, J.G., C.B. Cowey and A. Youngson. Rainbow trout liver microsomal lipid peroxidation. The effect of purified glutathione peroxidase, glutathione S-transferase and other factors. Biochim. Biophys. Acta 795:91-99, 1984. Bhuyan, D.K. and K.C. Bhuyan. Regulation of hydrogen peroxide levels in eye humors, effect of 3-amino-1H-1,2,4-triazole on catalase and glutathione peroxidase of rabbit eye. Biochim. Biophys. Acta 497:641-651, 1977. Bielski, B.H.J., H.W. Richter and P.C. Chan. Some properties of the ascorbate free radical. Ann. N.Y. Acad. Sci. 258:231-237, 1975. Boveris, A. and E. Cadenas. Production of superoxide radicals and hydro- gen peroxide in mitochondria. In: Superoxide Dismutase, edited by L.W. Oberly. Boca Raton, Fl: CRC Press, 1982, vol. II, p. 15-30. 115 116 Boveris, A. and B. Chance. The mitochondrial generation of hydrogen per- oxide. General properties and effect of hyperbaric oxygen. Biochem. J. 134:707-716, 1973. Bradford, M.M. A rapid and sensitive method for quantitation of micro- gram quantities of protein utilizing the principle of protein dye binding. Anal. Biochem. 72:248-253, 1976. Bresnick, G.H. Oxygen-induced visual cell degeneration in the rabbit. Invest. Ophthal. 9:372-387, 1970. Bridges, W.Z. Electroretinographic manifestations of hyperbaric oxygen. Arch. Ophthal. 75:812-817, 1966. ‘1 -L .1- Chance, B., H. Sies and A. Boveris. Hydroperoxide metabolism in mammali- an organs. Physiol. Rev. 59:527-605, 1979. Cohen, G. In defense of Haber-Weiss. In: Superoxide and Superoxide Dismutases, edited by A.M. Michelson, J.M. McCord and I. Fridovich. New York: Academic Press, 1977, p. 317-321. Cohen, 0., D. Dembiec and J. Marcus. Measurement of catalase activity in tissue extracts. Anal. Biochem. 34:30-38, 1970. Cooper, T.G. Gel Permeation Chromatography. In: The Tools of Biochemistpy. New York: Wiley & Sons, 1977, p. 169-193. Crapo, J.D. Morphologic changes in pulmonary oxygen toxicity. Ann. Rev. Physiol. 48:721-731, 1986. Crapo, J.D., B.E. Barry, H.A. Foscue and J. Shelburne. Structural and biochemical changes in rat lungs occuring during exposures to lethal and adaptive doses of oxygen. Am. Rev. Resp. Dis. 122:123-143, 1980. Crapo, J.D., K. Sjostrom and R.T. Drew. Tolerance and cross tolerance using N0 and O . I. Toxicology and biochemistry. J. Appl. Physiol.? Respi at. Environ. Exercise Physiol. 44:364-369, 1978. Crapo, J.D. and D.F. Tierney. Superoxide dismutase and pulmonary oxygen toxicity. Am. J. Physiol. 226:1401-1407, 1974. Czapski, G. Reaction of 'OH. In: Methods lp_Enzypology, edited by S.P. Colowick and N.0. Kaplan. New York: Academic Press, vol. 105, 1984, p. 209-215. Deneke, S.M. and B.L. Fanburg. Normobaric oxygen toxicity of the lung. N. Engl. J. Med. 303:76-86, 1980. Desrochers, P.E. and J.R. Hoffert. Superoxide dismutase provides protec- tion against the hyperoxia in the retina of the rainbow trout (Salmo gairdneri). Comp. Biochem. Physiol. 76B:24l-247, 1983. 117 Desrochers, P.E., K.A. Pratt, P.O. Fromm and J.R. Hoffert. Oxygen diffusion in the trout retina. Exp. Eye Res. 41:607-618, 1985. Eldred, G.E. Assay of superoxide dismutase in the hyperoxic retinal and choroidal tissues of the rainbow trout (Salmo gairdneri). Ph.D. Dissertation, Michigan State University, E. Lansing, MI, 1979. Eldred, G.E. and J.R. Hoffert. A test for endogenous interferences in superoxide dismutase assays. Anal. Biochem. 110:137-143, 1981. Fairbanks, M.B., J.R. Hoffert and P.0. Fromm. The dependence of the oxygen-concentrating mechanism of the teleost eye (Salmo gairdneri) on the enzyme carbonic anhydrase. J. Gen. Physiol. 