RAT LIVER MICROSOMAL ELECTRON TRANSPORT: MECHANISM OF NADPH-DEPENDENT UPID PEROXJDATION Dissmation for the Degree of Ph. D. MICHTGAN STATE UNIVERSITY THOMAS C. PEDERSON 1973 LI B R A R Y Michigan Sta 1*..- University .; .w___J g -* alumna BY ‘J ms 3. 8038’ JJtw 9mm! we. , “BT- ..v’.f BINDERS ‘I saw. : {-f- IICEEiiAj m _ —-‘ \‘ ABSTRACT RAT LIVER MICROSOMAL ELECTRON TRANSPORT: MECHANISM OF NADPH-DEPENDENT LIPID PEROXIDATION BY Thomas C. Pederson NADPH-cytochrome c reductase in rat liver microsomes was solubilized by bromelain digestion and purified to homogeneity. An antibody preparation obtained by immuniza- tion with this enzyme was found to inhibit the NADPH- cytochrome c reductase activity of both the purified enzyme and intact microsomes by more than 90%. This antibody also inhibited the NADPH-dependent peroxidation of microsomal lipid which occurs in the presence of ferric ion chelated by ADP (ADP-Fe). The bromelainrsolubilized reductase will also promote NADPH-dependent peroxidation of extracted nucrosomal lipid if the reaction mixture includes ferric ion chelated by ethylenediaminetetraacetate (EDTA—Fe)- These results demonstrate that the NADPH-dependent Peroxidation of microsomal lipid involves the activity of the microsomal flavoprotein, NADPH-cytochrome c reductase: and also show that a similar peroxidation reaction can be Emomoted in a model system containing extracted microsomal lipid and the purified reductase. The conditions required fbr Optimal NADPH-dependent peroxidation in the model “x r! b 7 33» Thomas C. Pederson system include high ionic strength (0.4 M NaCl) and 0.1 mM EDTA-Fe, which are very similar to the conditions required for optimal reduction of EDTA-Fe. A purified preparation of the microsomal enzyme NADH-cytochrome b 5 reductase, which reduced EDTA-Fe, will also promote the peroxidation of extracted microsomal lipid. Intact microsomes, in the presence of ADP-Fe, are specific for NADPH instead of NADH in promoting the peroxidation of microsomal lipid. However, in the presence of both EDTA-Fe and ADP-Fe, NADH will also promote lipid peroxidation. These results suggest that the NADPH-dependent peroxidation, promoted in the presence of only ADP-Fe, involves an additional electron transport component. This component is specifically reduced by NADPH-cytochrome c reductase, and its function in the peroxidation reaction can be replaced by EDTA-Fe. In aqueous solutions, at neutral pH, ferrous ions are raPidly oxidized by molecular oxygen which may involve the generation of a number of reactive intermediates including the superoxide anion (02-) , H202, and the hydroxyl radical. In the presence of Fe3+ and complexing anions, the peroxi- dation of unsaturated liver microsomal lipid in bOth intact microsomes and in a model system containing extracted microsomal lipid can be promoted by either NADPH and NADPH-cytochrome c reductase, or by 02— generated by xanthine and xanthine oxidase. The presence 0f erYthrocuprein effectively inhibits the activity PromOted by xanthine and xanthine oxidase but produces much less Thomas C. Pederson inhibition of NADPH-dependent peroxidation. The singlet oxygen trapping agent, l,3-diphenylisobenzofuran, had no effect on NADPH-dependent peroxidation but strongly inhibited the peroxidation promoted by 02-. Furthermore, it was found that the generation of H202 in the presence of Fe3+, complexing anions, and extractable microsomal lipid promoted no peroxidation activity. NADPH-dependent lipid peroxidation was also shown to be unaffected by hydroxyl radical scavengers. The addition of catalase was found to have no effect on NADPH-dependent lipid peroxida- tion, but it significantly increased the rate of nalondialdehyde formation in the reaction promoted by xanthine and xanthine oxidase. These results demonstrate that NADPH-dependent lipid peroxidation is promoted by a reaction mechanism which does not involve either 02-, Singlet oxygen, H202, or the hydroxyl radical. The peroxi- dation activity promoted by xanthine and xanthine oxidase involves 02- and the generation of singlet oxygen, Emesumably required to initiate the formation 0f hydroperoxides. It is concluded that NADPH-dependent lipid Peroxidation is promoted by the reduction of Fe3+ and also reQuires a mechanism for initiating the formation 0f hYdr0peroxides which likely involves some form of the Perferryl ion, [Fe02]2+. The rate of peroxidation was also found to be markedly influenced by the presence of the complexing anions, EDTA and ADP. Both NADPH-dependent peroxidation and the Thomas C. Pederson peroxidation promoted by 02- require the presence of complexing anions whereas ascorbic acid readily promoted lipid peroxidation in the presence of only FeCl3 and Tris- HCl buffer. NADPH-dependent peroxidation of extracted microsomal lipid promoted by a purified preparation of the reductase specifically requires the presence of both Fe3+ chelated by EDTA and unchelated Fe3+, but is unaffected by the presence of ADP, whereas NADPH-dependent peroxidation in intact microsomes requires the presence of ferric ion complexed by ADP. The complexing anions are apparently required for the reduction of Fe3+ and to prevent the precipitation of Fe(OH)3 or the binding of Fe3+ by other components of the reaction mixture. The formation of [Fe02]+ or other iron-containing species involved in the peroxidation reaction may also require specific complexing anions. RAT LIVER MICROSOMAL ELECTRON TRANSPORT: MECHANISM OF NADPH-DEPENDENT LIPID PEROXIDATION BY 6 Thomas C? Pederson A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Biochemistry 1973 To My Parents ii ACKNOWLEDGMENTS I would like to express my appreciation to Dr. Steven D. Aust for his many contributions as my academic advisor. I not only appreciate his guidance concerning my graduate work, but also his interest and discussions concerning numerous other subjects of mutual interest. I also wish to express my thanks to the other members of my guidance committee, Drs. Loran L. Bieber, Robert M. Cook, Clarence H. Suelter, and Willis A. Wood, and to Dr. Ahsan U. Khan for their helpful discussions. I would also like to thank Dr. Minor J. Coon of the University of Michigan for providing preparations of cytochrome Puso and detergent- solubilized NADPH-cytochrome Puso reductase; and Dr. Tokuji Kimura of Wayne State University for providing a purified preparation of adrenodoxin. Finally, I wish to acknowledge the help and friendship I received from all the people associated with the Department of Biochemistry. iii TABLE OF CONTENTS Page INTRODUCTION. . . . . . . . . . . . . . . . . . . . . 1 LITERATURE REVIEW . . . . . . . . . . . . . . . . . . 4 MATERIALS AND METHODS . . . . . . . . . . . . . . . . 22 Materials Sources. . . . . . . . . . . . . . . . . 22 Preparation of Microsomes. . . . . . . . . . . . . 23 Microsomal Lipid Peroxidation Assay. . . . . . . . 24 Measurement of Malondialdehyde by the TBA Reaction . . . . . . . . . . . . . . . . . . . . 25 Microsomal Drug Metabolism Assay . . . . . . . . . 26 Assay of Cytochrome c Reductase Activity . . . . . 29 Solubilization and Purification of NADPH- Cytochrome c Reductase . . . . . . . . . . . . . 29 Preparation of Antibody to NADPH- Cytochrome c Reductase . . . . . . . . . . . . . 3O Solubilization and Purification of NADH- Cytochrome bs Reductase. . . . . . . . . . . . . 31 Solubilization and Purification of Cytochrome b5 0 o o O o o o o o o o o o o o o o o o o o o o 31 Extraction of Microsomal Lipid . . . . . . . . . . 32 Preparation of Liposomes . . . . . . . . . . . . . 33 Peroxidation of Extracted Microsomal Lipid . . . . 38 Erythrocyte Hemolysis. . . . . . . . . . . . . . . 39 Purification of Erythrocuprein . . . . . . . . . . 40 iv Assay of 02- Production . . . . . . . . . . . Addition of l,3-Diphenylisobenzofuran to Aqueous Reaction Mixtures . . . . . . . . . Identification of o-Dibenzoylbenzene in Reaction Mixtures Containing 1,3- Diphenylisobenzofuran . . . . . . . . . . . Other Methods . . . . . . . . . . . . . . . . RESULTS AND DISCUSSION . . . . . . . . . . . . . CHAPTER I. THE INVOLVEMENT OF NADPH-CYTOCHROME C REDUCTASE IN LIVER MICROSOMAL NADPH- DEPENDENT LIPID PEROXIDATION. . . . . . . . . Purification of NADPH-Cytochrome c Reductase O O O C O O O O O C O O O C O O O Involvement of NADPH-Cytochrome c Reductase in Microsomal Lipid Peroxidation. The Relationship Between Microsomal Lipid Peroxidation and Drug Hydroxylation . CHAPTER II. NADPH-DEPENDENT LIPID PEROXIDATION IN A MODEL SYSTEM: EVIDENCE FOR AN ADDITIONAL MICROSOMAL ELECTRON TRANSPORT COMPONENT . . . Peroxidation of Extracted Microsomal Lipid by Purified NADPH-Cytochrome c Reductase. . The Use of Erythrocyte Hemolysis to Detect Reactive Radical Intermediates. . . . . . . + . . The Two Forms of Fe3 Required in the Model System. . . . . . . . . . . . . . . . The Role of EDTA-Fe in Lipid Peroxidation . . Evidence for an Additional Microsomal Electron Transport Component. . . . . . . . CHAPTER III. THE MECHANISMS INVOLVED IN NADPH- DEPENDENT LIPID PEROXIDATION. . . . . . . . . The Involvement of H202 in NADPH-Dependent Lipid Peroxidation. . . . . . . . . . . . . Page 41 41 42 43 44 44 47 59 67 74 76 87 91 100 105 117 122 Page Lipid Peroxidation Promoted by Xanthine and Xanthine Oxidase. . . . . . . . . . . . . . 126 The Generation of 02— by Microsomes and NADPH-Cytochrome c Reductase. . . . . . . . . . 137 The Involvement of 02- and Singlet 02 in NADPH-Dependent Lipid Peroxidation. . . . . . . 144 The Role of Complexing Anions . . . . . . . . . . 150 Does the Perferryl Ion Promote Peroxidation? . . 153 SUMMARY. . . . . . . . . . . . . . . . . . . . . . . 159 LIST OF REFERENCES . . . . . . . . . . . . . . . . . 162 APPENDIX . . . . . . . . . . . . . . . . . . . . . . 178 vi TABLE II III IV VI VII VIII IX XI LIST OF TABLES Solubilization of NADPH-Dependent Lipid Peroxidation from Liver Microsomes by Incubation in the Presence of Proteases and Lipases. . . . . . . . . . . . . . . . . The Effect of Antibody to NADPH-Cytochrome c Reductase on Microsomal NADH-Cytochrome c Reductase Activity . . . . . . . . . . . . The Effect of Carbon Monoxide on the Inhibition of Microsomal Lipd Peroxidation by 3,4-Benzpyrene. . . . . . . . . . . . . . Inhibition by 3,4-Benzpyrene of Ascorbate- Promoted Lipid Peroxidation in the Presence Of NADPH O O O O O O O C O I I O O O O O O 0 Requirements for Peroxidation of Lipid Catalyzed by Purified NADPH-Cytochrome c Reductase. O O O O O O O O O O O C O O O O O The Effect of EDTA-Fe and High Ionic Strength on NADPH-Dependent Lipid Peroxidation in Intact Microsomes. . . . . . Erythrocyte Hemolysis Promoted by the NADPH- Dependent Peroxidation of Microsomal Lipid . Requirement for Microsomal Lipid in the Erythrocyte Hemolysis Reaction Catalyzed by NADPH-Cytochrome c Reductase. . . . . . . The Requirement for Ferric Ion not Chelated by EDTA 0 O O I O O O O O O O O O O NADPH-Dependent Lipid Peroxidation in the Model System Using a Detergent-Solubilized Preparation of the Reductase . . . . . . . . Lipid Peroxidation Catalyzed by Purified NADPH-Cytochrome c Reductase with Other Ferric Ion Chelators used in Place of EDTA . vii Page 46 63 71 72 80 90 92 93 99 101 103 TABLE Page XII NADH-Dependent Lipid Peroxidation in the Presence of EDTA-Fe. . . . . . . . . . . . . . 104 XIII Comparison of Inhibition by Antibody of Lipid Peroxidation Promoted by Either Purified Reductase or Intact Microsomes. . . . 107 XIV Comparison of Inhibition by Antibody of Activity of NADPH-Cytochrome c Reductase with Either Cytochrome c or Ferricyanide as Terminal Acceptor . . . . . . . . . . . . . 