'IIII'II'II'I'F:, I} 2% l “j l 7/ (9 MICHIGAN STATE UNIVERSITY lIHpAmE ll ll lllllll lillllllllllliilflfll'lllll 3 1293 00569 1922 LIBRARY E" Michigan State University This is to certify that the dissertation entitled CHARACTERIZATION OF THE FACTORS REGULATING THE COUPLING AND RESPIRATORY CONTROL OF ISOLATED CHICK HEART MITOCHONDRIA presented by PETER PAUL TOTH has been accepted towards fulfillment of the requirements for Ph.D. Biochemistry degree in Major professor MS U i: an Affirmative Action/Equal Opportunity Institution 0-12771 IV1£31_} RETURNING MATERIALS: Place in book drop to LJBRAfiJES remove this checkout from -;‘_. your record. FINES will be charged if book is returned after the date stamped below. CHARACTERIZATION OF THE FACTORS REGULATING THE COUPLING AND RESPIRATORY CONTROL OF ISOLATED CHICK HEART MITOCHONDRIA By Peter Paul Toth A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Biochemistry 1988 ABSTRACT CHARACTERIZATION OF THE FACTORS REGULATING THE COUPLING AND RESPIRATORY CONTROL OF ISOLATED CHICK HEART MITOCHONDRIA By Peter Paul Toth This study is concerned with identifying and characterizing the factors regulating the degree to which chick heart mitochondria con- trol rates of respiration in both the presence and absence of ADP. In order to preserve the activity of outer membrane proteins, chick heart mitochondria are isolated with the use of collagenase rather than the non-specific protease Nagarse. Up to 70% of the mitochondria in the starting myocardium can be recovered using this method. This isolation procedure also preserves the activity of a novel soluble uncoupling protein. This protein binds to the chick heart mitochondrial outer membrane with high apparent affinity, has a molecular mass of 6-15 kD, and can be inactivated by heat or protease digestion. These mitochondria are more highly coupled than any yet reported. Under optimal conditions (0.05 nmole cyt aa 20 mM Pi’ 5 mM pyruvate/ 3! 2.5 mM malate) the organelles respire with infinite respiratory control. Factors which decrease rates of state A respiration include limited Nagarse digestion of the mitochondria, pre-incubating mitochondria with substrate prior to the addition of ADP, and EGTA. EGTA inhibits the bind- ing of Ca(II) and Mg(II) released from the mitochondria to two contamina- ting ATPases: actcmyosin and the F1FO-ATPase of broken mitochondria. The ADP:0 ratios of chick heart mitochondria are fractional and significantly greater than 3.0 during the oxidation of pyruvate/malate (P/M), B-hydroxybutyrate (BHB), or a-ketoglutarate (aKG). Results with these NAD-linked substrates support a 13, and possibly 1A, proton model for oxidative phosphorylation. During the oxidation of glutamate/malate (G/M) and succinate, ADP:O ratios approximate to the classic values of 3.0 and 2.0, respectively. The P1 metabolism of these mitochondria is complex. Based on rates of swelling monitored by 90° light scatter measurements, the P transport 1 protein has a Km for P S 50 pH. P transport thus cannot be rate- 1 i limiting for oxidative phosphorylation. P is inhibitory to respiration 1 during state A when either P/M or aKG are used as substrates. The data suggest that P inhibits the dehydrogenases for pyruvate and aKG in the i absence of ADP. In contrast, during state 3, P stimulates respiration. i This stimulation is hyperbolic in the presence of P/M and G/M, sigmoidal in the presence of aKG, and biphasic in the presence of BHB. To Karen, Ole Building and Loan Buddy Of Mine, for showing me that 1 + 1 = 1,000,000 iv ACKNOWLEDGEMENTS I thank my mentor Dr. Clarence Suelter, G.A.F.C. (Guardian Angel First Class) for giving me the freedom to pursue the type of education I believed I needed and for the privilege of having been one of his students. Thanks are also due to: Dr. Jack Holland (Commander of the Universe), for all the mud, the blood, the beer, and the marvelous lessons in spectroscopy; Dr. Loran Bieber, for a productive and very enjoyable collaboration on mitochondrial carnitine metabolism; Dr. John E. Sell, for his assistance in operating the computerized spectrometer and for making light scattering measurements all the more interesting to perform; my outstanding undergraduate assistants, Kendall Sumerix, Randall Kuntzmann, and Clarissa Stropp, for their diligence and their contributions to this work; Dr. Shelagh Ferguson-Miller, for her willingness to collaborate on parts of this project and for allowing me to use much equipment in her laboratory; Dr. John Wang, for providing me with the two antisera I was in need of; the other members of my guidance committee, Dr. Estelle McGroarty, Dr. Dale Romsos, and Dr. John Wilson for their advice and support; and Dr.'s Britton Chance (University of Pennsylvania) and Antonio Scarpa (Case Western Reserve University), for allowing me to perform experiments in their laboratories. Finally, I wish to thank the American Heart Association of Michigan for funding much of the the work contained in this dissertation. I shall always value the laughter, the difficulties, and the exhilaration embodied within this consuming endeavor. TABLE OF CONTENTS Page LIST OF FIGURES ................................................... x LIST OF TABLES .................................................... xv LIST OF ABBREVIATIONS ............................................. xvi INTRODUCTION ...................................................... 1 CHAPTER 1. LITERATURE REVIEW ...................................... A The Chemisomotic Theory ........................................... 5 Carrier Systems of the Mitochondrial Inner Membrane ............... 6 Substrate Dehydrogenases .......................................... 8 Electron Transfer Chain ........................................... 1 ATP Synthesis ..................................................... 1 The Proton, Oxygen, and ATP Stoichiometries of Oxidative Phosphorylation ................................................ 18 Alternative Coupling Schemes ...................................... 20 The Control of Respiration ........................................ 23 References ........................................................ 27 CHAPTER 2. ISOLATION OF HIGHLY COUPLED HEART MITOCHONDRIA IN HIGH YIELD USING A BACTERIAL COLLAGENASE ............ 36 Introduction ...................................................... 37 Experimental Procedures ........................................... 39 Materials ...................................................... 39 Enzyme Assays .................................................. A0 Transmission Electron Microscopy ............................... A3 Scanning Electron Microscopy ................................... AA Isolation of Mitochondria ...................................... AA Oxygen Consumption Assays ...................................... A8 Results ........................................................... A9 Mitochondria Recoveries ........................................ A9 Outer Membrane Enzymes ......................................... A9 Membrane Intactness ............................................ 53 Respiratory Control Ratios ..................................... 57 Back Diffusion of Oxygen ....................................... 57 General Comments ............................................... 62 Conclusions ....................................................... 68 References ........................................................ 69 CHAPTER 3. STUDIES OF THE FACTORS AFFECTING THE RESPIRATORY CONTROL AND ADPzO COUPLING RATIOS OF ISOLATED CHICK HEART MITOCHONDRIA IOOOOOOCOOOOOCOCOCOOCOOO0...... 71 vi Introduction ...................................................... Experimental Procedures ........................................... Materials ...................................................... Polyacrylamide Gel Electrophoresis ............................. Immunoblotting ................................................. Isolation of Chick Myocardial Myosin ........................... ATPase Assays .................................................. Oxygen Consumption Assays ...................................... Nucleotide Assays .............................................. Other Assays ................................................... Calculations and Statistical Analyses .......................... Results ........................................................... Rates of Respiration Vary with Mitochondria Concentration ...... ADP:O Ratios Also Vary with Mitochondria Concentration ......... Magnitude of Respiratory Parameters at the CMC ................. Effects of Pre-Incubating Mitochondria with Substrate .......... Oligomycin Sensitivity of State A Respiration .................. The Effect of EGTA on Respiratory Parameters ................... The Effect of Exogenous Mg(II) on Respiratory Parameters ....... Quantitation of ATPase Activity ................................ Identification of the Non-Mitochondrial ATPase ................. Ca(II) Uptake by Chick Heart Mitochondria ...................... Discussion ........................................................ Uncoupled Respiration of Chick Heart Mitochondria .............. ATPase Activity in the Extramitochondrial Space ................ The Critical Mitochondrial Concentration ....................... ADP:O Stoichiometries for Oxidative Phosphorylation ............ Conclusions .................................................... References O0....0......0.0......0.00000000000000000000000.0.0.0... CHAPTER A. THE ADVANTAGES AND LIMITATIONS OF USING 90° LIGHT SCATTER TO MONITOR CHANGES IN THE VOLUME AND PYRIDINE NUCLEOTIDE CONTENT OF THE CHICK HEART MITOCHONDRIAL MATRIX .0...IOU...0..O00.000.000.000...OOOOOOIOOIOOOOOOO Introduction ...................................................... Experimental Procedures ........................................... Instrumentation ................................................ Isolation of Mitochondria ...................................... Theoretical Considerations ........................................ The Effect of Photon Recovery on Light Scattering Measurements . Correlation of the Intensity of Light Scattered at 0° and 90° .. Results and Discussion ............................................ Relationship Between Light Scatter Intensity and Medium Osmolality .................................................. Relationship Between Light Scatter Intensity and Wavelength of Light .................................................... Measurement of Changes in Matrix NADH Levels ................... Sensitivity of 0° and 90° Light Scatter to Changes in Matrix VOlume 0....00.0.0.0...OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOIOOOO vii 151 152 156 156 159 161 161 163 170 170 170 173 188 conClUSionS O00....0.0...OO...0.00000000000000000000000000...... References 00......OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO CHAPTER 5. INTERACTION OF INORGANIC ORTHOPHOSPHATE WITH CHICK HEART MITOCHONDRIA. I. CHANGES IN VOLUME AND ION COMPOSITION OF THE MATRIX OOOOOOOIOOOICOOOOOOOOOOOOOI... Introduction ...................................................... Experimental Procedures ........................................... Materials ...................................................... Isolation of Mitochondria ...................................... Light Scattering Measurements .................................. Determination of Intramitochondrial (32)-Orthophosphate ........ Quantitation of Transmembrane Ionic Fluxes ..................... Other Assays ................................................... Results ........................................................... Volume Changes Induced by Inorganic Orthophosphate are Complex . Limitations on the Rate and Extent of Swelling ................. Factors Regulating the Amount of Inorganic Orthophosphate Transported into the Matrix ................................. Discussion ........................................................ Ca(II) is not Required for Pi-induced Swelling in Chick Heart Mitochondria .......................................... K+ and Mg(II) Fluxes Secondary to Pi Addition .................. Implications for Myocardial Ischemia ........................... Use of Light Scatter Methodology to Measure Pi Uptake Kinetics . Role of the Pi Transport Protein in Respiratory Control ........ Conclusions .................................................... References O0..O0.0...0..000......O0......0..OOOOOOOOOOOOOOOOOOOOOO CHAPTER 6. INTERACTION OF INORGANIC ORTHOPHOSPHATE WITH CHICK HEART MITOCHONDRIA. II. SUBSTRATE-DEPENDENT MODULATION OF STATE 3 AND STATE A RATES OF RESPIRATION ............... Introduction ...................................................... Experimental Procedures ........................................... Materials ...................................................... Isolation and Preparation of Mitochondria ...................... Assays ......................................................... Results OOOOOOOOOOOOOOOOOOOOOOOOO....0.00.0.00...OOIOOOOOOOOOOOIOOO The Inorganic Phosphate Requirements of State 3 Respiration .... The Inorganic Phosphate Requirements of State A Respiration .... Activation of Substrate Dehydrogenase Activities by Inorganic Phosphate ................................................... Inhibition of Substrate Dehydrogenase Activities by Inorganic Phosphate ................................................... Discussion ........................................................ Phosphate Control of Dehydrogenases During State 3 Respiration . Phosphate Control of Dehydrogenases During State A Respiration . viii 19A 195 197 198 201 201 201 203 20A 206 206 208 208 216 217 236 236 237 2A0 2A1 2A2 2A3 2A5 2A9 250 253 253 25A 25A 256 256 261 266 277 283 283 285 COflClUSions 00.......0...0.00......OOOOOOOOOOOOOOOOO00.0.0000... References OOOOOOOOOOOOOOOOOOOOOO0.00.0...O...IOOOOOOOOOOOOOOOOOOOO CHAPTER 7. IDENTIFICATION OF A SOLUBLE PROTEIN FACTOR THAT UNCOUPLES MITOCHONDRIAL SUBSTRATE OXIDATION FROM THE PHOSPHORYLATION OF ADP IN CHICK MYOCARDIUM ............. Introduction ...................................................... Experimental Procedures ........................................... Materials ...................................................... Isolation of Mitochondria ...................................... Collection of Crude Fraction Containing the Uncoupler Protein .. Stabilization of Concentrated 88000 in Glycerol ................ Ammonium Sulfate Precipitation of the Uncoupling Activity in S8000 .......................................................... Assay for Uncoupling Activity .................................. Polyacrylamide Gel Electrophoresis ............................. Other Assays ................................................... Results ........................................................... Stability of the Uncoupling Activity ........................... Stimulation of Respiration ..................................... Ammonium Sulfate Precipitation of the Uncoupler Protein ........ Molecular Weight Estimate for the Uncoupler Protein ............ Discussion ........................................................ References OOOOOOOOOOOOOOOI.0..0....OO...0.0...OOOOIOOOOOOOOOOOOIOO APPENDIX: Published Papers and Abstracts ix 287 288 291 292 29A 29A 29A 295 295 296 296 296 297 299 299 299 302 302 308 310 FIGURE LIST OF FIGURES CHAPTER 1 Schematic Depiction of the Mitochondrial Electron Transport Chain O0.0.0....OOOOOOOOOOO...IOOOOOOOOOOOOOOOO CHAPTER 2 Difference Spectrum Showing the a Bands of Chick Heart Mitochondrial Cytochromes c1 + c, b, and aa3 ............ Flow Diagram for Collagenase Facilitated Isolation of Heart MitOChondria .0.0...OOOOOOOOIOOOOOIOOOOOOOO0.000000 Oxygen Electrode Tracings used for Monitoring Enzyme ACtiVities O...I...0..O...0.0.0.0....IOOOOOOOIOOO0.0..0.0 Representative Scanning and Transmission Electron Micrographs of Chick Heart Mitochondria Isolated with COlla-genase OO.I.0.000....0.00000COOOOOOOOOOIOOOOOOOO0.0. Respiratory Control Ratios Measured as a Function of Mitochondria Concentration .............................. Oxygen Consumption Behavior of Chick Heart Mitochondria ISOlated With COllagenase 0......OOOOOOOOOOOOOIOOOOOOOIOO Quantitation of oxygen consumption by the Clark Electrode and Back Diffusion of Atmospheric Oxygen into the oxygraph Assay Vessel I.I.0.0.00000000000000000000000000. Chick Heart Mitochondrial Respiratory Control Measured as a Function of EDTA Concentration ........................ CHAPTER 3 Extraction of P from Suspensions of Mitochondria HYdPOlYZing ATP OO..0...O...0..OOOOOOOOOOOOOOOOOIOOOOO0.. Rates of State 3 and State A Respiration of Mitochondria Measured as a Function of Mitochondria Concentration .... Variation of ADP:0 Ratios as a Function of Mitochondria concentration 0......0.0.0.0...OOOOOOIOOIOOOOOOOOOOOOO... Effect of ADP on the Rate of Uncoupled Respiration Page 111 M2 ”6 52 56 59 61 6A 67 81 91 1O 11 12 13 1M 15 16 17 18 19 Induced by CCCP when Mitochondria Oxidize Different SUbStrates OOOOOOOOOOOOOOOOOOOOOOOOOOOO0.0.0.0.000...O... Rates of State 3 and State A Respiration at the Critical Mitochondrial Concentration Measured as a Function of Time of Preincubation with 5 mM pyruvate/2.5 mM malate Prior to the Addition of ADP ............................ Sensitivity of State A Respiration to Oligomycin A Measured as a Function of Mitochondria Concentration .... Inhibition of Chick Heart Mitochondrial State 3 Respiration with Increasing Concentrations of carboxyatractYlOSIde I0.0.0.0..0.0.0.000...0.00.00.00.00. Steady-state Concentrations of ADP Regenerated During StateuRespiratj-On O0.00000000000000000000.0.0.0.0000... Inhibition of State A Respiration by Increasing concentrations Of EGTA IOOOOOOOOOOOO00.000.000.000...O... Effect of Mg(II) on Chick Heart Mitochondrial Respiratory contrOl 0...0...0......COO...0.0.0....OOOOOOOOOOOOOCOOOOO Effect of Mg(II) on ADP:0 ratios and State A Respiration Sensitivity of Chick Heart Mitochondrial ATPase Activity to OligomYCinA 0.0.0.000...OOOOOIOOOOOOOOOO0.0.0.0000... Quantitation of Total Divalent Cation-Sensitive ATPase Activity Associated with Isolated Chick Heart MitOChondria II...O0......0..OOOOOOOOOIOOOOOOOOOOOO...0.. Titration of Chick Heart Mitochondrial Divalent Cation- Sensitive ATPase Activities with either Mg(II) or Ca(II) NaDodSOu-Polyacrylamide Gel Electrophoresis of the Chick Heart Myosin Preparation used in ATPase Kinetic Studies . Immunoblotting of Chick Myocardial Actin and Myosin ..... Titration of Chick Heart Myosin with Mg(II) in either the Absence or Presence of Ca(II) ........................... Titration of Chick Heart Myosin with Ca(II) ............. Kinetics of Ca(II) Uptake and Release by Isolated Chick Heart MitOChondria 0.......0...OOOOOOOOOOOOOOOOOOO0.00... CHAPTER A System Configuration of the Integrated Spectrofluorometer/ xi 96 100 102 10“ 108 110 112 115 118 120 123 126 129 131 133 137 1O 11 Spectrometer used for Light Scattering Measurements ..... Intensity of Scattered Light at 0° and 90° Measured as a Function of Mitochondria Concentration .................. Responsiveness of the Chick Heart Mitochondrial Matrix to Changes in External Osmolality .......................... Intensity of Light Scattered by Chick Heart Mitochondria Measured as a Function of Wavelength .................... Apparent Changes in the Absorbance of Chick Heart Mitochondrial Matrix Pyridine Nucleotides During State 3 to StateuTr‘anSitj-ons .00....OOOOOOOOOOOOOOOOOOOOOIOOOO. Apparent Changes in the Absorbance of Rat Liver Mito- chondrial Matrix Pyridine Nucleotides During State 3 to StateuTranSitions O0..OOIOOOOOIOOOOOOOOOOOOOOO...OI. Off-center cell Rotation 0..OOOOOOOOOOOOOOOOO0.00.0000... Fluorescence and Absorbance Measurements of Chick Heart Mitochondrial Matrix Pyridine Nucleotides ............... Changes in the Intensity of Light Scattered at 90° Subsequent to the Addition of Pi, Substrate, and ADP .... Relative Sensitivity of 0° and 90° Light Scatter to Pi- Induced Changes in Chick Heart Mitochondrial Matrix VOlume O0.0...00......O...I.OCOOOOOOICOOOOOOOOOOOOOCOO... Double Reciprocal Plot Relating Changes in Optical Density to Increases in the Concentration of Mitochondria CHAPTER 5 Perturbations in 90° Light Scattering Intensities Subsequent to the Addition of Inorganic Phosphate to Suspensions of Chick Heart Mitochondria ................. Determination of the First-Order Kinetic Constants of the Initial Mitochondrial Swelling Change Induced by P1 ..... Inhibition of Light Scattering Changes Secondary to the Addition Of Pi 0....00.0.0...O0.0.0.0000...0.00.00.00.00. Kinetic Constants of the Initial P -Induced Swelling Phase Measured as a Function of Pi Concentration ,,,,,,,, Total Change in Chick Heart Mitochondrial Light Scatter- ing Intensity Measured as a Function of Initial Extra- mitochondrial Pi Concentration .......................... xii 158 166 172 175 178 180 182 185 187 191 193 210 212 215 219 221 1O Mitochondrial Uptake of 32Pi measured as a Function of P1 concentration 00.0.00...0....0.0.0....OOOIOOCOOOOOOOOOOO. The Effect of P and Exogenous Substrate on the Rate of K Efflux from hick Heart Mitochondria ................. Mg(II) Concentration in the Extramitochondrial Space Measured as a Function of Mitochondria Concentration .... The Effect of P and Exogenous Substrate on the Rate of Mg(II) Efflux from Chick Heart Mitochondria ............. The Effect of P. and Exogenous Substrate on the Rate of Ca(II) Efflux from Chick Heart Mitochondria ............. CHAPTER 6 State 3 Respiration Measured as a Function of P Concentration When Mitochondria Oxidize Pyruvate/Malate, Glutamate/Malate’ or a-Ketoglutarate O O I O O I O O I O O O O O O O O O O 0 State 3 Respiration Measured as a Function of P Concentration When Mitochondria Oxidize B-Hydroxybutyrate State U Respiration Measured as a Function of P Concentration When Mitochondria Oxidize Pyruvate/Malate, Glutamate/Malate, a-Ketoglutarate, or B-Hydroxybutyrate . Efficacy of HEPES in buffering the INT reduction assay system OOOOOOOOOOOOOIOOOOO0.0.0.000...OOOIOOCOOOOOOOOOIOO Titration of Mitochondrial Dehydrogenase Activity with P. using a-Ketoglutarate and Glutamate as Substrates ....... Titration of Mitochondrial Dehydrogenase Activity with Ca(II) using B-Hydroxybutyrate as Substrate ............. Titration of Mitochondrial Dehydrogenase Activity with P in the Absence of ADP using Pyruvate/Malate and a-Ketoglutarate as Substrates ........................... Integrative Scheme Depicting the Reactions in and Coupled to the Tricarboxylic Acid Cycle ......................... CHAPTER 7 The Stimulation of Chick Heart Mitochondrial Oxygen consumption by 088000 0.00.00.00.00000000IOOOOOOOOIOOOOIO Effect of Increasing Concentrations of 088000 on the Rate xiii 223 226 229 232 235 258 263 265 268 270 27" 280 282 301 of Oxygen Consumption by Chick Heart Mitochondria ....... 30H Activity of the Uncoupler Protein in the Ammonium Sulfate cuts or 038000 0..I0......00....0.0.0.0...OOOOOOOOOOOOOOO 306 xiv LIST OF TABLES TABLE PAGE CHAPTER 2 1 Yield of Heart Mitochondria Isolated with or without - COllagenase 00....0..0......000......OOOOOOOOOOOOIOOOOOOOO 50 2 Recovery of Hexokinase in Heart Mitochondria Isolated WithC01lagenase 0......O..0...OOOOOOOOOOOOIOOOOOOOOOOOOOO 5)4 CHAPTER 3 1 Respiratory Parameters of Chick Heart Mitochondria - ISOlated With COllagenase O00.00.000.00...OOOOOOOOOOOOO... 92 2 Apparent Kinetic Constants of Divalent Cation Stimulated ATPases Sedimenting with Isolated Chick Heart Mitochondria 13H CHAPTER 5 1 Effect of Substrates on the Kinetic Parameters for the Stimulation of State 3 Respiration by Inorganic Phosphate 259 2 Apparent Kinetic Constants for the Stimulation of Chick Heart Mitochondrial Matrix Dehydrogenases by Inorganic Phosphate .0.0...0..0.0...0....IOIOOOOOIOOOOOOOOOOOIOO..0. 271 3 Apparent Michaelis Constants for the Stimulation of Chick Heart Mitochondrial NADH Production by Nucleoside Diphosphates when a-Ketoglutarate is Oxidized ............ 276 XV A1 A2 ADP ADP:O ATP ATPase BCIP BSA CCCP cyt aa EDTA EGTA F1 FA FAD AC AC OX G/M GDP GTE LIST OF ABBREVIATIONS conventional absorbance absorbance determined spectrofluorometrically adenosine 5'-diphosphate ratio of nmoles of ADP phosphorylated per ng-atom of oxygen consumed adenosine 5'-triphospahte adenosine 5'-triphosphatase S-bromo-H-chloro-3-indolyl phosphate bovine serum albumin carbonyl cyanide m-chlorophenylhydrazone cytochrome c oxidase (E.C. 1.3.9.1) ethylenediaminetetraacetic acid ethylene-glycol-bis(N-amino-ethyl ether)-N,N,N',N'-tetra- acetic acid fluorescence intensity from cell position 1 Fluorescence intensity from cell position A flavin adenine dinucleotide phosphorylation potential or Gibbs free energy change of ATP synthesis redox potential or Gibbs free energy change of respiration 5 mM glutamate and 2.5 mM malate guanosine 5'-diphosphate 50% (v/v) glycerol, 20 mM Tris, pH 7.”, 1 mM EGTA xvi HEPES A-(2-hydroxyethyl)-1-piperazineethanesulfonic acid B-HB 5 mM B-hydroxybutyrate IDP inosine 5'-diphosphate INT 2-(A-iodophenyl)-3-(A-nitrophenyl)-5-phenyltetrazolium chloride I.U. international unit of enzyme activity (conversion of one nmole of substrate to product per minute) Jo rate of respiration a-KG 5 mM a-ketoglutarate Aem emission wavelength Aex excitation wavelength L.S. light scatter MalNEt N-ethylmaleimide MOPS 3-(N-morpholino)propanesulfonic acid Mr molecular weight MS 225 mM mannitol and 75 mM sucrose M88 225 mM mannitol, 75 mM sucrose, and 0.2% BSA MSBE 225 mM mannitol, 75 mM sucrose, 0.2% BSA, and 1 mM EGTA, PH 7.A MSPi 225 mM mannitol, 75 mM sucrose, and 20 mM Tris-Pi, pH 7.