54:203-211, 1 1969. Fairbanks, M.B., J.R. Hoffert and P.O. Fromm. Short circuiting of the ocular oxygen concentrating mechanism in the teleost Salmo * gairdneri using carbonic anhydrase inhibitors. J. Gen. Physiol. 64:263-273, 1974. Fee, J.A. and J.S. Valentine. Chemical and physical properties of super- oxide. In: Superoxide and Sgperoxide Dismutases, edited by A.M. Michelson, J.M. McCord and I. Fridovich. New York: Academic Press, 1977, p. 19-60. Flohe, L., A. Azzi, G. Loschen and C. Richter. Superoxide radicals in mitochondria. In: Superoxide and Superoxide Dismutases, edited by A.M. Michelson, J.M. McCord and I. Fridovich. New York: Academic Press, 1977, p. 323-334. Flohe, L. and F. Otting. Superoxide dismutase assays. In: Methods 12 Enzymology, edited by L. Packer. New York: Academic Press, 1984, V01. 105, p. 93-104. Fonner, D.B., J.R. Hoffert and P.O. Fromm. The importance of the countercurrent multiplier mechanism in maintaining retinal function in the teleost. Comp. Biochem. Physiol. 46A:559-567, 1973. Foote, 0.8. Light, oxygen and toxicity. In: Pathology p£_0xygen, edited by A. Autor. New York: Academic Press, 1982, p. 21-44. Foote, C.S., F.C. Shook and R.B. Abakerli. Characterization of Singlet Oxygen. In: Methods p£_Enzymology, edited by L. Packer. New York: Academic Press, 1984, vol. 105, p. 36-47. Forman, H.J., and A.B. Fisher. Antioxidant enzymes of rat granular pneumocytes. Constitutive levels and effect of hyperoxia. Lab. Invest. 45:1-6, 1981. Freeman, B.A. and J.D. Crapo. Hyperoxia increases oxygen radical produc- tion in rat lungs and lung mitochondria. J. Biol. Chem. 256:10986- 10992, 1981. 118 Freeman, B.A. and J.D. Crapo. Biology of disease. Free radicals and tissue injury. Lab. Invest. 47:412-426, 1982. Freeman, B.A., R.J. Mason, M.C. Williams and J.D. Crapo. Antioxidant enzyme activity in alveolar type II cells after exposure of rats to hyperoxia. Exp. Lung Res. 10:203-222, 1986. Freeman, B.A., M.K. Topolosky and J.D. Crapo. Hyperoxia increases oxygen radical production in rat lung homogenates. Arch. Biochem. Biophys. 216:477-484, 1982. Fridovich, I. Quantitative aspects of the production of superoxide anion radical by milk xanthine oxidase. J. Biol. Chem. 245:4053-4057, 1970. Fridovich, I. Superoxide dismutases. In: Advances lp_Enzymology, edited by A. Meister. New York: Wiley and Sons, 1974, vol. 41, p. 35-97. Fridovich, I. Oxygen is toxic! Bioscience. 27:462-466, 1977. Fridovich, I. The biology of oxygen radicals. Science. 201:875-880, 1978. Fridovich, I. Superoxide radical: an endogenous toxicant. Ann. Rev. Pharmacol. Toxicol. 23:239-257, 1983. Fridovich, I. Oxygen: aspects of its toxicity and elements of defense. Current Eye Res. 3:1-2, 1984. Fridovich, I. Biological effects of the superoxide radical. Arch. Biochem. Biophys. 247:1-11, 1986a. Fridovich, I. Superoxide dismutases. In: Advances 12_Enzymology, edited by A. Meister. New York: Wiley and Sons, 1986b, vol. 58, p. 61-97. Fridovich, I. and B. Freeman. Antioxidant defenses in the lung. Ann. Rev. Physiol. 48:693-702, 1986. Gerschman, R., P.W. Nadig, A.C. Snell, Jr. and S.W. Nye. Effect of high oxygen concentrations on eyes of newborn mice. Am. J. Physiol. 179:115-118, 1954. Goldstein, A. Biostatistics: ép'lntroductorerext. New York: Macmillan, 1965. Green, M.J. and H.A.O. Hill. Chemistry of dioxygen. In: Methods 12 Enzymology, edited by L. Packer. New York: Academic Press, 1984, vol. 105, p. 3.22. Halliwell, B. Superoxide and hydroxylation reactions. In: Superoxide and Superoxide Dismutases, edited by A.M. Michelson, J.M. McCord and I. Fridovich. New York: Academic Press, 1977, p. 335-349. 