LIPID PEROXIDATION: ENZYME - CATALYZED PEROXIDATION 0F MEMBRANE LIPIDS AND THE ROLE OF CHELATED IRON Dissertation for the Degree of Ph. D, MICHIGAN STATE UNIVERSITY JOHN A. BUEGE 1976 b— 3:? 2 5 Mt}; ABSTRACT LIPID PEROXIDATION: ENZYME-CATALYZED PEROXIDATION 0F MEMBRANE LIPIDS AND THE ROLE OF CHELATED IRON By John A. Buege NADPH-cytochrome c reductase from rat liver was solubilized from the microsomal membrane with either bromelain or a crude pancreatic lipase. Lipase-solubilization of the enzyme was shown to be catalyzed by a protease contaminant of the lipase preparation, since protease inhibitors prevented solubilization. When the crude lipase preparation was subjected to gel filtration, the solubilizing activity was found only in the fractions containing protease activity. The bromelain solubilized enzyme was purified to homogeneity. An antibody directed against the purified NADPH-cytochrome c reductase inhibited both the enzyme's cytochrome c reducing activity and NADPH—dependent liver micro- somal lipid peroxidation. NADPH-dependent lipid peroxidation in lung microsomes was shown to occur much slower than NADPH-dependent peroxidation in liver micro- somes. Since NADPH-cytochrome c reductase catalyzes NADPH-dependent lipid peroxidation in liver microsomes, the corresponding lung enzyme was purified and compared to the liver enzyme. Both enzymes had John A. Buege identical ionic strength optima and were equally inhibited in the reduc- tion of cytochrome c and ferricyanide by antibody directed against the liver enzyme. Double diffusion immunoprecipitation on Ouchterlony plates of the detergent-solubilized liver and lung microsomes resulted in converging precipitin lines indicating similar antigenic sites. The molecular weights of the detergent-solubilized and bromelain-solubilized enzymes from both liver and lung microsomes were 79,000 and 7l,000 daltons, respectively. Both enzymes were equally effective in catalyz- ing the peroxidation of liposomal membranes in the presence of NADPH and chelated Fe+++. Peroxidation in microsomes catalyzed by NADPH-cytochrome c reduc- tase requires the presence of ADP chelated Fe+++. However, ADP-Fe++ is much more effective than ADP-Fe+++ in catalyzing lipid peroxidation in the absence of enzyme, thus suggesting that the enzyme functions during lipid peroxidation to reduce App-Fe+++ During the course of reductase catalyzed lipid peroxidation, lipid hydroperoxide levels build up early in the reaction, followed by a decrease in hydrOperoxide levels, indicating that net breakdown of hydroperoxides is occurring. Lipid peroxidation in liposomes catalyzed by NADPH-cytochrome c reductase and ADP-Fe+++ is greatly enhanced by the addition of Fe+++ chelated by EDTA. However, EDTA-Fe++, unlike ADP-Fe++, has no ability to initiate lipid peroxidation in the absence of enzyme. In the presence of NADPH-cytochrome c reductase, EDTA-Fe+++ catalyzed the rapid break- down of cumene hydroperoxides, probably through a lipid radical gener- ating process. It is proposed that the enzyme-catalyzed reduction of John A. Buege ADP-Fe+++ causes the initiation of NADPH-dependent lipid peroxidation and that enzymatically reduced EDTA-Fe+++ greatly enhances lipid perox- idation via its interaction with lipid hydroperoxides to produce new lipid radicals. Lactoperoxidase, in the presence of H202 and I', also catalyzes the peroxidation of microsomal and liposomal membranes. Fe+++ is not required for the initiation of lipid peroxidation in this system, but does function to assist in the breakdown of lipid hydroperoxides when the membranes are exposed to heat. The free radical trapper BHT abo- lishes the appearance of hydroperoxide breakdown products in liposomes and decreases their appearance in microsomes. Lactoperoxidase-catalyzed lipid peroxidation is not initiated by the formation of singlet oxygen, superoxide, hydroxyl radicals or tri- iodide, since scavengers of these species do not inhibit enzyme- catalyzed lipid peroxidation. An oxidized product of I' may be respon- sible for lactoperoxidase-catalyzed lipid peroxidation. The antibacterial activity of the lactoperoxidase, H202, 1' system is not mediated by the peroxidation of bacterial membrane lipids since antioxidants do not prevent lactoperoxidase-catalyzed bacterial killing. In addition, other lipid peroxide forming enzyme systems do not cause bacterial killing. Soybean lipoxygenase catalyzes lipid hydroperoxide formation in y-linolenic acid micelles and in detergent—solubilized microsomes and liposomes. In liposomes, the breakdown of lipoxygenase formed lipid hydroperoxides occurs primarily when the membranes are heated in the presence of Fe+++. BHT blocks the appearance of lipid hydroperoxide John A. Buege breakdown products. In microsomes, a factor is present which facilitates the breakdown of lipid hydroperoxides without the requirement for heat and Fe+++. Lipid hydroperoxides formed by lipoxygenase in microsomes inter- act with cytochrome P-450 to catalyze the metabolism of the drug, amino- pyrine. Recent evidence suggests that cytochrome P-450 is the microsomal factor responsible for the breakdown of lipid hydroperoxides during enzyme-catalyzed lipid peroxidation. The reduction of EDTA-Fe+++ by NADPH-cytochrome c reductase in liposomes may act to mimic the hydroper— oxide breakdown function proposed for cytochrome P-450. LIPID PEROXIDATION: ENZYME-CATALYZED PEROXIDATION 0F MEMBRANE LIPIDS AND THE ROLE OF CHELATED IRON By John A. Buege A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Biochemistry 1976 To Barbara And My Parents ii ACKNOWLEDGMENTS I would like to express my sincere appreciation to Dr. Steven 0. Aust, my graduate research advisor, for his guidance, encouragement, and understanding, both within and outside of my scientific career. I would also like to express my thanks to the other members of my guidance committee, Drs. David D.McConnell, Michael N. Rathke, John C. Speck, Jr. and Willis A. Wood for their assistance throughout my graduate work at Michigan State University. My special thanks also goes to Dr. John E. Wilson for his help and encouragement at the beginning of my graduate studies. I would also like to acknowledge the collaborative assistance from my research colleagues, including Dr. Thomas Pederson, Dr. Ann Welton, Mr. Robert Moore, Mr. Fred O'Neal, Mr. Bruce Svingen, and Mrs. Poonsin Olson. Finally, I wish to express my thanks for the help and friendship I have received from the members of the Department of Biochemistry. TABLE OF CONTENTS Page INTRODUCTION .......................... l LITERATURE REVIEW ....................... 4 MATERIALS AND METHODS ..................... 2l Material Sources .................... . . 2l Preparation of Rat Liver and Lung Microsomes ........ 22 Lipase Solubilization of NADPH-Cytochrome c Reductase . . . 22 Assay of Protease Activity ................. 23 Assay of Lipase Activity ................ . . 24 NADPH-Dependent Peroxidation of Membrane Lipids ...... 24 Measurement of Lipid Hydroperoxides by the Iodometric Assay 25 Isolation and Measurement of Liver and Lung Microsomal NADPH-Cytochrome c Reductase ............. . . 26 Preparation of Antibody to Liver NADPH-Cytochrome c Reductase ..................... . . . 29 Ouchterlony Double Diffusion Analysis of the Antibody- Antigen Reaction ..................... 30 Immunoprecipitation of Liver and Lung Microsomal NADPH- Cytochrome c Reductase by Antibody ............ 30 Sodium Dodecylsulfate-Polyacrylamide Gel Electrophoresis . . 3l Assay of LactOperoxidase Activity . . . . . . ..... . . 3l Lactoperoxidase-Catalyzed Peroxidation of Membrane Lipids . 32 Purification of Superoxide Dismutase . . . . ...... . . 32 Preparation of Bacteria . . . ............. . . 33 Determination of Bacterial Growth . . . . . . . . . . . . . 34 iv Lipid Peroxidation Catalyzed by Lipoxygenase ........ Microsomal Drug Metabolism Assay .............. Other Methods ....................... CHAPTER I. CHARACTERIZATION OF NADPH-DEPENDENT MICROSOMAL LIPID PEROXIDATION ........................ The Solubilization of NADPH—Cytochrome c Reductase ..... Preparation of an Antibody to NADPH-Cytochrome c Reductase . Lung Microsomal Lipid Peroxidation ............. Isolation and Characterization of Lung Microsomal NADPH- Cytochrome c Reductase .................. The Mechanisms Involved in NADPH-Dependent Lipid Peroxidation ..................... CHAPTER II. CHARACTERIZATION OF LACTOPEROXIDASE-CATALYZED LIPID PEROXIDATION ........................ Lactoperoxidase-Catalyzed Lipid Peroxidation During Iodina- tion of Membrane Protein ................. Optimum Condition for Lactoperoxidase-Catalyzed Peroxidation of Microsomal Membranes ................. Lactoperoxidase-Catalyzed Lipid Peroxidation in the Model System .......................... The Effect of Fe+++ on Lactoperoxidase-Catalyzed Lipid Peroxidation ....................... The Effect of BHT on Lactoperoxidase—Catalyzed Lipid Peroxidation ....................... Reaction Mechanisms Capable of Promoting Lactoperoxidase- Catalyzed Lipid Peroxidation ............... The Antimicrobial Activity of the Lactoperoxidase Lipid Peroxidation System ................... CHAPTER III. CHARACTERIZATION OF LIPID PEROXIDATION CATALYZED BY LIPOXYGENASE .................... Lipid Peroxidation Catalyzed by Lipoxygenase ...... Lipid Hydroperoxide Mediated Drug Metabolism ........ Page 34 35 36 37 43 43 48 63 87 87 89 TOT 107 114 118 127 I33 I33 I4I Page DISCUSSION ........................... l47 The Prooxidant Activity of Cytochrome P-450 ........ 147 The Peroxidase Activity of Cytochrome P-450 ........ l50 A Model for the Enzyme-Catalyzed Peroxidation of Membrane Lipids .......................... l5l SUMMARY ............................ l56 LIST OF REFERENCES ....................... l59 APPENDIX ............................ I75 vi TABLE I I0 II I2 I3 I4 LIST OF TABLES The Effect of Soybean Trypsin inhibitor (STI) and Phenyl- methylsulfonylfluoride (PMSF) on the Solubilization of NADPH-Cytochrome c Reductase by Crude Pancreatic Lipase . . NADPH-Dependent Lipid Peroxidation in Liver and Lung Microsomes ........................ Effect of Microsomal Protein Concentration of Lipid Peroxidation ....................... Lipid Peroxidation in Liposomes Catalyzed by Purified Liver and Lung NADPH-Cytochrome c Reductase ........... NADPH-Dependent Microsomal Lipid Peroxidation as a Function of Added ADP-Fe+++ .................... The Effect of ADP-Fe+++ on Ascorbate-Catalyzed Lipid Peroxidation in Liposomes ................. Microsomal Lipid Peroxidation Promoted by ADP-Fe++ or ADP-Fe ......................... The Effect of EDTA-fe+++ on NADPH-Dependent Lipid Peroxi- dation in Liposomes .................... The Effect of ADP-Fe+++ and EDTA-fe+++ on Ascorbate- Catalyzed Lipid Peroxidation in Liposomes ......... Effect of Various Fe+++ Chelators on NADPH-Dependent Lipid Peroxidation in Liposomes ................. Effect of Nashing Microsomes on the Formation of Malondial- dehyde and Hydroperoxides in NADPH-Dependent Lipid Peroxidation ....................... The Effect of H 02 Generation on Lactoperoxidase-Catalyzed Lipid Peroxidation in Microsomes ............. The Effect of ADP-re+++ and H202 on Liposomes ....... Effect of Catalase on Non-Enzymatic Lipid Peroxidation in Microsomes and Liposomes . . . .............. Page 40 47 62 69 7I 72 77 78 80 93 IIO I14 TABLE 15 16 I7 18 19 20 21 22 23 24 25 26 27 The Effect of Adding ADP-Fe+++ During or After Incubation of Membranes with the Lactoperoxidase Lipid Peroxidation System ......................... The Effect of BHT on Membrane Lipid Peroxidation Catalyzed by Lactoperoxidase When Added After the Reaction Period . . The Effect of Hydroxyl Radical Scavengers on Lactoperoxidase-Catalyzed Lipid Peroxidation ..... The Effect of Singlet Oxygen Scavengers on Lactoperoxidase- Catalyzed Lipid Peroxidation ............... Inhibition of Bacterial Growth by Lactoperoxidase, KI, and H 0 ........................... 2 2 The Effect of Peroxide Forming Systems on Bacterial Growth Lipoxygenase-Catalyzed Lipid Peroxidation of Linolenic Acid The Effect of Detergent on Lipoxygenase-Catalyzed Lipid Peroxidation in Microsomes and Liposomes ......... Conditions for Lipoxygenase-Catalyzed Lipid Peroxidation in Microsomes ........................ The Effect of Fe+++ and BHT on the Breakdown of Lipid Hydroperoxides formed by Lipoxygenase in Microsomes . . . . The Effect of Fe+++ and BHT on the Breakdown of Hydroper- oxides Formed by Lipoxygenase in Liposomes ...... The Effect of Detergent-Solubilization of Microsomes on Both NADPH and Hydroperoxide-Dependent Drug Metabolism Microsomal Drug Metabolism Catalyzed by Membrane Lipid Hydroperoxides ...................... viii Page 115 117 123 125 130 131 136 I37 138 140 142 M4 146 FIGURE 1 10 11 12 LIST OF FIGURES Standard Curve for Measuring Lipid Hydroperoxides by the Iodometric Assay ..................... Gel Filtration of Crude Pancreatic Lipase on a Sephedex G-lOO Column ....................... Effect of Ionic Strength on the Activity of Purified NADPH-Cytochrome c Reductase from Rat liver and Lung Microsomes ........................ Inhibition of NADPH-Cytochrome c Reductase by Antibody to the Bromelain-Solubilized Liver Reductase ........ Inhibition by Antibody of the Activity of NADPH-Cytochrome c Reductase with Ferricyanide as the Terminal Electron Acceptor ......................... ImmunOprecipitation of the Antibody to Purified, Bromelain-Solubilized NADPH-Cytochrome c Reductase from Rat Liver Microsomes with the Purified Reductase and with Detergent-Solubilized Microsomes from both Rat Liver and Lung on an Ouchterlony Double Diffusion Agar Plate . . . . Sodium Dodecylsulfate-Polyacrylamide Gel Electrophoresis Protein Profiles of Immunoprecipitates Formed from Sodium Deoxycholate-Solubilized Rat Liver and Lung Microsomes and the Reductase Antibody .................. Time Course of Hydroperoxide and Malondialdehyde Formation During NADPH-Dependent Microsomal Lipid Peroxidation . . . Time Course of Hydroperoxide and Malondialdehyde Formation During NADPH-Dependent Liposomal Lipid Peroxidation The NADPH-Dependent Decomposition of Cumene Hydroperoxide in the Presence of Chelated Fe+++ ........... Lactoperoxidase-Catalyzed Lipid Peroxidation in Microsomes Lactoperoxidase-Catalyzed Lipid Peroxidation of Microsomal Lipid as a Function of Enzyme Concentration ..... ix Page 28 42 51 53 55 58 6O 67 75 83 91 FIGURE 13 14 15 16 17 18 19 Lactoperoxidase-Catalyzed Lipid Peroxidation of Microsomal Lipids as a Function of K1 Concentration ......... Lactoperoxidase-Catalyzed Lipid Peroxidation in Microsomes as a Function of Added H202 ............... Lactoperoxidase-Catalyzed Lipid Peroxidation in Liposomes Lactoperoxidase-Catalyzed Lipid Peroxidation of Liposomes as a Function of pH ................... Lactoperoxidase-Catalyzed Lipid Peroxidation in Liposomes as a Function of H202 Concentration ........... Sodium Dodecylsulfate Polyacrylamide Gel Electrophoresis of Commercial Lipoxygenase ................ Reaction Mechanisms Involved in Microsomal and Liposomal Lipid Peroxidation Catalyzed by the NADPH-Cytochrome c Reductase, Lipoxygenase, and Lactoperoxidase Systems . . . Page 97 99 104 106 109 121 154 ADP BHA BHT CHP DBB DEAE- DOC DPIF EDTA IgG NADH NADPH NDHG PMSF SDS STI TBA TCA TMPD Tris ABBREVIATIONS adenosine-S'-diphosphate 2,6-di-tert-butylanisole butylated hydroxytoluene cumene hydroperoxide o-dibenzoylbenzene diethylaminoethyl- sodium deoxycholate l,3-diphenylisobenzofuran ethylenediaminetetracetate immunoglobulin-G reduced nicotinamide adenine dinucleotide reduced nicotinamide adenine dinucleotide phosphate nordihydroguarietic acid phenylmethylsulfonylfluoride sodium dodecylsulfate soybean trypsin inhibitor thiobarbituric acid trichloroacetic acid N,N,N',N'-tetramethyl-p-phenylenediamine Tris(hydroxymethyl) aminomethane xi INTRODUCTION Lipid peroxidation is an example of a biological, free radical process that is deleterious to living organisms. Damage to biological components, particularly unsaturated fatty acids, through a peroxida- tive attack by molecular oxygen has been implicated in a wide variety of pathological states and may be an integral part of the overall degenerative processes involved in aging. Both enzymatic and nonenzy- matic initiators of lipid peroxidation have been discovered. My interest in the processes involved in lipid peroxidation began in an effort to continue the work begun by Dr. Thomas Pederson on the characterization of NADPH-dependent lipid peroxidation in micro- somes. It had been suggested that NADPH-dependent peroxidation of microsomal membrane lipids was enzymatically catalyzed by the flavo- protein, NADPH-cytochrome c reductase. Chapter I deals with the study of the reaction mechanism involved in microsomal lipid peroxidation catalzyed by NADPH-cytochrome c reductase. This study was greatly faciliated by the development of an assay that would permit the detection of membrane lipid hydroper- oxides. Prior to the development of this assay, the standard metho- dology for monitoring membrane lipid peroxidation permitted only the detection of the lipid peroxidation breakdown product, malondialdehyde. It could be demonstrated that malondialdehyde production was not always a valid indicatorcrfthe peroxidative events occurring during membrane lipid peroxidation. Extensive use was also made of a model peroxidation system composed of artificial membranes containing lipids extracted from microsomal membranes. The use of the model peroxida- tion system permitted the precise examination of the factors involved in enzyme-catalyzed lipid peroxidation without interference from components present in the microsomal membrane. A major problem in characterizing the mechanism involved in the initiation of enzyme-catalyzed lipid peroxidation was the catalytic, radical propagating interference demonstrated by free and chelated forms of metal ions. The NADPH-dependent peroxidation system appeared to require chelated forms of ferric ion for the initiation of membrane lipid peroxidation. The ferric ions also contributed to the free radical, chain prepagation processes involved in lipid peroxidation, thus making the study of the initiation phase of the peroxidation reaction difficult. Chapters 11 and III deal with the characteriza- tion and the subsequent use of two new lipid peroxidation catalyzing enzyme systems to study the initiation phase of enzyme-catalyzed lipid peroxidation. Neither enzyme system appeared to require chelated metals for the initiation of membrane lipid peroxidation. It was possible, through the use of free radical trappers and metal chelators, to separate the events occurring in the initiation phase of lipid peroxidation from events occurring in later phases of lipid peroxida- tion. Recent evidence is examined concerning the possible involvement of a major hemoprotein constituent of microsomes, namely cytochrome P-450, in the overall control and propagation of microsomal lipid peroxidation. A comparative study of the enzyme-catalyzed peroxidation of microsomal membranes versus liposomal membranes permits the assign- ment of a new, participating role for cytochrome P-450 in microsomal membrane lipid peroxidation. LITERATURE REVIEW A characteristic property of unstaturated fatty acids is that in the presence of free radical initiators and molecular oxygen, they undergo oxidative deterioration (l-4). Membranes are particularly susceptible to peroxidative attack since they contain a large amount of unsaturated lipid and are bathed in an oxygen-rich, metal contain- ing fluid (5, 6). Oxidation of unsaturated lipid involved the forma- tion of semistable peroxides from the reaction between lipid radicals and molecular oxygen (7). The simplified scheme below illustrates the reactions leading to the formation of lipid peroxides, starting with the abstraction of a hydrogen from the lipid chain (8, 9). -H' 02 LH LH ________9, L' ———————-° LOO' -———-—-—9 LOOH + L' Unsaturated fatty acids are usually considered the site of initiation of lipid radical formation since the divinyl methane group of the fatty acid chain is particularly susceptible to hydrogen abstraction (9-11). / H\c c — H -C u c H H H The lipid radical resulting from hydrogen abstraction readily reacts with oxygen to yield a peroxy radical. Uri (7) points out that this reaction proceeds so quickly that termination of the peroxy radical brought about by a bimolecular collision is unlikely, if oxygen is present in sufficient amount. The peroxy radical formed can abstract a hydrogen atom from a neighboring lipid to produce a lipid hydroper- oxide plus another lipid radical, thus propagating the radical chain reaction (7). Lipid hydroperoxides can also form radicals either by homolytic scission (l2-l4); LOOH ———-—————+ LO' + °OH or by bimolecular recombination (15-17). 2LO0H —-—-—-———+ L00° + L0° + HOH The latter two reactions are less likely to occur in biological tissues because of competition for the hydroperoxide by other oxidizable species (18). One group of oxidizable species commonly associated with lipid membranes is free or chelated metals. Metals undergo univalent redox coupling with hydroperoxides to produce radicals (12, 19-22). Once initiated, lipid peroxidation in biological membranes could become autocatalytic and proceed until the lipid substrate was exhausted, except for the presence of tissue components which prema- turely terminate the chain reaction (23-24). Water and lipid soluble radical scavengers, proteins, and nucleic acids all react with lipid radicals to interrupt the radical chain reaction. In addition, lipid radicals interact to produce non-radical recombination products (9, 25-28). 2L° > L - L L- + LOO“ t LOOL 2L00' > LOOL + O 2 Therefore, the extent of lipid peroxidation depends on the factors governing the initiation, prOpagation, and termination of the radical reactions involved. Initiation of lipid peroxidation may occur through an autoxida- tive attack on unsaturated fatty acids by molecular oxygen, or by for- mation of free radical initiators by enzymatic or non-enzymatic reactions (28-35). Compounds with allylic or benzylic hydrogens are readily oxidized by molecular oxygen to form hydroperoxides in fairly good yields (l8). 00H ,__ .____I /"V\/ 2 >/\/v This is not surprising since molecular oxygen has the electronic dis- tribution of a diradical (36). In addition to direct interaction of oxygen with unsaturated lipids, non-enzymatic mechanisms exist that result in initiation of lipid peroxidation via radical formation. Ionizing radation produces cation radicals and electrons as primary products, which decay to pro- duce a variety of charged or neutral radicals (37-40). Irradiation can initiate radical reaction in vivo, which lead to peroxidation of unsaturated fatty acids (9, 4l, 42). The primary initiation reaction in vivo appears to be one in which electrons, hydrogen atoms and hydroxy radicals produced by radiolysis of water react with cell constituents (43). Photolysis of chemical bonds can also produce radicals (44). In animals, individuals with light pigmentation may be particularly sensi- tive to sunlight. The accelerated aging of human skin in areas exposed to sunlight is a result of direct photolysis and photoinitiated autox- idation of collagen (45). Alcohols and ethers are photooxidized fairly readily by ketonic sensitizers. These reactions are probably initiated through hydrogen abstraction by sensitized ketones, resulting in free radicals (48). The resulting radicals can then react with olefins containing allylic hydrogens, as in unsaturated lipids, to give lipid radicals (49). In addition to light stimulated radical production, lipid hydroperoxides can be formed by another photodynamic process. Fatty acids and cholesterol have been shown to give photo- oxidation products, including hydroperoxides, in which the double bond shifts to the allylic position during an attack on the lipid by photo- dynamically produced singlet oxygen (50, 51). ii i" ‘0 iii 90“ l —CH2—C=£—CH2—C=C— 2 {Hie-c—c-cl-c— H Lipid peroxidation catalyzed by either the photoproduction of radicals or singlet oxygen is probably the cause of damage to membranes commonly reported as a result of photooxidation (52-54). Ozone, at levels present in normal air (0.01-0.02 ppm), initiates autoxidation of polyunsaturated fatty acids in vivo (30, 55). The peroxidation of linolenic acid exposed to 0.02 ppm ozone occurs sub- stantially faster than control lipids exposed to pure air (43). The observed autoxidation was inhibited by radical