“ AUH+ electrochemical potential difference of protons n mechanistic stoichiometry NAD nicotinamide adenine dinucleotide (oxidized form) NADH nicotinamide adenine dinucleotide (reduced form) NaDodSOu sodium dodecyl sulfate NBT nitro blue tetrazolium ApH transmembrane pH difference xvii P/M Aw RCR TEMED TBS TTBS Tris 5 mM pyruvate and 2.5 mM malate inorganic phosphate transmembrane electrical potential photon recovery coefficient respiratory control ratio (state 3/state A) relative mobility N,N,N',N'-tetramethylethylenediamine tris-buffered saline tris-buffered saline enriched with Tween-20 tris(hydroxymethyl)aminomethane xviii INTRODUCTION 2 The nature of mitochondrial respiratory control and the proton, oxygen, and ATP stoichiometries for oxidative phosphorylation have been subjects of intense experimental investigation for the past thirty five years. An understanding of these processes is fundamental to formulating integrative models of how tissues conserve and effectively distribute the energy of substrate oxidation in the form of ATP. Much of our current insight into bioenergetics has been obtained from studies performed on isolated mitochondria. Yet, it is clear from inspection of the available literature that workers in the field of mitochondrial bioenergetics have been somewhat unsuccessful in identifying and con- trolling many of the factors which compromise the degree to which isolated mitochondria can couple the complex process of substrate oxidation to the phosphorylation of ADP. Consequently, apart from the use of lipophilic protonophores and various types of inhibitors, few systematic methods for non-toxically modulating the magnitude of mitochondrial respiratory control have appeared. Because most published studies have used poorly coupled mitochondria, there is widespread dis- greement over what enzymes participate in the phenomenon of respiratory control and what the precise stoichiometries of oxidative phosphoryla- tion can and should be. This dissertation is concerned with characterizing previously unrecognized factors responsible for modulating the coupling and respiratory control of isolated chick heart mitochondria. This work is divided into seven chapters. The first chapter is a review of the literature relevant to the work presented herein. It is divided into seven sections and is intended to provide the reader with a brief but comprehensive understanding of how mitochondria synthesize ATP at rates 3 commensurate with the energy demands of the cell. Chapters 2-7 are presented as separate papers detailing the experimental findings of this investigation. Each paper is comprised of separate Introduction, Experimental Procedures, Results, and Discussion sections. Chapter 2 details the isolation of highly coupled chick heart mitochondria with intact outer membrane enzymes using collagenase. The data presented in Chapter 3 demonstrates that uncoupled respiration can be minimized by controlling a number of factors and identifies the extramitochondrial ATPases responsible for the regeneration of ADP under particular conditions. Values of the ATP:0 stoichiometries for oxidative phos- phorylation during the oxidation of a variety of substrates are proposed. Chapter A is an investigation of the light scattering properties of chick heart mitochondria. It details a theoretical basis for the deviation of mitochondrial suspensions from Beer's law and shows that light scattering measurements at 90° are much more sensitive than those at 0°. In Chapter 5 the light scattering properties of these mitochondria are used to obtain estimates for the various kinetic constants of P uptake. The effect of P on matrix 1 i cation content is also explored. Chapter 6 is concerned with the effect of Pi on the activity of various substrate dehydrogenases during state 3 and state A respiration and argues that at least two of the substrate dehydrogenases can be rate-limiting for uncoupled respiration. Finally, Chapter 7 presents data which strongly suggests that chick myocardium contains a low molecular weight uncoupler protein which has an apparent high affinity for mitochondria and can completely uncouple substrate oxidation from the phosphorylation of ADP. Chapter 1 LITERATURE REVIEW I. THE CHEMIOSMOTIC THEORY Elucidating the detailed mechanism of mitochondrial oxidative phos- phorylation is of concern to many biologists and has been for years. An early description of this process, the chemical coupling hypothesis (Slater, 1953). assumed that, just as in glycolytic substrate-level phosphorylation, a high energy phosphorylated chemical intermediate is formed during electron transport. This compound could then be hydrolyzed and the free energy released used to drive ATP synthesis. No such comp- ound could, however, be found. The inability to identify chemical inter- mediates in oxidative phosphorylation stimulated the formulation of a novel approach, the chemiosmotic theory. The chemiosmotic theory was a revolutionary and deeply insightful view of how mitochondrial ATP synth- esis is driven (Mitchell, 1961, 1975. 1985). The four postulates of chemiosmosis are as follows (Mitchell, 1979 a and b): (a) The mitochondrial inner membrane contains proton- or hydroxyl- coupled (these are indistinguishable) solute translocases. The carriers are responsible for transport of solutes from the cytosol into the mitochondrial matrix and for osmotic stabilization. (b) Electron transport chains are intramembrane enzyme systems comprised of an alternating sequence of electron and proton carriers with a characteristic H /2e- stoichiometry. The respiratory chain and the ATPase translocate protons in the same direction (i.e., out of the matrix). (0) The ATP synthetase is a reversible, intramembrane ATPase with a characteristic H /ATP stoichiometry. (d) The mitochondrial inner membrane is a proteo-lipid osmotic barrier that has a low permeability to solutes in general and to protons and hydroxyl ions in particular. According to chemiosmosis, the role of the electron transport chain is 6 to commit the free energy of scalar substrate oxidation reactions toward the vectorial translocation of protons out of the matrix. These oxida- tion reactions create an electrical potential (Aw, negative inside) and a pH gradient (alkaline inside) across the mitochondrial inner membrane. The resulting electrochemical potential gradient of protons is defined by the following relationship: AuH+ = FAw + RT lnECHln/c OUtJ (1) H where F is the Faraday constant (96,N90 coulombs/mole of electrons), R is the gas constant, (8.31M joules/°K mole), T is the absolute temper- ature, and c is the concentration of protons internal or external to H the matrix, as indicated. The subject matter of this review will be discussed within the context of chemiosmosis. The current evidence which supports its validity as well as its shortcomings will be discussed, and the various steps of mitochondrial energy metabolism will be traced in order to lay a foundation upon which the experiments presented in this dissertation may be interpreted. II. CARRIER SYSTEMS OF THE MITOCHONDRIAL INNER MEMBRANE The mitochondrial inner membrane is essentially impermeable to charged solutes and, as a consequence, it ensures that the cytosol and mitochondrial matrix are maintained as chemically distinct compartments. In order for the two compartments to communicate and exchange metabolites, the mitochondrial inner membrane contains a system of transmembrane pro- teins which Specifically transport a broad variety of molecules in and out of the matrix (reviews: LaNoue and Schoolwerth, 1979; Meijer and Van Dam, 1981). Three types of facilitated transport are known to exist: 7 uniport, in which a single solute is transported by the carrier; symport, two or more different solutes are simultaneously transported in the same direction by the carrier; antiport, two or more different solutes are simultaneously transported in opposite directions by the carrier. The carriers that are relevant to the research presented herein are briefly surveyed below. The carriers for P1 (Palmieri g£_al., 1970; Coty and Pederson, 1974). pyruvate (Papa gt_al., 1971), and glutamate (Hoek and Njogu, 1976) symport protons with their respective substrates in order to maintain electroneu- trality. The pyruvate carrier also appears to be responsible for the transport of B-hydroxybutyrate (Pande and Parvin, 1978). The adenine nucleotide translocase and the dicarboxylate carrier are proteins that facilitate antiport. On the former, extramitochondrial ADP is exchanged 1:1 for intramitochondrial ATP (Pfaff and Klingenberg, 1968; Pfaff 25 al., 1969; Klingenberg and Rottenberg, 1977; Kramer and Klingenberg, 1982); on the latter, dicarboxylic acids (e.g., a-ketoglutarate, malate, and succinate) are exchanged for one another or for HPOHZ- (Palmieri g£_al., 1971). There is an elaborate transport system for Ca(II), some aspects of which are still poorly understood. Uptake proceeds via electrophoretic uniport (Rottenberg and Scarpa, 197M; Lotscher_g£;al., 1980), while efflux may proceed on a Ca(II)/Na+ antiporter (Haworth §£_al., 1980), a Ca(II)/H+ antiporter (Lehninger gt_al., 1979; Riley and Pfeiffer, 1986), or through an energy-dependent but as yet uncharacterized mechanism (Puskin g£_§l., 1976). K+, the major cation and osmotic stabilizing agent of the matrix, is transported by a K+/H+ antiporter (Martin gt $1., 198”) whose activity is believed to be regulated by Mg(II) (Nakashima et al., 1982). III. SUBSTRATE DEHYDROGENASES Once transported into the mitochondrial matrix, substrates are channeled into precisely regulated catabolic pathways designed to meet the fluctuating energy demands of myocardium. Fatty acids (Oram §£_§l., 1973; Lysiak et_al., 1986 and 1988) and pyruvate (Williamson, 197“) are both important sources of fuel for myocardium. This review, however, will focus on the oxidation of pyruvate and tricarboxylic acid cycle intermediates because the experiments to be detailed in this disserta- tion are not concerned with fatty acid metabolism. The dehydrogenases for pyruvate, threo-Ds-isocitrate, and a-ketoglutarate catalyze non- equilibrium reactions and are all highly regulated. Because of their importance in determining the rate at which matrix NADH is produced, the known mechanisms regulating the activity of each enzyme will be discussed in turn. The pyruvate dehydrogenase complex (PDH) catalyzes the irreversible conversion of pyruvate to acetyl CoA. The PDH complex is comprised of three distinct enzymes: pyruvate decarboxylase (E1), dihydrolipoate acyltransferase (E2), and dihydrolipoyl dehydrogenase (E3) (Reed, 197A). The cofactors for E1, E2, and E3 are thiamin pyrophosphate, lipoic acid, and FAD, respectively. In the terminal step of the reaction, E3 transfers electrons from the reduced flavin to NAD+. PDH is interconverted between a nonphosphorylated active form (PDHa) and a phosphorylated inactive form (PDHb) (Linn 3t_a_l_., 1969 3 and 33). PDHa is phosphorylated by PDH kinase, an enzyme that uses ATP as phosphate donor. The kinase is associated with E2 (Yeaman, 1986) and phosphorylates a serine residue in 9 E1 (Yeaman g£_al., 1978). thereby inactivating the enzyme. PDH kinase is inhibited by pyruvate (Yeaman, 1986; Budde g£_§l., 1988), ADP (Hucho 32 al., 1972), and NAD+ and CoA (Pettit §t_gl., 1975); it is activated by NADH and acetyl CoA (Kerbey g£_§l., 1976; Hansford, 1977; Latipaa 32 al., 1985), the two products of the reaction catalyzed by PDH. PDHb is converted back to PDHa via a PDH phosphate phosphatase. This enzyme is activated by Ca(II) (Denton and McCormack, 1980; McCormack and Denton, 1985 and 1986) and inhibited by NADH (Pettit §£_gl., 1975). In addition to covalent modifications, PDH is also subject to product feedback inhibition. NADH and acetyl CoA prevent the binding of NAD+ and CoA, respectively, to binding sites on the enzyme (Blass and Lewis, 1973; Kerbey §£_al., 1976; Hansford, 1976). To summarize, increases in the ratios of NADH/NAD+, acetyl CoA/CoA, and ATP/ADP result in an increase in the ratio of PDHb/PDHa. NAD-isocitrate dehydrogenase is comprised of identical subunits and its activity is subject to allosteric control. ADP increases the affinity (i.e., decreases the Km) of the enzyme for threo-DS-isocitrate in both rat (Goebell and Klingenberg, 1963) and bovine (Chen and Plaut, 1963) heart mitochondria. NAD+ (Goebell and Klingenberg, 1963; Lowenstein, 1967) and Ca(II) (McCormack and Denton, 1986) also decrease the Km of the rat heart enzyme for threo-DS-isocitrate. NADH is a competitive inhibitor of NAD+ binding (Lowenstein, 1967). Consistent with these findings, it is now well established that as the ratios NADH/NAD+ and ATP/ADP decrease, the activity of the enzyme increases (Hansford, 1980; Bulos §t_al., 198A). a-Ketoglutarate dehydrogenase is an enzyme that is structurally and mechanistically similar to PDH (Yeaman, 1986). Unlike PDH, however, the 10 activity of this enzyme is not modulated by phosphorylation/dephosphoryl- ation reactions. The Km of the mammalian enzyme for a-ketoglutarate is decreased by either an increase in matrix Ca(II) concentrations or a decrease in the matrix ATP/ADP ratio (McCormack and Denton, 1979 and 1981; Hansford and Castro, 1981). The mechanism by which Ca(II) acti- vates this enzyme is as yet unknown (McCormack and Denton, 1986). The activity of a-ketoglutarate dehydrogenase is negatively modulated by increases in the ratios of NADH/NAD+ and succinyl CoA/CoA, due presumably to feedback inhibition (Hansford, 1980). It is also not known whether ADP activates the dehydrogenase directly (i.e., by inducing a structural transition) or if it activates indirectly by stimulating nucleoside diphosphokinase. The latter enzyme will relieve feedback inhibition by stimulating the hydrolysis of succinyl CoA via the succinate thiokinase reaction (see Fig 8 in Chapter 6 of this dissertation). The striking parallels in the regulation of these dehydrogenases allows for coordinated stimulation and inhibition, depending upon the metabolic conditions prevailing within the cell. Thus, increases in the redox poise (NADH/NAD+) or energy state (ATP/ADP) adversely affect the rate at which oxidative reactions proceed within the mitochondrial matrix. The importance of dehydrogenase regulation by Ca(II) is gaining increasing recognition. It is now doubtful that mitochondria simply buffer cytosolic concentrations of Ca(II) with their sophisticated set of carriers for Ca(II) uptake and efflux. Rather, it is believed that Ca(II) acts as a second messenger (McCormack and Denton, 1986), media- ting the coupling of hormonal stimulation of cardiac contraction to increases in the rate at which dehydrogenases produce NADH. The enhanced rates of NADH output result in higher rates of ATP production by mito- 11 chondria. It is therefore appropriate to envision the stimulation of cardiac mitochondrial oxidative metabolism as dependent upon the intramitochondrial concentrations of both ADP and Ca(II) (Hansford, 1985). In view of the structural and mechanistic complexity of these dehydrogenases, it would be of interest to ascertain whether additional forms of chemical regulation exist. Elucidating the full range of possible regulation would allow for a more comprehensive model of how mitochondria can so precisely meet the energy demands of myocardium. IV. ELECTRON TRANSFER CHAIN The reduced flavin (linked to succinate dehydrogenase) and pyridine nucleotides resulting from oxidative reactions in the matrix are used to transfer protons and electrons from substrates to the electron transfer chain of the mitochondrial inner membrane. The electrons (e-) enter the chain with a high redox potential. The energy of these electrons is defined by the following relationship: AGOx = -nFAEh (2) where AGOX is the Gibbs free energy of oxidation, n is the number of e- transferred during an oxidation-reduction reaction, F is the Faraday constant, and AEh is the observed difference in potential of the redox couple. AE is quantitated with use of the Nernst equation: h 2.303RT [electron acceptor] = : AEh E0 + log (3) nF [electron donor] where E0' is the standard redox (or midpoint) potential. The energy of 12 the electron is progressively released as it is transferred from one component of the respiratory chain to the next (Dutton g£_§l., 1970). The electron transfer chain is comprised of NADH:ubiquinone oxidoreduc- tase (complex I), coenzyme Q, succinatezubiquinone oxidoreducatse (complex II), ubiquinol:ferricytochrome c oxidoreductase (complex III), cytchrome c, and ferrocytochrome c: oxygen oxidoreducxtase (cytochrome oxidase or complex IV). A schematic diagram indicating the basic consti- tution of the respiratory chain is given in Fig 1. Comprehensive reviews detailing the sequence of electron carriers (Chance and Williams, 1956; Ernster and Lee, 196“; Chance, 1967), the nature of transition-metal mediated electron transfer (Williams, 1973). and the molecular consti- tution and structure of the complexes (Hatefi, 1985) and of cytochrome c (Ferguson-Miller §t_al., 1979; Margoliash and Bosshard, 1983) have appeared. The reader is referred to these if detail beyond that given here is desired. The primary biochemical function of heart mitochondria is to produce ATP at steady-state concentrations sufficient to sustain the beat-to-beat energy requirements of myocardium. Electron transport plays a fundamental role in harnessing redox free energy to the synthesis of ATP. Electrons undergo large decreases in redox potential (£300 mV) during transport through complexes I, III, and IV (Erecinska §t_al., 197A). The oxidative energy released is coupled to the transport of protons from the matrix to a variety of proposed locations (reviewed below). Thus, the AGox generated from substrate oxidation in the matrix is used to drive the transport of protons against their concentration gradient. Although it is not known how complex I translocates protons, Mitchell (1979) proposes that the flavin moiety of this complex macromolecular assembly ejects 13 Figure 1. Schematic Depiction of the Mitochondrial Electron Transport Chain FeS and cyt denote iron-sulfur centers and the cytochromes, respectively. The numerical subscripts on the b cytochromes represent the absorption maxima of their a bands. Complexes I, II, III, and IV and coenzyme Q are located within the mitochondrial inner membrane. Cyt c is a soluble protein found in the inter-membrane space. The FeS cluster in com- plex III is also known as the Rieske iron-sulfur protein. H+ on top of each box delineating complexes I, III, and IV are understood to be inside the matrix; H+ on the bottom of these boxes are understood to be transported out of the matrix to one or more possible locations, as described in the text. +Ii< #1 imam. Emfiimiegémmizn. .92 H .8358 J +1 +1 now .80 +I 3.6.950 E 3368 Emma“. m.“ :8 may. . mom , N104 m GP... 1066.2 9088 @658: a m 586 3.38 <8 15 protons from the matrix as it is sequentially oxidized and reduced. Proton transport by complex II occurs via either a Q cycle (Mitchell, 1976) or b cycle (Wikstrom et_al., 1976), in which 000 or the b cyto- chromes are believed to release protons from the inner membrane, respectively. In contrast, both Chance (1972) and Papa (1976) suggest that, by analogy with hemoglobin, the proton translocation catalyzed by complex II is achieved by membrane Bohr effects. Rigorous proof has not yet been promulgated for any of these theories. Cytochrome oxidase, the terminal component of the electron transport chain and the enzyme responsible for transferring e- to oxygen, has been definitively shown to be a redox-driven proton pump (Wikstrom, 1977; reviews: Wikstrom and Krab, 1979; Malmstrom, 1985). V. ATP SYNTHESIS Mitochondrial ATP synthesis is catalyzed by the FIFO-ATP synthetase. The F0 component of this enzyme is a transmembrane protein comprised of three subunits known as a, b, and c (Alonzo and Hacker, 1979). F1 is a soluble protein located in the mitochondrial matrix (Racker, 1968) and is made up of 5 subunits (Kagawa, 1978; Baird and Hammes, 1979; Maloney, 1982) with the following stoichiometric ratio (Amzel and Pedersen, 1983): Yde. F is connected to F via the Oligomycin sensitivity a383 1 O conferring protein (Senior, 1979) and the 6 subunit of F (Pedersen and 1 Carafoli, 1987a). The F moiety of the synthetase is a proton channel 0 (Alonzo and Racker, 1979; Seren et al., 1985); F can catalyze both ATP 1 synthesis and ATP hydrolysis (Boyer et al., 1982). It is currently believed that expression of the ATP hydrolytic mode by the enzyme is blocked in vivo by an ATPase inhibitor peptide (Schwerzmann and Pedersen, 16 1986). The F1FO-ATP synthetase can be irreversibly inhibited by treatment with Oligomycin A (Bertina et al., 197“), an antibiotic known to bind to a subunit of the FO moiety (Senior, 1979). As to be expected from an enzyme that is structurally this complex, the ATP synthetase is mechanistically quite sophisticated. The energy "stored" in the AuH+ produced by the electron transport chain is used indirectly to drive ATP synthesis on the ATP synthetase (Hutton and Boyer, 1979). ADP and P bind to high affinity catalytic sites on the B 1 subunits or at sites located at a-B interfaces (Pedersen and Carafoli, 1987b). Because the equilibrium constant for ATP synthesis from the bound substrates approximates to unity, the synthesis of ATP proceeds on the enzyme with no external need for free energy (Grubmeyer §£_§l., 1982; Eisenberg and Hill, 1985). Protons flowing down their concentration gradient and back across the mitochondrial inner membrane are believed to protonate functional groups within F0, particularly glutamate 61 in subunit 0 (Penefsky, 1985). This induces a conformational change in F O which in turn induces a conformational change in F This conformational 1. change in F concomitantly facilitates the dissociation of ATP at one 1 catalytic site and the binding of ADP and P with high affinity at 1 another (Boyer, 1975; reviews: Boyer, 1979; Boyer §£_§l., 1977; for a discussion of the thermodynamics of this process, see O'Neal and Boyer, 198”). Thus, in this "alternating site catalysis" model, ATP synthesis occurs by the reversal of ATP hydrolysis with ADP displacing a hydroxyl group from P1 (Boyer, 1984). It is currently believed that the free energy for phosphoanhydride bond formation in ATP results from the capacity of the ATP synthetase to conserve the energy released during the binding of ADP and P1 to the F protein (Jencks, 1975). Protons play 1 17 no part in the actual chemistry; rather, H+ are used to mediate the coupling of conformational changes to changes in the binding of subs- trates and products on F1. In order for oxidative phosphorylation to occur, a sufficient AUH+ must be avaialable to drive other processes as well. The net internaliz- ation of oxidizable substrate and of Pi will dissipate part of the ApH component of AuH+. The exchange of ADP3- in for ATPu- out is electrogenic (Klingenberg and Rottenberg, 1977). Therefore, the magnitude of the transmembrane electrical potential (Aw) decreases during adenine nucleo- tide translocation. It is clear that AuH+ is consumed by transport pro- cesses and by the need for continuously stimulating conformational changes in or about the F proton channel. The free energy of oxidation 0 is conserved in the sense that ATP, a compound with a relatively large (*7.3 Kcal/mole) negative free energy of hydrolysis, is produced. The continuous synthesis of ATP ensures that the cell's energy requirements can be met. The free energy change associated with the synthesis of ATP from ADP and Pi is defined by the cytosolic phosphorylation potential: [ATP] = o AGP AGP + RTln (u) [ADPJCPIJ where AGP° is the free energy of ATP synthesis under standard conditions and all other terms have their usual meaning. The indirect coupling between oxidation and phosphorylation via a proton gradient may be depicted as AGox AGP (5) AuH+ 18 In equation 5 the arrows are bidirectional because: (a) during the hydrolysis of ATP, the F1FO-ATP synthase translocates protons out of the matrix (Nicholls, 197A); and (b) during reverse electron transfer through complex I, NADH is regenerated (Chance, 1961; Chance and Hollunger, 1961). VI. THE H+, OXYGEN, AND ATP STOICHIOMETRIES OF OXIDATIVE PHOSPHORYLATION The stoichiometry of proton ejection by the electron transport chain (H+/O) and the number of protons consumed per molecule of ATP synthesized (H+/ATP) are still not established. The major obstacle to determining these stoichiometries is the fact that in those mitochondria preparations yet studied, substrate oxidation is not completely coupled to the phosphorylation of ADP. Imperfect coupling is a consequence of: (a) secondary proton transport pathways and (b) redox "slips" in the electron transport chain. Mitchell (1961) predicted that the mitochon- drial inner membrane has a finite permeability to H+. This prediction was later substantiated (Mitchell and Moyle, 1967; Nicholls, 197A; Stucki, 1976). Because proton transport out of the matrix is an energy dependent process, the non-productive circulation of these particles constitutes a futile cycle. The leak characteristics of the membrane depend upon whether the driving force for proton permeation is an electric field (matrix negative) or a pH gradient (matrix alkaline). As either ApH or Aw are increased, the membrane leak conductance is ohmic (Krishnamoorthy and Hinkle, 198A) and non-ohmic (Zoratti e£_al., 1986), respectively. In the former case, protons flow down an increasingly steeper concentration gradient back into the matrix; in the latter case, protons are drawn electrophoretically into the matrix at a rate that 19 increases exponentially as the electrical potential increases. Two types of proton "slips" have also been identified (Pietrobon et al., 1983, 1986; Pietrobon and Caplan, 1986 a and b; Pietrobon, 1986; Zoratti et al., 1983, 1986). Protons may flow through F without inducing ATP 0 release or may traverse the protein domain of a redox pump without stimulating reverse electron transfer. As discussed above, a substantial transmembrane proton flux is associated with the electroneutral uptake of various substrates. Decreases in coupling are also attributable to occasional errors in the redox pumps, which may transfer electrons without extruding protons from the matrix (Pietrobon and Caplan, 1985). These leaks and slips lead to underestimates of H+/O ratios by increas- ing the amount of oxygen that must be consumed in order to produce a pH gradient of a given magnitude. In view of the complications inherent to any analysis of proton flow, it comes as no surprise that a broad range of proton and ATP stoichiometries have been reported. Estimated values for the H+/O ratio of electron flow coupled to NADH oxidation in rat liver mitochondria include 6 (Mitchell and Moyle, 1967), 8 (Brand, 1979), 10 (Hinkle and Yu, 1979), 11 (Beavis and Lehninger, 1986 a_and b; Beavis, 19873 and b), 12 (Vercesi gt_§l., 1978; Lehninger, 198A), and 13 (Lemasters, 198“). As an example, in the 13 proton model of Lemasters (198A), the proton stoichiometries for complexes I, III, and IV are 5, A, and A, respectively (Freedman and Lemasters, 198A; Lemasters and Fleischman, 1987). It is now widely accepted that the H+/ATP ratio is equal to 3 (Brand g£_al., 1976). This value increases to A if the proton required for P1 transport is taken into account (Alexandre et al., 1978). From knowledge of the H+/O and ATP/H+ ratios, it is possible to estimate the ATP/O stoichiometry 20 for oxidative phosphorylation. The variable proton pumping stoichiometries obtained by different workers using different experimental methodologies underlies the widespread controversy raging over this issue. ATP/O ratios obtained during the oxidation of NADH-linked substrates are believed to equal 2.0 (Hinkle and Yu, 1979), 2.75 (Beavis and Lehninger, 1986b). and 3.25 (Lemasters, 198“; see also Lemasters §£_§l., 198A). It is important to note here that the values reported in the latter two works are theore- tical, not empirical. Due to high rates of respiration in the absence of ADP, these workers had to apply sophisticated mathematical corrections to their data in order to obtain estimates for what the mechanistic stoichiometry of oxidative phosphorylation could be if the mitochondria used had been perfectly coupled. Whatever the final verdict on these stoichiometries will be, it must be kept in mind that the amount of ATP that can be produced is thermodynamically limited by the degree to which mitochondria can couple the oxidation of substrates to the phosphorylation of ADP. This limitation is described by the relationship: -AGox/AGP a ATP/O. (6) VII. ALTERNATIVE COUPLING SCHEMES There are an increasing number of findings which suggest that the chemiosmotic mechanism as first envisioned by Mitchell (1961) may not correctly describe the path followed by protons during oxidative phos- phorylation. It is generally accepted that protons are transported out of the matrix during respiration and that the gradient formed is likely the sole and obligatory intermediate of oxidative phosphorylation. In addition to the studies discussed above, this conclusion is supported by 21 the observations that an artifically generated AHH+ can drive ATP synthe- sis (Thayer and Hinkle, 1975) and that the Au and ApH components of AUH+ are kinetically equivalent in that either can drive ATP synthesis indepen- dent of the other (Jensen §£_§l., 1986). The chemiosmotic theory cannot, however, accomodate certain properties of the AuH+ generated across mito- chondrial membranes (reviews: Fillingame, 1980; Ferguson and Sorgato, 1982; Ferguson and Parsonage, 198A; Westerhoff g£_al., 1983, 198Aa). First, the rates of electron transport and of ADP phosphorylation are relatively insensitive to the magnitude of AUH+ (Padan and Rottenberg, 1973; Azzone et_al., 1978; Branca §t_§l., 1981; Duszynski §£_§l., 198M; for especially lucid demonstrations of this, see Sorgato gt_§l., 1978 and Zoratti gt_§l., 1982). Second, at equilibrium, AGP is not proportional to AuH+ (Nesterhoff et_§l., 1981 and references therein). The ratio of these thermodynamic potentials defines the proton stoichiometry of the F1FO-ATP synthetase: ATP AGP/AuH+ - nH (7) That AGP varies independent of AuH+ suggests that the proton stoichiometry of the ATP synthetase varies as a function of the magnitude of the proton electrochemical potential gradient. This is considered to be unlikely. Third, a variety of inhibitor studies (reviewed in Westerhoff gt_§l., 198MB) suggest that the proton pumps of the electron transport chain and the ATP synthetase are directly coupled. The explanation most often used to account for these anomalies is to postulate that protons are not transported from one bulk aqueous phase (the matrix) to another (the intermembrane space). Although practically untestable, Williams (1961; 198”; review: 1978) suggests that the high 22 energy intermediate of oxidative phosphorylation is an ensemble of energized protons contained within the highly controlled environment of the mitochondrial inner membrane. Kell (1979) has recently proposed that protons are pumped by the electron transport chain into the Gouy-Chapman unstirred layer along the outer face of the inner membrane. A somewhat refined version of this theory is that advanced by Westerhoff §t_§l. (198A b and 9). These authors argue that the three anomalies discussed above may be reconciled with a chemiosmotic-type mechanism if it is assumed that a primary and secondary proton pump are spatially closely situated so as to create a functional proton compartment or "proton domain." Thus, protons would be transferred directly from a redox pump to the FO channel of the ATP synthetase. Because the total volume of the proton domain would be small, very few protons would have to be trans- ported within the coupling unit in order to attain an electrochemical potential sufficient to drive ATP synthesis. A convincing statistical mechanical analysis of such a coupling scheme has been presented (Westerhoff and Chen, 1985). Seen in this light, the bulk phase AHH+ would therefore be essentially irrelevant to the coupling of oxidation to phosphorylation. The theme common to these three theories is that "energized" protons are localized either within or on the membrane. In contrast, according to Mitchell's theory, protons are delocalized because they are allowed to equilibrate with the bulk aqueous phase of a given compartment. Precise, convincing identification of the locale into which the electron transport chain pumps protons is still an unresolved problem. If proton translocation is indeed localized, it will be of interest to determine if this requires further revision in the experimentally determined stoichiometries of oxidative phosphorylation. 