119 Halliwell, B. Oxygen free radicals in living systems: dangerous but useful? In: Strategies p£_Microbial Life lp_Extreme Environments, edited by M. Shilo. Berlin: Dahlem Konferenzen, 1979, p. 195-221. Halliwell, B. and J.M.C. Gutteridge. Oxygen toxicity, oxygen radicals, transition metals and disease. Biochem. J. 219:1-14, 1984. Hassan, M.M. and I. Fridovich. Regulation of the synthesis of superoxide dismutase in Escherichia coli. J. Biol. Chem. 252:7667-7672, 1977. Haugaard, N. Oxygen poisoning. XI. The relation between inactivation of enzymes by oxygen and essential sulfhydryl groups. J. Biol. Chem. 164:265-270, 1946. Haugaard, N. Cellular mechanisms of oxygen toxicity. Physiol. Rev. 48:311-373, 1968. Hill, H.A.O. The chemistry of dioxygen and its reduction products. In: Oxygen Free Radicals and Tissue Damage. Ciba Foundation Symposium pp: New York: Excerpts Medica, 1979, p. 5-17. Hiramitsu, T., Y. Hasegawa, K. Hirata, I. Nishigaki and K. Yagi. Forma- tion of lipoperoxide in the retina of rabbit exposed to high con- centrations of oxygen. Experientia 32:622-623, 1976. Horn, R.S. and N. Haugaard. Inhibition of carbohydrate metabolism by oxygen and N-ethylmaleimide in rat heart homogenates. J. Biol. Chem. 211:3078-3082, 1966. Janoff, A., C. George-Nascimento and S. Rosenberg. A genetically engineered, mutant human alpha-l-proteinase inhibitor is more resistant than the normal inhibitor to oxidative inactivation by chemicals, enzymes, cells, and cigarette smoke. Am. Rev. Respir. Dis. 133:353-356, 1986. Jamieson, D., B. Chance, E. Cadenas and A. Boveris. The relation of free radical production to hyperoxia. Ann. Rev. Physiol. 48:703-719, 1986. Jenkinson, S.G., R.A. Lawrence, R.F. Burk and P.E. Gregory. Non-selenium dependent glutathione peroxidase activity in rat lung: association with lung glutathione S-transferase activity and the effects of hyperoxia. Tox. Appl. Pharm. 68:399-404, 1983. Joenje, H., J.J.P. Gille, A.B. Oostra and P. Van der Valk. Some charac- teristics of hyperoxia-adapted HeLa cells. A tissue culture model for cellular oxygen tolerance. Lab. Invest. 52:420-428, 1985. Junod, A.F., A. Clement, L. Jornot and H. Petersen. Differential effects of hyperoxia and hydrogen peroxide on thymidine kinase and adenosine kinase activities of cultured endothelial cells. Biochim. Biophys. Acta 847:20-24, 1985. 120 Junod, A.F., H. Petersen and L. Jornot. Thymidine kinase, thymidylate synthase, and endothelial cell growth under hyperoxia. J. Appl. Physiol. 62:10-14, 1987. Kimball, R.E., K. Reddy, T.H. Pierce, L.W. Schwartz, M.C. Mustafa and G.E. Cross. Oxygen toxicity: augmentation of antioxidant defense mechanisms in rat lung. Am. J. Physiol. 230:1425-1431, 1976. Lawrence, R.A., and R.F. Burk. Glutathione peroxidase activity in selenium-deficient rat liver. Biochem. Biophys. Res. Comm. 71:952- 958, 1976. Luck, H. Catalase. In: Methods 21 Enzymatic Analysis, edited by H.U. Bergmeyer. Weinheim7Bergstrasse, Germany:Verlag Chemie, 1965, p. 885-894. Mannervik, B., C. Guthenberg, H. Jensson, M. Warholm and P. Alin. Isoenzymes of glutathione S-transferases in rat and human tissues. In: Functions of Glutathione: Biochemical, Physiological, Toxicological,_3nd Clinical Aspects, edited by A. Larsson et al. New York: Raven Press, 1983, p. 75-88. Matkovics, B., R. Novak, H.D. Hanh, L. Szabo, Sz.I. Varga and G. Zalesna. A comparative study of some more important experimental animal peroxide metabolism enzymes. Comp. Biochem. Physiol. 