scavengers, such a 2,6-di-tert-butylanisole (BHA) and a-tocopherol. Although ozone, unlike oxygen, does not display the characteristics of a diradical, it does react with virtually every type of organic molecule to produce radicals (43, 56). The reaction mechanism by which ozone reacts with aldehydes, alkanes, and amines consistently includes the proposal of a radical intermediate. The reaction of ozone with polyunsaturated fatty acids yields an esr signal indicating the presence of a radical (57). Nitrogen dioxide, like ozone, is present in normal air and is an important pollutant in smog (58, 59). It is a free radical suffi- ciently stable to exist in high concentrations but reactive enough to initiate free radical chain reactions. It can initiate autoxidation of olefins at levels as low as 0.l ppm (60). Like ozone, it appears to exert its physiological effect, at least in part, by initiating radical reactions (6l). Vitamin E protects rats against damage caused by exposure to nitrogen dioxide (62) and the esr signal observed in radical reactions is present when olefins react with nitrogen dioxide (63, 64). Chlorinated hydrocarbons, such as chloroform and carbon tetra- chloride, cause toxic reaction in vivo that are mediated by radical formation (6, 65, 66). Although these compounds are considered stable and not likely to undergo unimolecular homolysis at biological temper- atures, they do appear to react with enzymes of the liver to produce radicals. Carbon tetrachloride, when added to microsomes in the pre- sence of NADPH, results in an increase in lipid peroxidation. Reck- nagel suggests that the microsomal NADPH-dependent drug metabolism system forms CC130 radicals via the metabolism of carbon tetrachloride. These radicals subsequently attack lipid molecules and initate lipid peroxidation (67). Ethanol metabolism may also result in free radical production. DiLuzio demonstrated that concomitant oral administration of ethanol and an antioxidant mixture containing BHT, BHA, and propyl gallate, protected rats against increased liver fat (68, 69). Since ethanol intoxication results in increased peroxide content in liver extracts, it was concluded that ethanol damage to liver occurs via lipid peroxidation (70, 71). 2, Fe+2), particularly in the presence of Metal ions (e.g. Cu+ reducing agents like ascorbate and cysteine, catalyze an increase in lipid peroxidation (73-75). This non-enzymatic reaction may proceed via the production of perhydroxy radicals (H02-) or its anion (0%) via the interaction of reduced metal ions and molecular oxygen (6). Enzymatic reactions are also known to produce radicals, which can diffuse from the enzyme's surface before they are oxidized or reduced to an even electron species (76). Sequential, one electron transfers resulting in a transient radical, may be an intermediate step in many two electron transfer reactions (77). The NAD- radical can be produced by one electron oxidation of NADH as well as the one electron reduction of NAD+ (78, 79). Oxygen may react with NAD- by a simple one electron transfer to produce superoxide (78). + - NAD + 02 > NAD + 02 A one electron internal transfer has been postulated for the oxidation of reduced cytochrome c. The one electron oxidation results in the formation of a ferrocytochrome II cation radical with the odd electron located in the porphyrin ring (80). An electron transfer fol- lows, resulting in ferricytochrome c. 10 II III . [Fell Cyt ci‘f ~-~~--—>—-——- Fe C“ C Fe Cyt c '53-“- Cyclic oxidation and reduction of the flavin coenzymes of the electron transport chains in intracellular organelles may result in flavin semiquinone radicals (Bl-83). H. FH —-———-———+ FH -————H—.——-—>FH 2 3 Evidence exists that indicates that chelated metals effect the equili- brium between quinones and hydroquinones, and semiquinone radicals. Chelated metals strongly favor the existence of the semiquinone radical (84, 85). Enzymes that have been demonstrated to mediate lipid peroxidation in vitro included NADPH-cytochrome c reductase (86-89), NADH-cytochrome b5 reductase (87), xanthine oxidase (88-92), lipoxygenase (93-95), superoxide dismutase (96) and lactoperoxidase (97). Although the mechanisms of action of these peroxide forming reactions are not clearly understood, it is likely that the transfer of electrons result— ing in the formation of radicals in involved in all these enzyme systems. Once initiation of lipid peroxidation has begun, autocatalytic radical chain reactions occur. The radical propagation involved in lipid peroxidation, in which the net number of radicals in conserved, can be divided into three reaction types (30, 55, 98). atom transfers R. + R'H ———-————» RH + R'- additions R - + R2-—C -CH2-—————-——+ R2-—C—-CH2—-R R R B scission R—-C-—0-i———————-»R-C=~O + R" RI 11 Atom transfer is the most common radical reaction and virtually all radical reaction systems demonstrate this reaction (43). In autox- idation of lipids, hydrogen abstraction is an important part of the chain propagation reaction (27, 43). L00' + LH-———-————+ LOOH + L' Under conditions where oxygen is abundant, the concentration of lipid peroxy radicals in high. In the autoxidation of methyl linolenate, the rate of hydrogen abstraction by methyl linolenate peroxy radical is seven orders of mag- nitude greater than the rate for the bimolecular, self-annihilation reaction between two peroxy radicals (27). Metals also contribute to radical chain propagation by reacting with lipid hydroperoxides and lipid radicals to produce more radicals. Metals accelerate autoxidation in a complex series of reactions lead- ing to the transfer of electrons between lipid molecules. The ferrous ion catalyzed decomposition of hydrogen peroxide (Fenton's system) illustrates this point (l8, 35, 98-100): fe++ + H202 ——-———-—+ Fe+++ + OH' + OH' Fe+++ + H 0 ——————-——+ Fe++ + H0 - + H+ 2 2 2 fe++ + OH- ——-——————> Fe+++ + OH- ++ +++ - Fe + H02- —-—————+ Fe + H02 Fe+++ + H0. ————-——+ Fe++ + 07 + H+ 2 2 OH' + H202 -———————+ H02' + H20 Oé + H202 --+ 02 + OH + OH' 12 Hematin and hemoproteins also accelerate lipid peroxidation, but are not thought to be involved in initiation of peroxidation (55, 101-103). Many transition metal complexes which occur in cells in vivo catalyze the decomposition of lipid hydroperoxides (104, 105). In vitro studies of the autoxidation of cell membranes also indicate that metal contain- ing compounds act as prooxidants in lipid peroxidation (106, 107). Finally, many of the symptoms of iron toxicity are similar to those conditions produced by conditions leading to in vivo lipid peroxida- tion (108, 109). Metals may also enhance lipid autoxidation in a manner not directly dependent upon radical propagation. Experiments with autoxi- dized blood plasma lipids indicate that chemiluminescence results upon addition of transition metals to the peroxidizing system (110). The resulting chemiluminescence is also observed in microsomes and mito- chondria when NADPH plus ferric ion or ferrous ion are used to promote lipid peroxidation (111, 112). These findings have led to the belief that metals react with lipid hydroperoxides to produce electronically excited species. Singlet oxygen (88) and excited, triplet carbonyl functions (113) have been proposed as the light emitting species. Both of these excited species are capable of interacting with unsaturated lipids to produce more hydroperoxides (114). In view of the damaging effect radicals can have on biological systems, it is not surprising that mechanisms exist for the termination of radical-catalyzed chain reactions. As in all radical reactions, termination of free radicals may occur through collision of two radical species (25, 26). In membrane lipid peroxidation, collision of two lipid radicals would result in cross-links between lipid chains. This 13 could be particularly destructive, since covalent bonds between lipid chains would drastically effect the structureanutproperties of the membrane (115-118). A number of endogenous substances exist in vivo that, at phy- siological concentrations, act to terminate radical processes. Water- soluble radical scavengers include vitamin C, thio containing compounds, proteins, and purine bases (119, 120). Vitamin C is commonly used as a food preservative and acts to trap two radicals in a two electron sequential reduction to form dihydroascorbate (121, 122). Thio compounds are well known scavengers of radicals (6). Thio compounds interrupt radical processes by acting as hydrogen donors, followed by dimerization. 2RSH + 2L'-—-—————e-2RS--—-—————+ RSSR + 2LH Cysteinamine, glutathione, ergothioneine and related thios all impart radical scavenging activity (6). The sulfhydryl groups present in pro- teins may function in the same manner. It is noteworthy that gluta- thione also functions to inhibit lipid peroxidation in the presence of glutathione peroxidase. Glutathione peroxidase reduces lipid peroxides to alcohols by oxidizing reduced glutathione (23). Purine bases may also play a role in controlling radical reac- tions. Uric acid, inosine and RNA have all been shown to inhibit lipid autoxidation of linolenic acid in vitro (124, 125). The high concen- tration of inosine found in the endoplasmic reticulum suggests that inosine functions to protect this membrane from peroxidative attack (126). 14 A number of antioxidants present in plants may act as radical scavengers when ingested by animals. Gallic acid, eugenol, and nordi- hydroguarietic acid (NDHG) all act as plant-derived radical scavengers (6). NDHG is commonly used as a food preservative in lard, oil, and baking mixes (6). Animals also possess a number of lipid-soluble antioxidants. Vitamin E (a tocopherol) is perhaps the most studied lipid-soluble antioxidant (127-128). It has been shown to inhibit in vitro lipid peroxidation. Vitamin E deficiency in vivo increases lipid peroxida- tion in erythrocytes, adipose and adrenals exposed to hyperoxia (129). Ubiquinone, from the mitochondrial respiratory chain, acts to terminate free radical reactions. Like many of the quinones, it acts as an antioxidant and, when administered to rats, protects against ethanol induced fatty livers (130). Ubiquinol-6 is as effective as a tocopherol at inhibiting free radical reactions (131). Carotenes also protect against radical mediated reactions. Forbes and Taliaferro (132) demonstrated that rats maintained on diets rich in carotenes were protected against carbon tetrachloride-induced liver injury, which is mediated by lipid peroxidation. B Carotene has been used to inhibit in vitro, enzymatic lipid peroxidation in microsomes (88). Exogenous antioxidants are commonly ingested as additives to food stuffs. BHT, BHA, propyl gallate, ethoxyquin, and NDHG are commonly used as food preservatives to prevent oxidative deterioration in lipid- rich foods (6). Some foods are particularly susceptible to changes in flavor, color and odor due to peroxidation, because of the high fat and chelated metal content (133). 15 Drug antioxidants like chlorOpromazine are very effective lipid radical scavengers, since they are lipid soluble and likely to pene- trate lipid membranes (134, 135). The inhibition of lipid peroxidation by the previously mentioned antioxidants is generally accomplished through the antioxidants‘ abi- lity to form stable, non-reactive radicals or by hydrogen donation by one of the following pathways (6). L00’ + AH —-————-——+ LOOH + AH- 2 2AH- —————————+ AH 01" LOO' + AH- ————————+ A + LOOH 2+A Lipid hydroperoxides are intermediate products of lipid peroxi- dation, and their breakdown leads to a variety of secondary products, such as volatiles, polymers, and oxygenated compounds (136). Hydro- peroxide breakdown can proceed by either homolytic or heterolytic mechanisms. Metal-catalyzed homolytic cleavage of hydroperoxides is common in foods, where transition metals and metaloproteins are abun— dant (137). The ferrous ion induced decomposition of purified linoleic acid hydroperoxides results in the production of at least 20 detectible compounds, including alcohols, ketones, and ethers (12). When polyun- saturated fatty acids with three or more methylene interruped double bonds undergo autoxidation, a breakdown product is formed which reacts with thiobarbiture acid (TBA) to give a colored chromophore (10, 138, 139). This product has been identified as malondialdehyde and is derived from lipid endoperoxides (11, 140). 