23 VIII. THE CONTROL OF RESPIRATION Consistent with the notion that the teleological aim of oxidative phosphorylation is to maximize the efficiency with which free energy intrinsic to metabolites can be released and conserved, the rate of respiration must be a highly controlled process. By control is meant the capacity of mitochondria to limit the rate of electron transport in the absence of ADP. If this were not the case, large amounts of energy would simply be released into the environment in the form of heat. This type of free energy dissipation has been shown with the use of such uncouplers as lipophilic protonophores (Westerhoff and Kell, 1985) and a transmem- brane protein in brown fat mitochondria known as thermogenin (Nicholls, 1979). Under normal metabolic conditions, however, it is of the essence that organisms commit the free energy of substrate oxidation to the synthesis of ATP. The reason for this is especially clear in a tissue such as myocardium: ATP is absolutely required to mediate cross-link formation between actin and myosin, and for the regulation of cytosolic cation concentrations by the Ca(II)- and Na+/K+-ATPases located in the sarcoplasmic reticulum and the plasmalemma, respectively. A considerable amount of experimental work indicates that ADP is the metabolite primarily responsible for regulating the rate of respi- ration. Early work (Lardy and Wellman, 1952; Chance and Williams, 1956) on isolated mitochondria shows that maximal rates of respiration could only be elicited with the addition of ADP. In the absence of ADP, mito- chondria respire at a much lower "resting" rate. In order to quantify the relative stimulation of respiration by ADP, Chance and Williams (1956) formulated the respiratory control ratio. State 3 respiration is the rate of respiration obtained in the presence of ADP, Pi’ and 2N oxidizable substrate, whereas state A is the respiratory rate observed in the presence of P and oxidizable substrate only. The definition of 1 states 1, 2, and 5 may be found in the monograph by Nicholls (1982). The respiratory control ratio (RCR) is the quotient of state 3 to state A respiration. It follows that the higher the magnitude of the RCR, the greater is the extent of control by ADP on the rate of respiration. Low RCR values suggest that the mitochondria are poorly coupled since rates of respiration are high in the absence of added ADP. The theory that ADP controls respiratory rate has been substantiated in vivo in skeletal 3TP-nuclear magnetic resonance spectroscopy (Chance et al., muscle using 1982, 1985, and 1986). Similar experiments in working heart muscle, however, do not support this conclusion (Balaban gt_§l., 1986). Instead, these authors conclude that the redox poise (NADH/NAD+) regulates respi- ratory rate. Consistent with the latter view are the findings of Hansford (1980) and of Denton and McCormack (1985) which indicate that since Ca(II) and ADP regulate the activity of the sub-strate dehydrogenases, these enzymes limit respiration by controlling rates of NADH production. Additional theories describing the molecular basis for respiratory control have been advanced. The work of Wilson and colleagues (Wilson gt_gl., 197A 3 and b; Holian et_§l., 1977; Erecinska gt_§l., 1977; Forman and Wilson, 1983; review: Erecinska and Wilson, 1982) suggests that the rate of respiration varies with the magnitude of AG . According P to this theory, the first two coupling sites of the respiratory chain (complexes I and III) are at near equilibrium with AGP, such that AG 1,111 :3 AG ' . ox p- AS the magnitude 0f AGP changes, the rate of respi- ration will either increase or decrease depending upon what the system must do in order to reestablish equilibrium (for a theoretical treatment 25 of this, see Bohnensack, 1981). Other workers (Slater et_2l., 1973; Davis and Lumeng, 1975; Kuster gt_§l., 1976) suggest that the rate of respiration depends on the ATP/ADP ratio because ATP is a competitive inhibitor of ADP binding sites on the adenine nucleotide translocase, the enzyme believed by some (Lemasters and Sowers, 1979; Klingenberg, 1980) to limit the rate of oxidative phosphorylation. The notion that the adenylate "energy charge" (Atkinson, 1971) regulates the rate of respiration was entertained for some time, but was later disproved (Erecinska §£_al., 1977). Clearly, because they are the substrates of the ATP synthetase and are used to regulate the activity of a number of mitochondrial enzymes, the adenine nucleotides and Pi must play a role in the regulation of respiratory rate. However, whether this control is kinetic, thermodynamic, or incorporates elements of both is an issue that is yet to be resolved. In addition to the adenine nucleotide translocase and the substrate dehydrogenases, other enzymes have been shown to play a role in the phenomenon of respiratory control. Among these are NADH dehydrogenase (complex I) and the F1FO-ATP synthetase (Doussiere §t_§l., 198A), cyto- chrome oxidase (complex IV) via allosteric modulation (Kadenbach, 1986), and the Pi transport protein (Mazat §t_§l., 1986). Given the fact that mitochondria are extremely complex both kinetically and thermodynamically, it is entirely possible that some or all of these different mechanisms of control are utilized by these organelles in order to attain the necessary level of responsiveness to rapid, fluctuational demands for energy in the cytosol. By applying the principles of control theory (Kacser and Burns, 1973; Heinrich and Rapoport, 197” a and b; Groen, 198“), it has been possible to show that the control of respiration is distributed 26 among a number of enzymes and transport proteins. In control theory, inhibitors are used to quantitate the control exerted by different enzymes over the rate at which mitochondria respire in either the presence or absence of ADP. When mitochondria oxidize succinate during state 3, the rate of respiration appears to be limited by the activity of the dicarboxylate carrier, bc complex, cytochrome oxidase, and the 1 adenine nucleotide translocase (Groen gt_§l., 1982; Tager §£_gl., 1983). Though not determined by control theory, it has been shown that during state A the rate of respiration is limited by proton leaks (Groen 32 al., 1982; Bohnensack §£_gl., 1982), Ca(II) cycling (Stucki and Ineichen, 197A), and the rate at which ADP is regenerated from ATP by ATPases in the extramitochondrial space (Masini §£_§l., 1983 a and 2; 198“). It is obvious from the number of different theories put forth that much remains to be elucidated about the nature of respiratory control. A likely reason for the diversity of results obtained is that there are many differences in experimental design. For instance, depending upon which substrate is used to drive respiration and at what concentration it is available, different enzymes will play roles in respiratory control. Moreover, it is not yet known whether all of the enzymes involved in energy transduction have been identified. It is also unknown whether all of the relevant forms of enzyme regulation have been elucidated. Before these gaps are filled in, no unifying and comprehensive model of respira- tory control can be formulated. 27 REFERENCES Alexandre, A., Reynafarje, B., and Lehninger, A.L. (1978) Proc. Nat. Acad. Sci. (U.S.A.) 75: 5296-5300. Alfonzo, M., and Racker, E. (1979) Can. J. Biochem. 57: 1351-1358. Atkinson, D.E. (1971) In Metabolic Pathways (ed. G.M. Grrenberg) 3rd ed., Vol. V, pp. 1-21. Academic Press, New York. Azzone, G.F., Pozzan, T., Masari, 8., and Bragadin, M. (1978) Biochim. Biophys. Acta 501: 296-306. Baird, B.A., and Hammes, 0.0. (1979) Biochim. Biophys. Acta 5N9: 31-53. Balaban, R. S., Kantor, H. L., Katz, L. A., and Briggs, R. W. (1986) Science (Washington, D. C. ) 232. 1121- 1123. Beavis, A.D., and Lehninger, A.L. (1986a) Eur. J. Biochem. 158: 307-314. Beavis, A.D., and Lehninger, A.L. (1986b) Eur. J. Biochem. 158: 315-322. Beavis, A.D. (19873) J. Biol. Chem. 262: 6165-6173. Beavis, A.D. (1987b) J. Biol. Chem. 262: 617A-6181. Bertina, R.M., Steenstra, J.A., and Slater, E.C. (197A) Biochim. Biophys. Acta 368: 279-297. Bohnensack, R. (1981) Biochim. Biophys. Acta 634: 203-218. Bohnensack, R., Kuster, U., and Letko, G. (1982) Biochim. Biophys. Acta 680: 271-280. ' Boyer, P.D. (1975) FEBS Lett. 50: 91-94. Boyer, P.D., Chance, 8., Ernster, L., Mitchell, P., Racker, E., and Slater, E.C. (1977) Ann. Rev. Biochem. #6: 955-1026. Boyer, P.D. (1979) In Membrane Bioenergetics (Lee, C.P., Schatz, G., and Ernster, L., edS) pp. A61-u79. Addison-Wesley, Reading, Massachusetts. Boyer, P.D. (198A) Symposium on Frontiers in Bio-Organic Chemistry and Molecular Biology (Ovchinnikov, Y.A.), pp. 1-8. Elsevier, Amsterdam. Branca, D., Ferguson, S.J., and Sorgato, C.M. (1981) Eur. J. Biochem. 116: 3A1-3N6. ‘ Brand, M.D., Reynafarje, B., and Lehninger, A.L. (1976) Proc. Nat. Acad. Sci. (U.S.A.) 73: u37-uu1. 28 Campo, M.L., Zhang, C.J., and Tedeschi, H. (1984) Biochem. Soc. Trans. 12: 384-386. Chance, 8., and Williams, G.R. (1956) Adv. Enzymol. 17: 65-134. Chance, B. (1961) J. Biol. Chem. 236: 1534-1543. Chance, 8., and Hollunger, G. (1961) J. Biol. Chem. 236: 1577-1584. Chance, B. (1967) Biochem. J. 103: 1-18. Chance, B. (1972) FEBS Lett. 23: 3-20. Chance, 3., Eleff, 8., Bank, W., Leigh, J.S., and Warnell, R. (1982) Proc. Nat. Acad. Sci. (U.S.A.) 79: 7714-7718. Chance, 8., Leigh, J.S., Clark, B.J., Maris, J., Kent, J., Nioka, S., and Smith, D. (1985) Proc. Nat. Acad. Sci. (U.S.A.) 82: 8384-8388. Chance, 8., Leigh, J.S., Kent, J., McCully, K., Nioka, S., Clark, B.J., Maris, J.M., and Graham, T. (1986) Proc. Nat. Acad. Sci. (U.S.A.) 83: 9458-9462. Chen, R.F., and Plaut, G.W.E. (1963) Biochemistry 2: 1023-1032. Denton, R.M., and McCormack, J.G. (1985) Am J. Physiol. 249: E453-8554. Doussiere, J., Ligeti, E., Brandolin, 0., and Vignais, P.V. (1984) Biochim. Biophys. Acta 766: 492-500. Duszynski, J., Bogucka, K., and Wojtczak, L. (1984) Biochim. Biophys. Acta 767: 540-547. Dutton, P.L., Wilson, D.F., and Lee, C.P. (1970) Biochemistry 9: 5077-5082. ‘ Eisenberg, E., and Hill, T.L. (1985) Science (Washington, D.C.) 227: 999-1006. ‘ Erecinska, M., Veech, R.L., and Wilson, D.F. (1974) Arch. Biochem. Biophys. 160: 412-424. Erecinska, M., Stubbs, M., Miyata, Y., Ditre, C.M., and Wilson, D.F. (1977) Biochim. Biophys. Acta 462: 20-35. Erecinska, M., and Wilson, D.F. (1982) J. Membrane Biol. 70: 1-14. Ernster, L., and Schatz, G. (1981) J. Cell Biol. 91: 2273-2558. Ferguson, S.J., and Parsonage, D. (1982) Biochem. Soc. Trans. 12: 416-420. Ferguson, S.J., and Sorgato, C. (1982) Ann. Rev. Biochem. 51: 185-217. 29 Ferguson, S.J. (1985) Biochim. Biophys. Acta 811: 47-95. Freedman, J.A., and Lemasters, J.J. (1984) Biochem. Biophys. Res. Comm. 125: 8-13. Fillingame, R.H. (1980) Ann. Rev. Biochem. 49: 1079-1113. Forman, N.0., and Wilson, D.F. (1983) J. Biol. Chem. 258: 8649-8655. Goebell, R., and Klingenberg, M. (1964) Biochem. Z. 340: 441-464. Groen, A.K., Wanders, R.J., Westerhoff, H.V., van der Meer, R., and Tager, J.M. (1982) J. Biol. Chem. 257: 2754-2757. Groen, A.K. (1984) Dissertation, University of Amsterdam. Grubmeyer, 0., Cross, R.L., and Penefsky, H.S. (1982) J. Biol. Chem. 257: 12,092-12,100. Hansford, R.0. (1976) J. Biol. Chem. 251: 5483-5489. Hansford, R.0. (1980) Curr. Top. Bioenergetics 10: 217-278. Hansford, R.0., and Castro, F. (1981) Biochem. J. 198: 525-533. Hansford, R.G. (1985) Rev. Physiol. Biochem. Pharmacol. 102: 2-72. Hassinen, I.E., and Hiltunen, K. (1975) Biochim. Biophys. Acta 408: 319-330. Hassinen, I.E. (1986) Biochim. Biophys. Acta 853: 135-151. Hatefi, Y. (1985) Ann. Rev. Biochem. 54: 1015-1069. Haworth, R.A., Hunter, D.R., and Berkoff, H.A. (1980) FEBS Lett. 110: 216-218. Heinrich, R., Rapoport, T.A. (1974a) Eur. J. Biochem. 42: 89-95. Heinrich, R., Rapoport, T.A. (1974b) Eur. J. Biochem. 42: 97-105. Hoek, J.B., and Njogu, R.M. (1976) FEBS Lett. 71: 341-346. Hinkle, P.C., and Yu, M.L. (1979) J. Biol. Chem. 254: 2450-2455. Holian, A., Owen, 0.8., and Wilson, D.F. (1977) Arch. Biochem. Biophys. 181: 164-171. Hill, T.L., and Eisenberg, E. (1981) Quart. Rev. Biophys. 14: 463-511. Hucho, F., Randall, D. D., Roche, T. E., Burgett, M. W., Pelley, J. W., and Reed, L. J. (1972) Arch. Biochem. Biophys. 159: 328- 340. 3O Hutton, R.L., and Boyer, P.D. (1979) J. Biol. Chem. 254: 9990-9993. Jencks, W.P. (1975) Adv. Enzymol. 51: 75-106. Kacser, H., and Burns, J.A. (1979) Biochem. Soc. Trans. 7: 1149-1160. Kadenbach, B. (1986) J. Bioenerg. Biomemb. 18: 39-54. Kagawa, Y. (1978) Biochim. Biophys. Acta 505: 45-93. Kauppinen, R.A., Hiltunen, J.K., and Hassinen, I.E. (1983) Biochim. Biophys. Acta 725: 425-433. Kauppinen, R.A., Hiltunen, J.K., and Hassinen, I.E. (1986) In Myocardial and Skeletal Muscle Bioenergetics (Bratbar, M., ed), pp. 331-341. Plenum, New York. Kell, D.E. (1979) Biochim. Biophys. Acta 549: 55-99. Kerbey, A.L., Randle, P.J., Cooper, R.H., Whitehouse, S., Pask, H.T., Denton, R.M. (1976) Biochem. J. 154: 327-348. Klingenberg, M., and Rottenberg, H. (1977) Eur. J. Biochem. 73: 125-130. Klingenberg, M. (1980) J. Membr. Biol. 56: 97-105. Klingenberg, M. (1984) Biochem. Soc. Trans. 12: 390-393. Kramer, R.S., and Pearlstein, R.D. (1983) Proc. Nat. Acad. Sci. (U.S.A.) 80: 5807-5811. Kramer, R., and Klingenberg, M. (1982) Biochemistry 21: 1082-1089. Kuster, U., Bohnensack, R., and Kunz, W. (1976) Biochim. Biophys. Acta 440: 391-402. LaNoue, K.F., and Schoolwerth, A.C. (1979) Ann. Rev. Biochem. 48: 871-922. Latipaa, P.M., Peuhkurinen, K.J., Hiltunen, J.K., and Hassinen, I.E. (1985) J. Mol. Cell. Cardiol. 17: 1161-1171. Lehninger, A.L., Vercessi, A., and Bababunmi, E.A. (1978) Proc. Nat. Acad. Sci. (U.S.A.) 75: 1690-1694. Lehninger, A.L. (1984) Biochem. Soc. Trans. 12: 386-388. Lemasters, J.J., and Sowers, A.E. (1979) J. Biol. Chem. 254: 1248-1251. Lemasters, J.J., Grunwald, R., and Emaus, R.K. (1984) J. Biol. Chem. 259: 3058-3063. 31 Lemasters, J.J. (1984) J. Biol. Chem. 259: 13,123-13,130. Lemasters, J.J., and Fleishman, K.E. (1987) Biochem. BiOphys. Res. Comm. 142: 176-182. ‘ Linn, T.C., Pettit, F.H., and Reed, L.J. (1969a) Proc. Nat. Acad. Sci. (U.S.A.) 62: 234-241. Linn, T.C., Pettit, F.H., Hucho, F., and Reed, L.J. (1969b) Proc. Nat. Acad. Sci. (U.S.A.) 64: 227-234. Lotscher, R., Winterhalter, K.H., Carafoli, E., and Richter, C. (1980) Eur. J. Biochem. 110: 211-216. Lowenstein, J.M. (1967) In Metabolic Pathways (ed. G.M. Greenberg) 3rd ed., vol I, pp. 146-270. Academic Press, New York. Lysiak, W., Toth, P.P., Suelter, C.H., and Bieber, L.L. (1986) J. Biol. Chem. 261: 13,698-13.703. Lysiak, W., Lilly, K., DiLisa, F., Toth, P.P., and Bieber, L.L. (1988) J. Biol. Chem. 263: 1151-1156. Malmstrom, 8.0. (1985) Biochim. Biophys. Acta 811: 1-12. Maloney, P.C. (1982) J. Membr. Biol. 67: 1-12. Malpress, F.H. (1982) Biochem. Soc. Trans. 12: 399-401. Margoliash, E., and Bosshard, H.R. (1983) Trends Biochem. Sci. 8: 316-320. Martin, W.H., Beavis, A.D., and Garlid, K.D. (1984) J. Biol. Chem. 259: 2062-2065. Masini, A., Ceccarelli-Stanzani, D., and Musctatello, U. (1983a) FEBS Lett. 160: 137-140. ' Masini, A., Ceccarelli-Stanzani, D., and Musctatello, U. (1983b) Biochim. Biophys. Acta 724: 251-257. Masini, A., Ceccarelli-Stanzani, D., and Musctatello, U. (1984) Biochim. Biophys. Acta 767: 130-137. Mazat, J.P., Jean-Bart, E., Rigoulet, M., and Guerin, B. (1986) Biochim. Biophys. Acta 849: 7-15. McCormack, J.G., and Denton, R.M. (1979) Biochem. J. 180: 533-544. McCormack, J.G., and Denton, R.M. (1981) Biochem. J. 196: 619-624. Meijer, A.J., and Van Dam, K. (1981) In Membrane Transport (Bonting, S.L., and de Pont, J.J., eds) pp. 235-256. Elsevier/North Holland 32 Biomedical Press. Mela-Hiker, L.M., and Bukoski, R.D. (1985) Ann. Rev. Physiol. 47: 645-663. ' Mitchell, P. (1961) Nature (London) 191: 144-148. Mitchell, P., and Moyle, J. (1967) Biochem. J. 105: 1147-1162. Mitchell, P. (1975) FEBS Lett. 50: 95-97. Mitchell, P. (1976) J. Theor. Biol. 62: 327-376. Mitchell, P. (1979a) Science (Washington, D.C.) 206: 1148-1159. Mitchell, P. (1979b) Eur. J. Biochem. 95: 1-20. Mitchell, P. (1985) J. Biochem. 97: 1-18. Nakashima, R.A., Dordick, R.S., and Garlid, K.D. (1982) J. Biol. Chem. 257: 12,540-12,545. Nicholls, D.0. (1979) Biochim. BiOphys. Acta 549: 1-29. Nicholls, D.G. (1982) Bioenergetics: An Introduction to the Chemiosmotic Theory. Academic Press, New York. O'Neal, 0.0., and Boyer, P.D. (1984) J. Biol. Chem. 259: 5761-5767. Oram, J.F., Bennetch, S.L., and Neely, J.R. (1973) J. Biol. Chem. 248: 5299-5309. ' Padan, E., and Rottenberg, H. (1973) Eur. J. Biochem. 40: 431-437. Palmieri, F., Quagliariello, E., and Klingenberg, M. (1970) Eur. J. Biochem. 17: 230-238. Palmieri, F., Prezioso, 0., Quaglieriello, E., and Klingenberg, M. (1971) Eur. J. Biochem. 22: 66-74. Pande, S.V., and Parvin, R. (1978) J. Biol. Chem. 253: 1565-1573. Papa, S., Francavilla, A., Pardies, 0., and Meduri, B. (1971) FEBS Lett. 12: 285-288. Papa, S. (1976) Biochim. Biophys. Acta 456: 39-84. Pedersen, P.L., and Carafoli, E. (19873) Trends Biochem. Sci. 12: 146-150. Pedersen, P.L., and Carafoli, E. (19872) Trends Biochem. Sci. 12: 186-189. 33 Penefsky, H.S. (1985) Proc. Nat. Acad. Sci. (U.S.A.) 82: 1589-1593. Peuhkurinen, K.J. (1984) J. Mol. Cell. Cardiol. 16: 487-495. Pettit, F.H., Pelley, J.w., and Reed, L.J. (1975) Biochem. Biophys. Res. Comm. 65: 575-582. ‘ Pfaff, E., and Klingenberg, M. (1968) Eur. J. Biochem. 6: 66-79. Pfaff, E., Heldt, H.W., and Klingenberg, M. (1969) Eur. J. Biochem. 10: 484-493. ’ Pietrobon, D., Zoratti, M., and Azzone, 0.F. (1983) Biochim. Biophys. Acta 723: 317-321. Pietrobon, D., and Caplan, S.R. (1985) Biochemistry 24: 5764-5770. Pietrobon, D. (1986) Bioeletrochem. Bioenerg. 15: 193-209. Pietrobon, D., Zoratti, M., Azzone, 0.F., Caplan, S.R. (1986) Biochemistry 25: 767-775. Pietrobon, D., and Caplan, S.R. (1986a) Biochemistry 25: 7682-7690. Pietrobon, D., and Caplan, S.R. (1986b) Biochemistry 25: 7690-7696. Puskin, J.S., Gunter, T.E., Gunter, K.K., and Russell, P.R. (1976) Biochemistry 15: 3834-3842. Racker, E. (1968) Scientific American 218: 32-39. Reed, L.J. (1974) Acc. Chem. Res. 7: 40-56. Riley, W.W., and Pfeiffer, D.R. (1986) J. Biol. Chem. 261: 28-31. Rottenberg, H., and Scarpa, A. (1974) Biochemistry 13: 4811-4819. Rottenberg, H. (1986) Methods Enzymol. 125: 3-15. Schwerzmann, K., and Pedersen, P.L. (1986) Arch. Biochem. Biophys. 250: 1-18. Senior, A.E. (1979) Methods Enzymol. 55: 391-397. Seren, S., Caporin, 0., Galiazzo, F., Lippe, 0., Ferguson, S.J., and Sorgato, M.C. (1985) Eur. J. Biochem. 152: 373-379. Slater, E.C., Rosing, J., and M01, A. (1973) Biochim. Biophys. Acta 292: 534-553. ' Sorgato, M.C., Ferguson, S.J., Kell, D.B., and John, P. (1978) Biochem. J. 174: 237-256. 34 Stucki, J.W. (1976) Eur. J. Biochem. 68: 551-562. Tager, J.M., Wanders, R.J.A., Groen, A.K., Kunz, W., Bohnensack, R., Kuster, U., Letko, 0., Bohme, 0., Duszynski, J;, and Wojtczak, L. (1983) FEBS Lett. 151: 1-9. Tanford, C. (1983) Ann. Rev. Biochem. 52: 379-409. Thayer, W.S., and Hinkle, P.C. (1975) J. Biol. Chem. 250: 5330-5335. Van Dam, K., Westerhoff, H.V., Rutgers, M., Bode, J.A., de Jonge, P.C., Bos, M.M.,‘and van den Berg, 0. (1981) In Vectorial Reactions in Electron and Ion Transport in Mitochondria and Bacteria (Palmieri, F., ed), pp. 389-397. Elsevier/North Holland Biomedical Press. Van Dam, K., Woelders, H., Colen, A.M., and Westerhoff, H.V. (1984) Biochem. Soc. Trans. 12:401-402. ‘ ' Van Der Bend, R.L., Petersen, J., Berden, J.A., Van Dam, K., and Westerhoff, H.V. (1985) Biochem. J. 230: 543-549. Vary, T.C., Reibel, D.K., and Neely, J.R. (1981) Ann. Rev. Physiol. 43: 419-430. Vercesi, A., Reynafarje, B., and Lehninger, A.L. (1978) J. Biol. Chem. 253: 6379-6385. Wang, J.H. (1983) Ann. Rev. Biophys. Bioeng. 12: 21-34. Westerhoff, H.V., Simonetti, A.L.M., and Van Dam, K. (1981) Biochem. J. 200: 193-202. Westerhoff, H.V., Colen, A.M., and Van Dam, K. (1984a) Biochem. Soc. Trans. 11: 81-85. ' Westerhoff, H.V., Melandri, B.A., Venturoli, 0., Azzone, 0.F., and Kell, D.B. (1984b) FEBS Lett. 165: 1-5. Westerhoff, H.V., Melandri, B.A., Venturoli, 0., Azzone, 0.F., and Kell, D.B. (1984c) Biochim. Biophys. Acta 768: 257-292. Westerhoff, H.V., and Chen, Y.D. (1985) Proc. Nat. Acad. Sci. (U.S.A.) 82: 3222-3226. Westheimer, F.H. (1987) Science (Washington, D.C.) 235: 1173-1178. Wikstrom, M. (1977) Nature (London) 266: 271-273. Wikstrom, M., and Krab, K. (1979) Biochim. Biophys. Acta 549: 177-222. Wikstrom, M., Krab, K., and Saraste, M. (1981) Ann. Rev. Biochem. 50: 623-655. 35 Williams, R.J.P. (1961) J. Theor. Biol. 1: 1-17. Williams, R.J.P. (1973) Biochem. Soc. Trans. 1: 1-26. Williams, R.J.P. (1978) Biochim. Biophys. Acta 505: 1-44. Williams, R.J.P. (1984) Biochem. Soc. Trans. 12: 396-399. Williamson, J.R. (1974) Biochem. J. 93: 97-106. Williamson, J.R. (1979) Ann. Rev. Physiol. 41: 485-506. Williamson, J.R., and Cooper, R.H. (1980) FEBS Lett. 117: K73-K85. Wilson, D.F., Stubbs, M., Veech, R.L., Erecinska, M., and Krebs, H. (1974) Biochem. J. 140: 57-64. Wilson, D.F., Stubbs, M., Oshino, N., and Erecinska, M. (1974) Biochemistry 13: 5305-5311. Yeaman, S.J., Hutchinson, E.T., Roche, T.E., Pettit, F.H., Brown, J.R., Reed, L.J., Watson, D.C., and Dixon, G.A. (1978) Biochemistry 17: 2364-2370. ' Zoratti, M., Pietrobon, D., and Azzone, 0.F. (1983) Biochim. Biophys. Acta 723: 59-70. Zoratti, M., Favaron, M., Pietrobon, D., and Azzone, 0.F. (1986) Biochemistry 25: 760-767. Chapter 2 ISOLATION OF HIGHLY COUPLED HEART MITOCHONDRIA IN HIGH YIELD USING A BACTERIAL COLLAGENASE 36 37 INTRODUCTION Numerous methods have been developed for the isolation of mitochon- dria from a variety of tissues (Nedergaard and Cannon, 1979; Ernster and Schatz, 1981). These procedures usually involve homogenization of the tissue by mechanical means, and separation of the mitochondria containing fraction by differential centrifugation. The consistency of myocardium often necessitates the use of a nonspecific proteinase such as Nagarse to facilitate the disruption process (Chance and Hagihara, 1963; Palmer g£_al., 1977; Mela and Seitz, 1979). However, as demonstrated by Pande and Blanchaer (1970), mitochondria isolated in the presence of Nagarse lose all long chain fatty-acyl-CoA synthase (E.C. 2.3.1.86) activity, and likely lose other outer membrane associated enzymes as well. Heart mitochondria isolated according to currently available methods also show significant rates of state 4 respiration. Although Oligomycin (Masini 33 al., 1983a, 1984) and an enzymatic ADP sink (Bishop and Atkinson, 1984) have been successfully used to reduce the rate of oxygen consumption in state 4, it is recognized that state 4 respiration is phenomenologically complex (Stucki and Ineichen, 1974; Nicholls, 1974; Stucki, 1976; Lemasters and Hackenbrock, 1980; Groen e£_gl., 1982; Masini gt_al., 1983b) and difficult to suppress or eliminate. Because collagen plays a major role in the organization of cells into tissues (Hay, 1981), and myocardium is known to have a highly developed collagen infrastructure (Caulfield and Borg, 1979), a colla- genase (clostridiopeptidase A, E.C. 3.4.24.3) was selected to facilitate 38 degradation of intercellular connections. The use of collagenase in place of a nonspecific proteinase has the advantage of being specific toward collagen, with no activity toward other proteins (Seifter and Harper, 1971). One potential obstacle to the use of collagenase for the disruption of heart muscle is that the enzyme has an absolute require- ment for Ca(II) (Bond and Van Wart, 1984), and addition of this divalent cation might be expected to damage the mitochondria (Lehninger e£_§l., 1967; Lehninger, 1970; Stadhouders, 1981). This difficulty is circum- vented by allowing the collagenase to be activated by endogenous Ca(II) released from the myocardium during homogenization. In this chapter a collagenase-facilitated isolation of highly coupled heart mitochondria in high yield with morphologically intact inner and outer membranes is described. 