108 XV NADPH-Dependent Lipid Peroxidation in the Model System Using Partially Purified Preparations of Cytochrome Puso. . . . . . . . 110 XVI NADPH-Dependent Lipid Peroxidation in the Model System Using a Purified Preparation of Adrenodoxin . . . . . . . . . . . . . . . . 113 XVII Lipid Peroxidation in NEM-Treated Microsomes . . . . . . . . . . . . . . . . . . 115 XVIII Inhibition of NADPH-Dependent Lipid Peroxidation by Commercial Preparations of Erythrocuprein and Catalase . . . . . . . . 121 XIX Peroxidation of Extracted Microsomal Lipid in the Presence of a H202 Generating systen‘ O O O O O O C O O C C O O O O O I O O O 123 XX Effect of Catalase and Hydroxyl Radical Scavengers on NADPH-Dependent Peroxidation of Extracted Microsomal Lipid. . . . . . . . . 125 XXI Lipid Peroxidation Promoted by Xanthine Oxidase. . . . . . . . . . . . . . . . . . . . 128 XXII The Effect of Erythrocuprein on Lipid Peroxidation Promoted by 02 . . . . . . . . . 129 XXIII The Effect of Catalase on Lipid Peroxidation Promoted by 02 . . . . . . . . . . . . . . . . 131 XXIV The effect of 1,3-Dipheny1isobenzgfuran on Lipid Peroxidation Promoted by 02 . . . . . . 133 XXV The Effect of l,3-Dipheny1isobenzofuran on Superoxide Production by Xanthine Oxidase as Measured by the Rate of Cytochrome c Reductase. . . . . . . . . . . . . . . . . . . 134 viii TABLE Page XXVI The Effect of Erythrocuprein on Lipid Peroxidation . . . . . . . . . . . . . . . . . 145 XXVII The Effect of 1,3-Dipheny1isobenzofuran on Lipid Peroxidation. . . . . . . . . . . . . 147 XXVIII Lipid Peroxidation in Intact Microsomes Promoted by Xanthine and Xanthine Oxidase. . . 149 XXIX The Effect of ADP-Fe and EDTA-Fe on Lipid Peroxidation . . . . . . . . . . . . . . 151 ix Figure 10 11 12 LIST OF FIGURES Page Standard Curve for Measurement of Malondialdehyde by the TBA Reaction . . . . . 28 Technique for Removal of Solvents from Microsomal Lipid Extracts Using a Bfichi Rotary Evaporator . . . . . . . . . . . 35 Technique for Preparing Liposomes by Sonication under Anaerobic Conditions . . . . 37 Sephadex G-100 Column Chromatography. . . . . 50 DEAE-Cellulose Affinity Chromatography. . . . 52 Sodium Dodecyl Sulfate Polyacrylamide Disc Gel Electrophoresis of Purified NADPH-Cytochrome c Reductase. . . . . . . . . 55 Absorption Spectrum of Purified NADPH- Cytochrome c Reductase. . . . . . . . . . . . 58 Inhibition of NADPH-Cytochrome c Reductase Activity by Antibody to the Enzyme. . . . . . 62 Inhibition of Microsomal NADPH-Dependent Lipid Peroxidation and Aminopyrine Demethylation of Antibody to NADPH- Cytochrome c Reductase. . . . . . . . . . . . 66 Inhibition by 3,4-Benzpyrene of Microsomal Lipid Peroxidation Promoted by Either NADPH or Ascorbate. . . . . . . . . . . . . . 70 Time Course of Lipid Peroxidation Promoted by Purified NADPH-Cytochrome c Reductase. . . 78 NADPH-Dependent Lipid Peroxidation in the Model System as a Function of Enzyme Concentration . . . . . . . . . . . . . . . . 82 Figure 13 14 15 16 17 18 19 20 21 NADPH-Dependent Model System as Concentration . NADPH-Dependent Model System as NADPH-Dependent Model System as Concentration . NADPH-Dependent Model System as Concentration . NADPH-Dependent Model System as Concentration . Lipid Peroxidation in the a Function of the Lipid Lipid Peroxidation in the a Function of pH. . . . . Lipid Peroxidation in the a Function of Salt Lipid Peroxidation in the a Function of the EDTA Lipid Peroxidation in the a Function of the EDTA-Fe A Tracing of the Thin-Layer Chromatogram Demonstrating the Formation of o-Dibenzoylbenzene in Peroxidation Reaction Mixtures . . . . . . . . . . . . Inhibition of Epinephrine Oxidation by Erythrocuprein. Inhibition of Epinephrine-Stimulated NADPH Oxidation by Erythrocuprein . . . . Reaction Mechanisms Involved in the Peroxidation of Microsomal Lipid Promoted by Either Xanthine and Xanthine Oxidase or NADPH and NADPH-Cytochrome c Reductase xi Page 84 86 89 95 98 136 140 142 156 ADP-Fe DBB DEAE- DPIF EDTA EDTA-Fe FAD NADH NADP NADPH NEM TBA Tris ABBREVIATIONS adenosine-5'-diphosphate ferric ion chelated by a 17-fold molar excess of ADP o-dibenzoylbenzene diethylaminoethyl- l,3-diphenylisobenzofuran ethylenediaminetetraacetate ferric ion chelated by 1 mole of EDTA flavin adenine dinucleotide reduced nicotinamide adenine dinucleotide nicotinamide adenine dinucleotide phosphate reduced nicotinamide adenine dinucleotide phosphate N-ethylmaleimide thiobarbituric acid Tris(hydroxymethyl)aminomethane xii INTRODUCTION The objective of this thesis is to present a discussion, based on experimental results, of the mechanism by which NADPH oxidation in liver microsomes promotes the peroxidative destruction of endogenous unsaturated lipid. The first two chapters are concerned primarily with determining which electron transport components of the microsome are involved in promoting the NADPH-dependent peroxidation activity. My interest in this subject grew out of my earlier studies of microsomal drug metabolism promoted by the microsomal NADPH-dependent electron transport system and catalyzed by cytochrome Pu5°.* My original intent was to specifically investigate microsomal electron transport and determine the relation- ships between the various microsomal activities promoted by the oxidation of reduced pyridine nucleotides. NADPH- dependent peroxidation in microsomes is not catalyzed by cytochrome Pkso, but it has been commonly believed that it is closely associated with the electron-transport compo- nents which promote drug metabolism. As described in *See: Pederson, T. C. (1969) Masters Thesis, Michigan State University, East Lansing, Michigan; and Pederson, T. C., and Aust, S. D. (1970) Biochem. Pharmacol. 12, 2221. ‘ 1" .71' "a. .-n . ,.uu Nil uvc 'U‘I Is. in. Chapter I of this thesis, my early experiments demonstrated that the primary experimental evidence on which the belief in this association was based was misleading and had been erroniously interpreted. Consequently, I became primarily concerned with determining the reaction mechanism and microsomal components involved in the peroxidation activity. There is a growing body of evidence which suggests that lipid peroxidation does occur in yivg, particularly under certain stress situations. It is known that many diseases involve peroxidative damage to tissues, but it is often not known whether the peroxidation is a cause or a conse- quence of the initial develOpment of the disease. The process of aging is also thought to involve accumulative damage to cellular components caused by the free radicals generated during the peroxidation of unsaturated lipid. As a consequence, there is increasing interest in the mechanisms by which lipid peroxidation is either promoted or prevented in_yiyg. During the course of my research, I have become very interested in the mechanisms by which the formation of lipid peroxides is initiated in biological systems. Since molecular oxygen is in a triplet state, spin barriers forbid a direct reaction between oxygen and unsaturated lipid to form hydroperoxides, and although oxygen contains unpaired electrons, it will not initiate radical chain oxidation. Therefore, either the lipid or molecular oxygen must be activated to initiate the formation of M!" "1:85 '3'...“ .gnf’nr; n- 0' n M' 5 . A :0 F 'e~ "on ' ”I . . I -.p on“ L... poi v .“-'V: y ”to. u A .u. i“ ' .4" tin < .: v. '- I-‘~. I . a. .1 F I. . peroxides. It appears that a common property of oxygen incorporating enzymes is a catalytic site which avoids the spin restriction by activating oxygen prior to its reaction with the intended substrate. This activation frequently occurs by interaction of oxygen with a transition metal ion in a reduced state to form a reactive complex. Oxygen can also be activated by excitation to one of the excited singlet states or by reduction to produce reactive species such as 02- or the hydroxyl radical. In the final chapter of this thesis, I will present evidence which suggests that NADPH-dependent lipid peroxidation does involve activation of molecular oxygen and demonstrate that the peroxidation of microsomal lipid is promoted by the generation of other activated states of oxygen. LITERATURE REVIEW Unsaturated fatty acids present in lipids exposed to atmospheric oxygen can be readily destroyed by autoxida- tion. It has long been recognized that this process is a peroxidative reaction promoted primarily by a free radical chain oxidation (1-4). This reaction is commonly described by the following simplified scheme. Initiation: RH 4%—R- + (H-) Propagation: R- + 02 -> R02- R02- + RH + ROOH + R- Termination: (R-, RO', or ROO') + (R-, RO-, or ROO’) + inactive product. The reaction is catalyzed by trace amounts of transi- tion metal ions (5-7) which promote the formation of radicals by the following reactions: ROOH + Mn+ + R0- + 0H" + M(n I 1)+ (n + 1)+ ROOH + M + ROO- + 3+ + Mn+ Once autoxidation begins in the presence of trace metal ions, the reaction is autocatalytic; but the presence of substances referred to as antioxidants, which react with the radicals to terminate the chain reaction, decrease the amount of oxidation which takes place (8-10). One of the well known characteristics of lipid autoxidation is the formation of a product which reacts with thiobarbituric acid, TBA, to form a red colored compound (11). Lipids from a variety of tissues all pro- duced a similar TBA-reactive product during their oxidation (12). It was soon demonstrated that the TBA-reactive product appeared to be identical with malondialdehyde which formed a compound with TBA having an extinction of 1.5 x 105 M"1 cm”1 at 535 nm (13). HEN; N OH H+ HMH 2 N I SIWH OH Since nanomolar quantities of malondialdehyde can be easily and accurately determined, the TBA assay has long been the most common method for determining the extent of lipid peroxidation occurring in biological systems. The mechanism by which malondialdehyde is formed is not fully understood. Studies of fatty acid autoxidation bur Dahle gt $1., (14) demonstrated that only fatty acids With three or more methylene-interrupted olefin bonds Produce any malondialdehyde while undergoing oxidation. They concluded that malondialdehyde formation requires formation of a fatty acid hydroperoxide radical which can subsequently attack an olefin bond at the 8 position to form a five membered endoperoxide. The decomposition of the endOperoxide may then result in malondialdehyde formation. The role of the five membered endoperoxide I I if [T .1 I i in the formation of malondialdehyde has been supported by the observed production of malondialdehyde by the enzymatic system which catalyzes prostaglandin biosynthesis via the formation of a similar 5 membered endoperoxide (15-17). The observed incorporation of tritium, from arachidonate labeled on all the olefinic carbons, into carbons l and 3 cm malondialdehyde is also in agreement with the proposed mechanism (18). Polyunsaturated fatty acid moieties are a ubiquitious component in biological lipids, particularly in the Eflnospholipids located in membranes (19, 20). These lipids ‘Viall readily form peroxides ig_yit£g, particularly when 'tdlee tissue is disrupted by techniques such as homogenization (21-23). In addition to destroying unsaturated fatty acids, lipid peroxidation also alters other constituents in the biological sample. The destruc- tion of sulfhydryl groups is one of the first consequences of lipid peroxidation (24). Other components, including proteins and nucleic acids, suffer