00- 00. /O , IVA/N ——»/"v/i\/'\ —~-» 0 9. .0. H’ V11 16 Detection of malondialdehyde via the reaction with thiobarbituric acid is a commonly used assay for lipid peroxidation. An intermediate in the formation of malondialdehyde from lipid endoperoxides is a prosta- glandin-like compound. In purified lipid emulsions, this prostaglandin- like compound decomposes to yield malondialdehyde both thermally and under the mild acid conditions of the TBA assay (11). In biologically active materials, hydroperoxides are also degraded enzymatically by mechanisms which are little understood (136). The liver cytosol and the mitrochondrial matrix in animals contain the previously mentioned hydroperoxide metabolizing enzyme, glutathione peroxidase (123, 141). Hydroperoxide isomerase in plants metabolizes hydroperoxides to a ketols (142, 143). Flour-water suspensions contain an enzyme that converts lipid hydroperoxides to epoxides (144, 145), while an enzyme found in potato tubers converts peroxides to vinyl ethers (146). The major sites within a cell that are susceptible to lipid peroxidation are the biomembranes, especially those of the subcellular organelles. Membrane associated enzymes and proteins are particularly susceptible to oxidative damage during lipid peroxidation. Among the consequences of lipid peroxidation on membrane associated proteins are: loss of enzyme activity (147), loss of solubility due to aggregation or complex formation (148), chain scission (149), as well as loss of specific amino acids. Cysteine, lysine, histidine, and methionine are particularly susceptible (150, 151). The inactivation of sulfhydryl containing amino acids usually occurs through the formation of cross- links (152). Polymers of cross-linked protein are formed when human serum albumin is stored (153). This is not surprising since 25% of the 17 fatty acids bound to human serum albumin are unsaturated and suscepti- ble to autoxidation. The cross-linked serum albumin also exhibits a fluorescence absent in freshly prepared samples. This fluorescence has been observed in lipid-protein systems undergoing lipid peroxidation (154). When freshly prepared serum albumin was exposed to either per- oxidized linoleic acid or to carbonyl compounds similar to those pro- duced during the breakdown of lipid peroxides, the fluorescent spectra resulted. It was subsequently demonstrated that amines and imines pre— sent in lysine and histidine react with lipid peroxidation breakdown products, including malondialdehyde, to produce fluorescent chromo- phores. Biologically important amines like RNA, DNA, and phospholipids form cross-links with malondialdehyde and other carbonyl compounds produced during lipid peroxidation to form fluorescent chromophores with Shiff base structures (155). R ——-N ===CH -— CH == CH -—-NH ——-R These fluorescent lipid-protein pigments accumulate in animals and are called lipofusin, ceroid, or aging pigments (156, 157). Lipofusin pigments are especially found in heart and brain and their formation is a function of age, oxidative stress, and antioxidant deficiency (8). Reichel studied the correlation between the appearance of lipo- fusin pigments and age in rat testes (158). The increased appearance of lipofusin pigments in testes was found to be proportional to age. When rats were fed a diet supplimented with vitamin E, methionine, and BHT, a 44% reduction in the appearance of lipofusin pigments was observed. In vitro, factors which promote lipid peroxidation can be shown to cause damage to lipid-protein membrane preparations. When ascorbate 18 and ferrous ions were incubated with erythrocytes a rapid breakdown of the membranes occurred (6). Ionizing radiation and the radical initia- tor, dialuric acid, also increased erythrocyte hemolysis (159, 160). Hemolysis paralleled the formation of malondialdyhyde. Antioxidants like a tocopherol and BHT prevented hemolysis. These findings have led others to suspect that the use of hyperoxia as a clinical treatment may increase peroxidative damage to erythrocytes, since lipid peroxidation increases in vitro with increased oxygen partial pressure (129). Erythrocytes are also sensitive to light. When erythrocytes were incubated with small amounts of photosensitizers, like Rose Bengal, Eosin, or Neutral red, in the presence of light of a wavelength greater than 350 nm, hemolysis occurred within minutes (161). Isolated mitochondria, like erythrocytes, are adversely effected by ferrous ions and reducing agents. Mitochondrial swelling, which parallels lipid peroxidation, produces irreversible damage under these conditions (162). Waldschmidt gt_al. (163) demonstrated that freshly prepared mito- chondria contained a large number of free radical centers, as indicated by esr signals. The number of free radical centers increased with increasing age of the animal and could be decreased in numbers by inhi- bitors of mitochondrial respiration. Such centers probably arise from the transfer of electrons from the respiratory chain to radical forming species. Pacer gt_al. (162) demonstrated that NADPH increased mitochon- drial lipid peroxidation and that cyanide and azide inhibited lipid peroxidation. Hyperbaric pressure also increases the appearance of malondial- dehyde in mitochondria, either in organ cultures or in suspension (164). 19 The increase in lipid peroxidation was largely prevented by addition of EDTA to the medium. The endoplasmic reticulum is particularly susceptible to lipid peroxidation owing to the high concentrations of C20:4 and C22:6 unsaturated fatty acids. Microsomes from liver, kidney, and brain all undergo lipid peroxidation at 37° C, especially in the presence of cell sap or ferrous ions (6). Microsomal peroxidation is inhibited by metal chelators, antioxidants, and inhibitors of cytochrome P-450 (75). NADPH greatly stimulates microsomal lipid peroxidation (165) and carbon tetrachloride further increases lipid peroxidation in the presence of NADPH (166). Recent evidence indicates that paraquat (methyl viologen) induced pulmonary lesions in mammals result from paraquat mediated lipid per- oxidation (167, 168). NADPH is required and both superoxide dismutase and singlet oxygen scavengers inhibit peroxidation. NADPH-cytochrome c reductase was shown to mediate the paraquat-induced peroxidation (169, 170). When microsomal membranes are incubated with ascorbate and ferrous ions, glucose 6 phosphatase activity decreases as lipid peroxidation increases (171). The loss of activity of the enzyme results as a con- sequence of the loss of membrane structure essential for the enzyme's activity. Amino acid incorporation into rough endoplasmic reticulum in vitro is also inhibited concurrent with lipid peroxidation (172). Hatefi and Hansten (173) have studied the effect of chaotrophic agents on mitochondria and microsomes. Their results indicate that the destructive processes that promote lipid peroxidation are built into the machinery of these two organelles. Under normal conditions, 20 molecular oxygen cannot find access to the membrane lipids to catalyze autoxidation. However, under conditions which lead to the destabiliza- tion of the membrane, such as addition of chaotrophic agents, lipid peroxidation occurs. a Toc0pherol and metal chelators protect the destabilized membranes from peroxidative damage. Desai has studied the effectcniultra violet radiation on lysosomal membranes. A correlation was observed between lipid peroxidation, free radical production, and release of lysosomal enzymes (174). A similar release of enzymes was observed with exposure of lysosomes to y-radia- tion. Fong gt_al, (91) have demonstrated that incubation of lysosomes with enzyme systems known to promote lipid peroxidation results in release of acid phosphatase. The xanthinezxanthine oxidase peroxidation system and the NADPH-cytochrome c reductase system both cause peroxida- tion damage to lysosomes and the release of acid phosphatase. Lysosomes contain large amounts of lipofusin pigments. It has been proposed that lysosomes engulf lipid-rich particles and that these particles slowly peroxidize. Lysosomes particularly rich in lipofusin pigments have low levels of acid phosphatase, suggesting that hydroper- oxide breakdown products cross-link with lysosomal enzymes to produce lipofusin pigments. Lysosomal lipofusin pigments stain histochemically for acid phosphatase. Goldfischer et_gl, (176) have suggested that lipid peroxidation in lysosomes is accelerated when heme-rich mitrochon- dria or microsomes are engUlfed, owing to the prooxidant properties of hemes and chelated metals. METHODS AND MATERIALS Material Sources: Male Sprague-Dawley rats (225-250 g) were obtained from Spartan Research Animals, Inc., Haslett, Michigan. Rats were fasted for 24 hours before sacrifice. Lactoperoxidase (milk), lipoxygenase (soybean), glucose oxidase (Sigma type II), and catalase (beef liver) were obtained from Sigma Chemical Company, St. Louis, Missouri. Bromelain was obtained as a gift from the Dole Company, Honolulu, Hawaii. Lipase (pancreatic) was obtained from Nutritional Biochemistry Company, Cleveland, Ohio. A stock culture of Escherichia coli was obtained from Mr. William Litchfield, Department of Biochemistry, Michigan State University, East Lansing, Michigan. ADP (Sigma fermentation grade), cytochrome c (Sigma type IV), ascorbate, butylated hydroxytoluene, o-dianisidine (dihydrochloride), EDTA, Coomassie blue, dithiothieitol, heparin (Grade I), y-linolenic acid, D-mannitol, NADPH, sodium dodecyl sulfate, phenylmethylsulfonyl- fludride, trypsin inhibitor (soybean), thiobarbituric acid, Tween 20, and Tris base were obtained from Sigma Chemical Company, St. Louis, Missouri. l,3-Diphenylisobenzofuran was obtained from Aldrich Chemical Company, Milwaukee, Wisconsin. Freund's adjuvant (complete) and agar (Special Noble grade) were obtained from Difco Laboratories, Detroit, Michigan. Trypticase soy broth and trypticase soy agar were obtained 21 22 from BBL, Division of Becton-Dickerson Company, Cockeysville, Maryland. Phenobarbital was obtained from Merck and Company, Inc., Rahway, New Jersey. AminOpyrine and cumene hydroperoxide were obtained from K and K Laboratories, Plainview, New York. Casein (Hammersten) was obtained from Schwartz Bioresearch Inc., Orangeburg, New York. Phenol reagent was obtained from Harleco, Philadelphia, Pennsylvania. Preparation of Rat Liver and Lung Microsomes: Liver microsomes were isolated from perfused rat livers as pre- viously described (177). Lung microsomes from the same animals were obtained by a similar procedure with slight modifications. Lung from male rats (225-250 g) were removed, blotted dry, and weighed. The weighed lung tissue was cut into fine pieces with scissors and homogen- ized in 4 volumes of 1.15% KCl, 0.2% nicotinamide, using 9 passes of a Potter-Elvehjem homogenizer with Teflon pestle. The homogenate was centrifuged at 9,000 x g for 20 minutes. The supernatant was carefully decanted and centrifuged at 105,000 x g for 90 minutes. The microsomal pellets from both liver and lung were washed by resuspending the pellets in 0.3 M sucrose, 0.1 M sodium pyrophosphate, pH 7.5 and recentrifuging at 105,000 x g for 90 minutes. Microsomes thus obtained were resus- pended by homogenization and stored at -20° C under N2 in 0.05 M Tris- HCl, pH 7.5, containing 50% glycerol. The average yield of microsomal protein per 250 g rat was 128 mg for liver tissue and 4 mg for lung tissue. Lipase Solubilization of NADPH-Cytochrome c Reductase: A crude pancreatic lipase preparation has previously been used to release NADPH-cytochrome c reductase from the microsomal membrane 23 surface (178). Crude pancreatic lipase was subjected to gel filtration and the eluting fractions tested for their ability to solubilize NADPH- cytochrome c reductase. Lipase was dissolved in 0.1 M sodium phosphate, pH 7.5 at a concentration of 20 mg/ml. One ml of the lipase was applied to a Sephadex G-100 column (20 x 250 mm) and 1.2 ml fractions were collected. Aliquots (0.05 mls) from each column fraction were mixed with 10 mg microsomal protein in 1.95 ml of 0.05 M Tris-HCl, then diluted with ice-cold 0.05 M Tris-H01, pH 7.5 and centrifuged at 105,000 x g for 90 minutes. The supernatant was decanted and the pellet resus- pended in Tris buffer. NADPH-cytochrome c reductase activity was assayed in both supernatant and microsomal fractions and the percent of the enzyme solubilized was calculated. NADPH-cytochrome c reductase was assayed as described under "Methods." Assay of Protease Activity: Protease activity was measured by following the increase in TCA- soluble peptides released from casein. Casein (3.0 mg) in 1.0 ml of 0.1 M sodium phosphate, pH 7.5, was mixed with enzyme (in 0.05 ml) and incubated for six minutes in a 37° C water bath. The reaction was ter- minated by the addition of an equal volume of 10% TCA. The mixture was mixed and centrifuged for 20 minutes at 1,000 x 9. One ml of the super- natant was neutralized by the addition of 2.0 ml of 0.5 N NaOH followed by 0.6 m1 of l N phenol reagent. The solution was mixed well and allowed to stand for 30 minutes at room temperature. The absorbance of the solution at 750 nm was determined against a blank containing all the reagents except enzyme. Protease activity was expressed as umoles of TCA-soluble tyrosine equivalents released per mg casein per minute. 