568:31-34, 1977. Mattenheimer, H. Clinical Enzymology. Ann Arbor: Ann Arbor Science Publishers Inc., 1971, p. 142-143. Mazeaud, F., J. Maral and A.M. Michelson. Distribution of superoxide dismutase and glutathione peroxidase in the carp: erythrocyte manganese SOD. Biochem. Biophys. Res. Comm. 86:1161-1168, 1979. McCord, J.M. Oxygen-derived free radicals in postischemic tissue injury. N. Engl. J. Med. 312:159-163, 1985. McCord, J.M. and I. Fridovich. Superoxide dismutase. An enzymic function for erythrocuprein (hemocuprein). J. Biol. Chem. 244:6049-6055, 1969. Morris, S.M. and J.T. Albright. Superoxide dismutase, catalase, and glu- tathione peroxidase in the swim bladder of the physoclistous fish, Opsanus tau. Cell Tissue Res. 220:739-752, 1981. Morris, S.M. and J.T. Albright. Catalase, glutathione peroxidase, and superoxide dismutase in the rete mirable and gas gland epithelium of six species of marine fishes. J. Exp. 2001. 232:29-39, 1984. Nangano, T. and I. Fridovich. Superoxide radical from xanthine oxidase acting upon lumazine. J. Free Rad. Biol. Med. 1:39-42, 1985. Neter, J. and W. Wasserman. Applied Linear Statistical Models. Homewood, IL: Richard D. Irwin, Inc., 1974. 121 Nishiki, K., D. Jamieson, N. Oshino and B. Chance. Oxygen toxicity in the perfused rat liver and lung under hyperbaric conditions. Biochem. J. 160:343-355, 1976. Nishikimi, M. Oxidation of ascorbic acid with superoxide anion generated by the xanthine-xanthine oxidase system. Biochem. Biophys. Res Comm. 63:463-468, 1975. Noell, W.K. Visual cell effect of high oxygen pressures (Abstract). Fed. Proc. 14:107-108, 19553. Noell, W.K. Metabolic injuries of the visual cell. Am. J. Ophthal. 40:60-70, 1955b. Noell, W.K. Effects of high and low oxygen tension on the visual system. In: Environmental Effects pp_Consciousness, edited by K. E. Schaeffer. New York: Macmillan, 1962, p. 3-18. Nohl, H. and D. Hegner. Do mitochondria produce oxygen radicals in vivo? Eur. J. Biochem. 82:563-567, 1978. Oyama, V.I. and H. Eagle. Measurement of cell growth in tissue culture with a phenol reagent (Folin-Ciocalteau). Proc. Soc. Expt'l. Biol. Med. 91:305’307 , 19560 Packer, J.E., T.F. Slater and R.L. Willson. Direct observation of a free radical interaction between vitamin E and vitamin C. Nature, 278:737-738, 1979. Paglia, D.E. and W.N. Valentine. Studies on the quantitative and quali- tative characterization of erythrocyte glutathione peroxidase. J. Patz, A. Retrolental fibroplasia. Am. J. Ophthal. 36:1511-1522, 1954. Patz, A. Current concepts of the effect of oxygen on the developing retina. Curr. Eye Res. 3:159-163, 1984. Patz, A., L.E. Hoeck and E. De La Cruz. Studies on the effect of high oxygen administration in retrolental fibroplasia: nursery observa- tions. Am. J. Ophthal. 35:1248-1253, 1952. Rafael, J. Cytochrome c oxidase. In: Methods p£_Enzymatic Analysis, 3rd edition, edited by H.U. Bergmeyer. New York: Academic Press, 1983, p. 266-273. Rosen, G.M., E. Finkelstein and E.J. Rauckman. A method for the detec- tion of superoxide in biological systems. Arch. Biochem. Biophys. 215:367-378, 1982. Roy, R.S. and J.M. McCord. Superoxide and ischemia: conversion of xan- thine dehydrogenase to xanthine oxidase. In: Oxy Radicals and their Scavenger Systems, edited by R.A. Greenwald and G. Cohen. New York: Elsevier, 1983, vol. II, p. 145-153. 122 Sawyer, D.T. and J.S. Valentine. How super is superoxide? Acc. Chem. Res. 14:393-400, 1981. Sedlak, J. and R.H. Lindsay. Estimation of total, protein-bound, and nonprotein sulfhydryl groups in tissue with Ellman's reagent. Anal. Biochem. 