39 EXPERIMENTAL PROCEDURES I. MATERIALS .Water. In order to isolate highly coupled mitochondria, it is critical that the water used be as pure as possible. For the prepar- ations described herein, water is glass distilled, deionized by passage over a column of Dowex MR-3 (Sigma Chemical Co., St. Louis, MO) mixed bed ion-exchange resin, and then redistilled from alkaline permanganate. The last step of this procedure may be essential for removing organic contaminants. Collagenase. Collagenase (Sigma type VII, lot 33F-6819) is reconsti- tuted prior to each experiment in a solution containing 0.225 M mannitol and 0.075 M sucrose. Reagent . The following substances were reagent grade or better, used without further purification, and obtained from the sources noted: sucrose (RNase free), coenzyme A (CoA; type III-L), fatty acid free bovine serum albumin (BSA; fraction V), ethylene glycol bis(B-aminoethyl ether)-N,N,N',N'-tetra-acetic acid (EGTA), 2-(4-iodophenyl)-3-(4-nitro- phenyl)-5-phenyltetrazo-lium chloride (INT), EDTA, rotenone, and the sodium salts of ADP and pyruvic, L-malic, L-glutamic, DL-B-hydroxybutyric, a-ketoglutaric, and succinic acids (Sigma); Tris (Boehringer Mannheim Biochemicals, Indianapolis, IN); mannitol, potassium permanganate, and phosphoric acid (Mallinckrodt, Paris, KY); palmitic acid (Nutritional Biochemical Corp., Cleveland, OH); and L-carnitine (gift of L.L. Bieber, Michigan State University). 7-Deoxycholic acid (Sigma, grade III) was 40 recrystallized three times from 80% acetone. Cytochrome 0 (horse heart, Sigma type VI) was purified and reduced as described by Thompson and Ferguson-Miller (1983). Animals. Fertile single comb white leghorn eggs were purchased from a local farm and allowed to hatch in incubators at Michigan State University's Department of Animal Science. Chicks were maintained on a diet of Chick 00125 feed (Kent Feeds, Inc., Muscatine, IA) and were not starved prior to sacrifice. Glassware. It is critical that all of the glassware used be as clean as possible. As described by Mela and Seitz (1979), the glassware used to isolate mitochondria must never be washed with detergent. Rather, glassware is cleaned with absolute ethanol and rinsed exhaustively with water distilled from permanganate. Departures from this regimen ensure that the mitochondria one isolates will have average to poor coupling parameters. II. ENZYME ASSAYS Cytochrome c oxidase (cyt aa E.C. 1.9.3.1) concentrations are 3. quantitated from difference spectra (Fig 1) of solubilized mitochondria (dithionite reduced minus ferricyanide oxidized) obtained with a Perkin- Elmer 559-A spectrOphotometer. Mitochondria (=2 mg/ml) are solubilized in a buffer containing 40 mM sodium phosphate, pH 7.8, 1% (w/v) deoxycho- late. A Ae(605-630 nm) = 24 mM”1 cm-1 is used (von Jagow and Klingenberg, 1972). Succinate dehydrogenase (succinate:INT reductase, E.C. 1.3.99.1) is quantitated by the endpoint assay of Pennington (1961) after 7-10 min incubations at 37°C. The percentage yield of mitochondria is estimated 41 Figure 1. Difference Spectrum Showing the a Bands of Chick Heart Mitochondrial Cytochromes Samples of solubilized mitochondria were reduced with excess ferricyanide or oxidized with dithionite. The absorbance of cyt aa was used to quantitate mitochon- 3 dria in respiration experiments. A Absorbance 42 003 560 550 660 wavelength, nm 650 43 from the fraction of succinate:INT reductase activity recovered from the crude homogenate. One unit of succinate:INT reductase catalyzes the synthesis of one nmol of INT-formazan per min at 37°C. Hexokinase (E.C. 2.7.1.1) is assayed according to the method of Polakis and Wilson (1982). The hexokinase activity remaining in the mitochondrial fraction after five rinses with 0.225 M mannitol/0.075 M sucrose is solubilized by incubation in 0.1 M sodium phosphate, pH 6.5, 1% Triton X-100 for 20 min. One unit of hexokinase catalyzes the syn- thesis of one nmol of glucose 6-phosphate per min at 30°C. Long Chain Fatty-Acyl-CoA Synthase. To test for the presence of this enzyme, mitochondria (20.78 nmol cyt aa3) are incubated at 30.5°C in a 0.225 M mannitol/0.075 M sucrose solution (1.75 ml) containing 450 pM 00A, 5 mM ATP, 5 mM magnesium acetate, 50 uM palmitate, 2 mM L-carni- tine, and 10 mM Tris-buffered Pi’ pH 7.4. In addition, 0.2% (w/v) fatty acid free BSA is included in the assay medium in order to control the concentration of free fatty acid in the mitochondrial suspension by creating an equilibrium between the free and bound forms (Spector gt al., 1971), thereby safeguarding against the uncoupling effect of fatty acid (Van den Bergh, 1967), the detergent effect of acylcarnitine derivatives (L.L. Bieber, personal communication), and inhibition of the adenine nucleotide translocase (Woldegiorgis et_§l., 1982). B-oxidation of palmitoylcarnitine is monitored with an oxygen polarograph. III. TRANSMISSION ELECTRON MICROSCOPY Mitochondria were embedded in 4% (w/v) warm Bacto-Agar (Difco Laboratories, Detroit, MI), and fixed at 4°C for 1 hr in a buffer that contained 4% (v/v) glutaraldehyde and 0.1 M sodium phosphate, pH 7.2. 44 After two washes in a 0.1 M sodium phosphate buffer, mitochondria were post-fixed in 1% (w/v) 030“ at room temperature for 1 hr. Samples were dehydrated in a graded series of ethanol and embedded in a mixture of epon-araldite-Spurrs resin. Ultra-thin sections (70-80 nm) made with an MT-2 Ultramicrotome were stained with aqueous 2% (w/v) uranyl acetate and then counterstained with Reynolds lead (Dawes, 1971). IV. SCANNING ELECTRON MICROSCOPY Mitochondria were fixed in 4% (v/v) glutaraldehyde buffered with 0.1 M sodium phosphate, pH 7.2. Fixed organelles were then mounted on a coverslip covered with 1% (w/v) poly-L-lysine hydrobromide (Sigma, type 1-B). The mounted organelles were washed three times with deionized water and dehydrated in a graded series of ethanol. The mitochondria were then critical point dried, sputter-coated with gold, and examined with a Jeol JSM-35C Scanning Electron MicroscOpe operated at 20 kV. V. ISOLATION OF MITOCHONDRIA Mitochondria were isolated from ventricular myocardium according to the scheme shown in Fig 2. For a typical isolation of mitochondria, hearts were removed from three, 14-20 day old chickens. The pericardium, major vessels, and atria were excised, and the ventricles of each heart were immediately minced with scissors into 5 ml of fresh ice-cold osmo- tic support medium containing 0.225 M mannitol, 0.075 M sucrose, and 0.2% (w/v) fatty acid free BSA (MSB medium). In order to maintain opti- mal mitochondrial respiratory control, it is critical that the heart tissue from each animal be isolated and minced within 30 sec after decapitation and that all subsequent operations be performed at 0-4°C. 45 Figure 2. Flow Diagram for Collagenase Facilitated Isolation of Heart Mitochondria Designations are as described under "Experimental Procedures." 46 Myocardial Mince A 200 IU Collagenase B 4.5 Strokes of P/E Homogenizer Crude Homogenate A Incubate on ice for train 8 Add EGTA to mm C Centrifuge for 5min at 4809 A Incubate on ice for 3-10min 8 Add EGTA to lmM C Apply 2 strokes of PIE Homogenizer D Centrifuge for 5min at 4809 P S B A A (Mitochonmd' +L (Cell Debns' ) (Cell Debris) (Mitochond'ia-o-L sosorne Fraction . Fraction A Filter through A Rehomogenize pellet A filter through cheesecloth B Centrifuge for 5min cheesecloth B Centrifuge for 10 at 480g 8 Centrifuge for IO min at 80009 min at BOOOg 53: PP: 8": P“I (Discard) / (Discard) A Resuspend PC 5c A Resuspend B Centrit e for 10min - B Centrif for 10min at 8066:; (mm) A 7'7"" WW9“ at 800060;; cheesecloth B Centrifuge for ID min at 8000g P ' P 82 I32 5c c 5‘2 A2 (Discard) l I (Discard) AResuspend J L. ________ B Centrifuge for 10min 833 33 3 (Discard) Final Mitochondria (Discard) Final Mitochondria Pellet Pellet 47 The combined mince is washed with three 30 ml volumes of M88 medium, resuspended in approximately 10 ml of M88 per gram of myocardium, and 200 I.U. of collagenase is added. Homogenization was performed in a glass vessel with four to five up and down strokes of a motor driven Potter-Elvejhem (P/E) pestle set at 500 rpm with a clearance of 0.117 mm (caliper measurement). The homogenate is allowed to incubate on ice for 1 additional min before EGTA is added to a final concentration of 1 mM to halt collagenolysis and prevent mitochondrial uptake of Ca(II). The volume of the homogenate was measured and a sample (0.5 ml) for yield determination withdrawn. The homogenate was centrifuged at 480g for 5 min in 15 ml Corex tubes with a Sorvall SS-34 rotor in order to sediment cellular debris. The supernatant liquid was filtered through cheesecloth and centrifuged at 8,000g for 10 min to sediment the mitochondria. The pelleted mito- chondria were gently resuspended with a Pipetman (Gilson Medical Instruments, Middleton, WI) in M88 medium containing 1 mM EGTA (MSBE medium). The mitochondria were washed twice in 10 ml of MSBE by centri- fuging at 8,000g for 10 min and resuspended in this medium at a final concentration of 15-20 mg/ml. The above protocol should take no more than 1 hr to complete. The resulting mitochondrial pellet is designated PA 3. If enhanced yields are needed the following protocol should be followed. For a 10-15% increase in yield, the pellet resulting from low- speed (480g) centrifugation is rehomogenized, centrifuged once again at 480g, and then subjected to high speed (8,000g) centrifugation. The resulting mitochondrial pellet is washed once and pooled with the first pellet. This mitochondrial suspension corresponds to PA3,. For maximal 48 yield, the crude homogenate is allowed to incubate on ice for 3 min in the presence of collagenase, made 1 mM in EGTA, further homogenized with two strokes of the P/E pestle, and centrifuged at 480g for 5 min. The mitochondria in the supernatant liquid are pelleted and washed and then pooled with mitochondria recovered from the crude pelleted material. The combined pellet (P83) is washed and resuspended as above in MSBE medium. VI. OXYGEN CONSUMPTION ASSAYS Mitochondrial coupling parameters were assessed at 30.5°C with a Gilson model K-IC oxygen polarograph (Gilson Medical Electronics, Middleton, WI), equipped with a Yellow Springs Instruments (Yellow Springs, 0H) Clark oxygen electrode (Model 5331) and a water-jacketed, 1.75 ml glass reaction chamber. Constant temperature was maintained with a Haake (Saddle Springs, NJ) circulating water bath. Mitochondrial respiration experiments were conducted in an assay medium containing 0.225 M mannitol, 0.075 M sucrose, and 20 mM Tris-Pi, pH 7.4. State 3 respiration was induced with approximately 400 nmoles of ADP after a 1 min incubation with oxidizable substrate. Respiratory control ratios were calculated as previously described (Chance and Williams, 1956; Estabrook, 1967). Mitochondria and all reaction components were added to the assay medium with Hamilton glass syringes (Hamilton Co., Reno, NE). Oxygen back diffusion into the assay medium at 30.5°C was estimated across a range of oxygen saturation levels. Oxygen was displaced by bubbling the assay medium with argon gas. 49 RESULTS I. MITOCHONDRIA RECOVERIES The percent recoveries of succinate:INT reductase for the indicated number of preparations are shown in Table I. Compared to control samples, the yield of heart mitochondria is enhanced nearly two-fold when isolated in the presence of collagenase. Yields are enhanced an additional 10-15% when the crude pelleted cell debris are rehomogenized and the remaining mitochondria are recovered. Subjecting the crude homogenate to further mechanical disruption after it has been incubated with collagenase for 3 min and recovering mitochondria lost in the crude pelleted material gives yields for PB of 70 i 8%. Importantly, the mitochondria in 3 pellets PA3 and P83 have comparable respiratory parameters (data not shown). II. OUTER MEMBRANE ENZYMES Chick heart mitochondria isolated in the presence of collagenase retain the activity of outer membrane enzymes. Considerable long chain fatty acyl CoA synthetase activity remains associated with mitochondria (Fig 3A). Adding L(-)carnitine to mitochondria incubated with CoA, ATP, and palmitic acid stimulates state 3 respiration, indicating that free palmitic acid is converted to its CoA thioester and is available for transport by carnitine palmitoyl-transferase I into the mitochondrial matrix for B-oxidation. The ADP necessary to stimulate state 3 respira- tion is formed by the adenylate kinase-catalyzed transphosphorylation of 50 TABLE I. YIELD OF HEART MITOCHONDRIA ISOLATED WITH OR WITHOUT COLLAGENASE. Chicken heart mitochondria were isolated with collagenase as described under "Experimental Procedures." The designations for mitochondrial pellets correspond to those described in the text. Yields were determined from the percent recovery of succinate INT reductase from the crude homogenate. The values are the mean t S.D. of the indicated number of preparations. Mitochondria Pellet Collagenase Yield No. Preps % PA3 - 17 i 4 7 PA3 + 31 t 8 29 PA3' + 41 1 3 S P + 70 t 8 4 133 51 Figure 3. Oxygen Electrode Tracings used for Monitoring Enzyme (A) (B) Activities Activity of long chain fatty acyl CoA synthetase associated with chick heart mitochondria isolated with collagenase. Assay conditions were as described under "Experimental Procedures." The rate of state 3 respiration subsequent to the addition of 2 mM L-carnitine was 235 ng-atom O min-i (nmol cyt aa3)-1. Activity of adenylate kinase. 0.069 nmol cyt aa (ml)-1 3 were assayed for respiration in 0.225 M mannitol, 0.075 M sucrose, 5 mM Mg(II), and 10 mM Tris-buffered Pi’ pH 7.4, at 30.5 °C. After the phosphorylation of 400 nmol ADP, 200 nmol of AMP was added. The rate of state 3ADP was 678 ng-atom ATP+AMP 0 min.1 (nmol cyt aa )-1 and that for state 3 was 239 3 ng-atom 0 min”1 (nmol cyt aa3)-1. The total oxygen consumed was 127 and 146 ng-atom 0 following the addition of ADP and AMP, respectively. Oxygen consumption is approximately con- sistent with the reaction AMP + ATP—>2ADP. 52 Carnltine 53 AMP, which is produced during the fatty acid activation reaction. That these mitochondria contain the adenylate kinase necessary to produce ADP is demonstrated in Fig 3B. The intactness of outer membrane enzymes is further supported by the observation that over 50% of the hexokinase, whose reversible binding to the mitochondrial outer membrane is meta- bolically regulated (Wilson, 1980), remains with the mitochondrial fraction after exposure to collagenase (Table II). The I.U. of hexoki- nase per mg of mitochondrial protein is in good agreement with studies on rabbit heart mitochondria (Aubert-Foucher et al., 1984). III. MEMBRANE INTACTNESS Heart mitochondria isolated in the presence of collagenase have intact inner and outer membranes as evidenced by the inability of exogenous 1 mM NADH to support the phosphorylation of ADP and by the lack of respiratory stimulation in the presence of 2 mM reduced cyto- chrome c (data not shown). Scanning and transmission electron microscopy also show intact mitochondrial membranes (Fig 4 A and C). As shown in Fig 4C, chick heart mitochondria are comprised of a highly convoluted matrix and have easily distinguishable inner and outer membranes sepa- rated by an inter-membrane space. Approximately 3-3.5 hrs after the isolated mitochondria are suspended, the membranes become perforated (Fig 4B) and begin to extrude matrix material (Fig 4D). The disintegra- tion of mitochondrial membranes coincides temporally with progressive increases in state 4 rates of respiration and decreases in state 3 rates (data not shown). The cause of this structural disintegration was not investigated, although the activation of proteases endogenous to the mitochondria is one possibility (Dean, 1983). Most preparations of chick 54 TABLE II. RECOVERY OF HEXOKINASE IN HEART MITOCHONDRIA ISOLATED WITH COLLAGENASE. Heart mitochondria isolated in the presence of collagenase were washed five times and assayed for hexokinase activity as described under "Experimental Procedures." Trial Crude Homogenate Mitochondria Recovery Specific Activity Mitochondrial Hexokinase unitsa unitsb 1 units mg-1 1 11.1 6.0 54.1 0.092 2 8.6 5.0 58.1 0.10 a Indicates total units in the crude homogenate of a preparation. b The units of hexokinase activity in the mitochondrial fraction were corrected for the yield of mitochondria. Figure 4. (A) (B) (C) (D) 55 Representative Scanning and Transmission Electron Micrographs of Chick Heart Mitochondria Isolated with Collagenase Scanning electron micrograph (24,000x) of mitochondria at the time of suspending the final pellet in M/S medium. Note smooth, undisrupted membrane surfaces. The bar denotes 1 micron. Scanning electron micrograph (24,000x) of mitochondria 5 hrs after isolation. Perforated membranes are a common feature of the mitochondria at this time. The bar denotes 1 micron. Transmission electron micrograph (99,360x) of mitochondria immediately after suspending the final pellet. Mitochondria present with smooth inner and outer membranes, an intermem- brane space, and a highly convoluted matrix. Transmission electron micrograph (54,000x) of mitochondria 5 hrs after isolation. Broken membranes and matrix extrusion are widespread at this time. 57 heart mitochondria maintain optimal RCR and ADP:O ratios for 3-3.5 hrs. IV. RESPIRATORY CONTROL RATIOS The respiratory control ratio (RCR) of chick heart mitochondria is strikingly dependent on the concentration of cyt aa3 (Fig 5). The basis for this behavior is discussed in detail in Chapter II. For all prepara- tions tested there is a narrow range of mitochondria concentration (0.050 i 0.005 nmoles cyt aa /ml) where respiratory control is highest. 3 Figure 6 is an example of an oxygen electrode tracing obtained at this concentration of mitochondria. ADP ADPZ, and ADP designate the first, 1’ 3 second, and third addition of 400 nmoles of ADP, respectively. These mitochondria typically have a significant state 4 rate of respiration after phosphorylating ADP . However, after phosphorylating ADP and 1 2 ADP3, state 4 rates approach and even attain zero. This demonstrates that chick heart mitochondria can exert extraordinary control over the rate at which electrons are transferred from the respiratory chain to oxygen in the absence of ADP. In keeping with the definition of RCR values (i.e., the ratio of the rates of state 3 and state 4 respiration), a zero rate of state 4 implies that these mitochondria, when optimized for concentration, can respire with infinite respiratory control. V. BACK DIFFUSION OF OXYGEN Because state 4 rates frequently approach zero and near-zero values, it was of interest to determine whether the diffusion of atmospheric oxygen into the respiration chamber could artifically depress state 4 rates. Oxygen back-diffusion measurements were made by recording the change in oxygen concentrations as a function of time at different 58 Figure 5. Respiratory Control Ratios Measured as a Function of Mitochondria Concentration Increasing concentrations of mitochondria (expressed as nmoles cyt ea /1.75 ml) were assayed for respiratory 3 control ratios using a Gilson oxygen polarograph as described under "Experimental Procedures." RCR1 and RCR2 values were calculated from the rates of state 3 and state 4 respiration following two sequential addi- tions of 400 nmoles of ADP. Mitochondria oxidized 5 mM pyruvate/2.5 mM malate in the presence of 20 mM Tris-Pi, pH 7.4, at 30.5°C. RESPIRATORY CONTROL RATIO 4o 36 32 28 24 20 16 59 o RCR. o RCR2 l l l l l 0.1 0.2 0.3 0.4 0.5 NANOMOLE CYTOCHROME OXIDASE 0.6 Figure 6. 60 Oxygen Consumption Behavior of Chick Heart Mitochondria Isolated with Collagenase Three state 3 to state 4 transitions at the critical mitochondrial concentration are shown. M, addition of mitochondria at 0.050 nmol cyt aa3 ml-i to M/S medium containing 20 mM Tris-buffered Pi’ pH 7.4, at 30.5°C; P/M, addition of 5 mM pyruvate/2.5 mM malate; ADP1_3, sequential additions of 400 nmoles of ADP. 61 56 6:5. _ m _ “Q" 1 OO— uoliolnios luamad ‘ [uebflxo] 62 concentrations of oxygen. The rate of change of oxygen concentration in the respiration chamber depends on the oxygen concentration in the oxygen polarograph's assay vessel (Fig 7). At 40% saturation there is no change in oxygen concentration. Above 40% saturation there is a net decrease of oxygen presumably because the Clark electrode reduces oxygen to water at a rate that is more rapid than that at which oxygen diffuses back into the respiration chamber. Below 40% saturation diffusion into the respi- ration chamber is more rapid than uptake by the electrode. Since the changes in respiration induced by ADP were monitored at oxygen con- 1-3 centrations above 40% saturation, it is concluded that the rates of state 4 respiration reported herein are not artificially low. No corr- ections in rates of respiration were made for oxygen consumption by the Clark electrode. VI. GENERAL COMMENTS At the concentration of mitochondria where optimum respiratory control is observed, a zero or very low rate of state 4 respiration subsequent to ADP is observed between 22 and 33°C. Above 33°C, the 2 rate of state 4 respiration increases as a function of temperature. The respiratory control parameters of heart mitochondria isolated according to this method are extremely sensitive to micromolar concen- trations of Mg(II), Mn(II), and Ca(II). The rate of state 4 respiration is increased and ADP:0 ratios are decreased by these cations as a func- tion of concentration. Mn(II) and Ca(II) also decrease the rate of state 3 respiration, indicating that chick heart mitochondria undergo some degree of damage subsequent to exposure to these cations. Consequently, these divalent cations should not be constituents of the assay medium if Figure 7. 63 Quantitation of Oxygen Consumption by the Clarke Electrode and Back Diffusion of Atmospheric Oxygen into the Oxygraph Assay3Vessel 1.75 ml samples of M/S medium containing 20 mM Tris-buffered Pi were degassed to varying oxygen tensions by bubbling with argon gas. Oxygen consumption and back diffusion were measured using either a Yellow Springs "standard" (.,0) or "high sen- sitivity" (A) Teflon membrane stretched very tightly over the Clark electrode. Samples were allowed to run for 20-30 min so as to enhance the accuracy of estimated rates. The line drawn through the data was calculated by least squares linear regres- sion analysis. 64 I I i I I 1 1 l l _ I.O — _7 .E E O Crossover Point .5. *5 O l E O ,_ _ c» C o -I.O 1 1 1 1 1 1 1 1 1 1 o 10 20 so 40 so so 70 [Oxygen], percent saturation 80 90 100 65 optimal respiratory parameters are desired. Concentrations of EDTA in excess of 100 uM decrease RCR values (Fig 8). This chelating agent stimulates state 4 respiration but has no effect on state 3 respiration. In contrast, EGTA concentrations up to 5 mM stimulate neither state 3 nor state 4 rates of respiration. However, as to be discussed more fully in Chapter 4, EGTA can be used to decrease state 4 rates of respiration by chelating divalent cations released from the mitochondria. This prevents the regeneration of ADP from ATP by divalent cation sensitive ATPases which cosediment with the mitochondria. The age of the chicks from which mitochondria are isolated exerts a critical influence over the magnitude of respiratory control that one may expect to observe. RCR values greater than 50 are only found with mitochondria isolated from chicks less than 20 days old (ex ovo). The RCRs of heart mitochon- dria from older chicks progressively decrease as a function of age. All solutions must be kept rigorously free of microbial contamina- tion. For optimum respiratory control, mannitol/sucrose solutions should be freshly made for each mitochondria preparation and stock solutions of Pi and substrate should be made every 7-10 days. Of all the respiratory substrates studied (5 mM pyruvate/2.5 mM malate, 5 mM glutamate/2.5 mM malate, 5 mM a-ketoglutarate/2.5 mM malate, 5 mM B-hydroxybutyrate, 5 mM succinate/SuM rotenone), the combination of pyruvate and malate sup- port the highest state 3 rates and the lowest state 4 rates when the concentration of Pi is held fixed at 20 mM. Collagenase can also be used to facilitate the isolation of highly coupled heart mitochondria from 14-20 day old Sprague-Dawley rats. For Optimal respiratory control, 0.12 M KCl should be used as the osmotic support medium in both the isolation and assay media. 66 Figure 8. Chick Heart Mitochondrial Respiratory Control Measured as a Function of EDTA Concentration Mitochondria were suspended at 0.074 nmoles cyt aa /ml in 3 an osmotic support medium comprised of 0.225 M mannitol, 0.075 M sucrose, and 10 mM Tris-Pi, pH 7.4. EDTA was added to the indicated concentrations, and respiratory control ratios were measured. (0), RCR1; (.), RCRZ. 67 5.0 '.o 4. 2 3 40 020m. .9200 tofieammm [EDTA] ,mM 68 VII. CONCLUSIONS The results of this investigation indicate that collagenase facili- tates the release of mitochondria from myocardium. The resulting mito- chondria are highly coupled, have enzymatically intact outer membranes, morphologically intact inner and outer membranes, and can be isolated in high yield. 69 REFERENCES Aubert-Foucher, A., Font, B., and Gautheron, D.C. (1984) Arch. Biochem. Biophys. 232: 391-399. ' Bishop, P.D., and Atkinson, D.E. (1984) Arch. Biochem. Biophys. 230: 335-344. ‘ Bond, M.D., and Van Wart, H.E. (1984) Biochemistry 23: 3085-3091. Caulfield, J.B., and Borg, T.K. (1979) Lab. Invest. 40: 364. Chance, 8., and Hagihara, B. (1963) Proc. Int. Congr. Biochem. 5: 3-37. Chance, 8., and Williams, G.R. (1956) Adv. Enzymol. 17: 65-134. Chou, A.C., and Wilson, J.E. (1972) Arch. Biochem. Biophys. 151: 48-55. Dean, B. (1983) Arch. Biochem. Biophys. 227: 154-163. Ernster, L., and Schatz, G. (1981) J. Cell Biol. 91: 2273-2553. Estabrook, R.W. (1967) Methods Enzymol. 10: 41-47. Groen, A.K., Wanders, R.J.A., Westerhoff, H.V., van der Meer, R., and Tager, J.M. (1982) J. Biol. Chem. 257: 2754-2757. Hay, E.D. (1981) J. Cell Biol. 91: 2058-2238. Lehninger, A.L., Carafoli, E., and Rossi, 0.8. (1967) Adv. Enzymol. 29: 259-320. ‘ Lehninger, A.L. Biochem. J. 119: 129-138. Lemasters, J.J., and Hackenbrock, C.B. (1980) J. Biol. Chem. 255: 5674-5680. Markwell, M.A.K., Haas, S.M., Tolbert, N.E., and Bieber, L.L. (1981) Methods Enzymol 72: 296-303. ' ' Masini, A., Ceccarelli-Stanzani, D., Muscatello, U. (1983a) FEBS Lett. 160: 137-143. Masini, A., Ceccarelli-Stanzani, D., Muscatello, U. (1983b) Biochim. Biophys. Acta 724: 251-257. Masini, A., Ceccarelli-Stanzani, D., Muscatello, U. (1984) Biochim. Biophys. Acta 767: 130-137. 70 Mela, L., and Seitz, S. (1979) Methods Enzymol. 55: 39-46. Nedergaard, J., and Cannon, B. (1979) Methods Enzymol. 55: 5-28. Nicholls, D. (1974) Eur. J. Biochem. 