extensive damage which has been shown to be caused primarily by the free radical intermediates generated during the peroxidation reaction (25). Hematin compounds accelerate lipid peroxidation and are destroyed in the process (26). Hemoproteins promote the most rapid rate of peroxidation when they are partially denatured, exposing the heme group (27). The presence of inorganic iron and a reducing agent, such as ascorbate or cysteine, also increases the rate of peroxidation (7, 28). As a consequence of lipid peroxidation occurring in biological preparations, alterations in the activity of many enzymes and in the structural integrity of cellular and subcellular organelles will occur. A well known example of such peroxidative damage is the swelling and lysis of mitochondria in the presence of ascorbic acid and ferrous ion which occurs as a result of lipid peroxidation (29-31). In liver homogenates, a considerable portion of the lipid subject to peroxidative oxidation is found in the Microsomal membranes. The microsomal fraction is composed <3fi membrane vesicles, both with and without bound ribosomes, ‘VIIiLch are derived from the endoplasmic reticulum of the liver cell (32, 33). Ottolenghi (34) showed that lipid peroxidation occurs in the microsomal fraction just as readily as in the mitochondrial fraction when incubated in the presence of ascorbate or ferrous salts. In 1963, Hochstein and Ernster (35) reported that when rat liver microsomes were incubated in the presence of ADP, the oxidation of NADPH promoted rapid peroxidation of endogenous lipid as evidenced by the formation of malondialdehyde. Beloff-Chain 33 21., (36) reported simultaneously that ADP stimulated both NADPH oxidation and oxygen uptake by liver microsomes. Both groups soon discovered that the ADP- stimulated activity also required the presence of ferric ion which had previously been present as a contaminant in the ADP (37, 38). In addition to ADP, pyrophosphate or any of the nucleoside di- and triphosphates would stimulate NADPH-dependent peroxidation in the presence of ferric ion, but chelating agents such as EDTA or o-phenanthroline inhibited peroxidation (39). Wills (40, 41) later reported that liver microsomal NADPH-dependent lipid peroxidation was promoted in the absence of ferric ion and ADP or similar perphosphates, however, the peroxidation activity was increased considerably by the presence of phosphate and still more by pyrophosphates containing trace levels 0f ferric ion. Poyer and McCay (42) have recently demon- strated that NADPH-dependent microsomal lipid peroxidation has an absolute requirement for ferric ion which can only be demonstrated using microsomes which have been washed with Chelex-treated buffer. Liver microsomes contain two electron transport systems. One is linked to the oxidation of NADPH (43, 44) and the other to the oxidation of NADH (45, 46). The NADPH-dependent electron transport system is associated with cytochrome P“50 which catalyzes the mixed-function oxidase activity responsible for the hydroxylation of a large variety of drugs, steroids, carcinogens, and other lipid soluble compounds (47, 48). The lipid peroxidation promoted in the presence of ADP and ferric ion has a specific requirement for NADPH; NADH promotes little or no peroxidation activity (35, 40). The oxidations catalyzed by cytochrome Puso are characterized by the inhibition observed in the presence of carbon monoxide (49). NADPH- dependent lipid peroxidation is not inhibited by carbon monoxide (39); therefore, cytochrome Puso is apparently not involved in the peroxidation activity. However, it was observed that the presence of drug metabolism substrates undergoing hydroxylation inhibited NADPH-dependent lipid peroxidation (50-52). With some of the substrates, it was shown that the inhibition of lipid peroxidation could be partially reversed by the presence of carbon monoxide (39). It was therefore concluded that lipid peroxidation and drug hYdroxylation compete for reducing equivalents from some <3Cummon electron transport component. The microsomal flavoprotein, NADPH-cytochrome c reductase, was originally 10 believed to be the reductase responsible for catalyzing the reduction of cytochrome Puso, because its activity increases in parallel with the drug hydroxylation activity in animals treated with phenobarbital (53). It was therefore suggested that this flavoprotein may also be responsible for promoting NADPH-dependent lipid peroxida- tion (39, 42, 46). Orrenius, Berg, and Ernster (54) observed that when microsomes were incubated with trypsin, both drug metabolism and lipid peroxidation activities were lost in parallel with the solubilization of the NADPH-cytochrome c reductase activity from the microsomal membrane. In more recent work, the role of NADPH-cytochrome c reductase in the drug hydroxylation activity has been established by demonstrating that a reconstituted drug hydroxylation system requires a fraction containing this flavoprotein (55), and antibody to a purified preparation of the flavoprotein inhibits both the reduction of cytochrome Paso and the hydroxylation of drugs in intact microsomes (56-59). McCay and co-workers (60-63) have demonstrated that during NADPH-dependent lipid peroxidation, the long chain polyunsaturated fatty acids, primarily arachidonic and docosahexaenoic acids, disappear from the phospholipids in liver microsomes. The altered phospholipids were more ENDIar but still subject to the action of phospholipase A, Slilggesting that a semialdehyde remains bound to the IPIlcospholipid molecule following oxidative cleavage of the 11 carbon chain. Numerous degradation products were produced including malondialdehyde (42). Since there was some controversy as to whether malondialdehyde was the only thiobarbituric acid-reactive product produced during the oxidation of unsaturated fatty acids (64-66), Niehaus and Samuelsson (18) demonstrated that the only thiobarbituric acid-reactive product formed during microsomal NADPH- dependent lipid peroxidation was malondialdehyde since it also condensed with urea to form 2-hydroxypyrimidine (67). It had been previously shown that malondialdehyde was only a minor product of the peroxidation of unsaturated fatty acids (14). Ernster and Nordenbrand (39) found that the stoichiometry of NADPH—dependent lipid peroxidation involved the oxidation of NADPH, the uptake of oxygen, and the formation of malondialdehyde in a molar ratio of approximately 6:23:l. Wills (40) also found that the formation of malondialdehyde accounted for only about 5% of the oxygen uptake. May and McCay (62) found that during the initial stage of the reaction, the ratio between fatty acid loss and oxygen consumption was about 1:1, but the final ratio was about 1:4; however, the ratio of oxygen uptake to double bonds lost remained at 1:1 throughout the reaction. They also found that malondialdehyde accounted for only about 12% of the fatty acid loss or 3% of the Oxygen uptake . NADPH-dependent peroxidation of microsomal lipid has been shown to be readily inhibited by a variety of free 12 radical trapping agents (35, 62, 68), suggesting that the microsomal peroxidation activity involves a radical chain oxidation. Tam and McCay (63) have also shown that there is a transient formation of lipid hydroperoxides which are rapidly decomposed producing a variety of carbonyl com- pounds. The formation of a reactive intermediate with radical-like properties has been indicated by the rapid lysis of either erythrocytes or lysosomes included in the microsomal peroxidation reaction mixture (68, 69). It has also been demonstrated by Nilsson et al,, (70, 71) that NADPH oxidation in liver microsomes generates a reactive intermediate which promotes the chemiluminescent oxidation of luminol. The chemiluminescence was greatly increased by the addition of ferric ion and pyrophosphate suggesting that the reactive intermediate may also be involved in NADPH-dependent lipid peroxidation. NADPH-dependent lipid peroxidation causes a number of structural and functional changes in the liver microsome. Ernster and co-workers (39, 72) observed a decrease in the enzyme activities of glucose-G-phosphatase and NADH- cytochrome c reductase as well as lysis of the microsome and release of microsomal proteins. The changes occurring in the microsomal membrane during the peroxidation reaction have been further studied by a number of investigators (63, 73-76) who concluded that the degradative effects involve both alteration of enzymes and other microsomal components by the reactive radical intermediates and physical changes 13 in the membrane structure as a result of the conversion of phospholipids to more polar species. Wills (52) observed that drug hydroxylation activity was markedly reduced as a result of the peroxidation activity. Hycray and O'Brien (77, 78) subsequently showed that cytochrome Pl.50 reduces peroxides, including lipid peroxides, and is destroyed in the process. The decrease in the cytochrome Pu5° content of liver microsomes as a result of lipid peroxidation pro- moted by either NADPH oxidation or other methods has been described by other investigators (76, 79-81), and occurs by a mechanism which destroys the heme, producing carbon monoxide and a green discoloration of the microsomes. The extent to which lipid peroxidation occurs i2.¥£!2 is not well understood and remains a subject of controversy. However, it is quite evident that biological systems have several protective mechanisms to prevent the peroxidation of endogenous unsaturated lipid from occurring. One of these mechanisms is afforded by the presence of vitamin E (o-tocopherol), a lipid soluble antioxidant capable of scavenging free radicals and terminating radical-promoted chain oxidations (8-10, 82-86). It has been shown that animals deficient in vitamin E contain evidence of i2 gizg peroxide formation, and the tissues from these animals are subject to much more rapid peroxidative damage in_yitgg (87-89). It was also found that mitochondrial functions were impaired in the livers of vitamin E deficient animals (22). The dietary requirement for vitamin E has also been 14 shown to increase when the level of polyunsaturated fatty acids in the diet is increased (90). There have been numerous other studies which indicate that vitamin E is required as an antioxidant, particularly in times of stress, but there are also a considerable number of investigators who believe that this is not the only function of vitamin E (91). One of the trace metals known to be an important dietary requirement is selenium, which when deficient produces effects similar to those seen in vitamin E deficient animals, suggesting that selenium (or some biological component containing selenium) must function as an antioxidant (92-94). There is a large volume of literature concerning the antioxidant properties of a variety of seleno compounds; however, it has recently been demonstrated that glutathione peroxidase, obtained from erythrocytes, is a selenoenzyme containing 4 gram atoms of selenium per mole of native tetrameric enzyme (95, 96). It had previously been shown that glutathione peroxidase, in the presence of reduced glutathione, will catalyze the reduction of fatty acid hydroperoxides (97-99). The role of glutathione peroxidase appears to be particularly important in the erythrocyte. It had been shown that pro- tection against oxidative damage and lysis of erythrocytes induced by hyperbaric oxygen or hydrogen peroxide is afforded by both the antioxidant, vitamin E, and the presence of glucose or reduced glutathione (100, 101). It 15 was consequently found that erythrocytes from selenium deficient animals were more sensitive to oxidative damage, practically devoid of glutathione peroxidase activity, and were no longer protected by the presence of added glutathione (95). The amount of glutathione peroxidase is increased in some tissues when subjected to peroxidative stress. It was found that the lungs of animals exposed to an atmosphere containing ozone showed evidence of peroxida- tive damage and elevated activities of glutathione peroxidase, glutathione reductase, and glucose-6-phosphate dehydrogenase (102). It was also observed that these enzyme activities were increased in adipose tissue and muscle, but not liver, lung, and kidney of animals fed a vitamin E-free diet (103). The most characteristic evidence for the occurrance of in viyg lipid peroxidation is the accumulation of ceroid and lipofusion pigments which are complexes of lipid and protein substances with characteristic fluorescence spectra (104). These pigments accumulate in all animal tissues, particularly adipose fat, heart, and nervous tissue as a function of age (105-107). Tappel and co- workers have shown that similar pigments are formed during lipid peroxidation in_!i££g (108) and appear similar to the Shiff-base products formed from the reaction of malondialdehyde with the amino groups of protein, the bases of nucleic acids, and the ethanolamine and serine residues of phospholipids (109, 110). The importance of .