25:192-205, 1968. Smith, J. and A. Shrift. Phylogenetic distribution of glutathione perox- idase. Comp. Biochem. Physiol. 633:39-44, 1979. Stadie, W.C., B.C. Riggs and N. Haugaard. Oxygen poisoning. IV. The effect of high oxygen pressures upon the metabolism of liver, kidney, lung and muscle tissue. J. Biol. Chem., 160:209-216, 1945. Steel, R.G.D. and J.M. Torrie. Principles and Procedures 21 Statistics. New York: McGraw-Hill, 1960. Stevens, J.B. and A.P. Autor. Induction of superoxide dismutase by oxygen in neonatal rat lung. J. Biol. Chem. 252:3509-3514, 1977. Tappel, A.L. Vitamin E and free radical peroxidation of lipids. Ann. N.Y. Acad. 8C1. 203:12-28, 1972. Tappel, M.E., J. Chaudiere and A.L. Tappel. Glutathione peroxidase activities of animal tissues. Comp. Biochem. Physiol. 373:945-949, 1982. Thomas, C.E., L.A. Morehouse and 8.0. Aust. Ferritin and superoxide- dependent lipid peroxidation. J. Biol. Chem. 260:3275-3280, 1985. Turrens, J.F. and A. Boveris. Generation of superoxide anion by the NADH dehydrogenase of bovine heart mitochondria. Biochem. J. 191:421-427, 1980. Turrens, J.F., B.A. Freeman and J.D. Crapo. Hyperoxia increases H 02 release by lung mitochondria and microsomes. Arch. Biochem. Biophys. 217:411-421, 1982. Turrens, J.F., B.A. Freeman, J.G. Levitt and J.D. Crapo. The effect of hyperoxia on superoxide production by lung submitochondrial particles. Arch. Biochem. Biophys. 217:401-410, 1982. Tyler, D.D. A protective function of superoxide dismutase during respir- atory chain activity. Biochim. Biophys. Acta 396:335-346, 1975. Ubels, J.L. and J.R. Hoffert. Ocular+oxygen toxicity: the effect of hyperbaric oxygen on retinal Na -K ATPase. Exp. Eye Res. 32:77-84, 1981. Ubels, J.L., J.R. Hoffert and P.O. Fromm. Ocular oxygen toxicity: the effect of hyperbaric oxygen on the lp_vitro electroretinogram. Comp. Biochem. Physiol. 57A:29-32, 1977. 123 Umbreit, W.H., R.H. Burris and J.F. Stauffer. Manometric techniques. Minneapolis:Burgess, 1964, p. 131-133. Weiss, S.J. Oxygen, ischemia and inflammation. Acta Physiol. Scand., Suppl. 548:9-37, 1986. Weiss, S.J. and A.F. LoBuglio. Phagocyte-generated oxygen metabolites and cellular injury. Lab. Invest., 47:5-18, 1982. Wdzieczak, J., G. Zalesna, E. Wujec and G. Peres. Comparative studies on superoxide dismutase, catalase and peroxidase levels in erythrocytes and livers of different freshwater and marine fish species. Comp. Biochem. Physiol., 73B:361-365, 1982. Willson, R.L. Hydroxyl radicals and biological damage in vitro: what revelance in vivo? In: Oxygen Free Radicals and Tissue Damage. Ciba Foundation Symposium 65. New York: Excerpta Medica, 1979, p. 19-42. Winkler, B.S. and F.J. Giblin. Glutathione oxidation in retina: effects on biochemical and electrical activities. Exp. Eye Res. 36:287-297, 1983. Wittenberg, J.B. and B.A. Wittenberg. Active secretion of oxygen into the eye of fish. Nature. 194:106-107, 1962. Wolf, K. Physiological salines for freshwater teleosts. Prog. Fish Cult. 25:135-140, 1963. Yagi, K. and N. Ohishi. Mechanisms of degeneration of the retina of animals exposed to a high concentration of oxygen. In: Biochemical and Medical Aspects p£_Active Oxygen, edited by 0. Hayaishi and K. Asada. Baltimore: University Park Press, 1977, p. 299-307. Yanoff, M., W.W. Miller and J.A. Waldhausen. Oxygen poisoning of the eyes. Comparison in cyanotic and acyanotic dogs. Arch. Ophthal. 84:627-629, 1970. ER "11111111111111filtflllllllllwiflljlll“