50: 305-315. Palmer, J.W., Tandler, B., and Hoppel, C.H. (1977) J. Biol. Chem. 252: 8731-8739. Pande, S.V., and Blanchaer, M.C. (1970) Biochim. Biophys. Acta 202: 43-48. Pennington, R.J. (1961) Biochem. J. 80: 649-654. Polakis, P.C., and Wilson, J.E. (1982) Biochem. Biophys. Res. Comm. 107: 937-943. Spector, A.A., Fletcher, J.E., and Ashbrook, J.D. (1971) Biochemistry 10: 3229-3232. Stadhouders, A.M. (1981) In Mitochondria and Muscular Diseases. (R.F.M. Busch, F.G.I. Jennekens, and H.R. Scholte, eds). P.77. Mefar b.v. Beetswerg, The Netherlands. Stucki, J.W., and Ineichen, E.A. (1974) Eur. J. Biochem. 48: 365-375. Stucki, J.W. (1976) Eur. J. Biochem. 68: 551- 562. Thompson, D.A., and Ferguson-Miller, S.M. (1983) Biochemistry 22: 3178-3187. von Jagow, 0., and Klingenberg, M. (1972) FEBS Lett. 24: 278-282. Van den Bergh, 8.0. (1967) Methods Enzymol. 10: 749-755. Wilson, J.E. (1980) Curr. Top. Cell. Regul. 16: 2-54. Woldegiorgis, 0., Yousufzai, S.Y.K., and Shrago, E. (1982) J. Biol. Chem. 257: 14,783-14,787. Chapter 3 STUDIES OF THE FACTORS AFFECTING THE RESPIRATORY CONTROL AND ADP:O COUPLING RATIOS OF ISOLATED CHICK HEART MITOCHONDRIA 71 72 INTRODUCTION Mitochondrial substrate oxidation, electron transfer, and ATP syn- thesis together comprise an integrated metabolic process designed to sensitively maintain cellular energy balance. The overall reaction of oxidative phosphorylation may be summarized with the following equation: + + + nADP + nPi + NADH (FADHZ) + H + 1/202 + nATP + NAD (FAD ) + H20. (1) In general, it is believed that mitochondria couple substrate oxidation to the phosphorylation of ADP through an electrochemical potential gradient of protons (Mitchell, 1979). Protons are electrogenically ejected from the mitochondrial matrix into the intermembrane space by redox pumps located at the three coupling sites of the respiratory chain: NADH-ubiquinone oxidoreductase (site I), ubiquinol-cytochrome g oxidoreductase (site II), and cytochrome c oxidase (site III). The loss of energy from proton extrusion is coupled to a decrease in the oxidation-reduction potential of electrons as they pass through each of the coupling sites (Chance, 1977; Wilson, 1980). How mitochondria regulate the rate of respiration is the subject of much study. The classic investigations of Chance and Williams (1956) show that ADP exerts predominant control over the extent to which respiration is stimulated in isolated mitochondria. More recent studies, however, have demonstrated that the control of respiration depends upon a number of other components of the mitochondrial matrix. Evidence shows that rates of respiration are modulated by changes in the activity of 73 various substrate dehydrogenases (Denton and McCormack, 1985; Hansford and Castro, 1981), electron transfer chain components (Forman and Wilson, 1982; Erecinska and Wilson, 1982; Chance, 1972), and inner mitochondrial membrane transport proteins (Lemasters and Sowers, 1979; Klingenberg, 1980; Doussiere et_al., 1984; Groen §£_al., 1982). The widespread distribution of control insures that respiration is poised to rapidly respond to changes in metabolite transport, thermodynamic potentials, and dehydrogenase activities. Respiratory control is a measure of the mitochondrion's capacity to limit the rate of electron transport in both the presence and absence of the phosphoryl group acceptor ADP. The ratio of the rate of oxidation during ATP synthesis (state 3) to that during "resting" (state 4) respiration is the respiratory control ratio (RCR; Chance and Williams, 1956). RCR values are regarded as an important index of mitochondrial function in vitro and it is known that mitochondria demonstrate respi- ratory control in vivo (Hassinen and Hiltunen, 1975; Wilson gt_al., 1974). Differences in RCR values among heart mitochondria preparations arise largely from differences in state 4 respiration which, from inspection of the literature, may be quite substantial. A variable percentage of the respiration during state 4 arises from residual ATP synthesis (Lemasters and Hackenbrock, 1980; Masini gt_al., 1983). The remaining respiration is due to the oxidation of pyridine and flavin nucleotides in the absence of phosphorylation (i.e., the respiration is uncoupled). This uncoupled respiration is accounted for in chemi- osmotic terms by two mechanisms. The first assumes that protons are able to permeate the mitochondrial inner membrane via non-specific leak pathways. This constitutes a non-ohmic conductance of protons (Nicholls, 74 1974; Pietrobon §£_al., 1983). The second proposes that the redox pumps and the F0 moiety of the ATP synthase are intrinsically uncoupled (Pietrobon and Caplan, 1985; Zoratti gt_al., 1986; Pietrobon §t_al., 1986). These proteins are believed to "slip", whereby the redox pump may transfer electrons without ejecting a proton and the FO protein may conduct protons without concomitantly inducing the release of ATP from F The depletion of endogenous substrate in rabbit heart mitochondria 1. decreases respiratory control (Tarjan and von Korff, 1967), while ATP loading in human placental mitochondria increases respiratory control (Illsley g£_§l., 1985). In both cases, changes in rates of state 3 respiration accounted for changes in RCR values; state 4 rates were unaffected. Apart from complete inhibition of the ATP synthase (Masini et_al., 1983, 1984) or the addition of an enzyme that traps ADP gener- ated from ATP in the extramitochondrial space (Bishop and Atkinson, 1984), few means for decreasing state 4 rates of respiration have been elucidated. The efficiency with which mitochondria transduce redox free energy into the free energy of hydrolysis of ATP is assessed from the ADP:0 ratio (corresponding to n of equation 1). For many years it was believed that one ADP molecule is phosphorylated at each of the coupling sites per pair of electrons oxidized (Ernster, 1977). Such a stoichiometry would yield whole number ADP:O ratios of 3.0 for mitochondria oxidizing NAD-linked substrates. However, because currently debated models of chemiosmosis (Mitchell, 1979; Westerhoff gt_al., 1984) regard the proton as the intermediate between oxidation and phosphorylation, fractional ADP:O ratios are also acceptable stoichiometries. A careful reexamina- tion of oxidative phosphorylation in rat liver mitochondria by Lemasters 75 et_al. (1984) showed that ATP/site stoichiometries are fractional and can yield an ideal upper limit for the ADP:O ratio of 3.37 when 8- hydroxybutyrate is used as substrate. This stoichiometry is in sharp contrast to those reported by others (2.0, Hinkle and Yu, 1979; 2.75, Beavis and Lehninger, 1986; reviews: Lemasters g£_§l., 1984; Flatt gt al., 1984; Ferguson, 1985; Beavis, 1988). Clearly, although ADP:O ratios likely vary as a function of the tissue source and the actual degree to which the mitochondria are coupled, there is significant disagreement as to what the values can and should equal in isolated mitochondria phos- phorylating nmole pulses of ADP. The focus of the present investigation was to characterize the factors affecting the respiratory parameters of isolated chick heart mitochondria. Because of so many other such studies performed in the past, such an undertaking may appear mundane. However, this study is of particular interest for two reasons. First, these mitochondria dem- onstrate an extraordinary capacity to control the rate of respiration in the absence of ADP. Practical consequences of such exquisite coupling are that these mitochondria have RCR values approaching infinity and ADP:O stoichiometries that equal the theoretical limits proposed by Lemasters gt_al. (1984). Second, a number of factors which regulate the rate of state 4 respiration are identified and characterized. The control of these factors makes it possible to systematically vary the respiratory parameters of chick heart mitochondria to fit the needs of particular types of experiments. Preliminary accounts of this work have been presented (Toth et al., 1985, 1986, and 1988). 76 EXPERIMENTAL PROCEDURES I. MATERIALS The water used for these experiments was purified as previously described (Toth g£_al., 1986). Mitochondria were isolated from the hearts of 14-21 day old chicks with collagenase according to the method of Toth §t_al. (1986). Corn mitochondria were isolated by the method of Day and Hanson (1977). Antisera and Arsenazo III were kindly provided by Dr. John Wang (Michigan State University) and Dr. Antonio Scarpa (Case Western Reserve University), respectively. Lactate dehydrogenase (type II), pyruvate kinase (type III), and collagenase (lots 27F-6824 and 47F- 6829) were from Sigma (St. Louis, MO). All of the chemicals used were reagent grade or better and used without further purification. II. POLYACRYLAMIDE GEL ELECTROPHORESIS SDS-polyacrylamide gel electrophoresis was performed using the discontinuous buffer system of Laemmli (1970) in slab gels (16 cm x 18 cm x 0.75 mm). The resolving gel was 8% acrylamide/0.21% bisacrylamide, and was polymerized by adding 0.05% (w/v) ammonium persulfate and 0.071 M TEMED. The stacking gel (1 cm) contained 3% acrylamide/0.08% bisacryl- amide. This gel was polymerized with 0.05% (w/v) ammonium persulfate and 0.014 M TEMED. Samples were solubilized in sample buffer (0.125 M Tris, pH 6.8, 4% NaDodSou (w/v), 20% glycerol (v/v), 0.1 M DL-dithiothreitol, and 0.002% bromophenol blue (w/v)) and incubated in a boiling water bath for 5 min. Gels were run for 1 hr at 100 V, and then for 3.5 hrs at 77 300 V. The electrophoresis apparatus was cooled with tap water during the entire course of a run. Gels were fixed overnight in a fixative solution comprised of 50% methanol, 40% water, and 10% glacial acetic acid (v/v). Gels were stained for 3-4 hrs in fixative that contained 0.25% (w/v) Coomassie brilliant blue R250, and destained with two changes of a solvent that was 5% methanol, 7.5% glacial acetic acid, and 87.5% water (v/v). Plots of log Mr versus R for the molecular weight standards used F were non-linear. Data were fitted to the following relation (Peterson and Hokin, 1981): log R = log a - m log(1 + Mr/C)' (2) F using the iterative least-squares linear regression methods specified by Bates and MaCallister (1974). Once the constants (a, c, and m) for the standard curve were calculated, the molecular weight of other bands were estimated from knowledge of their relative mobility. For the gel shown in Figure 15, a = 1.47, m = 2.96, and c = 121,461. In calculating the relationship between Mr and R it was consistently found that BSA F9 migrated anomalously. The error of curve fitting was reduced markedly by excluding BSA from the analysis. Consequently, this was routinely done. III. IMMUNOBLOTTING Actin and myosin were identified by immunoblotting. Samples were electrophoresed in NaDodSOu-polyacrylamide gels as described above. Proteins were transferred from the gel and bound to nitrocellulose membranes as described by Towbin §£_al. (1979), for 1120 mA-hrs in transfer buffer (0.025 M Tris, 0.19 M glycine, pH 8.3, and 20% methanol 78 (v/v)). Nitro-cellulose membranes were then blocked in 20 mM Tris, pH 7.5, 0.5 M NaCl, 3% BSA for 6 hrs on a shaker platform. Excess BSA was removed by washing membranes twice in 20 mM Tris, pH 7.5, 500 mM NaCl, 0.05% Tween 20 (TTBS) for 10 min. Membranes were then incubated over- night with one of two antisera: (a) rabbit anti-calf thymus actin, or (b) rabbit anti-chicken gizzard myosin. Antisera were diluted 1:200 with 20 mM Tris, pH 7.5, 500 mM NaCl (TBS) prior to use. Excess antibody was removed by washing membranes twice for 10 min in TTBS. The membrane was incubated with a 1:3000 dilution of goat anti-rabbit IgG alkaline phos- phatase conjugate for 1 hr (Bio Rad), and washed twice with TTBS and once with TBS. Blots were stained in 0.3 mg/ml nitro blue tetrazolium (NBT; Sigma, grade III), 0.15 mg/ml 5-bromo-4-chloro-3-indolyl phosphate (BCIP; Sigma, p-toluidine salt), 0.1 M Tris, pH 9.5, 0.1 M NaCl, and 2 mM MgCl . Purified rabbit skeletal muscle actin and chick heart myosin 2 heavy chain were used as positive controls. IV. ISOLATION OF CHICK MYOCARDIAL MYOSIN Myosin was isolated from homogenized chick ventricular myocardium using the method of Wikman-Coffelt g£_al. (1973). After ammonium sulfate precipitation (35-42% saturation), the myosin fraction was suspended in 50 mM Tris-H01, pH 7.5, 1 mM DL-dithiothreitol, and 0.5 M KCl. The myosin was then dialyzed overnight against two changes of this buffer, made 50% (v/v) in glycerol, and stored at -20°C. V. ATPase ASSAYS (A). Myosin ATPase activity. 100 01 (0.36 or 0.4 mg/ml) of the myosin preparation was suspended in 0.9 ml of assay buffer (0.225 M mannitol, 79 0.075 M sucrose, 20 mM Tris-H01, pH 7.4, varying concentrations of either Ca(II) or Mg(II), and 25 mM KCl). Reactions were initiated with the addition of 840 nmols of ATP and allowed to proceed for 8 min at 30.5°C. The kinetic parameters for the inhibition of myosin with Mg(II) were estimated by a least squares non-linear regression of the data fitted to the following equation assuming a constant absolute error: V [Mg(II)] v0 = b _ max (3) K50 + [Mg(II)] In equation 3, b represents the rate of ATP hydrolysis in the absence of Mg(II), and the other terms have their usual meaning. The amount of Pi released during the course of an incubation was assayed by the method of Pollard and Korn (1973), with one minor modification. Prior to denaturing protein, the assay mixture was diluted 3:1 with water. Diluting the assay 3:1 with water increased the efficiency of extracting the phosphomolybdate complex by 90% (Fig 1A). It is likely that dilution is necessary in order to decrease the concentration of mannitol, a compound known to have a high affinity for phosphomolybdate (Hagihara and Lardy, 1960). The amount of ATP hydrolyzed was estimated from the standard curve shown in Fig 18. (B). Mitochondrial ATPase activity. The conditions for assaying mito- chondrial ATPase activity are described in the legends to the pertinent Figures. The inorganic phosphate produced was assayed as described above. N-ethylmaleimide, an inhibitor of the P transport protein 1 (Kaplan et al., 1986), was included in the assay medium to prevent mitochondria from internalizing P released during ATP hydrolysis. 1 Electron flow was blocked with rotenone in order to inhibit the uptake Figure 1. (A) (B) 80 Extraction of Inorganic Phosphate from Suspensions of Chick Heart Mitochondria Hydrolyzing ATP Optimization of conditions for the extraction of P1 from suspensions of chick heart mitochondria. 940 nmols of Pi were added to the indicated solvents. The P1 was then extracted with an acidified isobutanol-benzene mixture. The relative amount of P1 in samples was determined by the method of Pollard and Korn (1973). M/S indicates 0.225 M mannitol and 0.075 M sucrose. Standard curve used for estimating the amounts of Pi released during ATPase assays. Calibrated amounts of Pi were added to M/S medium, diluted 3:1 with water, and then extracted and assayed as described in (A). 81 Absorbance, 720 nm .0 .o .0 :- -b a: a) O I fl I I > .0 N I l 1 l l O l M/S ws:1-1,o. ms: 11.0, ms: 11.0. 11,0 3:1 Ill 1:3 Medium from which P, Extracted I I I I Absorbance, 720 nm 0 :- :- i“ in m m o I I I I .0 .p. r O l. L 0 500 1000 1500 2000 n moles Pi 82 of Ca(II) and the phosphorylation of ADP. VI. OXYGEN CONSUMPTION ASSAYS Mitochondrial coupling parameters were assessed at 30.5°C with a Gilson model K-IC oxygen polarograph (Gilson Medical Electronics, Middleton, WI), equipped with a Yellow Springs Instruments (Yellow Springs, OH) 0.5 cm diameter Clark oxygen electrode (Model 5331) and a water-jacketed, 1.75 ml glass reaction chamber. Constant temperature was maintained with a Haake (Saddle Springs, NJ) circulating water bath. Reaction mixtures were stirred with a Teflon-coated metal bar (0.9 x 0.3 cm) and a Micro V (Cole Parmer) magnetic stirring device. The magnetic stir bar was operated at 600 rpm. Mitochondrial respiration experiments were conducted in an assay medium containing 0.225 M mannitol, 0.075 M sucrose, and 20 mM Tris-Pi, pH 7.4. State 3 respiration was induced with approximately 400 nmoles of ADP after a 1 min incubation with oxidizable substrate. State 4 respiration was typically allowed to proceed for 2 min after the completion of state 3 respiration. Respiratory control ratios were calculated as previously described (Chance and Williams, 1956; Estabrook, 1967). ADP:O ratios were calculated using "total" oxygen as described by Lemasters (1984). The concentration of oxygen atoms in the air-saturated assay medium at 30.5°C was 405 uM as determined by measuring the oxygen consumption of corn (Zea mays, cultivar W64ATMS) mitochondria respiring in the presence of a known amount of NADH. Unlike mammalian mitochondria, the inner membrane of corn mitochondria is freely permeable to exogenous NADH (Miller et al., 1970; Douce, 1985). The concentration of oxygen was calculated by assuming that the added NADH was completely oxidized and 83 that one atom of oxygen was consumed for each molecule of NADH oxidized. Mitochondria and all reaction components were added to the assay medium across the capillary (5.7 x 0.2 cm) of a glass stopper with Hamilton glass syringes (Hamilton Co., Reno, NE). VII. NUCLEOTIDE ASSAYS ADP concentrations were assayed in a 1.0 ml reaction mixture con- taining 25 mM MOPS, pH 6.8, 5 mM magnesium acetate, 1 mM EDTA, 1.8 mM phosphoenolpyruvate, 0.32 mM NADH, 44 I.U. lactate dehydrogenase, and 12 I.U. pyruvate kinase. NADH concentrations were assayed in a 1.0 ml mixture containing 50 mM sodium acetate, pH 5.0, 3 mM pyruvate, and an excess of lactate dehydrogenase. For both assays, the oxidation of NADH was measured at 340 nm using a Beckman DU spectr0photometer. A molar extinction coefficient for NADH of Ae(340 nm) = 6.23 x 103 M.1 cm"1 was used. In order to ascertain whether the mitochondria described herein converted all exogenous ADP to ATP, both intra- and extra-mitochondrial ADP remaining after oxidative phosphorylation were assayed using the lactate dehydrogenase-pyruvate kinase coupled enzyme system noted above. Samples of mitochondria at the critical mitochondrial concentration (0.05 nmol cyt aa /ml) after phosphorylating three sequential additions 3 of 400 nmoles of ADP using a-ketoglutarate as substrate were acidified with 0.5 ml of 2.0 N perchloric acid and allowed to incubate on ice for 10 min. Denatured samples were neutralized with 0.4 ml of 3.0 M sodium carbonate and centrifuged at 8,000g in an Eppendorf 3200 microcentrifuge for 10 min. The resulting supernatant liquid was assayed for ADP. The capacity of pyruvate kinase to prevent Mg(II)-induced stimula- 84 tion of respiration during state 4 was assayed in a reaction mixture containing 50 mM KCl, 150 mM mannitol, 50 mM sucrose, 20 mM Tris-Pi, pH 7.4, 1 mM phosphoenolpyruvate, 1 mM Mg(II), and 5 mM pyruvate/2.5 mM malate. The inhibition of respiration during state 4 was monitored using an oxygen polarograph. VIII. OTHER ASSAYS Cytochrome oxidase was quantitated as previously described (Toth gt al., 1986). The P contained in the assay medium used for oxygen consump- 1 tion assays was determined by the method of Bencini §£_al. (1983), using potassium phosphate as the standard. Mitochondrial protein was determined by the modified Lowry method of Markwell §t_al. (1981), using BSA as the standard. For the experiments described herein mitochondria concentrations were expressed in terms of the amount of cyt aa in oxygen consumption 3 experiments and by protein in ATPase assays. The former was done in order to minimize the error in estimating respiratory parameters due to contaminating protein. ATPase activities were quantitated on a per mg protein basis because these enzymes are not associated with respiration per se. IX. CALCULATIONS AND STATISTICAL ANALYSES Hyperbolic Michaelis-Menten type kinetic data were analyzed for apparent Vmax and Km values using the computer program Wilman IV (Brooks and Suelter, 1986). The kinetic data fitted to equation II were analyzed using a nonlinear regression program provided by Dr. Stephen Brooks (Carleton University, Ottawa, Canada). Data were analyzed assuming a constant absolute error as specified by Wilkinson (1961), and were 85 assumed to fit the specified kinetic model if the residuals of points from the calculated regression line were randomly distributed (Mannervik, 1982). The least significant differences at the p < 0.01 and p < 0.05 levels between the respiratory parameters associated with successive additions of ADP were calculated by performing an analysis of variance using the F-test with software provided by Dr. Stanley Ries (Michigan State University). 86 RESULTS I. RATES OF RESPIRATION VARY WITH MITOCHONDRIA CONCENTRATION Rates of mitochondrial oxygen consumption vary with the concentration of mitochondria in a complex manner (Fig 2). For the experiments shown in Fig 2, mitochondria oxidized 5 mM pyruvate and 2.5 mM malate. The state 3 rates of respiration for chick heart mitochondria isolated with collagenase are independent of mitochondria concentration. In contrast, state 4 rates initially decrease toward a minimum value [515 ng-atom 0 min-1 (nmol cyt aa )-i] and then increase toward a higher approximately 3 constant value [70 ng-atom O min-i (nmol cyt aa3)-i] as the concentration of mitochondria is increased. The amount of mitochondria necessary to minimize state 4 rates corresponds to 0.05 i 0.005 nmoles of cyt aa3/ml (n = 30). The highest RCR values for a given mitochondria preparation are obtained at this concentration of mitochondria (see Fig. 5 of Chapter 1). It will, therefore, be referred to as the "critical mitochondrial concentration." The concentration dependence of state 4 respiration is striking. Insight into the basis for this relationship is provided by repeating these measurements with chick heart mitochondria isolated in the presence of the non-specific protease Nagarse. For mitochondria isolated by this method, state 4 respiration at or below the critical mitochondrial con- centration (CMC) is completely abolished. In this concentration range the availability of ADP becomes rate-limiting for respiration by Nagarse treated mitochondria. As mitochondria concentrations are increased above the CMC, state 4 rates of respiration rapidly increase and reach a plateau. Figure 2. 87 Rates of State 3 and State 4 Respiration of Mitochondria Measured as a Function of Mitochondria Concentration Rates of state 3 (A) and state 4 (B) respiration were measured for mitochondria isolated with either collagenase (.) or Nagarse ((:)). Mitochondria were quantitated on the basis of cytochrome oxidase concentrations. All rates were obtained subsequent to the addition of ADP The substrates for 2. respiration were 5 mM pyruvate/2.5 mM malate and 20 mM Tris- buffered Pi’ pH 7.4. Other assay conditions were as described under "Experimental Procedures." Rate of Respiration ng - atom 0- min"- (nmol cyt 003)" 88 1200 IIOO 1000 900 800 ;\ \\' 80 70 60 50 4o 30 0.2 l 0.3 cyt 003, nmoles 04 ii 0.5 06 89 Mitochondria isolated with Nagarse respire at significantly higher rates during state 3 respiration than mitochondria isolated with collagenase. At the CMC, Nagarse and collagenase treated mitochondria respire at 1144 )‘1. i 75 (n 10) and 789 i 93 (n = 30) ng-atom 0 min.1 (nmol cyt aa 3 respectively. Above the CMC, rates of state 3 respiration decrease toward values approximating those for collagenase mitochondria. These results suggest that Nagarse destroys a factor that stimulates state 4 respiration. In addition, limited Nagarse digestion appears to release some type of control over the Vmax of oxidative phosphorylation. II. ADP:O RATIOS ALSO VARY WITH MITOCHONDRIA CONCENTRATION The empirical ADP:O ratios of chick heart mitochondria oxidizing pyridine nucleotide linked substrates also vary as a function of mito- chondria concentration (Fig 3). ADP:0 ratios are highest at or below the CMC. At the CMC, ADP:O ratios range from 3.2-3.5 when 5 mM pyruvate/2.5 mM malate (P/M), 5 mM B-hydroxybutyrate (BHB), or 5 mM a-ketoglutarate (0K0) are oxidized during the phosphorylation of a second addition of approximately 400 nmoles of ADP (ADPZ). As the concentration of mito- chondria is increased, ADP:O ratios decrease to approximately 3.0. No difference is observed in the magnitude nor in the progressive fall in ADP:O ratios when mitochondria isolated with collagenase are compared with those isolated with Nagarse. III. MAGNITUDE OF RESPIRATORY PARAMETERS AT THE CMC. The respiratory parameters of chick heart mitochondria at the CMC oxidizing a variety of substrates are summarized in Table I. ADP:O ratios increase with each successive addition of ADP during the oxida- Figure 3. (A) (B) (C) (D) 90 Variation of ADP:O Ratios as a Function of Mitochondrial Concentration The substrate(s) oxidized and the enzyme used to facilitate mitochondrial isolation were as follows: 5 mM pyruvate/2.5 mM malate, collagenase. 5 mM pyruvate/2.5 mM malate, Nagarse. 5 mM B-hydroxybutyrate, collagenase. 5 mM a-ketoglutarate, collagenase. These ADP:O ratios were obtained after a second addition of 418 nmoles of ADP. The concentration of ADP and oxygen atoms in these assays was determined as described under "Experimental Procedures." ADP= 0 91 3.5 - 3.0J; 4p M 3.0 L ‘l 0A D N _M_/ I 1 l I 2 .3 .4 .5 .[cyt 003] , nmoles l .6 92 mm.F m.m we? _F: m cecoeoeom :1 m m Fm._ s.m mom mm: _ \eemeaoosm ze m emo.o A mm.m m._ n w.o_ *m.m A m.» *oms a wow m eemamz 2a m.m m ms.o e =A.m m.F n z.m_ m.,m H w.ee me. A sea _ \mememesao :5 m *F_.o H mm.m m.o A m.ml *e.o_ H 3.3m *AA A _em m o o_.o H eo.m o._ A A.o_ m.