24 Assay of Lipase Activity: Lipase activity was assayed by monitoring the enzyme-catalyzed hydrolysis and release of fatty acids from a solution of Tween 20. Five ml of a solution of 2.0% Tween 20 in 0.05 M sodium acetate, pH 7.7 at 25° C, was placed in the reaction vessel of a Radiometer, Type III titrigraph with mechanical stirrer. Enzyme (0.05 ml) was added to the Tween 20 and the release of fatty acids recorded by titrating the solu- tion automatically with 0.02 N NaOH. Lipase activity was expressed as pequivalents of free fatty acids released per minutes. NADPH-Dependent Peroxidation of Membrane Lipids: NADPH-dependent lipid peroxidation of microsomal membranes was assayed as follows. Microsomes (1.0 mg protein/ml) were incubated in a Dubnoff shaker at 37° C with 0.1 mM ADP, 0.1 mM FeCl and 0.2 mM NADPH 3, in 0.05 M Tris-HCl, pH 7.4. The reaction was initiated by the addition of NADPH. The extent of lipid peroxidation was determined by removing l.O aliquots at various time intervals and measuring either the amount of hydroperoxides present of the amount of malondialdehyde formed. NADPH-dependent lipid peroxidation in liposomes was determined as follows. Liposomes, prepared from extracted microsomal lipids, were prepared as previously described (87). The amount of lipid phosphate present was determined according to Bartlett (179). Liposomes (0.5 umoles lipid phosphate/ml) were incubated in a Dubnoff shaker at 37° C with 0.1 mM ADP, 0.1 mM EDTA, 0.2 mM FeCl3, purified NADPH-cytochrome c reductase, and 0.2 mM NADPH in 0.05 M Tris-HCl, pH 7.4. The reaction was initiated by the addition of NADPH. Malondialdehyde formation in both microsomes and liposomes was determined as previously described (86). The malondialdehyde 25 concentration of the sample was calculated using an extinction coeffi- 5 1 -1 cient of 1.56 x 10 M' cm . Lipid hydroperoxide concentrations were determined as described below. Measurement of Lipid Hydroperoxides by the Iodometric Assay: Reduction of hydroperoxides by iodide is a useful tool in deter- mining the amount of lipid hydroperoxides present in a membrane sample. The procedure is basedtwithe ability of I' to reduce peroxides by the following reaction (23): ROOH + 31' + 2H+ -———-——-+ ROH + H20 + 13" Under the conditions of the assay used here, only lipid hydroperoxides react with I', thus excluding from the reaction the endoperoxides that break down to form malondialdehyde. The amount of hydroperoxides pre- sent in a membrane sample was determined by extracting the membrane lipids from the aqueous reaction mixture followed by reacting the lipids with iodide. One ml of reaction mixture was added to 5.0 ml of chloro- form: methanol (2:1) and vortexed for one minute. The mixture was centrifuged in a clinical centrifuge for one minute to accelerate the separation of the two layers. Most of the upper, aqueous layer was aspirated, and 3.0 ml of the lower, chloroform layer was removed using a syringe and placed in a small test tube. The chloroform was evapor- ated off under a stream of N2. While still under a stream of N2, 1.0 ml of acetic acid: chloroform (3:2) was added to the dried lipid residue, followed immediately by 0.05 ml of KI (1.2 g/ml H20). The solution was stoppered and quickly mixed and placed in the dark for exactly five minutes. At the end of this time, 3.0 m1 of 0.5% cadmium 26 acetate was added and mixed, followed by centrifugation at 1,000 x g for ten minutes. The absorbance at 353 nm of the upper layer was deter- mined against a reagent blank obtained in the same manner but minus lipid. Standardization of the reaction was accomplished by the use of known quantities of cumene hydroperoxide. The standard curve is shown in Figure 1. At hydroperoxide concentrations less than 150 nmoles per assay, the standard curve is linear with a slope of 0.0076 absorbance units per nmole hydroperoxide in the sample. Isolation and Measurement of Liver and Lung Microsomal NADPH-Cytochrome c Reductase: Liver and lung microsomes were isolated as described under "Meth- ods." The microsomal NADPH-cytochrome c reductase activity was isolated and purified by a method employing bromelain digestion of the micro- somes, followed by gel filtration on Sephadex G-100 and affinity chro- matography on DEAE cellulose (87). Both liver and lung reductase were assayed for their ability to reduce cytochrome c as follows. The reaction mixture contained 0.075 mM cytochrome c, 0.1 mM NADPH, and purified liver or lung NADPH-cytochrome c reductase in 0.3 M sodium phosphate, pH 7.5 at 25° C. Absorbance measurements at 550 nm were made on a Perkin-Elmer model 124 spectrophotometer using a reagent blank missing only the reductase. The rate is expressed as nmoles of cyto- chrome c reduced per minute using an extinction coefficient of 2.10 x 4 l 10 M' cm.1 (180). One unit of activity is defined as one nmole of cytochrome c reduced per minute. 27 .emuo:umz= Long: nmnwcummu vegums on“ use muwxocmqocuxc mamasu mo mmwpwpcmso czocx mcem: nm:?ELmumu me; E: mmm pm mucmncomnm mg» .>m mmonommaomo>I oHaHA wsz2mmsu oa: of 59C 33268 8 3858.3 -Iaowpum A<.|III. mo cowpwcum mcp An umzmwpnmpmm mm: zumcmcum owcow ace .H4 qu 20mm mm0izaaHH>Hhu< mzh zo Ihwzmka quoA no Humumm .m mesmwm 51 .8. 20.52.2823 N.— «(‘02 .0— AllMlDV SSVIDHOBU D-OIAD 52 .=mco;pmz= Lung: umnwgummu mm o meogzuouxu mo cowuuanme we» go» umxmmmm use mapscwe op cow me saw: 0 cc um umumnaucwimca mm: AAE\mn A.ov mmmpuzume newewcaa A._||ii..v m::_ go A. IIII..V cm>wg .mmH4 omNHJHmngomizHoomHHz< >m mmui:aow4 .mmH4 omNHJHmDJOmizH<4mzomm NIH ow >oomHHz< >m mmuizaoemmno mm: vacw #3th Ac coAAosumm Azuvmu v. :5 A.o use 3388.» A.I .v 9:: co A..|||I .V cm>AA nmAAAcaq Ao As\m: A.o uchmAcou macaprE mxmmmm AA< .mOAamuu< zomhumAm 4uHmmmn IPA: mmqhozomm u mzomIQOA>oizaoPA>AHU< MIA no >oomAAz< >m zoAHAmAIzA .m mesmAm 55 0:3 lgG mg [ml 011 g 2 8 3 2. AIMIIDV °lo ‘Nouanaaa ’(ND)°:I‘)I 56 to either reductase. The antibody inhibits the reduction of cytochrome c by both reductases by up to 93%. The antibody is not as effective in inhibiting the reduction of ferricyanide by either enzyme, as seen in Figure 5. However, both liver and lung reductases are inhibited to the same extent. Figure 6 shows the precipitin lines obtained after double diffu- sion immunOprecipitation was performed on Ouchterlony plates using both bromelain-solubilized, purified reductase and detergent-solubilized, native reductase from liver and lung microsomes. All wells containing reductase form precipitin lines that merge without forming spurs, sug- gesting that the antigenic sites on both liver and lung enzymes are the same. The gel scans in Figure 7 are the protein profiles observed by scanning Coomassie blue stained sodium dodecylsulfate-polyacrylamide gels containing immunoprecipitates from detergent-solubilized liver and lung microsomes. We have previously shown that the antibody directed against the bromelain-solubilized liver reductase could precipitate the native enzyme from sodium deoxycholate-solubilized microsomes (182). When 125I-labeled microsomes were solubilized in detergent and incubated with antibody, only a single protein containing radioactivity (79,000 dalton) was observed after subjecting the immunoprecipitate to sodium dodecylsulfate-polyscrylamide gel electrophoresis (182). Figure 7 (b) shows the protein profile of a gel that contained the immunoprecipitate from detergent-solubilized lung microsomes incubated with antibody. The two major bands are the heavy (52,000 dalton) and the light (25,000 dalton) chains of the reduced IgG molecules (211). The highest molecu- lar weight band corresponds to a protein of 79,000 daltons. Only a 57 .=msosuwz= ewss: swsAewmwu mw wuss wewz mwAwAs AserwAsuso .0 N as s «N E swesuuo soAAwAAaAwwesosssEA .wmwv -usswe ew>AA As an N msAwmwou a .oz Asz .wmwpusswe mssA swAAAess As on N msAwAsow m .02 Asz .AwAonsuzxowu saAsom AA sA Ae\me mAvsAonesAmeom -oewAE ew>AA As we N.A msAwAsow N .02 Asz .AwAonsuxxows EaAsom NA sA Ae\me my sAones AweomoewAe ew>AA As we v.0 msAwAsou A .02 Asz .AA .ozv sou wsp EoeA wmAzxwko ewAssou swewnsss wew mAsz ewpso wsA .AsAones we s.Av cmA wmwwwsswe w wEoesuopxwizsosziAAsw msAwAsou Asz ewuswo wsA .mh<4a mzogmmhxuso z< 20 @224 oz< mm>A4 AH4 Auuzqooom-z< mzh mo zoAA02+Fe 3) FeH + H202 —-——————-> OH- + 0H“ + Fe+++ 4) NADPH-cytochrome c reductase, found either in microsomes or in its puri- fied (bromalain digestion) state, has been shown to produce 0; upon oxidation of NADPH (212-214). Their scheme called for the peroxidation of polyunsaturated lipids by the abstraction of a hydrogen from a methylene carbon atom of the lipid by the OH‘. The direct extraction of a hydrogen by 05 was not considered feasible since superoxide dismutase cause no inhibition of lysozomal lysis. In addition to being partially inhibited by OH- trappers, the reaction was also inhibited by catalase. However, recent studies (215) have demonstrated that catalase does not inhibit NADPH-dependent lipid peroxidation. The discrepancy between these findings is probably due to the presence of thymol, a potent anti- oxidant, in commercial preparation of catalase. When catalase free of thymol is added to the NADPH-dependent peroxidizing system, lipid peroxidation is no longer inhibited. The inability of superoxide dismutase and catalase to inhibit NADPH-dependent lipid peroxidation has led to the proposal of a differ- ent reaction mechanism to explain microsomal lipid peroxidation. Peder- son and Aust (215) have suggested that reduced NADPH-cytochrome c reductase acts on Fe+++ to produce Fe++. Fe++ could either act on pre- existing hydroperoxides present in the membrane to produce more lipid radicals, or could combine with molecular oxygen to form the perferryl ion (Fe02)++. Oxygen is known to combine with Fe++ to form the perferryl 65 ion as an intermediate in the oxidation of Fe++ (216, 217). This spe- cies may be able to produce a peroxy radical or extract an electron from the lipid, yielding a lipid radical and H202. The possibility that NADPH-dependent lipid peroxidation is mediated by a reduced form of iron and molecular oxygen will now be examined. The detectable events that occur during NADPH-dependent lipid peroxidation include an uptake of oxygen (75), the oxidation of NADPH (138), the appearance of lipid hydroperoxides (218), a perturbation of the surface structure of microsomes resulting in a change in turbidity (218), a loss of unsaturated fatty acids (1), and the appearance of lipid peroxide breakdown products, including malondialdehyde (138). Figure 8 illustrates two of these observable changes in peroxidizing microsomes over a period of time. In the presence of NADPH and ADP- chelated Fe+++, microsomes produce an increasing amount of malondialde- hyde over the 60 minute incubation period. May and McCay (138), have demonstrated that the continued increase in malondialdehyde depend on the availability of reducing equivalents from NADPH. When NADPH is exhausted, lipid peroxidation ceases. Subsequent additions of NADPH cause further peroxidation, as judged by oxygen uptake and malondialde- hyde formation. In addition to measuring malondialdehyde production, lipid hydro- peroxide levels can also be measured through the use of the iodometric assay (6, 219) which is specific for easily reduced peroxides (peroxy acids, diacyl peroxides, and all hydroperoxides). This assay has been used extensively by nutritionist and oil and soap chemists to detect the presence of hydroperoxides in purified lipid samples (220-224). Reck- nagel and Ghoshal (225) have used a similar procedure on extracted 66 .