— .,..x v. (I' O o. . a o...' i... ‘la - ”up . . H ‘v f l6 lipid peroxidation in the aging process has also been indicated by studies showing that the lifespan of rats fed diets high in antioxidant content is increased (111). One of the early recognized effects of a vitamin E deficient diet was the degeneration of muscle tissue (112). The creatinuria and degeneration of muscle fibers are similar to the processes which occur in the inherited disease, myotonic muscular dystrOphy, and the vitamin E deficient effect has consequently been referred to as nutritional muscular dystrophy (113, 114). The muscles in the vitamin E deficient animals shown evidence of lipid peroxidation and release of lysosomal enzymes (115, 116), suggesting that the dystrophy may be the direct result of peroxidation of unsaturated lipids. However, it was observed that methionine was even more effective than vitamin E in promoting recovery from dystrophy even though it did not prevent lipid peroxidation from occurring (117). In addition, inherited muscular dystrophy has been shown to be unaffected by treatment with vitamins, including vitamin E (118). It now appears that myotonic dystrophy in both man and in laboratory animals is related to a fundamental defect in some aspect of membrane structure, because the permeability and functions of erythrocyte membranes and liver mitochondria, as well as the sarcolemma of skeletal muscle, are abnormal (119-121). The erthro- cytes from genetically dystrophic mice were found to have some surface alterations which causes irregular protrusions 17 from the cell surface (122). A similar alteration in the erythrocyte cell surface was found in rats with nutritional muscular dystrophy produced by a vitamin E deficient diet. Furthermore, the protein kinase activity in erythrocyte membranes from dystrophic patients, but not other patients, was inactivated during storage of the membrane (123), suggesting that peroxidative degradation of unsaturated lipid may be involved in the genetic disease. In animals deficient in vitamin E, one of the most severe effects may be the necrotic destruction of the liver. Schwarz and co-workers (124, 125) found that in addition to vitamin E, the dietary levels of selenium and of sulfur-containing amino acids were very important protective factors in the development of liver necrosis. They also found that even before gross macroscopic damage of the liver occurs, there is a decrease in the respiratory activity measured in yitgg and evidence of structural damage in the mitochondrial and microsomal fractions. Zalkin and Tappel (22) showed that these changes were apparently due to peroxidative destruction of membrane lipid. The hepatotoxic effect of several agents including ethanol and CCl“ have also been attributed in part to peroxidation of membrane lipids (90). The most studied hepatotoxin has been CCl“, and most of the evidence which supports the view that CCl“ toxicity is due to lipid peroxidation has been reviewed by Recknagel (126). Subse- quent work has shown that the metabolism of CCl“ in the 18 liver generates Cl and CCl3 free radicals which react primarily with membrane phospholipids (127). It has also been shown that CClu both in yitrg and in yiyg decreases liver microsomal drug metabolism activity and cytochrome P“5° by promoting lipid peroxidation (128). In 1946, Jaffe (129) observed that rats fed a diet containing wheat germ, known to be very high in vitamin E content, developed fewer tumors when injected intra- peritoneally with 3-methylcholanthrene. Haber and Wissler (130) also reported that the growth of subcutaneous tumors induced by 3-methylcholanthrene was much slower in animals on a vitamin E supplemented diet. More recently, Shamberger and co-workers (131, 132) have shown that vitamin E and selenium applied topically reduce tumor formation induced by application of dimethylbenzanthracene to mouse skin, and they also found the carcinogen-induced chromosome breakage was decreased by the antioxidants. Wattenberg (133) has also shown that the synthetic anti- oxidants, butylated hydroxyanisol (BHA), butylated hydroxytoluene (BHT), and ethoxyquin reduce the formation of gastric carcinomas induced by dimethylbenzanthracene. There is evidence which suggests that lipid peroxidation does take place during the formation and growth of tumors (131, 134, 135). Since it is the epoxides of polycyclic aromatic compounds which have mutagenic and carcinogenic properties (136, 137), it is conceivable that these epoxides, or other oxidized products with l9 carcinogenic properties, are formed during peroxidative lipid oxidation. Mitotic activity and the development of tumors involves a change in the acid-soluble thiol level in the cell (138), and it has been reported that tumors have higher levels of glutathione reductase and glucose-6- phosphate dehydrogenase than does non-cancerous tissue (139). Therefore, increases in the levels of these enzymes in response to lipid peroxidation may also be a mechanism for promoting tumor development. The marked decline in the incidence of gastric cancer in the U.S. since 1930 has been attributed to the introduction of wheat cereals and the use of BHA and BHT in prepared foods (140). It has also been established that the geographical variation in the cancer incidence rates is inversely related to the geographical variation of the selenium content in grain and forage crops (140-142). The synthetic antioxidants BHA and BHT appear to decrease occurrence of cancer in all parts of the gastrointestinal tract (140). It was found that the blood selenium level was signifi- cantly decreased in patients with gastrointestinal cancer (142, 143). There are other diseases in which peroxidative destruction of polyunsaturated fatty acids has been shown to be involved. People with diseases which seriously decrease the transport of lipid, such as cystic fibrosis and abetalipoproteinemia, develop vitamin E deficiencies and soon become necrotic unless additional vitamin E is 20 provided (144, 145). There is some evidence that lipid peroxidation may be involved in the development of atherosclerosis since atheromatous plaques are found to contain ceroid pigments and lipid peroxides (106). There has been more recent work by investigators in the U.S.S.R. showing that in rabbits, a vitamin E deficiency increases the incidence of atherosclerosis (146). Another interest- ing observation is that antioxidants inhibit the spread of mycobacterial infections (147). It was also found that in rats fed a vitamin E deficient diet high in polyunsat- urated fatty acids, Hansen bacilli grow rapidly and spread to the organs containing high levels of polyunsaturated fatty acids (4). One of the notable characteristics of chicks fed a diet low in vitamin E but high in polyunsat- urated fatty acids is the cerebellar encephalomalacin which is characterized by edema within the cerebellum, decrease in polyunsaturated fatty acids, degeneration in neural tissue and eventual disappearance of the Purkinje cells (148, 149). It has also been shown that there are several neurotoxins which promote peroxidation of brain lipids (90). Menke's disease, also known as the "kinky or steely hair syndrome", is a sex-linked genetic defect which results in degeneration of the central nervous system in newborn males. In the brains of victims of this disease, there is a marked decrease in the content of polyunsatu- rated fatty acids and accumulation of fluorescent pigments, most notably in the Purkinje cells, suggesting that it is 21 the peroxidation of lipid which accounts for the detrimental effects of this disease (150, 151). METHODS AND MATE RIALS Material Sources: Male rats of the Holtzman strain, weighing between 200 and 250 g, were obtained from Spartan Research Animal, Haslett, Michigan. Bromelain was obtained as a gift from the Dole Company, Honolulu, Hawaii. Catalase (Sigma 2X recrystal- lized), cytochrome c (Sigma type VI), glucose oxidase (Sigma type II), horseradish peroxidase (Sigma type I), NADP-isocitrate dehydrogenase (Sigma type IV), and xanthine oxidase (Sigma type I) were obtained from Sigma Chemical Company, St. Louis, Missouri. Erythrocuprein (Pentex) and bovine serum albumin (Pentex) were obtained from Miles Lab., Inc., Kankakee, Illinois. Partially purified preparations of cytochrome Puso and a deoxycholate solubilized preparation of NADPH- cytochrome c reductase were obtained as a gift from Dr. M. J. Coon at the Department of Biochemistry, Univer- sity of Michigan, Ann Arbor, Michigan. A purified preparation of adrenadoxin from bovine adrenal mitochondria was obtained as a gift from Dr. T. Kimura at the Department of Chemistry, Wayne State University, Detroit, Michigan. 