__ e o.em mo_ e m_o _ memeseeesxoeezmim :5 m esfi.o A em.m m.zm a z.mm *e.zr « 5.0m *o.mm H mm: m oF.o A m_.m 2.? A m.FF 0.: A e.mm 0.3m H soc _ mememezsmoeexie as m *m_.o H mm.m mo o.m A w.e 3.3m A mme m *mF.o e 0:.m om *e.e H N.m *_.mm H mm» N eemfimz :8 m.m om zl.o H mm.m s.m H F.m_ m.om H m.:m m.em H owe _ \eem>3e>a as m z msmem m eemsm Ammm uzo Hoecv\cws\o Boomismemc e m.oo"ao< emom oceaemeaamem Co msmm Dana Amvmsmeemesm m.mm .mooamnanoos owono>o oHpoenpfino no mnfixme anononu .Axuflnfionfi u momv onoN zaunosconm ono: mo< no muoscaao omono mo noHpoaznonamonq on» mnfizoaaom mono; z oumum onu oozooon onoo no: oflne .mouon : opmum ono m opmoo noos on» no oHpoL on» mnfixoo mo oopoasoaoo ono: oouonuonso ono opoaos onm opo>3n>q non: mmo< ono mma< no» oofiuon Honpnoo mnoamanoon ommno>o one .Aev mo.o v a no Axv _o.o v a no unoofluflnmfio onm mo< mo onofiuwouo o>Hmmooozo mnfizoaaou onoposonoa anooonwaoon noozpon moononopmfio .nfle P Lou ooonuonso no“: mfinononoOSAs mnfiuonsonfi miFmo< "moaosn wF: av mo< no mangoeo noaoefisco no onofiufiuom o>Hooooosm oonnu on 039 LmUMm mum—b ®Lm3 A .Aom u :v He\moo p>o Hos: moo.o H omo.o mo: ozoooo ooonu nH nOASonpnoonoo Hoanononooows nooe one .onoflponoaona no Lopes: nopmofinnfi onp non .m.m H onmos on» onm nzonm mosao> .Hm :8 ow mnflnfloonoo ssfiooe unoaaso oHposoo ooono:m\HOpHnnoE o nH nofloonflaoon now oo>oooo ono: mononuonsm osofino> mnHNHuon ofinononOOpfiz 94 tion of all of the NAD-linked substrates. These increases are signifi- cant at the p < 0.01 and p < 0.05 levels, as indicated. The data show that the ADP:O ratios are fractional and greater than 3.0 when P/M, BHB, or aKG support oxidative phosphorylation. These coupling ratios are not artifically inflated since: (a) both ADP and oxygen concentrations were enzymatically determined, and (b) after three sequential additions of ADP to mitochondria oxidizing aKG, no residual ADP is enzymatically assayable. ADP:O ratios are less than 3.0 when 5 mM glutamate and 2.5 mM malate (G/M) are used as oxidizable substrates. ADP:0 ratios obtained during the oxidation of 5 mM succinate approximate to the long accepted value of 2.0. The exogenous substrates are oxidized at different rates during states 3 and 4 when Pi concentrations are held fixed at 20 mM. The oxidation of P/M or G/M supports similar state 3 rates. The oxidation of aKG and BHB appears to generate reducing equivalents at equal rates, though more slowly than either P/M or G/M. When succinate is used as substrate, the rate of respiration is one-half that obtained with P/M. During the state 4 respiration following the first addition of ADP, the NAD-linked substrates are oxidized at comparable rates. Subsequent to the phosphorylation of ADP state 4 rates of respiration decrease 2- 2. to 10-fold, depending upon the substrate. The highest RCRs are obtained during the oxidation of P/M. As shown in Fig. 4A, the oxidation of P/M by chick heart mitochondria is subject to adenine nucleotide control. Under conditions in which the phosphorylation of ADP by the F1FO-ATP synthase is blocked by Oligomycin, maximal rates of uncoupled respira- tion are only obtained following the addition of ADP. Thus, ADP must regulate one (e.g., isocitrate dehydrogenase) or more catalytic steps 95 Figure 4. Effect of ADP on the Degree of Uncoupling Induced by CCCP when Mitochondria Oxidize Different Substrates Mitochondria at the critical mitochondrial concentration were suspended in an assay medium comprised of 0.225 M mannitol, 0.075 M sucrose, and 20 mM Tris-Pi, pH 7.4. Rates of oxygen consumption were measured with an oxygen polarograph as described under "Experimental Procedures." The abbreviations used are as follows: M, mitochondria; ADP, 420 nmoles per addition; 0, 1.7 ug Oligomycin A/ml; CCCP, 2.9 nmoles of this uncoupler/ml per addition. The substrate(s) used to support respiration were: (A) 5 mM Pyruvate/2.5 mM Malate (B) 5 mM B-Hydroxybutyrate (C) 5 mM Succinate 96 97 of the tricarboxylic acid cycle in chick heart mitochondria. During the oxidation of BHB, ADP does not increase rates of uncoupled respiration (Fig. 4B). The addition of an uncoupler to mitochondria oxidizing succi- nate during state 4 has no effect on the rate of respiration (Fig. 4C). This result suggests that succinate oxidation is already very poorly coupled to the phosphorylation of ADP. These observations correlate well with the substrate-dependent RCR values and state 4 rates shown in Table I, in that chick heart mitochondria exert greatest control over respira- tion with P/M and least control with succinate. IV. EFFECT OF PRE-INCUBATING MITOCHONDRIA WITH SUBSTRATE ON RESPIRATION As indicated in Table I, the rates of both state 3 and state 4 respiration decrease subsequent to each sequential addition of ADP. The decreases in rates of respiration are all statistically significant at the p < 0.01 level. It was of interest to investigate the cause of this behavior in greater detail. The rate of state 4 respiration following the addition of 840 nmoles of ADP [35-55 ng-atom O min-i (nmole cyt aa3)-1] does not correspond to that observed if two separate additions of 420 nmoles of ADP are interposed by a 2 min period of state 4 respi- ration [<10 ng-atom 0 min.1 (nmole cyt aa3)-i]. Since the state 4 period of respiration prior to the addition of ADP is necessary in order to 2 observe a high value for RCR it was reasoned that during state 4 some 2’ kinetic or thermodynamic pressure was established which would limit the rate of mitochondrial substrate oxidation in the absence of ADP. Because prolonged exposure of mitochondria to P1 in the absence of substrate 13 deleterious to the structure and function of these organelles (see Chapter 5) it was possible that continued exposure of the mitochondria 98 to substrate during state 4 was causing the decreases in respiratory rates. Indeed, as the time of pre-incubation with P/M prior to the initiation of oxidative phosphorylation with ADP is increased, the rate of state 3 respiration decreases linearly by 30 ng-atom 0 min”1 (nmole cyt aa3)-1/min (Fig 5). Similarly, the rate of uncoupled state 4 respi- ration also decreases, but it does so in a non-linear manner. On a percentage basis the rate of change in state 4 respiration is much greater than that for state 3; consequently, the capacity to control respiration becomes quite high (Fig 5, Inset). V. OLIGOMYCIN SENSITIVITY OF STATE 4 RESPIRATION In an effort to discern whether ADP is regenerated from ATP during state 4, increasing concentrations of mitochondria were treated with an amount of oligomycin A sufficient to completely inhibit ATP synthesis. Below the CMC the state 4 respiration of chick heart mitochondria is unaffected by oligomycin (Fig 6), indicating that the protease-sensitive factor is not an ATPase. Above the CMC state 4 respiration becomes increasingly sensitive to oligomycin in both collagenase and Nagarse preparations of mitochondria. Up to 45% of the total respiration at high mitochondria concentrations (>O.3 nmol cyt aa /ml) is inhibitable 3 by oligomycin. Consequently, at concentrations exceeding the CMC, a percentage of state 4 respiration is due to continuous ATP synthesis. The steady-state concentrations of ADP established during state 4 with different mitochondria concentrations were inferred from the above oligomycin inhibition studies. By assuming that the ADP requirements of oxidative phosphorylation conform to hyperbolic-type saturation kinetics, the following substitutions were made in the Michaelis-Menten equation: 99 Figure 5. Rates of State 3 and State 4 Respiration at the Critical Mitochondrial Concentration Measured as a Function of Time of Preincubation with 5 mM Pyruvate/2.5 mM Malate Prior to the Addition of ADP Respiration was monitored at 30.5 °C in an assay medium that contained 0.225 M mannitol, 0.075 M sucrose, and 20 mM Tris-buffered Pi. Least squares regression analysis was used to fit a line through the points for state 3 respiration. Insert: Respiratory control ratios measured as a function of time of preincubation of mitochondria with 5 mM pyruvate/2.5 mM malate. 100 11200 Mo IOUJU)°..U!UJ :0 wow -6u uoliaiidssa 17 aims l O O a) R 8 8 8 e I I i I I :0 ¢ “ o .‘c’.’ a m 6': :§: .. .83 vel- ( §S - o a U d§E "- 3 .83 I'- O l l l 1 1 o O Q A 11 E888 §§ c I_(‘00 Mo |0LIJU)-._ugu.1 - 0 wow -6u uoliaildssg g swig 600 500 300 TIME OF PREINCUBATION, sec 200 DO Figure 6. 101 Sensitivity of State 4 Respiration to Oligomycin A Measured as a Function of Mitochondria Concentration Increasing concentrations of chick heart mitochondria isolated with either collagenase (A) or Nagarse (B) were treated with 0.86 pg oligomycin A ml.1 during the period of state 4 respi- ration subsequent to ADP . (.), respiration before the 2 addition of oligomycin; ((:)), respiration after the addition of oligomycin. 102 -100 -80 ~60 ~40 dzo o 0 tn 0 2 2 a. a: g _9 O .2 8 N. _,., ‘9'. 1.. < ' 2» 3 s a .o 8 m 0 v; N I.("00 Mo |OUJU)-I_u!w .' 0 wow -5” ”04049598 17 albig [cyt 003], nmoles Figure 7. 103 Inhibition of Chick Heart Mitochondrial State 3 Respiration with Increasing Concentrations of Carboxyatractyloside Mitochondria (0.074 nmols cyt aa /ml) suspended in M/S oxidized 3 5 mM pyruvate and 2.5 mM malate in the presence of 20 mM Tris- buffered Pi’ pH 7.4. State 3 respiration was initiated by the addition of 2.4 nmoles of ADP. Mitochondria were allowed to phosphorylate ADP for 1 min. Varying concentrations of carbox- yatractyloside were added, and the state 3 rate subsequent to the binding of inhibitor was measured. 104 §§§§§8 ° “(9‘00 M0 |OLUU). "qu .0 mom-bu uo'uaigdseg 9; 91013 10 slog 3.0 4.0 5.0 -I (nmol cyt aa3) 2.0 1.0 nmoles carboxydtractyloside 105 (state 4 rate)(ADP:O) = [(state 3 rate)(ADP:O)][ADP] (4) Km + [ADP] where the product of respiratory rate and the ADP:O ratio is the rate of ATP synthesis, and Km is the Michaelis constant of the chick heart mitochondrial adenine nucleotide translocase for ADP (20 uM, Dr. S. Brooks, personal communication). After taking the reciprocal of both sides and rearranging, equation 4 reduces to: [ADPJSS = K ° - 1 (5) J 4,olig o where JO3 is the oxidative flux during state 3 respiration and J01"0118 is the oligomycin-sensitive oxidative flux during state 4. In this experiment, the adenine nucleotide translocase and ATPase(s) operate as a coupled enzyme system. Steady-state ADP concentrations could thus be maintained if the rate of ATP hydrolysis is equal to the rate of ATP synthesis. Because the measured parameter is an increase in the rate of respiration, it must be established that the translocase can be rate- limiting for oxidative phosphorylation. As shown in Fig 7, this appears to be the case. Although the plot shows slight sigmoidal character, carboxyatractyloside inhibits a percentage of state 3 respiration at all concentrations tested (KIapp = 3 uM). State 3 respiration is inhibited by over 99% when the molar ratio of carboxyatractyloside to cyt aa is 3 equal to 1.0. Since the adenine nucleotide translocase and cyt aa3 are present in the mitochondrial inner membrane at equimolar concentrations, carboxyatractyloside is clearly a very effective inhibitor. The calcu- 106 lations presented in Fig 8 show that the steady-state concentrations of ADP increase linearly as a function of mitochondria concentration. More- over, mitochondria isolated with either collagenase or Nagarse maintain approximately the same steady-state concentrations of ADP during "state 4" respiration, being 2.5 and 2.0 pM ADP (nmole cyt aa3)—i, respectively. VI. THE EFFECT OF EGTA ON RESPIRATORY PARAMETERS The data in the inset of Fig 9 show that the oligomycin-sensitive compo-nent of state 4 respiration can also be inhibited with EGTA. The inhibition is instantaneous and decreases the rate of state 4 respira- tion over two-fold. Titrating a fixed concentration of mitochondria with increasing concentrations of EGTA shows significant initial inhibition below 250 uM, followed by a subsequent slower rate of inhibition (Fig 9). This result suggests that the EGTA binds two types of divalent cation which have different affinities for this chelating agent. For the preparation titrated, 17 mM EGTA completely inhibited the oligomycin- sensitive respiration. EGTA (20 mM) also prevents ADP:O ratios from decreasing as mitochondria concentrations are increased (data not shown). VII. THE EFFECT OF Mg(II) ON RESPIRATORY PARAMETERS A direct implication of the inhibition of state 4 respiration by EGTA is that divalent cations released from mitochondria stimulate oxygen consumption by facilitating the hydrolysis of ATP. As a test of the tenability of this hypothesis, chick heart mitochondria at the CMC were titrated with increasing concentrations of Mg(II). Mg(II) causes a precipitous drop in RCRs from 200 down to 4 (Fig 10). State 3 rates are unaffected by Mg(II) (Fig 10, Inset). The decreases in respiratory Figure 8. 107 Steady-State Concentrations of ADP Maintained During State 4 Respiration The concentrations of ADP regenerated by increasing concentrations of mitochondria isolated with either collagenase (O) or Nagarse (.) were inferred from the oligomycin sensitivity of state 4 respiration using equation 3. The lines through the data were calculated by least squares linear regression. 108 I.O - [ADP] 55, FM A I _2 1 1 1 1 J .I .2 .3 .4 .5 [cyt 00,], nmoles Figure 9. 109 Inhibition of State 4 Respiration by IncreasingyConcentrations of EGTA Mitochondria above the critical mitochondrial concentration (0.2 nmol cyt aa ml-i) oxidized 5 mM pyruvate/2.5 mM malate 3 in M/S medium that contained 20 mM Tris-buffered Pi’ pH 7.4. The plot shows the percent inhibition of state 4 respiration by increasing concentrations of EGTA following a single addition of 334 nmoles of ADP. Insert: Oxygen polarograph trace showing that EGTA inhibits all of the oligomycin-sensitive respiration during state "4". M, mitochondria at 0.14 nmol cyt aa3 ml-1; ADP, 334 nmoles; EGTA, 27 mM; 0, oligomycin A at 1.7 ug ml-1. Rates of state 4 respiration shown are expressed as ng-atom 0 min“1 (nmol cyt aa3)-i. Percent Inhibition of State 4 110 30 - . ‘D" I 3.4-129 2 3.4““ - 62.4 20 I I 3 3.4“ - 62.4 Mite ‘ ma 0 10 § \/ I 88 2min EGTA ONO. ADP O l 1 1 o s 10 Is [EGTA], mM Figure 10. 111 Effect of Mg(II) on Chick Heart Mitochondrial Respiratory Control The critical mitochondrial concentration was titrated with increasing concentrations of Mg(II) under conditions described in "Experimental Procedures." (.), RCRs secondary to ADP - ((:)), RCRs secondary to ADP Insert: 1’ 2' Effect of Mg(II) on state 3 respiration. (. ), state 3 induced by ADP - (Q), state 3 induced by ADP 1’ 2° 112 ZOO, . ADP, O 120 L0 1 .4 Total [Mg (11)], mM 1 .2 mm m «L... 9 O 7 .118 to 365.752 .0 £20.? 8:235. n 22m m m 020m .2200 322.3% IIO - IOO so .. 4O .- II 5.0 In Total [Mg (11)], mM 113 control are due entirely to changes in state 4 respiration (Fig 11B). The manner by which Mg(II) stimulates state 4 respiration is clarified by the following experiments. First, the respiration stimulated by Mg(II) is completely inhibited by oligomycin, indicating that Mg(II) stimulates the hydrolysis of ATP synthesized during state 3 (data not shown). Second, ADP is regenerated extramitochondrially, as evidenced by the ability of pyruvate kinase to inhibit the respiration stimulated by Mg(II) (data not shown). Between 0 and 90 uM, the stimulation of ATP hydrolysis by Mg(II) is linear (Fig 11B). Above 90 uM Mg(II) no further stimulation is observed, suggesting that the ATPase(s) are saturated. The concentration of Mg(II) which gives half-maximal stimulation is approximately 34 uM. When the ATPase activity is saturated with Mg(II), oxygen consumption is stimulated to 20% of the Vmax of state 3 respi- ration. As to be shown in Chapter 5, the Mg(II) stimulated ATPase activity is not due to an ion-motive transmembrane Mg(II)-ATPase. Mitochondrial ADP:O ratios are also depressed by Mg(II) (Fig 11A). For example, the ADP:O ratio decreases from 3.10 to 2.54 following the addition of 1 mM Mg(II). In parallel with the above findings, ADP:O ratios are depressed because Mg(II) is stimulating the hydrolysis of ATP during state 3. This keeps the enzymes (adenine nucleotide trans- locase and F1FO-ATP synthase) regulating oxidative phosphorylation supplied with ADP for longer periods of time, thereby decreasing the empirical ADP:O ratio but not the "true" coupling of these organelles. Data presented in Chapter 5 show that chick heart mitochondria release Mg(II) and Ca(II). Divalent cations so released into the extramitochon- drial space will stimulate ATPase activity. Increasing the concentration of mitochondria in an assay therefore stimulates ATPase activity by Figure 11. 114 Effect of Mg(II) on ADP:O ratios and State 4 Respiration The critical mitochondrial concentration was titrated with increasing concentrations of Mg(II) and the effect on ADP:0 ratios (A) and state 4 rates of respiration (B) was quantitated. All data points were obtained subsequent to the phosphorylation of ADP1. 115 I l 1 I 1 ~9 .0, E ~_>_ 130 Figun;r'-’e~17. Titration of Chick Heart Myosin with Mg(II) in Either the Absence or Presence of Ca(II) Assays were performed as described under "Experimental Procedures." The curves through the data are theoretical, calculated with kinetic constants estimated by a least- squares non-linear regression of the data fitted to equation 1. The protein concentration in these assays was 0J4 mg/ml (0, +50 uM Ca(II)) and 0.36 mg/ml (., no Ca(II)). 131 0 10- TAE8>E 95. TEE. 85:06? 3.4. $65.: 200 300 fiflgCflfl,;lN| l00 132 Figure 18. Titration of Chick Heart Myosin with Ca(II) Myosin was suspended at 0.36 mg/ml and assayed as described under "Experimental Procedures." The theoretical curve drawn through the data was calculated with the kinetic constants estimated by a non-linear regression of the data fitted to the Michaelis-Menten equation. 133 - — p — b m m w mm 120 _ — p m m 7382: oEYTEE. 850.6? nr2. 868: O 2000 I500 200 400 600 800 1000 0 [Ca(II)] , pM 13“ TABLE II. APPARENT KINETIC CONSTANTS 0F CHICK HEART DIVALENT CATION STIMULATED ATPases ATPasea DIVALENT CATION(S) x b or K c v d m I max uM Mitochondria Mg(II) 35 i 5.” b 33 t 1.2 Ca(II) 315 t 61.8b 97 t 5.8 Myosin Mg(II) . 6 1 1.” c 37 t 2.6 Mg(II) + 50 uM Ca(II) 17 t 2.8 c u? t 1.9 Ca(II) u93 i 32.1b 187 1 1.5 As described under "Results," the term "ATPase" designates the preparation in which ATPase activity resides. It does not necessarily identify the ATPase(s) operating in the assay system. Concentration of divalent cation which gives half-maximal stimulation of total ATPase activity in the preparation assayed. Concentration of divalent cation which gives half-maximal inhibition of total ATPase activity in the preparation assayed. Maximal rates are expressed as nanomoles of ATP hydrolyzed/min/(mg of ATPase preparation). 135 site on myosin. Because the affinity of Mg(II) is so much higher than Ca(II), Ca(II) only weakly interferes with the binding of Mg(II) to myosin. The kinetic data alone do not provide strong support for the hypothesis that myosin is the second ATPase in these mitochondria preparations. The Mg(II) titrations for the two systems (mitochondria and myosin/actin) show different behavior. The ATPase activity of the mitochondria preparation is stimulated by Mg(II), whereas myosin ATPase activity is inhibited by this cation. Some correlation is seen between the Ca(II) titrations for the two systems. Both are stimulated by Ca(II) (Fig's 17 and 18) and the apparent Km's are similar (Table II). Immunoblotting afforded a more direct means of resolving this issue. As depicted in Fig 16, both actin and myosin are present in suspensions of mitochondria that were isolated with collagenase. This provides an unequivocal demonstration of the identity of the ATPases which contaminate suspensions of isolated heart mitochondria. The dis- crepancies between the kinetic data may be explained by differences in the degree to which the actin and myosin were associated in the samples of myosin and mitochondria that were tested. X. Ca(II) UPTAKE BY CHICK HEART MITOCHONDRIA Like most other mitochondria (Nicholls and Crompton, 1980), chick heart mitochondria transport Ca(II) from the extramitochondrial space into the matrix. As reported by the Ca(II) indicator Arsenazo III, uptake requires substrate and oxygen (Fig 19A). When the oxygen dissolved in the assay medium is exhausted, the Ca(II) is released back into the extramitochondrial space. In this experiment ADP was added to Figure 19. (A) (B) 136 Kinetics of Ca(II) Uptake and Release by Isolated Chick Heart Mitochondria The reaction mixture contained 0.225 M mannitol, 0.075 M sucrose, 20 mM Tris-buffered Pi’ pH 7.“, and 100 uM arsenazo III. Ca(II) transients were monitored by measuring the absorbance changes of arsenazo III at 685-675 nm with a dual wavelength spectrophotometer. M, mitochondria (0.25 mg/ml); P/M, 5 mM pyruvate/ 2.5 mM malate; ADP, H00 nmoles; R, 50 uM rotenone. M, mitochondria (0.25 mg/ml); P/M, 5 mM pyruvate/ 2.5 mM malate; R, 50 pH rotenone; S, 5 mM succinate. 138 ensure that anaerobiosis was achieved. Ca(II) release is unaffected by rotenone. The uptake of this cation, however, is inhibited by rotenone when P/M is used as oxidizable substrate (Fig 19B). By bypassing the rotenone-block with succinate, Ca(II) uptake is reinitiated. It is noted that in these experiments the mitochondria transported much more Ca(II) than they would in a typical oxygen consumption experiment because the commercial preparation of Arsenazo III used was contaminated with Ca(II) (A. Scarpa, personal communication). 139 DISCUSSION I. UNCOUPLED RESPIRATION 0F CHICK HEART MITOCHONDRIA Respiration that is not coupled to the phosphorylation of ADP is stimulated by Ca(II) and by a Nagarse-sensitive factor presumably bound to the outer membrane of chick heart mitochondria (for a description of some of the properties of this factor, see Chapter 7). From the data in Table I it is clear that there are significant decreases in both state 3 and state A rates of respiration subsequent to the phosphorylation of each sequential equimolar addition of ADP. The likely basis for this behavior is that the Ca(II) in the extramitochondrial space is taken-up into the mitochondrial matrix in an energy-dependent manner. As the extramitochondrial concentration of Ca(II) decreases, a greater percen- tage of AuH+ may be committed to ATP synthesis and, consequently, the rate of substrate oxidation is decreased. In addition, it is possible that as the concentration of ATP and Ca(II) in the extramitochondrial space increase and decrease, respectively, the Nagarse-sensitive uncoupling factor dissociates from the outer membrane. This suggestion makes two pre-suppositions: (a) that the binding of this factor to mitochondrial membranes is a regulated phenomenon; and (b) that ATP and Ca(II) may play a role in regulating the strength with which the uncoupling factor binds to mitochondrial membranes. Additional experi- ments must be performed in order to validate the latter two points. The stimulation of uncoupled respiration is attributable to other phenomena as well. The uptake of Pi pyruvate, B-hydroxybutyrate, and 1110 glutamate are energy-dependent processes. It is probable that until the concentrations of these solutes in the intra- and extramitochondrial spaces are equal, uptake into the matrix will continue. These substrate internalization processes constitute an energy drain. Differences in the rates of oxidative phosphorylation supported by different substrates are likely due to differences in the rates at which the dehydrogenases for these substrates can produce reducing equivalents. However, differences in the rates of uncoupled respiration during state A (assuming that they are corrected for Ca(II) transport and Nagarse-sensitive uncoupling activity) are probably due to two other factors: (a) the rate at which different substrates can be transported into the matrix; and (b) the degree to which different dehydrogenase activities can be regulated. Point (b) is supported by the experiments shown in Fig A, which indicate that succinate oxidation is a very poorly controlled reaction; in contrast, the oxidation of pyruvate is subject to regulation by ADP. II. ATPase ACTIVITY IN THE EXTRA-MITOCHONDRIAL SPACE These studies suggest that the Ca(II) and Mg(II) stimulated ATPase activity of isolated chick heart mitochondria is due to a myosin contami- nant and to the F1F0-ATPase molecules of broken mitochondria. The divalent cation sensitivity of respiration in isolated heart mitochondria is a well known phenomenon. By inhibiting part of the Mg(II)-stimulated ATPase activity of isolated bovine heart mitochondria with the F1FO-ATPase inhibitor protein (Cintron and Pedersen, 1979; Amzel and Pedersen, 1983), Barbour §t_al. (198A) concluded that their preparation was contaminated with a small population of broken mitochondria. Myosin contamination of many heart mitochondria preparations has been assumed but never demonstrated. 1N1 Though not quantitated in this study, it is also possible that the sponta- neous hydrolysis of ATP induced by the binding of Mg(II) to this nucleotide contributes to the total ATPase activity of this mitochondria preparation (Terada §£_al., 198A). The divalent cation-stimulated ATPase activity is an important cause of the concentration dependence of chick heart mitochondrial ADP:0 ratios and state A rates of respiration. The ATPases and the adenine nucleotide translocase operate as a coupled enzyme system. Increases in ATP cleavage rates result from increases in the concentrations of Ca(II), Mg(II), and of the ATPases themselves. Ca(II) is released from these mitochondria in the absence of exogenous substrate; Mg(II) is released during P uptake (see Chapter 5 for experimental details). As the con- i centration of mitochondria is increased during an oxygen consumption assay, increasing amounts of all three of these components will be present. Thus, the concentration of ATP that is hydrolyzed to ADP during state A will increase as the concentration of mitochondria is increased. Increasing rates of ATP hydrolysis will lead to progressive decreases in ADP:0 ratios and increases in state A rates of respiration. In most studies characterizing the respiratory parameters of mitochondria, high concentrations (21.0 mg/ml) of these organelles are used. This will, particularly with heart mitochondria, lead to severe underestimates of ADP:0 ratios and overestimates of state A rates. Depending upon the relative affinity of Ca(II) for the ATPases and the Ca(II) carrier systems of the mitochondria, ADP:0 ratios would be expected to increase with each successive addition of ADP as the Ca(II) is cleared from the extramitochondrial space. The data reported herein are consistent with this hypothesis. 