=muosuwz= ewus: uweAewmwu we uwAemme wewz soAAeEeoA quxoeweoeuxs use qusqueAuserz .weauxAE soAAwewe Aoeusou sA soAAeeeoA wuzsqueAuserE AD by use soAAeseoA quxoewsoeu»: 0V .wesuxAs soAAuewe AeusweAewaxw sA soAAeEeOA wuzsqueAusere A. .v use soAAeseoA quxoewsoeu»: A..|||I .v .Azm Amoo.o uwsAeA -sow wesprs soAAwewe Aoeusow wsA .emuofiwz= ewus: soApequoewe uAsAA AeEOmoewAE AswuswewuixsozmowI no mmmzou mzAH .w wesmAm 67 ugetmd Om/aplqeploipuolow salowu O -60 .8 e a 8 2 8 N 400 “ UIGWd blu/apgxoaadmpm salow u so 50 3O 20 Tlme(min) 68 lipids from the microsomal fractions of rats exposed to CCl4 to demon- strate an increase in membrane lipid hydroperoxides during liver necro- sis. Its use in detecting the appearance of hydroperoxides formed during in vitro enzymatically induced membrane lipid peroxidation is a comparatively new one. Tam and McCay (218) used a similar procedure to suggest that lipid hydroperoxides were transient species formed in the early stages of NADPH induced lipid peroxidation and that the hydroperoxides fell to low levels as their breakdown began to occur. Slater (6) demonstrated that microsomes allowed to autoxidize at 4°C had high levels of lipid hydr0peroxides, but low levels of malondialde- hyde. When identical microsomal samples were incubated at 37°C instead, hydroperoxide levels fell to a low level as malondialdehyde levels increased, indicating that malondialdehyde is derived from the breakdown of previously formed lipid peroxides. In the present study, it can be seen that lipid hydroperoxide and malondialdehyde levels increase simul- taneously during the early phase of NADPH-catalyzed lipid peroxidation (Figure 8). However, after reaching a peak at about 20 minutes, hydro- peroxide levels begin to decrease. This would suggest that the forma- tion of lipid hydroperoxides and the breakdown of lipid peroxides to form malondialdehyde may be two consecutive and separate events. It can also be seen that the free radical trapper, BHT, abolishes both hydro- peroxide and malondialdehyde formation. Therefore, at least the formation of hydroperoxides, and perhaps the breakdown of lipid peroxides to yield malondialdehyde, are dependent on the production of free radicals. ++ chelated by ADP in NADPH-dependent lipid The involvement of Fe+ peroxidation is demonstrated in Table 5. The presence of ADP in this system is thought to prevent Fe+++ from precipitating from solution as 69 TABLE 5 NADPH-QEEENDENT MICROSOMAL LIPID PEROXIDATION AS A FUNCTION OF ADDED ADP'Fe . The reaction mixtures contained 1.0 mg/ml microsomal protein and 0.2 mM NADPH as described for NADPH-dependent microsomal lipid peroxi- dation under "Methods". ADP-Fe or EDTA were added as indicated. NADPH was delected from the reaction mixture containing ascorbate. The reaction time equaled 5 minutes. Malondialdehyde formed Description nmoles/min/mg protein Microsomes and NADPH Pius 0.1 mM ADP-Fe+++ ...... 2.50 Pius 0.01 mM ADP-Fe+++ ..... 2.41 Pius 0.001 mM ADP-Fe+++ ..... 1.60 Minus ADP-Fe+++ ......... 1.07 Pius 0.1 mM EDTA ........ .03 Microsomes Plus 0.5 mM ascorbate ...... 1.60 7O Fe(0H)3 and to prevent Fe+++ from binding to components on the micro- somal surface (105, 218). It is apparent that small amounts of ADP-Fe+++ assist in NADPH-dependent lipid peroxidation as indicated by malondial- dehyde production. In the absence of added ADP-Fe+++, the addition of EDTA decreased the level of lipid peroxidation below the level observed when no ADP-Fe+++ is present. This would indicate that some metal is associated with the microsomal membrane and can be used to promote lipid peroxidation. This was confirmed by adding ascorbate to the microsomal incubation mixtures. The non-enzymatic oxidation of unsaturated membrane lipids catalyzed by ascorbate is known to be acce- lerated by inorganic or organic iron (226, 227). However, Table 6 demonstrates that ascorbate-catalyzed lipid peroxidation is completely dependent upon added iron in a purified liposomal system. Therefore, the peroxidation observed when microsomes and ascorbate are incubated together without exogenous Fe+++, must depend on iron associated with the microsomal membrane. If NADPH-dependent lipid peroxidation proceeds via the reduction of ADP-Fe+++, then ADP-Fe++ must be shown to catalyze lipid peroxidation. Table 7 demonstrates the results of incubating microsomes with either ADP-Fe++ or ADP-Fe+++ in the absence of NADPH. The concentration of chelated iron is high (0.7mM) to better illustrate the enhanced cataly- tic activity of ADP-Fe++ over ADP-Fe+++ in promoting lipid peroxidation. In peroxidation experiments using either purified NADPH-cytochrome c reductase or ascorbate, 10 u molar Fe“+ is sufficient to promote some + lipid peroxidation. However, in these systems, the reduction of Fe++ is continuous, thus replenishing the supply of Fe++. 71 TABLE 6. THE EFFECT OF ADP-Fe+++ 0N ASCORBATE-CATALYZED LIPID PEROXIDATION IN LIPOSOMES. The reacti0n mixtures contained 1.0 umoles lapid phosphate‘ml, 0.1 mM ascorbate and 0.05 M Tris-HCl, pH 7.4 at 37 C. ADP-Fe or EDTA were added as indicated. The reaction time equaled 5 minutes. Malondialdehyde formed Description nmoles/min umOTés lipid Pi Liposomes and ascorbate plus; 10 0M ADP—Fe+++ ..... 3.2 5 uM ADP-Fe+++ ..... 3.2 i uM ADP-Fe+++ ..... 3.1 0.5 uM ADP-Fe+++ . . . . 2.8 0.1 pM ADP-Fe+++ . . . . 2.2. No ADP-Fe+++ ...... 0.06 0.1 mM EDTA ....... 0.00 72 TABLE 7. MICROSOMAL LIPID PEROXIDATION PROMOTED BY ADP-Fe++ 0R ADP-Fe+++. The reaction mixtures contained 0.5 mg microsomal protein in 0.05 M Tris-Hcl, pH7.4 at 37° C. Iron was added as indicated. The reaction time equaled 10 minutes. Description Malondialdehyde formed nmoles/min/mg protein Microsomes Plus 0.7 mM ADP—Fe++ ......... 7.05 Pius 0.7 mM ADP—Fe+++ ........ 1.90 No additions ................. 0.15 73 When a model peroxidizing system composed of NADPH, purified NADPH- cytochrome c reductase, and ADP-Fe+++ was first used to promote lipid peroxidation in liposomes derived from purified microsomal lipid, it became apparent that an additional component was required for peroxida- tion activity (87, 215). Subsequently, others have confirmed that EDTA- chelated Fe+++, in addition to ADP-Fe+++, is required for peroxidation to occur in both liposomes and hemoprotein free lipoprotein particles (88, 111). Figure 9 demonstrates that both malondialdehyde and hydro- peroxides are produced in the model system in the presence of EDTA- Fe+++. As in microsomes, the hydroperoxide concentration rises early in the reaction and subsequently declines as hydroperoxides are appar- ently broken down. However, upon close examination of the iron requirements in the model peroxidizing system, it became evident that small amounts of both malondialdehyde and hydroperoxides were being produced when EDTA-Fe+++ was absent from the complete reaction mixture. This led to a re-examin- ation of the iron requirement in the model peroxidizing system. In microsomes, where ADP-Fe+++ alone is sufficient to catalyze NADPH- dependent lipid peroxidation, addition of EDTA-Fe+++ along with ADP- Fe+++ in equal amounts results in no more malondialdehyde production than an amount of ADP-Fe+++ equal to the total of the two chelated forms of iron. Yet in liposomes, deletion of EDTA-Fe”+ results in 60-70% decrease in malondialdehyde production. Various concentrations of ADP-Fe+++ and EDTA-Fe+++ were added to liposomes and incubated with NADPH and NADPH-cytochrome c reductase in Chelex-treated 0.05m Tris-HCl buffer. Chelex treatment of 0.05 M potassium phosphate buffer reduces total iron contamination to below 10 n molar (228). As demonstrated 74 .=muospwz= ewus: uwnAeumwu we uwzemme wew: A. .v soAAeeeoA wuzsqueAusere use A. .v soAAeEeoA quxoews -oeuxx .u Am Ae AuximAeA z mo.o sA wmepuauwe u wEoesquAUTIQoxmo4I no mmmzou mth .m wesmAs 1:! Win 910W rr/ apfluaploipuolow salow u 75 o 0 ID 0 Q 03 E to '6' r0 52 o a a 0- v N 0 1d pldl‘l slow in apixmadmpKH saiowu 30 4O 5O 60 20 IO Time (min) 76 in Table 8, EDTA-Fe+++ alone is not capable of supporting any lipid peroxidation in the model system, while ADP-Fe”+ alone will catalyze decreased levels of peroxidation. The greatest peroxidation occurs when + both ADP-Fe++ and EDTA-Fe+++ are present together. Even when the con- centration of ADP-Fe”+ is 100 times the total iron concentration of the two combined forms of iron, less peroxidation is evidenced. In order to verify that EDTA-Fe+++ undergoing reduction was not capable of supporting lipid peroxidation, ascorbate and EDTA-Fe+++ were incubated with liposomes. The ability of ascorbate to reduce EDTA-Fe+++ was tested by observing the reduction of cytochrome c by ascorbate in + the absence and presence of EDTA-Fe++ . A fourfold increase in the rate of cytochrome c reduction was observed in the presence of EDTA-Fe+++. It appears that ascorbate reduces EDTA-Fe“++ which in turn reduces cytochrome c. Table 9 demonstrates that ascorbate and EDTA-Fe+++ do not promote lipid peroxidation, whereas ascorbate plus ADP-Fe+++ or the + + combination of ADP-Fe++ and EDTA-Fe+++ do. If reduced EDTA-Fe+ I does not promote lipid peroxidation, then why does it assist so dramatically the peroxidation catalyzed by ADP-Fe+++ in the NADPH-dependent liposomal system (Table 8)? Chelation by EDTA greatly effects the reduction potential of Fe+++ (229). The NADPH oxidase activity of the bromelain-solubilized NADPH-cytochrome c reductase with no iron, 0.1 mM ADP-Fe+++, or 0.1 mM EDTA-Fe+++ are 0.09, 0.5, and 25,0 p equivalents of NADPH/min/mg enzyme, respectively (230). The oxidase activity of the enzyme in the presence + and EDTA-Fe+++ is not appreciably dif- of the combination of ADP-Fe‘L+ ferent from that observed with EDTA-Fe+++ alone (111). The effect of EDTA Chelation of Fe+++, therefore, appears to be to reduce the reduction 77 TABLE 8. THE EFFECT OF EDTA-Fe+++ 0N NADPH-DEPENDENT LIPID PEROXIDATION IN LIPOSOMES. All the reaction mixtures contained 1.0 nmoles lipid phosphate/ml, 0.2 mM NADPH, 0.12 pg/ml NADPH-cytochrome c reductase in 0.05 M Tris-HCl, PH 7-4 at 37° C. ADP-Fe+++ and EDTA Fe+++ were added in the amounts indicated. The reaction time equaled 15 minutes. Description Reaction mixture plus: Malondialdehyde formed mM ADP-Fe+++ mM EDTA-Fe+++ nmoles/min/pmoles lipid Pi 0.2 - 1.70 0.01 - 1.60 0.005 - 1.21 0.001 - 0.65 - 0.2 0.05 - 0.01 0.05 - 0.005 0.05 0.1 0.1 4.33 0.005 0.005 3.43 0.001 0.001 1.92 78 TABLE 9. THE EFFECT OF ADP-Fe+++ and EDTA-Fe+++ ON ASCORBATE-CATALYZED LIPID PEROXIDATION IN LIPOSOMES. The reaction mixtures contained 1.0 nmoles lipid Pi/ml and 0.1 mM+++ ascorbate in 0.05 M Tris-HCl. pH 7.4 at 37° C. ADP-Fe++f and EDTA-Fe were added as indicated. The reaction time equaled 10 minutes. Description Malondialdehde formed nmoles/min/umoles lipid Pi Liposomes and ascorbate Plus 0.1 mM ADP-Fe+++ .......... 1.94 Plus 0.1 mM ADP-Fe+++and 0.1 mM EDTA-Fe+++ ......... 2.14 Pius 0.1 mM EDTA-Fe"M ......... 