22 23 ADP (Sigma fermentation grade), ascorbic acid, butylated hydroxytoluene, o-dianisidine, dithiothreitol, EDTA, epinephrine, N-ethylmaleimide, D-mannitol, NADH, NADP+, NADPH, DL-sodium isocitrate, thiobarbituric acid, Tris base, and xanthine were obtained from Sigma Chemical Company, St. Louis, Missouri. Aminopyrine and o-phenanthroline were obtained from K & K Laboratories, Plainview, New York. Benzola]-pyrene, o-dibenzoylbenzene, l,3-diphenylisobenzofuran, malonic dialdehyde tetraethyl- acetal, and 2,4-pentanedione were obtained from the Aldrich Chemical Company, Milwaukee, Wisconsin. Metapyrone, also known as metopirone (l,2-bis-(3-pyridyl)- 2-methyl-l-propanone) was obtained from Ciba Pharmaceutical Company, Summit, New Jersey. Phenobarbital was obtained from Merck and Company, Inc., Rahway, New Jersey. All aqueous solutions were prepared with water which had been distilled and passed through a mixed bed resin ion exchange column. All other reagents used were analytical grade. Preparation of Microsomes: The microsomal fraction was isolated from the livers of male rats (200-250 9), fed water containing 0.1% phenobarbital 10 days prior to being killed, by the following method. The animals, which had been starved overnight, were killed and the livers perfused in situ by injection of 10 m1 of cold 1.15% KCl into the portal vein within 1 or 2 cm of the liver. The liver was then removed, 24 blotted, weighed, and minced by chopping with a pair of scissors. The minced tissue was homogenized in four volumes of 1.15% KCl containing 0.2% nicotinamide, added to inhibit NADP+-ase, with about 4 strokes in a Potter- Elvehjem homogenizer equipped with a Teflon pestle. The homogenate was centrifuged at 15,000 xg for 20 minutes and the precipitate containing the nuclear and mitochondrial fractions discarded. The microsomal fraction was isolated as a pellet by centrifuging the 15,000 xg supernatant at 105,000 xg for 90 minutes. The supernatant was discarded and the microsomes were either used immediately or resuspended in 0.05 M Tris-HCl (pH 7.5 at 0-5°) containing 50% glycerol at a concentration of about 50 mg of protein per ml. All operations were performed at 0-5° and the microsomes in glycerol were stored at -20° under N2. Microsomal Lipid Peroxidation Assay: The NADPH-dependent peroxidation of microsomal lipid was assayed by incubating microsomes in a Dubnoff shaker at 37° under air in reaction mixtures (usually of 5 m1 total volume) containing 0.05 M Tris-HCl (pH 7.5 at 37°), 2 mM ADP, 0.12 mM Fe(N03)3 (or FeNHu(SOu);),and 0.2 mM NADPH or an NADPH-generating system containing 7 mM MgC12, 2 mM D,L-isocitrate, 0.1 mM NADP+, and 0.05 unit of NADP- isocitrate dehydrogenase per ml. The reaction was initiated by the addition of NADPH (or the addition of isocitrate and NADP+) and the extent of lipid peroxidation 25 at various time intervals was determined by removing aliquots from the reaction mixture and measuring the amount of malondialdehyde present by the method described below. The rate of peroxidation is expressed as nanomoles of malondialdehyde formed per minute per m1. Measurement of Malondialdehyde by the TBA Reaction: The formation of malondialdehyde was measured by assaying the chromogen formation with thiobarbituric acid (TBA) by a method similar to that described by Bernheim, Bernheim, and Wilber (12). A stock TBA reagent was prepared containing 0.375% TBA and 15% trichloroacetic (w/v) acid in 0.25 N HCl. To prevent any additional peroxidation of lipid in the sample during color develop- ment with TBA, 0.01 volume of 2% butylated hydroxytoluene in ethanol was added to the thiobarbituric acid reagent just prior to use, producing a finely divided suspension of the antioxidant. The malondialdehyde content in a sample of 1.0 ml total volume was determined by mixing the sample with 2.0 m1 of the TBA reagent and heating the mixture for 15 minutes in a boiling water bath to develop the color. After cooling, the assay mixtures were centrifuged at 1000 xg and absorbance of the supernatant at 535 nm was determined using a Coleman Jr. Spectrophoto- meter equipped with a flow cell. The content of malondialdehyde was calculated using a standard curve prepared using known quantities of malondialdehyde 26 tetraethylacetal. The standard curve is shown in Figure 1. At absorbance values less than 0.3, the standard curve is linear with a slope of 0.0411 absorbance unit per nanomole malondialdehyde in the sample. In most instances, 0.5 ml aliquots were removed from the peroxidation reaction mixtures and increased to 1.0 m1 total volume by the addition of water. Microsomal Drug Metabolism Assay: The NADPH-dependent hydroxylation activity catalyzed by cytochrome Pl.5° was assayed as aminopyrine demethylase activity by measuring the formation of formaldehyde using the method of Nash (152). The reaction mixture (5 ml total volume) containing microsomes, 0.05 M Tris-HCl (pH 7.5 at 37°), 5 mM aminopyrine, and a NADPH-generating system including 7 mM MgC12, 2 mM D,L-isocitrate, 0.1 mM NADP+, and 0.05 unit of NADP-isocitrate dehydrogenase per ml, were incubated in a Dubnoff Shaker at 37°. The reaction was initiated by the addition of the isocitrate and the NADP+. The production of formaldehyde was measured by removing 1.0 ml aliquots from the reaction mixture at various time intervals and mixing the aliquot immediately with 1.0 ml of 10% trichloroacetic acid. After allowing time for the protein to precipitate (about 5 minutes), 2.0 m1 of Nash reagent (2 M ammonium acetate, 0.05 M acetic acid, and 0.02 M 2,4-pentanedione) were added and the mixtures were heated at 50° for 10 minutes. After 27 .=mamwumuoz cam moonu02= Hoods omnfluommo wonuoe on» ocm Houoomawnummuumu moxnmoaoflocone mo mmfluwucono czocx means omGHEHouwo mos Ec mmm um ooconuomno one .ZOHBUdmm dme mmB Nm mawmmoddHQZOde m0 BszmmDmde mom m>mDU Qm¢QZ¢Bm .H ousmflm 28 038mm _E\mo_oEc . oomroEmficofiz 9. 0.0 L I'll) I? R) p I, TILT ’ 4 4 ‘4 411 «is. I 111111 11 in lud‘fi. < 1 J 1. . 114414 J 1..) 1 Id . s . o u o . . l , n A u . n n M . hir‘lun a“! *1 4 i b ‘01--qu 4 'Lvih TAT. ' 1.14.1)! . 1+ 10.441. 0 u H v m m . . . “ ~ » . . . A . . . . . a . . .. . . . . i . T w , I” .. . . . . .11 N o . . .. TIM . . a. :4 . . . ( 1 . . 4| A . a o v . I .. , , m . . o . . A . o. .. . . . . . . . _ ‘r _ u . .o. W . M . y u o . . . v T O o . q I?! I'll...) > bl» > “Pi ifilrll L n. .. .T. . . . ,4; u. a 1 . . . u . . ‘ o . . N. . w . . . o . . . . . .. . o . . . p a p . .1 . I . . N . 2 J. h a . . ., . .H . ( . a <. 1 1.1 . . 14 . m .. m. ; - 3* A . . . u . . . . . . ‘, . o . . ... 4 ..¢ . o o . . s 0 41.0 11*AIT 'll 0...]. H s . h . a o . i.) I 11.1.“! 0 . . 1 . . H A a A q . . i . . u A , .m . w n . . . F b o . g . . .,u -.H . , ¢ H. o _ o N . J. u M . .u . u u N . ._ . . u _ . n. H . . . . A u . ” ..., ,1 . a .. .. .b,a .. u T.” . H .._ . .. M T 62:40.26-.. 4. .L. a 1.4 ,. . w 4 a w . . . . o ., . \ . o o H . . o . . H . .. s. . . . . . . . . . . . . . s . . i o 4 . .b. . N H “1. ...... W. i.) , . . M. “ . . . . . . _ . . . . . . . . v .. . . . . . . r . . . [It . . . W I» . . . . . -4- - («it -44! u . J1. .. . .. .1 . ., . . . . . u u . . s . . . . . . . .F b . M n .r» .w w . e w . 4‘ i .1 _ . . . . . . A N . . . j .. . . . . . . n . . . . _ ~ . . . . . _ _ . . .. . .. . . .. . a o, . H! o t lint: 4 101‘ fi LI» r “W 19.0-14 0| [0| . . . . . . .. . . N . n a M: ..... w .. w . . . . . . . . o . o . .. a . , x .. . . . . n u . .w . ... h . H n V n . . g A J . .3 .5 111. . . . . . . w . . a., . — . . . . u . . . . o. . . . . o u b .. H .. . . H . #w A . 00 , ll Ill... II. a I! .II III. pit .r 0 T vuoi Tn) IO. ‘1 VII! T .10 [A H i .. . 4 . .. . . .. . . . la . . . . o — p . . o , , . . ~ . A . . o .. r . o . . — . — A T. . m I W n . n . w J __ r . . n . . . .. . . . L . o . . m . . . . . o . M . . i r . . .. .. u , l . . . . . . . ‘ ‘0 +l¢§. nuLill O IBEmI. IT’LIO‘IO. H 'W . . I to T x l II:WII T .M I0 |.¢T£I|Io..|| 1.9.1 N . . . . a . - . . . o. * . . A . . L 1 . L . . LN » . . . . . 7’11 .4 A . f w u . _ _ . . . 1 1 .r 1 _ . .h b h I o i-.-‘ 4 1 I T IT: : |,Iv .C c n .0loltn 0 1-91 0 A . a i“ . n w I h _ h . . . . . " . . . A . _ . m I _ ii L. a i . s . - — 1 . . « . . . u . . . .. . o OTIL OI...T..| 1|- .o..u . u. a¢.. ¢- o .1 _ _ . . u . . . . . . . I . ._ u . . » . h i . N . M . w . - . - ~11: 1:1 L--1..--.-, ....... I- . q . To . . * . . u a . . . \m u o . — . . n > » w . . . .. A 4. .. T. . . i. . u o . . . . . . . o , . . . . l _ . *1 _ . . b A. 4 It .0! TV Iifili. :- u'o'll oooooo a .0. O ..n .3.) Oil .0) . 4 T . . . . . . . . _ _ m n . . z u . . T” u m W ». L III I). . . . . q .1 .. . . \ m . . . . u h u . e . . o . . s . . W . . . k _ ” v1 :93 .o v. -xol TIIO!‘ 1 ulplbbl leO,1.o.H..OOLA join-017091.00 I. ~-l1.ll.a¢i.0c0:l- uOITAol..I0. 1| '0 c.||.¢. .. T n . . . o. . 39.6.. ,. . . . . o . . o a “ . . L . . . .. . * . . . . L. . A .. . . o . . — u . . . . v .. k .4 w p . H H p - y L . p . _ N. O V: O LUU 99g ‘ ooueqdosqv QC 29 cooling, the mixtures were centrifuged at 1000 xg and the absorbance of the supernatant was measured at 412 mu using a Coleman Jr. Spectrophotometer equipped with a flow cell. The formaldehyde content was calculated using an extinction coefficient of 7.08 x 103 M”1 cm.1 and the dilution factor of 4. Assay of Cytochrome c Reductase Activity: The assays for the reduction of cytochrome c by NADPH were all made with 75 uM cytochrome c and 0.1 mM NADPH in 0.3 M phosphate buffer, pH 7.5, at 25°. The reduction of cytochrome c was measured by following the increase in absorbance at 550 nm on a Perkin—Elmer model 124 spectro- photometer. The rate is expressed as microequivalents of cytochrome c reduced per min using an extinction coefficient of 2.10 x 10“ M.1 cm”1 (153). Solubilization and Purification of NADPH-Cytochrome c Reductase: The method employed to isolate NADPH-cytochrome c reductase, using proteolytic digestion followed by gel filtration and DEAE-cellulose affinity chromatography, is similar to that of Omura and Takesue (154). Microsomes were first washed with 0.05 M Tris-HCl (pH 8.0 at 0-4°) containing 1 mM EDTA. About 2 to 3 g of washed microsomes were resuspended in the same buffer at a concentration of 10 mg per ml and incubated anaerobically with 0.1 mg of '0' .4r 30 bromelain per ml for 3 hours at 0-4°. The mixture was centrifuged at 105,000 xg for 60 min and the pellet was discarded. The supernatant fraction was concentrated by ultrafiltration on a Diaflo PM-30 membrane, and this concentrate was applied to a Sephadex G-100 column (27 x 825 mm) and eluted with the same buffer. The fractions containing the NADPH-cytochrome c reductase activity were adsorbed onto a DEAE-cellulose column (12 x 160 mm) and eluted with a linear 0 to 0.5 M KCl gradient (total volume, 100 ml) containing 0.05 M Tris-HCl (pH 8.0 at 0-4°) and 1 mM EDTA. The peak fractions containing reductase activity were combined and diluted about 3-fold with 0.05 M Tris-HCl (pH 8.0 at 0-4°) containing 20% glycerol and adsorbed onto a second DEAE-cellulose column (10 x 130 mm) pre-equili- brated with the above buffer. The enzyme was eluted with a linear 0.15 to 0.35 M KCl gradient (total volume, 30 ml) containing 0.05 M Tris-HCl (pH 8.0 at 0-4°), 20% glycerol, and 0.1 mM EDTA. The enzyme in the combined peak fractions from this column could be stored at 0-4° for several months with only moderate loss of activity. Preparation of Antibody to NADPH-Cytochrome c Reductase: The antibody was prepared by Mr. John Buege using the following methods. An adult male rabbit was immunized with the purified reductase by three weekly, cutaneous injections (abdomen and toe pads). Each injection contained 0.85 mg of the purified reductase administered 31 in 6, 2, and 1 ml of 50% Freund's complete adjuvant, in the let, 2nd, and 3rd weeks, respectively. A booster injection (1.0 mg) was given 1 month after the third injection, and 10 days later blood was collected from the ear vein and the serum was separated from the whole blood. The y- globulin fraction from both immune and preimmune serum was prepared by the method of Masters et al., (59). Solubilization and Purification of NADH-Cytochrome bS Reductase: The flavoprotein, NADH-cytochrome b5 reductase was purified from liver microsomes by Mr. John Buege using the method of Takesue and Omura (155), which involved lysosomal digestion to solubilize the enzyme. The only change made in the purification method was the addition of 0.1 mM dithiothreitol to all buffers used after (NHQ)ZSO“ fractionation to prevent loss of activity as a result of sulfhydryl oxidation (156). Solubilization and Purification of Cytochrome b5: Cytochrome b5 was also solubilized by digestion of microsomes with bromelain. During the purification of NADPH-cytochrome c reductase, the fractions from the Sephadex G-100 column containing cytochrome bs were lyophilized