142 III. THE CRITICAL MITOCHONDRIAL CONCENTRATION A plot relating state 4 rates of respiration to chick heart mito- chondria concentration is complex. When mitochondria are isolated with collagenase, there is a highly reproducible concentration of mitochon- dria at which state A rates of respiration are at a minimum. Below the cmc, all of the state 4 respiration is due to uncoupled respiration (Ca(II) and substrate transport, uncoupling factor activity). Above the cmc state A respiration is due to both uncoupled respiration and respi- ration induced by the regeneration of ADP. It is of interest that oligomycin-sensitivity of state A respiration is only observable above the cmc. At the cmc, ADP will be present in the steady-state at approx- imately 0.2 uM. This is 0.01Km of the adenine nucleotide translocase for ADP. Thus, at or below the cmc, respiration is probably stimulated by ADP to a degree that is below the sensitivity threshold of the oxygen polarograph. To our knowledge, these are the lowest state A rates of respiration yet reported for any preparation of mitochondria. Under the conditions of these experiments, the diffusion of atmospheric oxygen back into the oxygraph vessel does not artificially depress the measured state A rates of chick heart mitochondria (Toth g£_gl., 1986; see Fig 7 of Chapter 1). With respiratory control ratios approaching infinity during the oxidation of pyruvate and malate, it is apparent that these mitochondria must: (a) have an extremely low rate of flow through intrinsic proton leak pathways, and/or (b) some dehydrogenases can be rate-limiting for respiration under particular state A conditions. These data also suggest that at the cmc under the conditions prevailing during the period of 143 state A respiration subsequent to ADP (i.e., when ADP, bound uncoupling 2 factor, and extra-matrix Ca(II) concentrations are presumably minimal), the electron transport chain can come very close to equilibrium. IV. ADP:0 STOICHIOMETRIES FOR OXIDATIVE PHOSPHORYLATION There is continued disagreement over the ADP:0 (or ATP:0) stoichio- metries of heart and liver mitochondria. In the following analysis it is assumed that, like rat liver mitochondria (Lemasters, 198“), the H+/ATP ratio of chick heart mitochondria is equal to A. In the vast majority of studies, isolated mitochondria respire with RCRs ranging from 3-10. This indicates that the mitochondria are not well coupled. Therefore, in order to estimate the true mechanistic stoichiometry of oxidative phosphorylation, some workers have applied mathematical corrections to their data. These corrections take into account the degree of coupling (Lemasters, 198”) or allow for an extrapolation when the RCR is assumed to be infinite (Beavis and Lehninger, 1986). During the oxidation of pyruvate and malate, the RCRs of chick heart mitochondria at the cmc closely approach infinity (i.e., state A rates of respiration + 0). This indicates that, under these conditions, these mitochondria are almost perfectly coupled. No correction of the data should be necessary. The empirically obtained ADP:0 ratios approximate to 3.5 during the phos- phorylation of ADP and ADP3 (see Table I). Because ADP:0 ratios 2 approximate to 2.0 during succinate oxidation, this suggests that the H+/0 stoichiometry of complex I in chick heart mitochondria is 6. The total H+/0 stoichiometry for complexes III and IV would approximate to 8. Thus, based on a comparison of the ADP:0 ratios obtained during the oxidation of pyruvate/ malate and of succinate, the data support a 1R 1AM proton model for oxidative phosphorylation. Substrate-level phosphorylation is not taken into account by these studies. Therefore, it may be argued that the ADP:O ratios obtained with pyruvate/malate are inflated due to the phosphorylation of ADP by nucleo- side diphosphokinase. B-Hydroxybutyrate is an NAD-linked substrate whose oxidation does not involve substrate-level phosphorylation. During the oxidation of this substrate, the average ADP:0 ratio obtained subsequent to the addition of ADP2 is 3.33. This ratio would be consistent with a 13 proton model for oxidative phosphorylation (5 for complex I and 8 for complexes III and IV), and is in excellent agreement with the studies of Lemasters (198“). Of interest is that the oxidation of a-ketoglutarate also results in ADP:0 ratios consistent with a 13 proton model. The con- tribution of substrate-level phosphorylation to ATP synthesis during a- ketoglutarate and pyruvate/malate oxidation cannot be discerned from these data. Although 1-5 mM malonate is often included in the assay media used by other workers as a means of inhibiting substrate-level phosphorylation, assays on chick heart mitochondria show that even at 100 mM this compound does not completely inhibit succinyl CoA oxidation. A surprising result of these studies is that coupling is relatively insensitive to RCRs. The RCRs obtained during the oxidation of glutamate/ malate, a-ketoglutarate, and B-hydroxybutyrate are approximately equal. Yet, the ADP:0 ratios obtained with glutamate/malate (2.8-3.0) are sig- nificantly lower than those obtained with a-ketoglutarate or B-hydroxy- butyrate (z3.3). The reason for this is not immediately apparent. One point that can be made is that glutamate oxidation results in a lower yield of ATP per atom of oxygen consumed when compared with the other pyridine nucleotide-linked substrates. It is possible that the stoichi- 1N5 ometry for oxidative phosphorylation is depressed by the ammonia produced during the glutamate dehydrogenase reaction. This assumes that ammonia, if present at sufficient concentrations, is moderately toxic to mito- chondria. Substantiation of this possibility will require additional experimentation. Even though chick heart mitochondria are very poorly coupled during succinate oxidation, the empirical ADP:O ratios obtained with this substrate conform closely to the classically accepted value of 2.0. V. CONCLUSIONS These studies show that: (a) The ADP:O ratios and state A rates of respiration are highly dependent upon the concentration of mitochondria used to assay these parameters. (b) The isolated mitochondria are con- taminated with divalent cation-stimulated ATPases identified as myosin and the F1FO-ATPase of broken mitochondria. These ATPases are activated by the Ca(II) and Mg(II) released from mitochondria. They are completely inhibitable by EGTA. (0) Uncoupled state A respiration can be abolished by treating the mitochondria with Nagarse and by pre-incubating mitochon- dria with substrate. The rate of uncoupled respiration is highly depend- ent upon the substrates used to drive respiration. For chick heart mitochondria oxidation of the substrate couple pyruvate/malate results in optimal respiratory parameters. (a) RCR values approaching infinity can be obtained under certain conditions. (e) The ADP:O ratios obtained during the oxidation of pyruvate/malate, a-ketoglutarate, and B-hydroxy- butyrate are fractional and significantly greater than 3.0. These ATP/O stoichiometries suggest that the H+/O stoichiometry of chick heart mitochondrial oxidative phosphorylation is at least 13 and possibly as 1N6 high as 1D. The ADP:O ratios obtained with glutamate and succinate conform closely to the classically accepted values of 3.0 and 2.0, respectively. 147 REFERENCES Amzel, L.M., and Pedersen, P.L. (1983) Ann. Rev. Biochem. 52: 801-82N. Barbour, R.L., Ribaudo, J., and Chan, S.P. (198“) J. Biol. Chem. 259: 8286-8251. Beavis, A.D., and Lehninger, A.L. (1986) Eur. J. Biochem. 158: 315-322. Beavis, A.D. (1987) J. Biol. Chem. 262: 6165-6173. Bencini, D.A., Wild, J.R., and O'Donovan, G.A. (1983) Anal. Biochem. 132: 259-258. Bishop, P.D., and Atkinson, D.E. (1984) Arch. Biochem. Biophys. 230: 335'3u3. Bogucka, K., and Wojtczak, L. (1971) Biochem. Biophys. Res. Comm. “4: 1330’1337.‘ Boyer, P.D. (1979) In Membrane Bioenergetics (Lee, C.P., Schatz, G., and Ernster, L., eds) pp. M61—N79, Addison-Wesley, Reading, Massachusetts. Brooks, S.P.J., and Suelter, C.H. (1986) Int. J. Biomedical Computing 19: 89-99. Chance, 8., and Williams, G.R. (1956) Adv. Enzymol. 17: 65-13u. Chance, B. (1972) FEBS Lett. 23: 3-20. Chance, B. (1977) Ann. Rev. Biochem. ”6: 967-980. Cintron, N.M., and Pedersen, P.L. (1979) Methods Enzymol. 55: nos-u1u. Day, D.A., and Hanson, J.B. (1977) Plant Sci. Lett. 11: 99-104. Denton, R.M., and McCormack, J.G. (1985) Am. J. Physiol. 2H9: EM53-ESSH. Douce, R. (1985) In Mitochondria in Higher Plants: Structure, Function, and Biogenesis. Academic Press, New York. Doussiere, J., Ligeti, E., Brandolin, G., and Vignais, P.V. (198M) Biochim. Biophys. Acta 766: “92-500. Dow, D.S. (1967) Biochemistry 6: 2915-2922. Erecinska, M., and Wilson, D.F. (1982) J. Membr. Biol. 70: 1-1”. Ernster, L. (1977) Ann. Rev. Biochem. “6: 981-99”. 1N8 Estabrook, R.W. (1967) Methods Enzymol. 10: u1-u7. Ferguson, S.J. (1985) Biochim. Biophys. Acta 811: u7-95. Flatt, J.P., Pahud, P., Ravussin, E., and Jequier, E. (198M) Trends Biochem. Sci. 5: A66-u68. Forman, N.0., and Wilson, D.F. (1983) J. Biol. Chem. 258: 86u9-8655. Goren, A.K., Wanders, R.J., Westerhoff, H.V., van der Meer, R., and Tager, J.M. (1982) J. Biol. Chem. 257: 2753-2757. Hagihara, B., and Lardy, H.A. (1960) J. Biol. Chem. 235: 889-89”. Hansford, R.0., and Castro, F. (1981) Biochem. J. 198: 525-533. Hassinen, I.E., and Hiltunen, K. (1975) Biochim. Biophys. Acta N08: 319-333. Hinkle, P.C., and Yu, M.L. (1979) J. Biol. Chem. 25M: 2H50-2455. Illsley, N.P., Coade, S.B., and Harkness, R.A. (1985) Placenta 6: 187-198. Kaplan, R.S., Pratt, R.D., and Pedersen, P.L. (1986) J. Biol. Chem. 261: 12,767-12,773. ' Klingenberg, M. (1980) J. Membr. Biol. 56: 97-105. Laemmli, U.K. (1970) Nature (London) 227: 680-685. Lemasters, J.J., and Sowers, A.E. (1979) J. Biol. Chem. 25“: 12u8-1251. Lemasters, J.J., Grunwald, R., and Emaus, R.K. (198”) J. Biol. Chem. 259: 3058-3063. Lemasters, J.J. (198") J. Biol. Chem. 259: 13,123-13,130. Mannervik, B. (1982) Methods Enzymol. 87: 370-390. Markwell, M.A.K., Haas, S.M., Talbert, N.E., and Bieber, L.L. (1981) Methods Enzymol. 72: 296-303. ‘ Masini, A., Ceccarelli-Stanzani, D., and Muscatello, U. (1983a) J. Bioenergetics Biomembranes 15: 217-23U. ' Masini, A., Ceccarelli-Stanzani, D., and Muscatello, U. (1983b) Biochim. Biophys. Acta 724: 251-257. Masini, A., Ceccarelli-Stanzani, D., and Muscatello, U. (198M) Biochim. BiOphys. Acta 767: 130-137. Miller, R.J., Dumford, S.W., Koeppe, D.E., and Hanson, J.B. (1970) 149 Plant Physiol. 45: 649-653. Mitchell, P. (1979) Science (Washington, D.C.) 206: 1148-1159. Nicholls, D.C. (1974) Eur. J. Biochem. 50: 305-315. Nicholls, D.C., and Crompton, M. (1980) FEBS Lett. 111: 261-268. Peterson, G.L., and Hokin, L.E. (1981) J. Biol. Chem. 256: 3751-3761. Pietrobon, D., Zoratti, M., and Azzone, 0.F. (1983) Biochim. Biophys. Acta 723: 317-321. Pietrobon, D., and Caplan, S.R. (1985) Biochemistry 24: 5764-5770. Pietrobon, D., Zoratti, M., Azzone, 0.F., and Caplan, S.R. (1986) Biochemistry 25: 767-775. Pollard, T.D., and Korn, E.D. (1973) J. Biol. Chem. 248: 4682-4690. Pollard, T.D. (1982) Methods Cell Biol. 24: 333-371. Pullman, M.E., and Monroy, 0.0. (1963) J. Biol. Chem. 238: 3762-3769. Robertson, J.G., Thomas, J.E., Gowing, L.R., Boland, M.J. (1984) Arch. Biochem. Biophys. 232: 337-347. Tarjan, E.M., and Van Korff, R.W. (1967) J. Biol. Chem. 242: 318-324. Terada, R., Ikuno, M., Shinohara, Y., and Yoshikawa, K. (1984) Biochim. Biophys. Acta 767: 648-650. Toth, P.P., Ferguson-Miller, S.M., and Suelter, C.H. (1985) Proc. 13th Intn'l Cong. Biochem., vol 5, p.767. Toth, P.P., Ferguson-Miller, S.M., and Suelter, C.H. (1986) Fed. Proc. 45(6): 1922. Toth, P.P., Sumerix, K.J., Ferguson-Miller, S.M., and Suelter, C.H. (1988) FASEB J. 2(5): A1122. Toth, P.P., Ferguson-Miller, S., and Suelter, C.H. (1986) Methods Enzymol. 125: 16-25. Towbin, R., Staehelin, T., and Gordon, J. (1979) Proc. Nat. Acad. Sci. U.S.A. 76: 4350-4354. Tsien, R.Y. (1980) Biochemistry 19: 2396-2404. Webb, H.R., and Trentham, D.R. (1980) J. Biol. Chem. 255: 8629-8632. Westerhoff, H.V., Melandri, B.A., Venturoli, G., Azzone, 0.F., and Kell, D.B. (1984) Biochim. Biophys. Acta 768: 257-292. 150 Wikman-Coffelt, J., Zelis, R., Fenner, C., and Mason, D.T. (1973) Prep. Biochem. 3: 439-449. Wilkinson, G.N. (1961) Biochem. J. 80: 324-332. Wilson, D.F., Stubbs, M., Veech, R.L., Erecinska, M., and Krebs, H. (1974) Biochem. J. 140: 57-64. Zoratti, M., Favaron, M., Pietrobon, D., and Azzone, 0.F. (1986) Biochemistry 25: 760-767. Chapter 4 THE ADVANTAGES AND LIMITATIONS OF USING 90° LIGHT SCATTER TO MONITOR CHANGES IN THE VOLUME AND PYRIDINE NUCLEOTIDE CONTENT OF THE CHICK HEART MITOCHONDRIAL MATRIX 151 152 INTRODUCTION Like many other membrane-enclosed compartments, mitochondria are highly responsive to the osmolality of the environment in which they are maintained. These organelles are able to rapidly adjust their volume as osmotic conditions change because they: (a) contain a large number of transport proteins capable of mediating the transmembrane flux of many different ions (LaNoue and Schoolwerth, 1985); (b) are comprised of a highly convoluted inner membrane which may contract or unfold, depending upon the activity of water in the matrix space; and (c) can accomodate significant matrix swelling without bursting their outer membrane due to the presence of an inter-membrane space. A large number of studies (reviews: Lehninger, 1962; Brierly, 1973; Garlid and Beavis, 1987) have demonstrated that mitochondria obey the Boyle-van't Hoff law (Tedeschi and Harris, 1958; Beavis §£_al., 1985) and thus are well-behaved osmometers. The intensity of light that a suspension of mitochondria scatters is related to matrix volume (Tedeschi and Harris, 1955; Koch, 1961). For example, as the matrix shrinks, the intensity of light that is scattered away from the linear axis of observation (i.e., 0°) increases. This change is a consequence of an increase in the refractive index of the matrix as it becomes more compact. Because the transport of water across the mitochondrial inner membrane is essentially instantaneous (Beavis gt al., 1985), light scattering measurements have been used for some time to conveniently and continuously monitor the apparent transport rates of 153 a variety of electrolytes and non-electrolytes (Raaflaub, 1953; Harris and Tedeschi, 1955; Packer, 1960, 1969; Jung gt_al., 1977). Garlid and Beavis (1985) have recently refined light scatter methodology as applied to mitochondria by making these measurements more quantitative and by correcting for structure-dependent changes in apparent transport as measured by swelling rates. However, because these authors monitored changes at 0° (optical density), they concluded that the method was insensitive to the measurement of solute transport in isotonic media. The classical descriptions of light scatter by Rayleigh (1867), Mie (1909), and Debye (1947) have facilitated analysis of a number of physical systems. These and other more recent mathematical theories of light scatter phenomena (Bier, 1957; Koch, 1961; Melikhov §t_al., 1981; Twersky, 1983; Perrin and Chiapetta, 1985; Perrin and Lamy, 1986; Singham and Bohren, 1987), are highly sophisticated, but difficult to apply in meaningful ways to large biological systems such as mitochondria. There are three major reasons for this failure: 1.) Mitochondria are larger than the wavelengths of light with which they are typically irradiated during scatter experiments. ii.) Mitochondria are not opaque, ideal spheres. iii.) Concentrations of mitochondria used in experiments are far greater than those mathematically modeled in theoretically ideal situations. Two techniques are most often used to measure changes in mitochon- drial light scattering intensities. The first uses an optical system similar to that used to measure absorbance and views the intensity of light transmitted through the sample at a 0° angle (i.e., directly in- line with the collimated beam). This approach is called turbidimetry and 154 is a well-documented and popular method. The second technique allows one to view the intensity of light scattered at a 90° angle relative to the incident beam. This method is known as nephelometry. In this investiga- tion, the intensity of light scattered by mitochondria is measured at both 0° and 90°. Since the optical properties of mitochondrial suspen- sions makes them unamenable to analysis using classical theory, a more direct empirical theory for light scattering measurements is outlined. This empirical approach makes the following assumptions: 1.) Light scattering and absorption can both be defined as the loss of photons from a collimated beam as it traverses the sample. ii.) Measuring the scatter of light at an observation angle of 0° is similar to the process of measuring absorbance, except that in the former case there is a finite probability of recovering photons that at some point were deflected from the collimated beam. Clearly, in the case of absorbance, photons lost from the incident beam cannot be recovered. iii.) Measurements of light scattered at 0° or 90° have been typically used for different applications; however, they are dependent upon the same phenomena. Consequently, light scatter measurements at the two angles are coupled. In addition, it is shown that light scattering measurements at 90° are much more sensitive to changes in mitochondrial volume than those taken at 0°. The ability to obtain meaningful ratios of I and I0 (i.e., absorbance) using the 0° detector makes turbidimetric measurements reproducible and accurate. However, these measurements have a limited range (103 or A = 3.0) and their sensitivity is limited by the reference beam intensity. Nephelometric measurements, on the other hand, are apen- ended and have a wide range (106, not normalized). At 90° one starts from virtual darkness in the absence of scattering particles, thereby enabling a measurement of much greater sensitivity. This results in 155 kinetic measurements of increased signal-to-noise ratios. 156 EXPERIMENTAL PROCEDURES I. INSTRUMENTATION The instrument used for the spectroscopic investigations reported herein is a computerized spectrofluorometer (utilizing the optical com- ponents of a Perkin-Elmer Model 512) capable of performing simultaneous fluoresence and absorption measurements and of automatically correcting for the artifacts of fluorescence measurement. The system configuration is shown in Fig 1. Two photovoltaic cells (PVC, Hamamatsu Type 81337- 101080) were added to the cell compartment in order to measure the intensity of the incident beam of light before entering (R, reference) and after passing through (S, sample) the cuvette. A quartz plate is used to reflect approximately 4% of the light from a xenon lamp source toward the R detector. A concave mirror is employed to focus transmitted light onto the S detector. A photomultiplier tube (Hamamatsu R-446) is located at 90° to the collimated light source. The two PVC's and the photomultiplier tube provide the necessary means of detection in order to measure light intensities at 0° and 90°. In addition, the absorbance [log(R/S)] can be monitored simultaneously. Neutral density filters (Turner, 1% and 10% T) were used to adjust the intensities of light when required. A PDP/8e computer is dedicated to system control and data collection, processing, and outputting. The three detector outputs are amplified using a FET operational amplifier (Texas Instruments, TL081C), filtered, directed through a four channel CMOS mutliplexer (Datel Systems, MXD- Figure 1. (A). (B). (C). 157 Design Features of the Integrated Spectrofluorometer/ Spectrometer System configuration of the integrated spectrofluorometer/ spectrometer used for light scattering measurements. Optical diagram of instrumental components. Key: 1, xenon lamp/excitation source; 2, excitation monochromator; 3. quartz plate; 4, sample cell; 5, front-surface concave focussing mirror; 6, emission monochromator; 7, reference beam detector; 8, sample beam detector (or "S" detector); 9, fluorescence beam detector (photomultiplier tube). For light scattering measurements, component 8 is used to detect 0° scattering and component 9 is used to detect 90° scattering with components 2 and 6 set at the same wavelength. Off-center sample cell rotator device employed for absorp- tion corrected fluorescence and scattering measurements. 158 A A“ . .... A“ "' ‘ s“ I,’ \ ‘ ,‘o PERI!!! ‘, — 1 ' ‘. :1: ; convent: coupons \‘ o r; - \ s..- ‘ ' ,2 cm. nor-won I l l I I | EXPLODED VIEWS OF THE ELLIPTICAL CELL ROTATION SYSTEM 159 409), and sent to an A/D converter (Analogic, MP2112) for processing under computer control. Each reading represents the average of 40 points collected over a 250 msec period in order to minimize noise. A calibration table is utilized to correct for any imbalance in the Optical system and to convert measured light intensities into quanta. Real time output is achieved using a Tektronix 2000A graphics terminal. Data storage is accomplished by a floppy disk drive system (Data Systems, SA800/801), and a hard copy is obtained by a Houston Instruments plotter. A rotating cell pedestal is mounted in the spectrofluorometer's sample chamber. The design details of the system are described elsewhere (Adamsons, 1985). The off-axis rotation of the cell-rotator produces variations in the pathlength through which the incident beam of radiation must pass. II. ISOLATION OF MITOCHONDRIA Heart mitochondria were isolated from 14-21 day-old chicks using Nagarse, as previously described (Chance and Hagihara, 1961). Chick heart ventricular myocardium was minced into an ice-cold isotonic iso- lation medium that contained 225 mM mannitol, 75 mM sucrose, and 20 mM Tris, pH 7.4 (MST). The myocardium was then homogenized with five up-and-down strokes of a Potter-Elvejhem pestle (Operated at 500 r.p.m.) in the presence of approximately 0.5 I.U. Nagarse/g muscle. The homo- genized tissue was allowed to incubate on ice for 3 min, and was then subjected to two additional up-and-down strokes of the pestle. The crude homogenate was centrifuged for 5 min at 800g. The resulting supernatant liquid containing the mitochondria was filtered through two layers of cheesecloth and centrifuged at 8,000g for 10 min. The mitochondria were then rinsed twice by resuspending these organelles in 7-8 ml of MST and 160 centrifuging at 8,000g for 10 min. The final washed mitochondrial pellet was resuspended in MST at 15-20 mg/ml. All experiments were completed within 4-5 hrs after isolation of mitochondria. 161 THEORETICAL CONSIDERATIONS I. THE EFFECT OF PHOTON RECOVERY ON LIGHT SCATTERING MEASUREMENTS One may consider the efficiency of scatter to be related to the efficiency of absorption (absorptivity) and the probability of a photon encountering a surface from which it may be scattered. If one assumes that in very dilute solutions of the scattering species no multiple scatter encounters occur, then even in the absence of absorption it would be reasonable to expect that when measured at 0° the scattering species obeys the Beer-Lambert law, expressed as: log(IO/I) = abc (1) where I0 is the intensity of the incident beam, I is the intensity of the transmitted beam, 0 is the concentration of the scattering species, a is the absorptivity of the scattering species, and b is the pathlength of the sample through which the light must pass. However, as the concen- tration of the scattering species is increased, the likelihood for scat- tered photons to undergo sequential internal reflections and refractions also increases. This creates a probability of recovery whereby a fraction of the scattered radiation is redirected toward, and eventually reaches, the detector. Herein lies the major difference between absorption and scatter measurements. In the former case only one type of encounter occurs, namely the absorption and loss of a photon from the collimated beam, while in the latter sequential scattering events may take place and photons lost from the beam may be recovered. Thus, the actual 162 intensity of light measured by the detector in a mitochondrial light scattering experiment is greater than what would be expected from the Beer-Lambert relationship because of the recovery of increasing amounts of scattered light as the concentration of mitochondria is increased. For the purpose of the following considerations, it is assumed that the sample pathlength is equal to 1 cm and, therefore, b will be omitted from the equations which follow. The Beer-Lambert law (eqn 1) holds for dilute suspensions of scattering particles. However, as the concentra- tion of the scattering species is increased, the intensity of the trans- mitted beam increasingly deviates from the theoretical intensity that would satisfy the Beer-Lambert law. This deviation may be expressed as follows: IS = I x (IO/IS)aCR (2) where IS is the intensity of the observed beam in the presence of scattering, I is the theoretical intensity of light expected if beam attenuation is due to a single encounter only, I0 is the intensity of the incident beam, a is the probability of scatter, and R represents the probability of recovering photons. The value of R is a concentration- dependent factor but extrapolates to a constant at extremely high concentrations of mitochondria. The ratio I /IS may be interpreted as 0 the reciprocal of the light transmitted by the sample. For samples in which photons may have only one possible type of interaction with the suspended particle (i.e., absorption or a single scatter event), R = 0 and eqn 2 reduces to the Beer-Lambert law. Solving for I in equation 2 we obtain 163 I = IS x (IO/IS)'aCR. (3) This expression may be substituted into equation 1: log = ac. (4) -acR 18(10/13) Equation 4 may be rearranged to the following: log(I0/IS) + acR x log(IO/Is) = ac. (5) By substituting the measured absorbance, A, for log(IO/IS), the above may be rewritten as A + acRA = ac. (6) After factoring out A and taking the reciprocal of both sides, absorb- ance can be related to the recovery coefficient as specified by equation 7: 1/A = I/ac + R (7) Thus, in situations where absorption by the scattering species is negligible, a and R may be inferred from the slope and y-intercept, respectively, of the double reciprocal plot relating the concentration of mitochondria (mg/ml) to changes in optical density. II. CORRELATION OF THE INTENSITY OF LIGHT SCATTERED AT 0° AND 90° Since the detectors situated at 0° and 90° are measuring changes in the same scattering processes, one expects that the rate of photon recep- 164 tion for each will vary as a function of the concentration of scattering particles. Assuming no absorption occurs, the limits within which we can consider the scattering phenomena are from: (a) the completely focused beam, maintained only in the absence of scattering particles; and (b) the completely defocused beam, in which sequential internal reflections and refractions result in the uniform scattering of photons throughout the spherical dimensions of the observation compartment. It follows that the intensity of light reaching the 0° detector will be at a maximum when there are no scattering particles in the cuvette. As the concentra- tion of scattering particles increases, increasing amounts of light will be diverted from the focused beam. All of the scattered light will not reach the detector; hence, the detector sees less and less light (Fig 2). As the population of scattering particles increases, the probability for multiple sequential scattering of photons increases since the light, after the initial scattering, will encounter a greater and greater number of surfaces prior to exiting the cuvette. As this occurs a finite probability exists for a photon originally scattered from the focused beam to be redirected toward the 0° detector. Consequently, all of the photons removed from the focused incident beam are not lost, since a fraction of them will be recovered when sequential scattering processes occur. Eventually, a concentration of scattering particles is attained in which all of the light in the incident beam is subject to multiple scattering events. At this point the beam is completely defocused. In the ideal case, the light in all directions of the observation system will be of equal intensity. This means that the fraction of the scat- tered light striking the 0° detector can be calculated from the following relationship: 165 Figure 2. Intensity of Scattered Light at 0° and 90° Measured as a Function of Mitochondria Concentration Measurements at 0° (.) represent changes in the intensity of transmitted light or, simply, changes in optical density. Measurements at 90° ((:)) represent changes in the intensity of scattered light. For this experiment the sensitivity of the S detector and the photomultiplier tube were set equal. Samples were irradiated with light having a wavelength of 500 nm. Intensity of Scattered Light 166 4ooo .- a .8 (Arbitrary Units) 8 <3 0 0 i Complete Defocusment 1 1 L 1 1 1 1 1 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 mg Mitochondrial Protein -ml'I 1 9.0 Intensity of Scattered Light 166 4ooo — 8 8 S 0 0 (Arbitrary Units) 8 I Complete Defocusment l J l J L I I l l |.