0.03 79 potential of Fe+++. The NADPH oxidase activity of o-phenanthroline- chelated iron catalyzed by NADPH-cytochrome c reductase has been shown to be about four times greater than EDTA-FeI'++ (111). When o-phenanthro- line-chelated Fe+++ was tested for its ability to peroxidize liposomes in the presence of NADPH, it was able to support about 30% of the peroxidation catalyzed by ADP-Fe+++ alone (Table 10). Therefore, it would appear that the ability to promote lipid peroxidation is directly proportional to the reduction potential of Fe+++. When used alone, the more difficult to reduce forms of chelated iron initiate the greatest amount of peroxidation. Combining two forms of Fe+++, one with a higher (ADP-Fe+++ ) and one with a lower (EDTA-Fe+++) reduction potential appears to be the most effective way of promoting NADPH-dependent lipid peroxidation. Since ADP—Fe+++ alone is capable of initiating lipid peroxidation in the NADPH-dependent liposomal system as judged by malondialdehyde production, then the enhanced peroxidation observedirithe presence of ++ may be due to its ability to facilitate the breakdown of EDTA-Fe+ hydroperoxides to further propagate more lipid radicals. Earlier work in this section indicated that there were two phases in the lifetime of lipid hydroperoxides in membranes undergoing peroxidation. The early phase was characterized by a net build-up of lipid hydroperoxides while the latter phase showed a net breakdown of hydroperoxides. If the for- mation of hydroperoxides and their subsequent breakdown are catalyzed in two separate events, then it seems plausible that each of the two forms of chelated iron required in liposoaml peroxidation may play a predomi- nant role in only one of the two reactions. EDTA—Fe+++ may facilitate the breakdown and propagation of lipid peroxides by a mechanism similar 80 TABLE 10. EFFECT OF VARIOUS Fe+++ CHELATORS ON NADPH-DEPENDENT LIPID PEROXIDATION IN LIPOSOMES. The reaction mixtures contained 1.0 nmole lipid Pi/ml, 0.02 mM NAD H, 0.35 ug/ml NADPH-cytochrome c reductase in 0.05 M Tris-HCl, pH 7.4 at 37 C. Chelated Fe was added as indicated. The reaction time equaled 15 minutes. Description Reaction mixture plus: ADP-Fe+++ EDTA-Fe+++ o-phenanthroiine-Fe+++ Malondialdehyde formed nmoleslmin umo e ip 0.1 mM - - 0.1 mM 0.1 mM - - 0 1 MM - - 0.1 mM 0.1 mM - - 0.1 mM 0.59 2.58 0.03 0.33 0.19 81 to that proposed for the autoxidation of lipids. Lipid autoxidation is known to be facilitated by Fe+++ and Fe++ as seen below (6-9): ROOH + Fe+++ ——-—————-+ R00: + Fe++ + H+ ROOH + Fe++-————————+ RO' + 0H‘ + Fe+++ Cumene hydroperoxide was used to assess the ability of both ADP-Fe+++ and EDTA—Fe+++ to breakdown hydroperoxides. NADPH, NADPH- cytochrome c reductase, and cumene hydroperoxide were incubated in the + or EDTA-Fe+++. As seen in Figure 10, the presence of either ADP-Fe++ ability to catalyze the breakdown of the functional group of cumene hydroperoxide is inversely proportional to the reduction potential of Fe+++. EDTA-Fe+++ was found to be almost six times more effective than ADP-Fe+++ in the breakdown of cumene hydroperoxide. If the disappear- ance of the hydroperoxide function of cumene hydroperoxide proceeds as suggested above in the reaction of Fe++ with ROOH, then in membranes more lipid radicals would be formed, some leading to the formation of more lipid peroxides while others would lead to non-peroxide termination products (6-9): (R-, R0°, ROO’) +(R-, R0-, R00-) —————————+ breakdown products Since EDTA-Fe+++ is not required in the NADPH-dependent peroxida- tion of microsomal membranes, it has been proposed that microsomes pos- sess an electron transport component that is replaced by EDTA-Fe+++ in the model peroxidation system. This proposed component would only be reduced by NADPH-cytochrome c reductase. If NADH was used in place of NADPH, no peroxidation of microsomes was observed unless EDTA-Fe+++ was present in addition to ADP-Fe+++. Pederson et a1. (87) demonstrated 82 .smuossz= ewus: uwnAewmwu we xemme wAequoqu wsu 2s uwzemme wewz mAw>wA 828823: A: + $-58 :5 As use + £23 22. A.o .on oh. -58 2.5.0 .A.V +++wsimms wswszw monE s moA waAeEAxoesse uwsAeAsou Aonnexm uAAomv mweszAE soAAuewe AeuswsAewsxm . es swAeowIw so wuzwmwma MIA 2A wosxoswsosoez wzwssu so zoAAAmoszowwo Az%fl&wsws-xao4sz no zoAhuzsu < m< oAmAA 4g<~444Awwwe mweepre soAAwewe wsA .emuosszs ewus: uwsAeomwu me wmequoeweouweA use +++weiss< .Ax .Ae\As uAeAA onEn m.o uwsAeAsou mweeuer soAAuewe wsA .zoAAssAzwezow NONI so zoAszee s m<_mesom0ssszs onAesAxosws sAsAb ewNAomu<>goaimAumooo zsmaom .mA wees: 121 A53 woz<._.m_o zo_._. Fe++ > Fe+++ > Cu++. Decomposition of hydroperoxides by metals in their higher valence state was greatly facilitated by reducing agents, such as ascor- bate and cysteine. Therefore, the reduction of EDTA-Fe+++ by NADPH- cytochrome c reductase in the liposomal peroxidation syStem must dupli- cate the decomposition of hydroperoxides and the propagation of lipid radicals promoted by cytochrome P-450 in intact microsomes. It has previously been reported that the decomposition of lipid hydroperoxides catalyzed by heme containing compounds results in the destruction of the heme group (102). The incubation of hydroperoxides with microsomes also resulted in the destruction of cytochrome P-450, as indicated by the loss of drug metabolizing activity in microsomes (297). It is interesting to note that peroxidation of microsomal lipid cata- lyzed by NADPH (298, 299) and the lactOperoxidase protein iodinating system (248) both result in the destruction of cytochrome P-450. There- fore, the destruction of cytochrome P-450 during enzyme mediated lipid 150 peroxidation probably results from the catalytic interaction of cytochrome P-450 with lipid hydroperoxides during hydrOperoxide breakdown. The Peroxidase Activity of Cytochrome P-450: In a second series of papers, Hrycay and O'Brien (297, 300), ob- served another cytochrome P-450 catalyzed reaction with lipid hydroper- oxides. In the presence of a hydrogen donor, such as N, N, N', N'- tetramethyl-p-phenylenediamine, (TMPD), cytochrome P-450 acts as a peroxidase to decompose linoleic acid hydroperoxide. A peroxidase reac- tion results in the reduction of a peroxide with the simultaneous oxidation of a hydrogen donor (295). When microsomal “P-450 particles," containing cytochrome P-450 as the sole protoheme containing constituent, were incubated with linoleic acid hydroperoxide, in the presence of TMPD, TMPD was oxidized as the linoleic acid hydrOperoxide disappeared. During the course of the reaction, cytochrome P-450 was not destroyed. Therefore, in the presence of an oxidizable substrate and a lipid hydro- peroxide, cytochrome P-450 acts as a peroxidase without undergoing destruction. If the interaction of lipid hydroperoxides with cytochrome P-450 results in the formation of a peroxidase Compound I type species, (295), then the peroxidase activity of cytochrome P-450 is chemically reasonable. A peroxidase Compound I type species was also proposed for cytochrome P-450 as an intermediate in the lipid hydroperoxide- dependent hydroxylation of drugs. Of particular interest was the find- ing that linoleic acid hydroperoxide was three times more effective than cumene hydroperoxide in the cytochrome P-450 mediated oxidation of TMPD. It was demonstrated in Chapter III that endogenous, microsomal lipid hydroperoxides formed by lipoxygenase were more effective in promoting cytochrome P-450 mediated drug metabolism than cumene hydroperoxide. 151 It may be true that drug substrates, like peroxidase hydrogen donors, protect cytochrome P-450 from destruction. If lipid hydroper- oxides do not destroy cytochrome P-450 in the presence of hydrogen donors, then drug substrates may also afford protection. If this were not true, then lipid hydroperoxides would destroy cytochrome P-450 and hydroperoxide-dependent drug metabolism would not be possible. Mixed type reactions in which monooxygenases, peroxidases, and oxidases interchange activities under defined conditions have been observed (295). If peroxides are intermediate products of oxygen reduc- tion, then it may be said in many cases that peroxide metabolism is involved as a part of the overall metabolism of oxygen. Based on the evidence presented here, the following three reactions can be proposed for cytochrome P-450 in the presence of lipid hydroper- oxides. LOOH + AH2 9Yt° P'450a LOH + H20 + A l) LOOH + drug substrate Cyt. P-450f drug-OH + lipid products 2) LOOH Cyt. p-450; lipid radicals 3) (degraded Cyt. P-450) Reactions 1) and 2) probably do not result in the degradation of cyto- chrome P-450 and may not be radical producing. Reaction 3) is of inter- est in the present study, and available evidence indicates that both breakdown of hydroperoxides and promotion of lipid peroxidation occur via this reaction (297, 300). A Model for the Enzyme-Catalyzed Peroxidation of Membrane Lipids: Based on reaction 3), the following model can be proposed for the mechanisms involved in lipid peroxidation catalyzed by the NADPH- 152 cytochrome c reductase, lactoperoxidase, and lipoxygenase systems (Figure 19). NADPH-dependent lipid peroxidation is probably initiated by the interaction of reduced, ADP-chelated iron with molecular oxygen to pro- duce the perferryl ion. This species probably acts very much like the superoxide anion to abstract a hydrogen from unsaturated lipids. Lipid hydroperoxides subsequently result from the interaction of molecular oxygen with newly formed lipid radical. In microsomes, cytochrome P-450 acts to further propagate more free radicals via the catalytic breakdown of lipid hydroperoxides. Cytochrome P-450 is destroyed in the process. In liposomes, enzymatically reduced EDTA-chelated iron replaces cyto- chrome P-450 to catalyze the breakdown of lipid hydroperoxides and to propagate more free radicals. In both microsomes and liposomes, lipid endoperoxides (precursors to malondialdehyde) are formed during the incubation period. Lipoxygenase-catalyzed lipid peroxidation results in the formation of lipid hydrOperoxides without the intermediate production of free, lipid radicals. In microsomes, cytochrome P-450 catalyzes the breakdown of hydrOperoxides and the propagation of more lipid radicals. Lipid endoperoxides (and malondialdehyde) are formed during the incubation period. In liposomes, hydroperoxides are produced by the enzyme without their subsequent breakdown to form malondialdehyde. Only when lipid hydroperoxides are heated under mild acid conditions with Fe+++, do the hydroperoxides undergo breakdown. Malondialdehyde is formed only during the heating step. In lactoperoxidase-catalyzed lipid peroxidation, initiation of per- oxidation probably occurs via the formation of intermediate oxidation 153 Figure 19. REACTION MECHANISMS INVOLVED IN MICROSOMAL AND LIPOSOMAL LIPID PEROXIDATION CATALYZED BY THE NADPH-CYTOCHROME C REDUCTASE, LIPOXYGENASE, AND LACTOPER- OXIDASE SYSTEMS. 154 LH Ligtgggmfiigise : . (warm; -Aop-Fe+202) 2 L’ 02 ADP-Fe+2 .___.ADP-Fe+3 02 Li poxygenase 4., LOIIH NADPH + Cyt. c Reductase 02// \\ z... MICROSOMES LIPOSOMES A g “T V // \\ EDJA-Fe*2.——{DTA-Fe*3 LOOH LOOH Cyt. P-450 ———9 L0 ' LOO LO‘ LO L' L' Endoperoxides Endoperoxides (Malondialdehyde) (Malondialdehyde) HEAT + FE i LOO' LO- L. Malondialdehyde 155 products of 1'. Either I: or I+ would be capable of interacting with unsaturated lipids to effect lipid radical formation. Lactoperoxidase- catalyzed lipid peroxidation in microsomes is facilitated by lipid radi- cals produced through the interaction of lipid hydroperoxides with cytochrome P-450. In liposomes, neither lipid hydroperoxides nor malondialdehyde are detected during the incubation period. This probably indicates that the rate of initiation of hydroperoxide formation is slow in the lactoperoxidase system. Without a catalyst like cytochrome P-450 or reduced iron, peroxidation does not proceed very fast. Only under the influence of heat and mild acid does Fe+++ act catalytically to pro- mote increased hydroperoxide formation and breakdown. The involvement of cytochrome P-450 in the metabolism of membrane lipid hydroperoxides has some important implications on the regulation and control of the destructive processes associated with lipid peroxida— tion. If cytochrome P-450 acts as a true peroxidase in the presence of hydrogen donors, then lipid hydroperoxides should be reduced to lipid hydroxides and water. If this reaction does not proceed via the produc- tion of free lipid radicals, then lipid hydroperoxides would effectively be neutralized. Therefore, if suitable hydrogen donors are present in the endoplasmic reticulum, lipid peroxidation (either NADPH-dependent or non-enzymatically induced) would be inhibited by the peroxidase cata- lyzed reduction of hydroperoxides by cytochrome P-450. O’Brien has demonstrated that suitable hydrogen donors for the peroxidase activity of cytochrome P-450 include ascorbate, cysteine, ubiquinol, a tocopherol, reduced glutathione, tyrosine, and tryptOphan (301). In the absence of suitable hydrogen donors, however, cytochrome P-450 would act as a pro- oxidant, resulting in increased peroxidation and its own destruction. SUMMARY 1) The solubilization of NADPH-cytochrome c reductase from the microsomal membrane by either crude pancreatic lipase or protease results in the proteolytic cleavage and release of a catalytically active pro- tion of the native enzyme. Antibody to the purified, liver microsomal reductase inhibits the cytochrome c and ferricyanide reducing activity of both liver and lung microsomal NADPH-cytochrome c reductase equally. Ouchterlony analysis and SDS-gel electrophoresis of the immunoprecipi- tates obtained from detergent-solubilized liver and lung microsomes indicate that the liver and lung NADPH-cytochrome c reductase are extremely similar, if not identical. 