and saved for subsequent purification. The accumulated cytochrome bS was further purified by DEAE- cellulose affinity chromatography. The lyophilized -v w. c l I 32 cytochrome b5 was dissolved in 0.05 M Tris-HCl (pH 8.0 at O-4°) and dialyzed against the same buffer. The dialyzed protein was then applied to a DEAE-cellulose column (approximately 1 cm x 10 cm) and eluted successively with 50 m1 volumes of 0.05 M Tris-HCl, (pH 8.0), 0.1 M KCl in the Tris buffer, and 0.2 M KCl in the Tris buffer. The cytochrome bs was eluted in the buffer containing 0.2 M KCl and the peak fractions were combined. The content of cytochrome bS was determined by measuring A absorbance (424 nm - 409 nm) in the reduced vs. oxidized difference spectrum using an extinction of 1.85 x 105 M"1 cm.1 (157). Extraction of Microsomal Lipid: The extraction of total lipid from microsomal membranes was accomplished using the method of Folch et 31., (158). In order to minimize the oxidation of unsaturated lipid during the extraction process, all the solvents were flushed with nitrogen, nitrogen gas was passed over the lipid extract during the steps requiring transfer of the extract, and all operations were performed at 0-4°. Approximately 2-3 9 of washed microsomes (as pellet) were extracted with 200 ml of 2:1 chloroform-methanol (v/v) and the insoluble material removed by filtration using a Buchner funnel and a vacuum filter flask. The lipid extract was washed three times in a separatory funnel with a solution containing chloroform, methanol, and an aqueous salt solution (0.04% CaClZ, 0.034% MgClz, and 0.58% NaCl) in ,a 33 respective proportions of 3:48:47 by volume. The upper phase from the first two washes were removed by aspiration into a suction flask. The washed lipid extract was taken to dryness by rotary vacuum evaporation using a Bfichi evaporator, operated as shown in Figure 2. The extracted microsomal lipid was redissolved in approximately 5 ml of 2:1 chloroform methanol and cooled to -20° which caused precipitation of some proteinaceous material. The precipi- tate was removed by centrifugation and the lipid extract was stored at -20° under nitrogen in Teflon capped tubes wrapped in aluminum foil. The lipid stored in this fashion could be kept for several weeks with no evidence of deterioration. The content of lipid was measured as the amount of total lipid phosphorous by the method of Bartlett (159). Preparation of Liposomes: Aqueous suspensions of microsomal lipid were prepared by sonication under anaerobic conditions. This was accomplished by transferring an aliquot of the stock lipid solution to a thin walled plastic tube, removing the chloroform-methanol under a stream of nitrogen, adding nitrogen-saturated buffer, and then capping the tube under a stream of nitrogen. The technique used to disperse the lipid in aqueous solution is illustrated in Figure 3. The sealed tube was placed in a small glass vessel filled with water kept at the bath temperature. The probe of a Branson 34 .Eosoo> onu omooaou ou uououooo>o on» oune comonuwc mnwoooan an ooumnflsuou mos mmooouo one .oEoHo nonfio m nufl3 ooHHouu Icoo mo3 Moumuooo>o one ounw uoonuxo oHoHH onu mo 30am one .Edsoo> m neounfioe ou nonoufiomo Houo3 m on oouooncoo mos uououooo>o one .Eoumwm coHuouooflnmou o en coal waouoeflxouooo no oonflouCHmE moz mafloo mnflmnoocoo one ca unmaooo one .m084m0m¢>m wmdeom HmUDm d OZHmD mBU4mem DHqu AdzomomUHZ 20mm mezm>qom mo A¢>Ozmm mom MDOHszmB .N ousmem Figure 3. 36 TECHNIQUE FOR PREPARING LIPOSOMES BY SONICATION UNDER ANAEROBIC CONDITIONS. The stoppered tube containing lipid and buffer is inserted in one of the side necks of a 3-necked 50 m1 pear-shaped flask with 19 x 22 fittings. The flask contains water and is set in a polystyrafoam bucket containing ice. The probe of the sonicator (with large tip) is inserted through a rubber cushion into the flask so that it is located directly over the end of the tube containing the lipid. 38 model S 125 sonifier was placed in the outer vessel above the end of the plastic tube and a sonication current of 10 amps was applied for approximately 5 min. The final concen- tration of lipid in the suspension was 2.5 umoles of lipid phosphorus per ml, and the buffer routinely used was 0.25 M Tris-HCl (pH 6.8 at 37°) containing 0.25 M NaCl. However, the use of water in place of the buffer was also suitable. Peroxidation of Extracted Microsomal Lipid: The assays for the peroxidation of extracted micro- somal lipid promoted by the purified preparation of NADPH-cytochrome 0 reductase were performed in the following manner. Unless specified otherwise, reaction mixtures (usually of 5 ml total volume) containing 0.25 M Tris-HCl (pH 6.8 at 37°), 0.25 M NaCl, 0.5 umole of lipid phosphorus per ml, 2 mM ADP, 0.22 mM Fe(NOa) and 0.1 mM EDTA were 3: incubated at 37° under an atmosphere of air in a Dubnoff shaker. The formation of malondiadehyde was assayed by the method described for the peroxidation of lipid in intact microsomes. The rate of oxidation prior to the addition of the enzyme was determined after the addition of 0.2 mM NADPH to incubation mixtures which had been preincubated for 2 min at 37°. The rate of malondialdehyde production measured after the addition of enzyme was adjusted for this nonenzymatic rate. The peroxidation of extracted microsomal lipid pro- moted by NADH-cytochrome bs reductase was assayed under the 39 same condition described above with the exception of NADPH which was replaced by 0.2 mM NADH. The reaction mixtures for assaying the peroxidation of extracted lipid promoted by xanthine and xanthine oxidase contain the same buffer, salt, and amount of lipid as described above, plus 0.33 mM xanthine. The reaction mixtures also contain ferric ion added in the forms indicated in the text. The xanthine oxidase, which comes as a suspension in ammonium sulfate containing 10 mg per m1 of protein, was diluted to 5 mg per ml and passed through a Sephadex G-50 column (approximately 5 mm x 40 mm) immediately prior to use to remove the salicylic acid included as a stabilizing agent. The reactions were initiated by the addition of xanthine oxidase to reaction mixtures, preincubated for 2 minutes at 37°. The rate of formation of malondialdehyde was determined by the same methods previously described, removing aliquots at 30 second intervals during the first 2 minutes of the reaction. Erythrocyte Hemolysis: Radical-like intermediates produced during the lipid peroxidation reaction were detected with the use of an erythrocyte hemolysis assay (68). Erythrocytes, obtained from the blood of a young male goat, were washed and packed. The incubation mixtures for erythrocyte hemolysis contained 0.1 m1 of packed cells per ml in a reaction mixture containing 0.1 M Tris-HCl (pH 6.8 at 37°), 0.1 M 40 NaCl, 2.0 mM ADP, 0.12 mM Fe(N03) 0.05 mM EDTA, and 1 3, unit of isocitrate dehydrogenase per ml. The incubation mixtures were preincubated for 2 min at 37° in a Dubnoff shaker, and the reaction was initiated by the addition of the substrates necessary for the generation of NADPH: 2 mM sodium isocitrate, 7 mM MgC12, and 0.1 mM NADP+. The reaction mixture with intact microsomes contained approximately 1 mg of microsomal protein per ml. The reaction mixture with the purified reductase contained 2.4 ug of the reductase and 1.0 mmole of lipid phosphorus per ml. The percentage of hemolysis was assayed as described by Pfeifer and McCay (68) after 10 min of incubation in the presence of NADPH. Purification of Erythrocuprein: Erythrocuprein was purified from bovine erythrocytes, with the aid of Mr. Mark Ondrias and Dr. S. D. Aust, by the method of McCord and Fridovich (160). The purified protein was dialyzed against water, lyophilized and stored at -20°. The Cu++ content of the purified protein was determined by atomic absorption spectroscopy performed by Mr. Joseph Prohaska using a Perkin Elmer model 303 atomic absorption spectrophotometer. The preparation was found to contain 22,000 mg of protein per mmole of Cu++. The amount of erythrocuprein was calculated on the assumption of 2 gram atoms of Cu".+ per mole (161). 41 Assay of 02- Production: One method used to assay the production of 02- was the epinephrine oxidation assay. Misra and Fridovich (162) have demonstrated that 02- promotes the oxidation of epinephrine to form adrenochrome, and that the superoxide dismutase activity of erythrocuprein will effectively inhibit this reaction. The formation of adrenochrome was assayed by following the increase in absorbance at 480 nm. Adrenochrome has an extinction coefficient of 4.02 x 103 at this wavelength (163). The formation of 02- was also measured by assaying the reduction of cytochrome c in both the absence and presence of erythrocuprein, as described by McCord and Fridovich (160). Addition of l,3-Diphenylisobenzofuran to Aqueous Reaction Mixtures: The involvement of singlet molecular oxygen in lipid peroxidation mechanisms was investigated with the use of l,3-diphenylisobenzofuran, DPIF, which rapidly reacts with singlet oxygen to form o-dibenzoylbenzene (164). Since DPIF is completely insoluble in water, a method was required to insure that the compound was evenly distributed in the lipid in the reaction mixture. The addition of DPIF to the reaction mixtures containing extracted microsomal lipid was accomplished by adding the compound to the lipid extract in chloroform-methanol, removing the solvents under 42 a stream of nitrogen, and then preparing liposomes by the method already described. In the reaction mixtures containing intact microsomes, DPIF was added to the micro- somes prior to their addition to the reaction mixture by the following procedure. A finely divided suspension of the compound was prepared by rapidly diluting an appropriate aliquot of 0.1 M DPIF in acetone into a buffer solution. Microsomes were then added to the mixture at a final concentration of 5 mg of protein per m1. Care was taken in these operations to minimize the exposure to light. Identification of o-Dibenzoylbenzene in Reaction Mixtures Containing!ll3-Diphenylisobenzofuran: The formation of o-dibenzoylbenzene, DBB, in lipid peroxidation reaction mixtures containing DPIF was investigated using thin layer chromatography of a chloroform extract of the reaction mixture. To eliminate the photosensitized oxidation of DPIF to DBB, all operations were performed in a photographic dark room where the only source of illumination was a safelight with a Wratten series 1A filter. For chromatographic analysis, a 2.0 m1 aliquot was removed from the reaction mixture and extracted with 0.5 ml of chloroform. One- tenth m1 of the extract was applied to the Silica Gel G thin layer plate (20 cm x 20 cm, 250 u thick). The plate was developed with 3:1 heptane-dioxane (v/v). DPIF was visualized by its fluorescence, and DBB was visualized 43 after spraying the plate with 0.5% 2,4-dinitrophenyl- hydrazine in 2 N HCl. Other Methods: Protein was determined by the method of Lowry gt gt., (165) and standardized with bovine serum albumin using E1% 1 cm at 280 nm equal to 6.6 (166). Sodium dodecyl sulfate polyacrylamide gel electro- phoresis was performed by Ms. Ann Welton using the method of Fairbanks gt gt., (167). The polyacrylamide gels were stained with Coomassie blue and scanned in a Gilford spectrophotometer with a scanning attachment. The oxidation of reduced pyridine nucleotides was assayed by following the decrease in absorbance at 340 nm using an extinction coefficient of 6.2 x 103 M.1 cm-1 (168). The rate of hydrogen peroxide formation, catalyzed by glucose and glucose oxidase, was assayed by measuring the decomposition of o-dianisidine in the presence of horseradish peroxidase according to the method of Maehly and Chance (169, 170). RESULTS AND DISCUSSION CHAPTER I THE INVOLVEMENT OF NADPH-CYTOCHROME c REDUCTASE IN LIVER MICROSOMAL NADPH-DEPENDENT LIPID PEROXIDATION Most of the problems in the investigation of liver microsomal electron transport are related to the difficul- ties encountered in trying to separate the individual electron transport components from the microsomal membrane. Several of the microsomal electron transport components have been isolated by using incubation with proteases and lipases to release the components from the membrane (58, 154, 155, 171-176). It has been demonstrated that the solubilization produced by lipases is apparently due to protease activity present as a contaminant in either the lipase or in the microsomes (155, 175, 177). The micro- somal components, which can be solubilized by proteolytic digestion, appear to be amphiphathic proteins which are released into solution following the proteolytic cleavage of a hydrOphobic side chain (178-180). Although many of the catalytic functions are retained by the protease- solubilized components, they usually fail to interact with the remaining microsomal electron transport components (54, 44 45 180, 181). Consequently, most of the conclusions regarding microsomal electron transport have been based primarily on evidence obtained using selective inhibition or induction of the microsomal activities which depend on the electron transport system._ The initial experiment, in the attempt to identify the microsomal electron transport components which promote NADPH-dependent lipid peroxidation, was designed to deter- mine whether digestion of microsomes by proteases or lipases would produce a soluble fraction capable of catalyzing NADPH-dependent peroxidation of extracted microsomal lipid. A very fortunate aspect of this experi- ment, not understood at the time, was the use of a NADPH- generating system which contained EDTA (see Chapter II). The pertinent results from this experiment are shown in Table I. The soluble fractions obtained after digestion with the three proteases, trypsin, bromelain, and nagarse, and with crude pancreatic lipase were observed to promote some NADPH-dependent lipid peroxidation. Since the supernatant fraction from the bromelain-digested microsomes promoted the greatest rate of peroxidation and also clearly demonstrated a requirement for the addition of extracted microsomal lipid, the subsequent experimental pursuit of the microsomal components involved in NADPH-dependent lipid peroxidation was made with the use of the microsomal components solubilized by bromelain. 46 TABLE I THE SOLUBILIZATION OF NADPH-DEPENDENT LIPID PEROXIDATION FROM LIVER MICROSOMES BY INCUBATION IN THE PRESENCE OF PROTEASES AND LIPASES. Microsomes (10 mg per ml) were incubated overnight at 0-4° under anaerobic conditions in incubation mixtures containing 0.05 M Tris-HCl (pH 7.5), 20% glycerol (v/v) and 0.1 mg per ml of the enzyme indicated. The incubation mixtures was separated into a supernatant fraction and a pellet fraction following centrifugation at 105,000 x g for 90 minutes and resuspension of the pellet in an equivalent volume of .05 M Tris-HCl (pH 7.5). NADPH-dependent lipid peroxidation was assayed in both the absence and presence of added microsomal lipid (0.4 mg per ml) under the reaction conditions described under "Methods" for assaying NADPH-dependent lipid peroxidation in intact microsomes. Each peroxidation assay contained 0.25 ml of the indicated fraction in a reaction mixtures of 5.0 ml total volume. This experiment was performed with the assistance of Mr. Kurt Ivie. Reaction Mixture Malondialdehyde formed Enzyme in pre—incubation Fraction - lipid + lipid nmoles/min/ml None . . . . . . . . . . supernatant . . . 0.0 0.0 None . . . . . . . . . . pellet. . . . . . 2.1 4.7 Trypsin. . . . . . . . . supernatant . . . 0.5 0.7 Trypsin. . . . . . . . . pellet. . . . . . 0.5 0.9 Bromelain- . . . . . - . supernatant - . . 0.0 0.9 Bromelain- - - - - - - - pellet~ - - . . . 0.3 0.6 Nagarse. . . . . . . . . supernatant . . . 0.5 0.5 Nagarse. . . . . . . . . pellet. . . . . . 0.3 0.5 Phospholipase C. . . . . supernatant . . . 0.0 0.0 Phospholipase C. . . . . pellet. . . . . . 0.3 1.8 Phospholipase A. . . . . supernatant . . . 0.0 0.0 Phospholipase A. . . . . pellet. . . . . . 1.2 4.7 Pancreatic lipase. . . . supernatant . . . 0.0 0.4 Pancreatic lipase. . . . pellet. . . . . . 0.0 1.2 QFi _ 47 The supernatant fraction from the bromelain digested microsomes was found to contain essentially all the NADPH- cytochrome c reductase activity and a large portion of the cytochrome b5 originally present in the microsomes. Omura gt 31., (182) had previously reported similar results following tryptic digestion of rat liver microsomes and demonstrated that both components could be isolated by gel filtration and DEAE affinity chromatography. When the supernatant from the bromelain solubilized enzyme was passed through a Sephadex G-100 column, the NADPH-dependent lipid peroxidation activity co-eluted with the NADPH- cytochrome c reductase activity. When the reductase was further purified by DEAE-cellulose affinity chromatography, the peroxidase activity still eluted with the reductase activity indicating that the lipid peroxidation activity solubilized by bromelain was promoted solely by the activity of NADPH-cytochrome c reductase. However, it was still necessary to determine whether the NADPH- dependent peroxidation promoted by intact microsomes also involved this flavoprotein. Purification of NADPH-Cytochrome c Reductase: Horecker (183) first described a NADPH-cytochrome 0 reductase, obtained as an acetone powder from a liver homogenate and characterized as a flavoprotein. Williams and Kamin (171) subsequently demonstrated that this enzyme is located in the microsomal fraction and can be 48 solubilized by digestion of the microsomes with lipase. Phillips and Langdon (172) also reported that the same flaVOprotein could be purified following tryptic digestion of microsomes. It has recently been shown that the solubilization of this enzyme by digestion with lipase is apparently due to a proteolytic contaminant in the lipase (177). The NADPH-cytochrome c reductase solubilized by bromelain was purified by a method which is essentially the same as that described by Omura and Takesue (154) for the purification of the reductase and cytochrome bs from trypsin-treated microsomes. The first step in this purification accomplished the separation of the reductase from the cytochrome bS by gel filtration through a column of Sephadex G-100. The elution profile from the Sephadex column is shown in Figure 4. This elution profile was obtained from a column approximately 20 x 250 mm, but the enzyme was ordinarily purified using a larger column which resulted in better separation and greater purification. The second step in the purification was accomplished by adsorbing the reductase onto a DEAR-cellulose column and eluting the reductase with a KCl gradient. An elution profile from a column used in this step is shown in Figure 5. The final step in the purification involved adsorbing and then eluting the enzyme from a second DEAE column, as was suggested by Omura and Takesue (154) to insure obtaining a homogeneous preparation. 49 .=moocuwz= Home: confluommo mmocu mm wEMm may mum maofiuflocoo umnuo Had .EE omm x om hamumeflxondmm mums mcoflmcmsfic cEsHoo one .mmmuosomu o mEousoouwonmmodz UmNflHH95HOmICflwHwEouQ mo coflUMOHmwnsm map CH poms CESHoo may mo maflmonm coflusam mce .Nmmdmwoemzommu ZZDAOU ooalw xmofimmmm .e madman 50 --v- (wueov - wU vavwv ‘Sq 433 .. “1,01 x uuwwuu/vv "pea 31h: -HdOVN Q 0Q L0. v. N O. : <2 <2 <2 <9 0 10 q- 4|- MO r0 7" /on¢ .mflmmuonmouuomam on uoflum Houwounuoflcuflo mo mocwmmnm map ca sump HmumB maaawon m ca Umummc mm3 .muMMHSm axomooo EDfiGOm ma :H Am: oev madamm one .mmdaoDQmm O MZOMSUOBMUImmaflz QmHmHMDm m0 mHmmmommomeumqm Amw UmHQ MDH2¢meU¢wQOm mfidhqsm ANUMQOQ ZDHDOm .m wusmflm 55 56 markers run in parallel with the reductase. Welton EE,§l°v (185) have recently demonstrated that the minimal molecular weight of detergent-solubilized NADPH—cytochrome c reductase is 79,000 or greater, as determined by the same method. Therefore, the reductase must have been partially degraded during solubilization by bromelain. Similar differences in the molecular weights between detergent-solubilized and protease-solubilized proteins have been demonstrated for other microsomal components (178-180). NADPH-cytochrome c reductase has been described by other investigators as a flavoprotein containing 2 moles of FAD per mole of enzyme, and it is apparently devoid of any metal ion cofactors (186). The absorption spectrum of the bromelain-solubilized enzyme, shown in Figure 7, has a characteristic flavin absorption spectrum with absorption peaks at 455, 380, and 276 nm in the oxidized state. The absorption at 455 disappears upon reduction with either dithionite or NADPH. Furthermore, when the enzyme was aerobically reduced by NADH instead of NADPH, a half- reduced flavin spectrum was observed as well as an increase in the absorbance at longer wavelengths, suggesting that the flavin is primarily in the semiguinonoid oxidation state which has been shown to be the oxidized state of the enzyme normally present during the steady state catalytic cycle (186). The ratio of the extinction at 276 nm to that at 455 nm of the oxidized enzyme was 6.7. Passage of the exmyme through a short Sephadex G-25 column did not change 57 Figure 7. ABSORPTION SPECTRUM OF PURIFIED NADPH-CYTOCHROME C REDUCTASE. The spectrum was obtained with a solution containing 1.61 mg of protein per ml in 0.05 M Tris-HCl, pH 7.7, 20% glycerol, 0.1 mM EDTA, and about 0.25 M KCl at ambient temperature. , oxidized; - - -, reduced by addition of 0.1 mM NADH; . . . ., reduced by Nazszo“. AB SORBANCE 2.0-— 1.0- 0.0 WAVELENGYH (um) «7 :"' 0'. “‘1 !. 