2) Purified NADPH-cytochrome c reductase from lung microsomes is as effective as the corresponding liver enzyme in catalyzing lipid per- oxidation in model liposomal system. NADPH-dependent peroxidation in lung microsomes, however, occurs much slower than NADPH-dependent perox- idation in liver microsomes. The decreased rate of malondialdehyde production in lung microsomes is probably a function of the structure and composition of the lung microsomal membrane. 3) NADPH-dependent peroxidation of microsomes and liposomes requires ADP-chelated ferric ions. ADP-Fe++ is more effective in cata- lyzing lipid peroxidation than ADP-Fe+++. Therefore, the enzymatic reduction of ADP-Fe+++ by NADPH-cytochrome c reductase is probably responsible for the initiation of lipid peroxidation, presumably through 156 157 the interaction of a perferryl ion with unsaturated microsomal lipid. Neither the enzymatic nor non-enzymatic reduction of EDTA-Fe+++ results in the lipid peroxidation of microsomal lipid. 4) NADPH-dependent peroxidation in microsomes and liposomes results in an early build-up of lipid hydroperoxides followed by a net breakdown of hydroperoxides. Enzymatically reduced EDTA-Fe+++ is very effective in catalyzing the breakdown of cumene hydroperoxide and probably func- tions in the model, liposomal system to breakdown lipid hydroperoxide. The breakdown of lipid hydroperoxides by EDTA-Fe++ probably results in the production of peroxide forming lipid radicals. Neither ADP-Fe+++ nor ++ EDTA-Fe+ , in their oxidized states, catalyze lipid peroxide breakdown at 37° C. Since intact microsomes do not require EDTA-Fe+++ in NADPH- dependent peroxidation, EDTA-Fe+++ probably functions in the model lipo- somal system to replace a component of the microsomal membrane which catalyzes lipid hydroperoxide breakdown and radical propagation. 5) Lactoperoxidase, H202 and I’ catalyze both the iodination of protein and the peroxidation of microsomal and liposomal lipid membranes. Optimum concentrations of I" and H202 for lactoperoxidase-catalyzed - peroxidation of microsomes are 0.4 mM and 8.8 mM, respectively. The optimum concentration of H202 decreases for peroxidation of liposomes owing to elimination of the catalase contaminant present in microsomes. 6) Fe+++ is not required for the initiation of lactoperoxidase- catalyzed I'iP'idpe'roxidation, but does assist in the propagation of lipid radicals via the breakdown of lipid hydrOperoxides during heating of the membrane with TCA-TBA reagent. BHT added at the completion of the incu- bation period completely abolishes malondialdehyde production in lipo- somes and causes a small decrease in malondialdehyde production in 158 microsomes. Therefore, in liposomes, malondialdehyde produced from the breakdown of lipid peroxides, does not occur in appreciable amounts until the membrane lipids are heated in the presence of Fe+++. 7) Initiation of lipid peroxidation by the lactoperoxidase system is probably catalyzed by an oxidation product of I". Singlet oxygen, superoxide, hydroxyl radicals and 13’ are not involved in initiation of lactoperoxidase-catalyzed lipid peroxidation. 8) The antimicrobial activity of the lactoperoxidase, H202, 1' system is not apparently mediated by the peroxidation of bacterial mem- brane lipids. BHT does not block lactoperoxidase-catalyzed bacterial killing. Other peroxide forming systems do not cause killing of bac- teria. 9) Lipoxygenase catalyzes the production of lipid hydroperoxides in y-linolenic micelles and in detergent—solubilized microsomes and liposomes. The reaction is dependent on pH and enzyme concentration. 10) Hydroperoxides formed by lipoxygenase undergo breakdown to produce malondialdehyde in microsomes during the incubation period. In liposomes, breakdown of membrane lipid hydroperoxides occurs only upon heating of the membranes with Fe+++ and is radical mediated. ll) Lipoxygenase-catalyzed hydroperoxide formation in detergent- solubilized microsomes supports the cytochrome P-450 mediated metabolism of aminopyrine. Endogenous microsomal lipid hydroperoxides are more effective in cytochrome P-450 mediated drug metabolishm than exogenous cumene hydroperoxide. 12) Recent evidence suggests that cytochrome P-450 in microsomes is the factor responsible for the radical mediated breakdown of lipid hydroperoxides. REFERENCES 10. 11. 12. 13. 14. 15. 16. REFERENCES May, H.E., and McCay, P.B. (1968) J. Biol. Chem., 243, 2288. Khon, H.I., and Liversedge, M. (1944) J. Pharmacol. Exp. Therap. 82, 292. Bernheim, F., Bernheim, M.L.C., and Wilbur, K.M. (1948) J. Biol. Chem., leL, 257. Tappel, A.L., and Zalkin, H. (1959) Arch. Biochem. Biophys., g9, 333. McCay, P.B. (1966) J. Biol. Chem., 241, 2333. Slater, T.F. (1972) in Free Radical Mechanisms in Tissue Injury, p. 4, Pior Limited, London, England. Uri, N. (1961) in Autoxidations and Antioxidants, Vol. I, (Lund- berg, W.0., ed.) p. 55, Interscience Publishers, New York, N.Y. Tappel, A.L. (1975) in Pathobiology of Cell Membranes, Vol. I, (Trump, B.F., and Arstila, A.U., eds.) p. 146, Academic Press, Inc., New York, NY Mead, J.F. (1976) in Free Radicals in Biology (Pryor, w.A., ed.) p. 51, Academic Press, Inc., New York, NY. Dahle, L.K., Hill, E.G., and Holman, R.T. (1962) Arch. Biochem. Biophys., g, 253. Pryor, W.A., and Stanley, J.P. (1975) J. Org. Chem., 49, 3615. Gardng;é H.W., Kleiman, R., and Neisleder, D. (1974) Lipids, 9, Frankel, E.N., Evans, C.D., Cowan, J.C. (1960) J. Am. Oil Chem. Soc., 31, 413. Evans, C.D., List, G.R., Dolev, A., McConnell, 0.6., and Hoffman, R.L. (1967) Lipids, 2, 432. Hamberg, M., and Gotthammar, B. (1973) Lipids, 8, 737. Bateman, L., and Hughes, H. (1952) J. Chem. Soc., 159 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 160 Walling, C., and Heaton, L. (1965) J. Am. Chem. Soc., 82, 48. Walling, C., ed. (1957) in Free Radicals in Solution, p. 397, John Wiley and Sons, Inc., London, England. Maier5 V.g., and Tappel, A.L. (1959) J. Am. Oil Chem. Soc., 6, . O'Brien, P.J. (1968) Can. J. Biochem., 47, 485. Maier§6V.Pé, and Tappel, A.L. (1959) J. Am. Oil Chem. Soc., Sheng, M.N., and Zajacek, J.G. (1970) J. Org. Chem., 35, 1839. Swern, 0., ed. (1970) in Organic Peroxides, John Wiley and Sons, New York, NY. Scott, 0., ed. (1965) in Atmospheric Oxidations and Antioxidants, Elsevier Publ. Co., Amsterdam, Netherlands. Bateman, L., Gee, 6., Morris, A.L. and Watson, W.F. (1951) Discuss. Faraday Soc., _l_(_)_, 250. Batema3é6L., and Morris, A. L. (1953) Trans. Faraday Soc., 49, Howard, J.A. (1973) in Free Radicals, (Kochi, J.K., ed.) Vol. II, p. 3, John Wiley and Sons, New York, NY. Reich, L., and Stivala, 5.5. (1969) in Autoxidation of Hydrocar- bons and Polyolefins, Dekker, New York, NY. Schultz, H.W., Day, E.A., and Sinnhuber, R.0. (1962) in Lipids and Their Oxidations, Avi Publ., West Port, Connecticut. Pryor, W.A. (1971) Chem. Eng. News, 49, 34. Barber, A.A., and Bernheim, F. (1967) Advan. Gerontol. Res., g, 355. Samuelsson, B. (1970) in Lipid Metabolism, (Wakil, S.J., ed.) p. 107, Academic Press, New York, NY. Hamberg, M., Samuelsson, 8., Bjorkhem, 1., and Danielsson, H. (1974) in Molecular Mechanism of Oxygen Activation, (Hayaishi, 0., ed.) p. 29, Academic Press, New York, NY. Franksé F243Gent’ M., and Roberts, 8. (1965) J. Appl. Chem., 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 161 Siakotas, A.N., and Koppang, N. (1973) Mech. Ageing Develop., 2, 177. Linnett, J.W., ed. (1964) The Electronic Structure of Molecules, Methuen, London, England. Ausloos, P. (1968) Fundamental Processes in Radiation Chemistry, John Wiley and Sons, New York, NY. Bacq, Z.M., and Alexander, P. (1966) Fundamentals of Radio- biology, 2nd ed., Pergamon, Oxford, England. Errera, M., and Forssberg, eds. (1961) Mechanisms in Radiobiology, Vols. I and II, Academic Press, New York, NY. Meyersé89.5. (1973) Fed. Proc., Fed. Amer. Soc. Exp. Biol., 92, Feinstein, R.N., ed. (1963) Implications of Organic Peroxides in Radiobiology, Radiat. Res., Suppl. 3, Academic Press, New York, NY. Phillips, 6.0., ed. (1968) Energetics and Mechanisms in Radiation Biology, Academic Press, New York, NY. Pryor, W.A. (1976) in Free Radicals in Biology (Pryor, W.A., ed.) p. 1, Academic Press, New York, NY. Turro, N.J. (1965) Molecular Photochemistry, Benjamin, New York, NY. Labout, J.J.M. (1972) Int. J. Radiat. Bi01., 24, 483. La Bella, F.S., and Paul, G. (1965) J. Gerontol., 29, 54. Hochschild, R. (1971) Exp. Gerontol., 9, 153. Schenck, 6.0., Decker, H.D., Schulte, K.H., and Krauch, C.H. (1963) Chem. Ber., 99, 509. Smith, L.L., and Hill, F.L. (1972) J. Chromatogr., 99, 101. Clements, A.H., Van den Engh, R.N., Frost, D.J., Hougenhout, K., and Nooi, J.R. (1973) J. Am. Oil Chem. Soc., 99, 325. Nickon, A., and Bagli, J.F. (1959) J. Am. Chem. Soc., 84, 6330. Gallo, U., and Santamaria, eds. (1972) Research Progress in Organic, Biological, and Medicinal Chemistry, Vol. III, Amer. Elsevier, New York, NY. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 162 Spikes; J.D., and MacKnight, M.L. (1970) Ann. N.Y. Acad. Sci., 1 1, 149. Spikes, J.D. (1967) Annu. Rev. Phys. Chem., 19, 409. Pryor, W.A. (1973) Fed. Proc., Fed. Amer. Soc. Exp. Biol., 32, 1862. '—_ Goddard, W.A., Dunning, T.H., Hunt, W.J., and Hay, P.J. (1973) Accounts Chem. Res., 9, 368. Goldstein, 8.0., Balchum, O.J., Demopoulos, H.B., and Duke, P.S. (1968) Arch. Environ. Health, 11, 46. Leighton, P.A. (1961) Photochemistry of Air Pollution, Academic Press, New York, NY. Schuck, E.A., Ford, H.W., and Stephens, E.R. (1958) Air Pollu- tion Effects of Irradiated Automobile Exhaust as Related to Fuel Composition, Rep. No. 26, Air Pollut. Found. San Marino, California. Chao, S.C., and Jaffee, S. (1972) J. Chem. Phys., 99, 1987. Roehm, J.N., Hadley, J.C., and Menzel, 0.8. (1971) Arch. Environ. Health, 29, 142. Roehm, J.N., Hadley, J.C., and Menzel, 0.8. (1971) Arch. Intern. Med., 129, 88. Estefan, R.M., Guase, E.N., and Rowlands, J.R. (1970) Environ. Res., 9, 62. Jonkman, L., Muller, H., Kiers, C., and Kommandeur, J. (1970) J. Phys. Chem., 24, 1650. Recknagel, R.O. (1967) Pharmacol. Rev., 19, 145. 0i Luzio, N.R. (1973) Fed. Proc., Fed. Amer. Soc. Exp. Biol., 52, 1875. Ugazio, G., Koch, R.R., and Recknagel, R.O. (1973) Exp. Nol. Pathol . , 12;, 281. 0i Luzio, N.R. (1963) Physiologist, 9, 169. Di Luzio, N.R. (1964) Life Sci., 9, 113. Di Luzio, N.R. and Costales, F. (1965) Exp. Molec. Path., 4, 141. Kalish, C.H., and Di Luzio, N.R. (1966) Science, 192, 1390. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 86. 87. 88. 89. 90. 163 Nordenbrand, K., Hochstein, P. and Ernster, L. (1964) Sixth International Congress of Biochemistry, Vol. 8, New York. Abramson, H. (1949) J. Biol. Chem., 119, 179. Barber, A.A. (1966) Lipids, 1, 146. Hochstein, P., and Ernster, L. (1963) Biochem. Biophys. Res. Commun., 12, 388. Yamazaki, I. (1971) Advan. Biophys., 2, 33. Fisher, H.F., Conn, E.E., Vennesland, 0., and Westheimer, F.H. (1953) J. Biol. Chem., 292, 687. Swallow, A.J. (1972) Isr. J. Chem., 19, 999. Land, E.J., and Swallow, A.J. (1968) Biochem. Biophys. Acta, 192, 327. Dolphin, D. and Felton, R.H. (1974) Accounts Chem. Res., 1, 26. Mahler, H.R., and Cordes, E.H. (1966) Biological Chemistry, Harper International Edition, p. 576. Land, E.J., and Swallow, A.J. (1969) Biochemistry, 9, 2117. Massey, V., and Gibson, Q.H. (1964) Fed. Proc., Fed. Am. Soc. Exp. Biol., Q, 18. Malkig,5 Rii3and Rabinowitz, J.C. (1967) A. Rev. Biochem., Hemmerich, P., Devartanian, D.V., Veeger, C., and Vorst, J.D.W. (1963) Biochem. Biophys. Acta, 11, 504. Pederson, T.C., and Aust, 3.0. (1972) Biochem. Biophys. Res. Comm., _4_8_, 789. Pederson, T.C., Buege, J.A., and Aust, 5.0. (1973) J. Biol. Chem., 249, 7134. Sugio|2