'MM 1 A on" 'J- . . . "run ‘usfiaul 1 ”Mt“! d y—{V , -‘ RV. ‘fi'vg' I “51‘ in» 6“ -' 59 any of its spectral properties. These special character- istics are similar to those reported by Omura and Takesue (154). Involvement of NADPH-Cytochrome c Reductase in Microsomal Lipid Peroxidation: The preparation of antibody to purified NADPH- cytochrome c reductase, which also inhibited its enzymatic activity, was first described by Kuriyama gt al., (184) who demonstrated that the enzyme isolated from the liver microsomes of phenobarbital-treated rats was immunologically identical to the enzyme isolated from untreated animals. The ability of the antibody to this enzyme to inhibit its enzymatic activity has been used by other investigators to show that this enzyme is the initial component in the NADPH-dependent electron transport system associated with the cytochrome Puso-catalyzed hydroxylation of drugs and other compounds (57-59). Masters and co-workers (187) have shown that the NADPH-dependent mixed-function amine oxidase in liver microsomes is catalyzed by an FAD- containing flav0protein which is immunologically distinct from NADPH-cytochrome c reductase. They have also demonstrated that the cytochrome Pgso-catalyzed hydroxyla- tion activity in both liver and adrenal cortex microsomes can be inhibited by antibody to the reductase from liver microsomes, but there was no inhibition of the 6O NADPH-dependent, cytochrome Pnso-catalyzed hydroxylation activity in adrenal mitochondria (59, 188). Antisera to the bromelain-solubilized NADPH-cytochrome c reductase was prepared as described under "Methods". Since whole serum from preimmune blood as well as from immune blood was an effective inhibitor of lipid peroxidation, the y-globulin fractions were isolated by a method involving ammonium sulfate precipitation and DEAE- cellulose chromatography. The y-globulin from the serum of a rabbit immunized with the purified reductase had a high antibody titer against the reductase, as evidenced by the formation of a single immunoprecipitin line in double diffusion agar plates using both the purified reductase and detergent-solubilized microsomes (185). The antibody was also a very good inhibitor of the NADPH-cytochrome c reductase activity of the enzyme. Maximum inhibition required that the antibody and the reductase be incubated together a few minutes before the assay was initiated. As shown in Figure 8, the activities of both the purified reductase and intact microsomes were inhibited more than 90% by y-globulin from immune serum, and no inhibition was observed with y-globulin from preimmune serum. The specificity of the inhibition by the antibody was further demonstrated by showing that microsomal NADH-cytochrome c reductase activity was not inhibited. As shown in Table II, neither y-globulin from immune serum nor y-globulin 61 .:m©onuwzz Home: pmnHuommo mm omEuOMHmQ muw3 mammmd .HE mom camuoum Hmeomouowfi mo m: CH pmcwmucoo on mmEOmOHOHE uomucfl Suez whammm “HE Hod mE>Ncm may mo m: H.o Umcwmucoo on mwmuogpmu pmamausm suw3 m>mmw¢ .Eouwm mcsEEHmHm Eoum seasnoam I» . IIIIII “Enumm mcsEEH Eoum awasnonu> . .mzwwzm was 09 maomHez< um weH>Heom mmaeooomm Aon.ewov o mzommooewoummoeuom coauoawnuoaop oneuemocfleo one .HE Hod CHE Hod oohnooaoEHom mo moaoan m.a couscoum mnoepflonoo oEom onu noon: >ue>euom nofluoamnuoEoo onenhmonfiEm one .HE you see Hod opwnooaofionoaoe mo moHOEc N.H mo noeumEHom on» oouoeoum ceasnoamu> mo oonomno onu nfi wufi>fiuoo noflumo uwxouom ofimwa one .HE Mom neououm HMEOmouowE mo 08 mma.o nonwounoo mousuxee coeuomou Ham .A llllll V Esuom onseewoum Eouw neasnonl> no A v Eduom onsEEH Eonw nwasnoaml> Honuwo oneneopnoo monsuxfle coauomou ne non noeuoawnuofioo onwuwmonfleo can any noflumofixouom pfldfla mo unoouom one .mmeo waonflm m couscoum mane .HE Hod neououm 08 cm mo cofluouunoo Inoo m an .moEOmonoflE ou .onouooo mo poswfiam HHmEm m ca .ocouxmunon onu mcepom >n nouomoum ouoz poumoflonw onouwmucon mo unsOEo onu onenflounoo moEOmouoez .meomm¢ .ZOHBdeZWUZOU MEMNZM m0 ZOHBUZDm d md ZmBmMm Amooz mmB ZH ZOHBdDonmmm DHAHA Bzmozmmmalmma<2 .NH musmflm 82 o — o M o —o N o ~—-o ’ o T 1 °. °. 9 c N ° .m ugw/aplqomogpuolow sa|owu ENZYME (fig/ml) 83 Figure 13. NADPH-DEPENDENT LIPID PEROXIDATION IN THE MODEL SYSTEM AS A FUNCTION OF THE LIPID CONCENTRATION. The reaction mixtures contained 0.35 pg per m1 of a preparation of the reductase which had lost about one-half of its enzymatic activity. Assays were performed as described under "Methods". 84 0.50 0.25 0.00 2.0—4 _ 0 _E 55:110233332 3.02: 0.0 LIPID (pMoles P/ml) 85 .=mponuo2= Moon: ponfiuomoo mo UoEHOMuom oHoB mmomm< .hufi>euoo oeuoemuno one mo mamnuono uoonm umoH non nOHnB ommuosoou onu mo nowuoumoouo o no as mom on mm.o oonflounoo monsuxfle nowuomou one .md no oneozom m4 m4 Ememwm Ammo: mme 2H onenonommm onHq ezmozmmmoummonz .qa musmflm 86 8.0 7.0 l 9 2.0— 0.0 flu ugw/ aplqamogpuo'ow sapwu 87 activity occurs at about pH 6.8, but the activity decreases quite rapidly above pH 7, in contrast to lipid peroxidation in intact microsomes which still proceeds readily at pH 8 (40). A similar change to an acidic pH Optimum has been observed in the peroxidation of microsomal lipid promoted by ascorbic acid and ferric ion (40). The dependence of NADPH-dependent peroxidation in the model system on ionic strength is shown in Figure 15, which shows that Optimal peroxidation requires a NaCl concentration of 0.4 M or greater. Since the pH of the reaction is at the lower limit of the Tris-HCl buffering range, the use of a higher concentration of buffer was preferred. As shown in Figure 15, 0.25 M NaCl will allow Optimal peroxidation activity in reaction mixtures containing 0.25 M Tris-HCl. Kamin and Masters (198) have shown that the purified reductase, in contrast to intact microsomes, requires high salt concentrations for the NADPH-dependent reduction of numerous acceptors. As shown in Table VI, NADPH-dependent peroxidation in intact microsomes, in both the absence and presence of EDTA-Fe, is little affected by the ionic strength of the reaction mixture. The Use of Erythrocyte Hemolysis to Detect Reactive Radical Intermediates: Tappel (199) has suggested that the tissue necrosis caused by in vivo lipid peroxidation of microsomal or mitochondrial membranes may be due in large part to the 88 .HumImHue z m~.o nuas IOI “HumImwue z no.0 nufl3 IOI .=moonuo2= Moons ponflnomoo mo ooEuOmnoo ouoz mammmn .HE nod ommuosoou oowmwuom onu NO on mo.o coneoucoo monsuxfle noHuooou one .ZOHBdeZWUZOU BAém m0 ZOHBUZDm d m4 Zmemwm ammo: mme ZH ZOHBflOonmmm QHmHA Bzmozmmmolmmodz .mH ousmflm 89 V V O, In "' o' [In uguI /ap£qap|egpuolew salowu 0.4 . 0. Q0 0 [NaCfl M 90 TABLE VI THE EFFECT OF EDTA-Fe AND HIGH IONIC STRENGTH ON NADPH-DEPENDENT LIPID PEROXIDATION IN INTACT MICROSOMES. Each reaction mixture contained ferric ion and the complexing anions indicated, added as ADP-Fe (0.12 mM Fe3 and 2. 0 mM ADP) and EDTA-Fe (0.10 mM Fe3 and 0.10 mM EDTA).A11 reaction mixtures contained 0.2 mg of microsomal protein per m1 and buffer adjusted to pH 6.8 at 37°. Assays were performed as described under "Methods". Description Malondialdehyde formed nmo Zes/min/ml With 0.05 M Tris-HCl Plus ADP-Fe . . . . . . . . . . . . 1.41 Plus ADP-Fe and EDTA-Fe . , , , , , With 0.25 M Tris-HCl and 0.25 M NaCl Plus ADP-Fe . . . . . . . . . . . . 1.61 Plus ADP-Fe and EDTA-Fe . . . . . . 2.63 91 release of lysosomal hydrolytic enzymes. McCay and co- workers (68, 69) demonstrated that NADPH-dependent peroxidation of microsomal lipid in intact microsomes is accompanied by the production of a highly transient radical- like factor which will promote lysis of lysosomes or erythrocytes included in the reaction mixture. Further similarity between NADPH-dependent peroxidation in the model system and in intact microsomes is shown in Table VII, which demonstrates that erythrocytes were also lysed when included in the reaction mixture containing extracted lipid and the purified reductase. Neither system caused erythro- cyte hemolysis in the absence of NADPH. Pheifer and McCay (68) observed that erythrocyte hemolysis was promoted prior to the formation of malondialdehyde when using microsomes from vitamin E-supplemented rats and concluded that the factor causing erythrocyte hemolysis has no relation to the formation of lipid peroxide intermediates. However, as shown in Table VIII, the erythrocyte hemolysis in the peroxidation reaction mixture containing the purified reductase also required the addition of extracted microsomal lipid. The Two Forms of Fe3+ Required in the Model System: When the heat-stable factor in isocitrate dehydrogenase was identified as EDTA, the concentration of EDTA required for optimal activity was determined as shown in Figure 16. However, as the concentration of EDTA approached the ferric 92 TABLE VII ERYTHROCYTE HEMOLYSIS PROMOTED BY THE NADPH-DEPENDENT PEROXIDATION OF MICROSOMAL LIPID. Assays were performed as described under "Methods", and the percent hemolysis was assayed after 10 minutes incubation. Description % Hemolysis With intact microsomes . . . . . . . . - - 90 Minus NADPH . . . . . . . . . . . . . . . 0.5 With extracted microsomal lipid and purified NADPH-cytochrome c reductase 81 Minus NADPH . . . . . . . . . . . . . . . 0.5 93 TABLE VIII REQUIREMENT FOR MICROSOMAL LIPID IN THE ERYTHROCYTE HEMOLYSIS REACTION CATALYZED BY NADPH-CYTOCHROME C REDUCTASE. Assays were performed as described in the legend to Table VI, except each reaction mixture contained one-half as much of the reductase. Microsomal lipid concentration % Hemolysis nmoles of lipid P/ml 0.0 4 0.2 . . . . . . . . . . . . 39 0.4 - - 62 1,0 . . . . . . . . . . . . 57 Figure 16. 94 NADPH-DEPENDENT LIPID PEROXIDATION IN THE MODEL SYSTEM AS A FUNCTION OF THE EDTA CONCENTRATION. + All reaction mixtures contained 0.12 mM Fe3 , 2.0 mM ADP, 0.28 mg of reductase per ml, and the amount of EDTA indicated. All other conditions are the same as described under "Methods". 95 +— 200 150 5'0 160 [EDTA] pM _E c_E\OUmnOU_m_UcO_m.2 mO_O_2c 96 ion concentration (0.12 mM), inhibition occurred, suggesting that chelating all the ferric ion with EDTA inhibited the reaction. Therefore, the optimal level of EDTA was determined under conditions in which the concen- tration of ADP-Fe remained constant by adding ferric ion chelated with a molar equivalent of EDTA. Under these conditions, no inhibition was observed at higher concen- trations of EDTA, as shown in Figure 17. The concentration of EDTA-Fe routinely used in the lipid peroxidation system was 0.10 mM, although there is still some increase in the peroxidation activity observed at higher concentrations. The requirement for ferric ion not chelated by EDTA was investigated by adding ferric ion chelated with slightly less or slightly more than a mole equivalent of EDTA. The results in Table IX show that, although an EDTA concentra- tion in slight excess of the ferric ion concentration inhibits peroxidation activity completely, the concentration of EDTA-free ferric ion required must be at the umolar level. This would correspond to the optimal concentration of ferric ion required for NADPH-dependent peroxidation in intact microsomes which is about 3 umolar (42). The requirement for EDTA-Fe is the most significant difference between lipid peroxidation in intact microsomes and the reaction promoted by the purified reductase. Although the NADPH-dependent peroxidation in intact micro- somes is promoted by the activity of NADPH-cytochrome c reductase, the peroxidation promoted by the 97 .=mponuo2= moons oonwuomoo mo poEHOwuod ouo3 mummmd .coDMOflocH omIdeom mo noeuonucoonoo onu Ob nofiufloom ca .mo< :5 o.m sn omumfimno ”Amozvmm :5 ma.o new as you mmmuoswmu poemwuzd on» mo 0: mm.o conflounoo monsuxfle cofluomou Han .ZOHeflmeZMUZOO oml0NcoanIo mo moaozn omm .w* Homopfixo onfinunmx wanes ousuxee noflumnsonfl ouoaaeou .m# “ousuxfls n0euonsonw ouoHQEOU .~* “nonsmoNnonomH IahnonmenIm.H wanes ousuxHE nowuonoone ouonEOU .He .>Hxx oanoe ou pnomoa onu CH nonfluomop mm ommpflxo onenunmx new msouonm Imonm nemea mo oHOE moo unnumouconOmwaxconQAOIm.H mo oHoE o.a noneoucoo onsuxes noHuomou ouoameoo one .nBonm ouo Amoaoneo paaOmv onenmuomnawnonoonuecHan.m nufiz noxoudm mneon uouwm no Amoaouflo couponv oonoomoHOSHw neonu en oo>uomno muomm on» wano .=mnonuo2: noon: nonenomoo mo ooEHOMHom mmz nonsmounoQOmeamnonmflo Im.H new ononnonaeoNcoanIo mo coHuMOHweunooH oendoumoumeouno onu new mousuxfie nOfiuomon noflumowxonom mo nofluomnuxo one .mmmDBXHE ZOHBUdmm ZOHBéQonmmm ZH mZMszmANONmeHDIO m0 ZOHBHo>fluoommou HE mom moHOEc mm.o ono mo.o maouoEonuooo mo munsoso ca 0 can n Douonoo mBOuum onp en noumowonfl ouon3 ououxes cofluooou onu ou noonm mo3 nwouoso Iounumum .m Douonoo Bouum onu an powwowone ouonz ommuosnou o oEOunoou>0ImmoAO>ZH mzmHz¢mumZ ZOHBUflmm .Hm ousmem 156 cozoEeoe cot—02.8 52.0 828.22%: IA NO + 22.. coca--- H0631: C s\.. IIIIIIII IIIIIII~0 IIIIIIIII Non— :rlllflldsn m ...~o 18.22.. m m NO+NWO on; T NONI . IIIII .Ol . NO. < 22.. N +NOCII— INC 2542 llmluxnaqz 22: ifllaczeox 3288.... 0.23-1.5; $8_xo.2_£8x 9&3 4