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 mg Mitochondrial Protein -ml"' I = I (8) 4/311r'2 where I is the intensity of the detected radiation (photons per unit time), I is the intensity of the incident beam, A is the area of the 0 d detector, and r is the distance from the detector to the center of the observation sphere. The scenario for the 90° detector will be opposite to that of the 0° detector. In the absence of scattering particles there is virtually no light striking the detector. As the concentration of scattering particles increases, increasing amounts of light will be directed to the detector. When measured as a function of mitochondria concentration, the intensity of light scattered at 90° is non-linear (Fig 2). Above 0.5 mg mitochondria/ml, there is an abrupt increase in the slope of the light scatter function. In this concentration range, multiple sequential scattering becomes significant, increasing the overall scattering effici- ency. As for scatter measured at 0°, the probability of photon recovery increases with increasing concentrations of the scattering particles until a concentration of scattering particles is reached at which the incident beam is completely defocused. This concentration corresponds to the point where the intensity of light striking the detectors at 0° and 90° are equal. In the ideal case, as the concentration of scattering particles is further increased, the light intensity would remain the same for both detectors because the beam is completely defocused and additional scattering has no effect on the fraction of light striking either detector. In the real case this does not occur for two reasons: 168 i.) The scattering particles absorb a fraction of the incident beam. (Very small absorbances can become significant with multiple sequential scattering as each scattering event represents an additional opportunity for photons to be absorbed.) ii.) Back scatter of the incident radiation. As shown by the data presented in Fig 2, the light intensity corresponding to complete defocusing is not maintained with increasing concentrations of the scattering particle. As higher concentrations of mitochondria are added to the system, the probability of encounter between mitochondria and photons in the incident beam greatly increases, thereby reducing the mean distance into the solution that the individual photons travel prior to their first encounter with a scattering surface. This gives rise to a back scatter phenomenon which begins to redirect the light into the path of least resistance, i.e. in the direction of the light source. In the extreme, if sufficient mitochondria concentra- tions were attained, the wave-fronts of light would reflect off of mitochondria as from a plane surface, in accordance with Snell's law. An analysis of back scatter phenomena predicts that it would manifest itself first in the data received from the 0° detector and, subsequently, in the data received from the 90° detector. This is intuitively obvious from the geometrical relationship between the light source, the detector at 90°, and the detector at 0° (i.e., the incident light can travel a smaller mean distance in order to reach the 90° detector as compared to the one at 0°). Thus, as back scatter becomes significant, it will attenuate the intensity of light reaching the 0° detector more than that reaching the 90° detector. Careful consideration of Fig 2 indicates that this appears to occur in the actual experiment. What is useful information 169 from these data, however, is identification of the point of complete beam defocusment. This point is chosen as the maximum in the response curve of the 90° detector. This point can be a well-defined empirical quantity delineating the upper limit of a calibration scheme for light scatter measurements of dense suspensions. 170 RESULTS AND DISCUSSION I. RELATIONSHIP BETWEEN LIGHT SCATTER INTENSITY AND MEDIUM OSMOLALITY The intensity of light that chick heart mitochondria scatter depends upon the osmolality of the medium in which they are suspended (Fig 3). As the medium osmolality is decreased from 600 to 125 mosmol, mitochondrial matrix volume increases as indicated by a decrease in the intensity of scattered light. However, as the medium osmolality is decreased below approximately 125 mosmol, the extent of mitochondrial swelling increases sharply. Below 70 mosmol the amount of change in matrix volume induced by increasingly lower osmolalities decreases. These data are in close agreement with the findings of Beavis g£_al. (1985) which show that between 68 and 115 mosmol mitochondria undergo an irreversible structural transition. This transition involves rupture of the outer membrane and unfolding of the inner membrane. Thus, the volume changes above 115 mosmol correspond to matrix swelling proceeding within the constraints of the outer membrane; those occurring below =70 mosmol correSpond to matrix swelling unrestricted by the outer membrane. II. RELATIONSHIP BETWEEN LIGHT SCATTER INTENSITY AND WAVELENGTH OF LIGHT The efficiency with which chick heart mitochondria scatter light varies with the wavelength of light used to irradiate these organelles. In order to measure this scattering efficiency, mitochondria were sus pended in a solution that contained 0.225 M mannitol and 0.075 M sucrose. The excitation and emission monochromators of the computerized spectro- 171 Figure 3. Responsiveness of the Chick Heart Mitochondrial Matrix to Changes in External Osmolality Mitochondria (0.50 mg/ml) were suspended in 2.0 ml of M/S media of the indicated osmolality. The molar ratio of mannitol:sucrose was held constant at 3:1 in all assays. Once the photomultiplier tube output signal stabilized (indicating that the mitochondrial matrix volume had come to osmotic equilibrium), the intensity of light (500 nm) scattered at 90° was measured. 172 3000 .. m m 35:: 32:23 .8 8 22.. 3.968 a £825 ICC 200 300 400 500 600 0 Osmolality , mosmol 173 fluorometer were coupled, and the intensity of light scattered at 90° was measured. Between 500 and 700 nm, the intensity of 90° light scat- tered by these mitochondria varies exponentially with wavelength (Fig 4). Wavelengths smaller than 500 nm were not used because mitochondrial cytochromes and other components absorb significant amounts of such light. III. MEASUREMENT OF CHANGES IN MATRIX NADH LEVELS As demonstrated by Chance and Williams (1955), dual wavelength spectroscopy can be used to continuously monitor relative changes in mitochondrial matrix NADH levels. In this technique mitochondria are irradiated with two wavelengths of light: 340 nm (the absorption maximum of NADH) and 374 nm (the reference wavelength). Changes in the intensity of transmitted light result from increases or decreases in absorbance and the scattering of light. Clearly, the technique derives its sensi- tivity from being able to accurately measure changes in absorbance as matrix NADH concentrations change. If the intensity of light scatter remains uniform as the metabolic state of mitochondria is varied, then dual wavelength spectrOSCOpy will give reliable measurements of NADH levels. However, because mitochondria swell or shrink in response to various treatments, the efficiency with which these organelles scatter light will change at the same time NADH concentrations are changing. This combination of effects makes the interpretation of data obtained with dual wavelength measurements problematic. In most types of mitochondria yet tested, pyridine nucleotides undergo net oxidation during state 3 respiration. A notable exception to this is the blowfly flight muscle mitochondrion in which pyridine 174 Figure 4. Intensity of Light Scattered by Chick Heart Mitochondria Measured as a Function of Wavelength Mitochondria were suspended in 2.0 ml of an osmotic support medium comprised of 0.225 M mannitol and 0.075 sucrose at 0.35 mg/ml. The wavelength of light with which these organ- elles were irradiated was varied from 500-700 nm, and the intensity of light scattered at 90° was measured. The emis- sion and excitation monochromators were coupled (i.e., they were continuously adjusted in synchrony). 175 U Wavelength, nm 8.863. 8.8.8 .0 £5 ragga 176 nucleotides undergo net reduction after the addition of ADP (Hansford and Chappell, 1968; Hansford, 1972). As measured by dualwavelength spectrOphotometry (340-374 nm), chick heart mitochondrial pyridine nucleotides undergo an apparent net reduction during state 3 respira- tion when oxidizing a variety of NAD-linked substrates (Fig 5A). After the addition of ADP, absorbance at 340 nm relative to that at 374 nm increases. Once the ADP is phosphorylated to ATP, the differential absorbance decreases. As a control NAD was reduced with excess boro- hydride. As expected this reaction results in an upward deflection of the trace (Fig SB). In contrast to chick heart mitochondria, rat liver mitochondria experience a significant decrease in the apparent redox state of matrix pyridine nucleotides during state 3 (Fig 6). Because this difference in NADH production and utilization could have important metabolic consequences for chick heart muscle, additional experiments were conducted to affirm whether the information obtained with dual wavelength spectrometry could be interpreted in a straight- forward manner. The computerized spectrofluorometer was used to differentiate between absorbance and any artifacts due to changes in the intensity of light scatter. In order to do this, three types of measurements were made. The first was conventional absorbance of the excitation radiation or "A1". The two other measurements exploit the capability of this instrument to rotate the sample cell between three different viewing positions (Fig 7A). Rotation of the sample cell allows one to vary the solution thickness through which the incident radiation must pass. By measuring the emission of a sample at positions 1 and 4 (Fig 78), it is possible to measure the amount of light that is lost, due to absorption Figure 5. (A). (B). (C). (D). 177 Apparent Changes in the Absorbance of Chick Heart Mitochon- drial Matrix Pyridine Nucleotides During State 3 to State 4 Transitions Mitochondria (0.1 mg/ml) were suspended in 2.0 ml of a buffer that contained 0.225 M mannitol, 0.075 M sucrose, 20 mM Tris- buffered Pi' and the substrate(s) indicated below. 418 nmols of ADP were added at the times shown. Absorbance changes at 340-374 nm were measured with a dual beam spectrOphotometer. Reduction is indicated by an upward deflection in the trace. 5 mM pyruvate and 2.5 mM malate 5 mM a-ketoglutarate 5 mM B-hydroxybutyrate 5 mM glutamate and 2.5 mM malate The lower panel shows a control reduction of 1.2 umoles of NAD with excess sodium borohydride in the absence of mitochondria. 178 IAA'ODOI g ' a a-Ketofltaate J! ‘Am ‘ADP ‘8‘)" \ Hrfim a B-Hydmybtnyrate 179 Figure 6. Apparent Changes in the Absorbance of Rat Liver Mitochon- drial Matrix Pyridine Nucleotides During State 3 to State 4 Transitions Experimental conditions were as described in the legend to Fig 5. Oxidation is indicated by a downward deflection in the trace. The substrate(s) used are as indicated below. (A). 5 mM glutamate and 2.5 mM malate (B). 5 mM succinate 180 AA=0.00l I A. Glutamate / Malate 1-—,——1 I Mm ADP ADP P ADP ADP B. Succinate IO” Q—o bstrate ADP ADP ADP /ADP / hMitochmdr I Substrate 181 Figure 7. Off-Center Sample Cell Rotation (A). Positions 1, 2, and 4 of the off-center sample cell rotation system. (B). Schematic depiction of sample penetration by the incident beam of light in positions 1 and 4. The beam denoted "EX" is the excitation or incident light; the beam denoted "EM" is the sample emission or scattered light. 182 [___..___..__ r-_-_.____ [: EM EX POSITION 4 POSITION I 183 or scatter, between the two viewing positions. The latter measurement, designated "A2", is thus based on fluorescence and is described by the following relationship: A2 = G3log(F1/F4) (9) where F1 and F4 are the fluorescence intensities at positions 1 and 4, respectively, and G3 is a geometric factor defined as b/(d)2 - ¢1), where b is the pathlength of the sample cuvette and the g values are the cen- ters of the fluorescence observation windows. The third type of measure- ment was simply relative fluorescence intensity at position 1 (F1). The instrumental outputs of these functions are shown in Fig 8A-C. In agreement with results obtained with differential absorption spectra- metry, there is an increase in A1 after the addition of ADP to chick heart mitochondria pre-incubated with P and pyruvate/malate (Fig 8A). 1 This indicates that the mitochondria absorb a greater amount of 340 nm light during state 3 when compared to state 4. The F1 (Fig 8B) and A2 (Fig 8C) measurements both show, however, that the emission of 450 nm light decreases as the mitochondria undergo a state 4 to state 3 tran- sition. Fluorescence measurements therefore suggest that matrix NADH concentrations decrease during state 3 respiration. The question arises: Why do the results obtained with absorbance and fluorescence measurements conflict with one another? The lack of correlation is due to significant changes in the intensity of light scatter during a state 4 to state 3 transition. Changes in the intensity of light scatter were monitored by setting the excitation and emission monochromators at 500 nm. The addi- tion of ADP to mitochondria incubated as in Fig 9 results in a large increase in light scattered by the suspension. This is consistent with 184 Figure 8. Fluorescence and Absorbance Measurements of Chick Heart Mitochondrial Matrix Pyridine Nucleotides Mitochondria (0.1 mg/ml) were suspended in 2.0 ml of M/S osmotic support medium. The following reaction components were added to this suspension: Pi’ 20 mM (Tris-buffered, pH 7.4); P/M, 5 mM pyruvate/2.5 mM malate; ADP, 930 nmols per addition. Aex = 340 nm; Aem = 450 nm. Slit widths were set at 20 nm. (A). A1, changes in conventional absorbance (optical density). (B). 1F, fluorescence at 450 nm. (C). A2, fluorescence-based absorbance. (ADP 185 M110 1 Pi (P/M ADP m L I P L J .1111 L L . a .fi c A 8:28.... 88.28 a 2:5 32:24 IO Time, min 186 Figure 9. Changes in the Intensity of Light Scattered at 90° Subsequent to the Addition of Pi, Substrate, and ADP Mitochondria were suspended in 2.0 ml of an osmotic support medium that contained 0.225 M mannitol and 0.075 M sucrose at 0.80 mg/ml The following additions were made: Pi’ 20 mM; P/M, 5 mM pyruvate, 2.5 mM malate; ADP, 400 nmoles per addition. The excitation and emission monochromators of the integrated spectrofluormeter/spectrometer were set at 500 nm. 187 ADP [ADP P/M (M m) ,w E: 002 cozguom 8328 Lo £5 38.53 10 Time, min 188 the findings of Hackenbrock (1966) which showed that ADP induces a contraction of the mitochondrial matrix. Therefore, both conventional absorbance and differential absorption spectrometry register artifactual increases subsequent to the addition of ADP because the mitochondria become more efficient scattering particles (i.e., the matrix becomes more dense). With these measurements, changes in scattering obviate the possibility of accurately monitoring true changes in the absorbance of matrix pyridine nucleotides. In conclusion, chick heart mitochondrial pyridine nucleotides undergo net oxidation during state 3. It is likely that the redox state of rat liver mitochondrial pyridine nucleotides may be inferred from differential absorption measurements because these mitochondria do not contract to the same degree as chick heart mitochon- dria subsequent to the addition of ADP. IV. SENSITIVITY OF 0° and 90° LIGHT SCATTER TO CHANGES IN MATRIX VOLUME The major application of light scatter methodology to studies of mitochondria is to continuously monitor the uptake of various solutes into the matrix. The following properties of mitochondria make this possible: i.) Mitochondria are near perfect osmometers. ii.) The matrix undergoes rapid swelling/shrinking in response to perturbations in extramitochondrial osmolarity. iii.) As matrix configuration changes, the intensity of scattered light changes. iv.) As osmolarity +, matrix volume +, and light scatter t. v.) As osmolarity +, matrix volume 1, and light scatter +. vi.) Net inward transport (uptake) of a solute correlates 189 with matrix swelling. The sensitivity of 0° and 90° light scatter to Pi-induced swelling of the chick heart mitochondrial matrix is compared in Fig 10. Clearly, for kinetic measurements, nephelometry is the method of choice due to its much higher sensitivity and larger signal-to-noise ratio. The calculation of kinetic constants from the type of data presented in Fig IOB is discussed in Chapter 5 of this dissertation. It is of importance to note here that little if any quantitative information about the transport of Pi can be obtained from turbidimetric measure- ments alone. Turbidimetric measurements do, however, yield information about the magnitude of R and a. Double reciprocal plots relating the optical density of increasing concentrations of mitochondria before and after swelling is induced with the addition of 20 mM P1 are shown in Fig 11. The values for a are 4.07 and 3.24 before and after P -induced swelling, 1 respectively. These values are reasonable since one would expect swollen mitochondria to scatter light less efficiently than those which are at osmotic equilibrium with an isotonic support medium. Finally, although a slight difference in R values for the two conditions may be discerned from the plots, these are not significant. Small variations in R are observed in different preparations of mitochondria most likely due to errors in the determination of mitochondrial concentration. On a theo- retical basis R would be expected to be a constant since it depends on the geometry of the observation system and the number of scattering surfaces per unit volume, not their structural density. Figure 10. (A). (B). 190 Relative Sensitivity of 0° and 90° Light Scatter to Pi-Induced Changes in Chick Heart Mitochondrial Matrix Volume Mitochondria were suspended in 0.225 M mannitol, 0.075 M sucrose at 0.85 mg/ml. Inorganic phosphate (20 mM) and pyruvate (1 mM)/malate (0.5 mM) were added simultaneously at the indicated time to the mitochondrial suspension. Mitochondrial swelling is indicated by a downward deflection in the traces. Trace showing the change in absorbance (optical density) as a function of time. Trace showing the change in 90° light scatter intensity as a function of time. The initial drop in light intensity subsequent to the addition of Pi and substrate is due to a small dilution of the mitochondria and to the swelling which occurs during the time (2-4 sec) it takes to manually mix the sample. 191 P,+P/M b P,+P/M ./ 3;. _.. 12 T6 15 .4 0. 859034 0 558m 38st a £5 32:3 Time, min 192 Figure 11. Double Reciprocal Plot Relating Changes in Optical Density to Increases in the Concentration of Mitochondria Changes in optical density were monitored continuously from before the addition of 20 mM Tris-buffered Pi (O) to after mitochondrial volume had stabilized secondary to P1 uptake (CD). Mitochondria concentration was varied ten-fold, from 0.23 to 2.30 mg/ml. All assays were performed in M/S medium. Samples were irradiated with light having a wavelength of 500 nm. l / absorbance 193 C 1.5- O l.0- ' . o C . 0 .5- .. 0 1 1 1 1 1 0 IO 20 3.0 4.0 5.0 l/(mg Mitochondrial Protein-ml") 194 V. CONCLUSIONS The experiments and theoretical considerations presented in this study show that: (a) The deviation of mitochondrial suspensions from Beer's Law can be explained by assuming that much scattered light is recovered by the incident beam of light prior to striking the 0° detector; (b) Nephelometric measurements of mitochondrial P transport 1 are much more sensitive than those performed with turbidometric methods; (0) Turbidimetric measurements, although relatively insensitive for monitoring continuous changes in mitochondrial volume, are of value in that the magnitude of R (recovery coefficient) and a (scattering effici- ency) can be estimated from equilibrium measurements of mitochondrial suspensions. (d) Chick heart mitochondria can withstand large variations in external osmolarity (150-600 mosmolal) without undergoing appreciable apparent structural alteration or damage. (e) Dual wavelength spectropho- tometry yields artifactual information concerning chick heart mitochon- drial matrix pyridine nucleotide content during a state 4 to state 3 transition due to concomitant changes in the light scattering efficiency of these organelles. Fluorescence and fluorescence-based absorbance measurements provide much more accurate data on relative changes in the matrix pyridine nucleotide content. 195 REFERENCES Adamsons, K. (1985) Dissertation. Michigan State University. Beavis, A.D., Brannan, R.D., and Garlid, K.D. (1985) J. Biol. Chem. 260: 13424-13433. ‘ Bier, M. (1957) Methods Enzymol. 4: 147-166. Chance, B., and Williams, G.R. (1955) J. Biol. Chem. 217: 409-428. Debye, P. (1947) J. Phys. Colloid Chem. 51: 18-30. Garlid, K.D., and Beavis, A.D. (1985) J. Biol. Chem. 260: 13434-13441. Garlid, K.D., and Beavis, A.D. (1987) Biochim. Biophys. Acta 853: 187-204. Hackenbrock, C.R. (1966) J. Cell Biol. 30: 269-297. Halestrap, A.P., and Quinlan, P.T. (1983) Biochem. J. 214: 387-393. Hansford, R.G., and Chappell, J.B. (1968) Biochem. Biophys. Res. Comm. 30: 643-648. ' Hansford, R.G. (1972) Biochem. J. 127: 271-283. Jung, D.W., Chavez, E., and Brierly, G.P. (1977) Arch. Biochem. Biophys. 183: 452-459. Koch, A.L. (1961) Biochim. Biophys. Acta 51: 429-441. Melikhov, A.A., Storonkin, B.A., and Kuni, F.M. (1981) Opt. Spectrosc. (U.S.S.R.) 49: 321-324. ’ Packer, L. (1960) J. Biol. Chem. 236: 214-220. Perrin, J.M., and Chiapetta, P. (1985) Optica Acta 32: 907-921. Perrin, J.M., and Lamy, P.L. (1986) Optica Acta 33: 1001-1022. Quinlan, P.T., Thomas, A.P., Armston, A.E., and Halestrap, A.P. (1983) Biochem. J. 214: 395-404. Raaflaub, T. (1953) Helv. Physiol. Pharmacol. Acta 11: 142-156. Rayleigh, J.W. (1871) Phil. Mag. 41: 107, 274. Singham, S.B., and Bohren, C.F. (1987) Optics Lett. 12: 10-12. I96 Tedeschi, H., and Harris, D.L. (1955) Arch. Biochem. Biophys. 58: 52-67. Tedeschi, H., and Harris, D.L. (1958) Biochim. Biophys. Acta 28: 392-402. Twersky, V. (1983) J. Opt. Soc. Am. 73: 313-320. Chapter 5 INTERACTION OF INORGANIC PHOSPHATE WITH CHICK HEART MITOCHONDRIA I. CHANGES IN VOLUME AND ION COMPOSITION OF THE MATRIX 197 198 INTRODUCTION Mitochondria have evolved an elaborate system of carriers responsible for mediating the flux of inorganic phosphate across the inner membrane. Net inward transport and Pi-Pi transmembrane exchange is catalyzed by the Pi transport protein (Coty and Pederson, 1974; Ligeti §£_§l., 1985). Transport of Pi is a proton-coupled process (McGivan and Klingenberg, 1971), and at equilibrium Pi is distributed between compartments accord- ing to the following relation (Palmieri et al., 1970): log [Pi’Jm/[pi'Jc = ApH (1) where [Pi-Jm and [Pi-JC are the concentrations of monovalent P1 in the matrix and cytosol, respectively. The P1 transport protein has been isolated from bovine heart (Wohlrab, 1980) and rat liver mitochondria (Kaplan gp_gl., 1986), and molecular details concerning the actual transport mechanism of this carrier are beginning to emerge (Wohlrab, 1986). In addition to meeting the substrate requirements for oxidative phosphorylation, the inward transport of P compensates the charge 1 imbalance resulting from the electrogenic exchange of ADP3- for ATP“- on the adenine nucleotide translocase (McGivan et al., 1971). P1 is transported out of the matrix by the dicarboxylate carrier (Forman and Wilson, 1982) and the Pi transport protein (Coty and Pederson, 1975). In addition, other evidence suggests that P efflux may be catalyzed by i the adenine nucleotide translocase (Reynafarje and Lehninger, 1978) and an as yet undefined mechanism mediating a carboxyatractyloside-insensi- tive flux of Pi and adenine nucleotides (Austin and Aprille, I984). 199 Under conditions in which ATP is hydrolyzed in the matrix, the efflux of Pi has been attributed solely to the Pi transport protein (Tyler, 1980) or to the Pi transport protein and another carrier that is not the Pi/ dicarboxylate exchanger (Fonyo and Vignais, 1979; Kaplan and Pedersen, 1983). It has also been proposed that mitochondria contain an anion porter that conducts Pi out of the matrix when these organelles are osmotically stressed (Garlid and Beavis, 1987). It is apparent from these studies that mitochondria are highly responsive, both chemically and osmotically, to the intra- and extramatrix concentrations of Pi. Although criticized for its limitations (LaNoue and Schoolwerth, 1979). light scatter measurements have been used extensively to measure transmembrane fluxes in mitochondria (for reviews: Lehninger, 1962; Garlid and Beavis, 1987). This technique can be used to monitor mito- chondrial swelling changes induced by a variety of agents which include Pi (Packer, 1961), Ca(II) (Chappell and Crafts, 1965; Crafts and Chappell, 1965), K+ (Jung §p_al., 1977; Chavez g£_al., 1977; Martin g£_§l., 1984), and such hormones as thyroxine (Tedeschi, 1961), glucagon (Armston p£_al., 1982), and a-agonists (Halestrap §£_§l., 1986), among others. Light scattering measurements can be made at both 0° and 90°. The efficacy of 0° light scatter measurements was recently reevaluated and its quantitative applications expanded (Beavis gp_§l., 1985; Garlid and Beavis, 1985). It has been known for some time (Lehninger, 1962; see also Chapter 4), however, that light scattering measurements performed at 90° are more sensitive than those performed at 0°. Considering the fact that mitochondria swell during myocardial ischemia (Jennings and Ganote, 1976), studies of the ability of P to i induce mitochondrial swelling in various media (Izzard and Tedeschi, 200 1973; Jung et al., 1977 and references therein) are taking on increasing relevance. Previous studies suggest that P -induced swelling of heart 1 mitochondria is the result of a Ca(II)-dependent membrane transition (Hunter et al., 1976; Hunter and Haworth, 1979a). The rate of Pi-induced swelling has also been found to be dependent on the concentration of Pi (Izzard and Tedeschi, 1973; Vaghy et al., 1981). As part of our invest- igation of the nature by which Pi affects chick heart mitochondrial structure and function, we have examined the following: First, the effects of Pi on mitochondrial volume; second, the perturbation of matrix solute content subsequent to P1 uptake; third, whether Ca(II) is an obligatory component for swelling subsequent to the addition of Pi; and, fourth, the kinetic constants characterizing the Pi-induced swelling changes. The latter measurements also yielded an indirect estimate for the apparent Km of the P transport protein which is much 1 less than previous estimates of this parameter. 201 EXPERIMENTAL PROCEDURES I. MATERIALS Water was purified first by passage through a standard reverse osmosis system (Continental Millipore, Detroit, MI) which removes contaminants by carbon adsorption, microporous membrane filtration, and ion-exchange, and then by distillation from alkaline permanganate. The following substances were reagent grade or better, used without purification, and obtained from the sources noted: sucrose (RNase-free), carbonyl cyanide m-chlorophenyl hydrazone (CCCP), N-ethylmaleimide (MalNEt), ADP (grade X), and the sodium salts of pyruvic and malic acids (Sigma Chemical Co.); mannitol (Fisher Scientific); strontium chloride and toluene (J.T. Baker Chemical Co.); potassium permanganate and phosphoric acid (Mallinckrodt); Tris (Boehringer-Mannheim); carrier-free [32P1J-ortho- phosphoric acid and [1uC(U)]-sucrose (671 mCi/mmol) were from New England Nuclear; 2,5-diphenyloxazole and Triton-X100 (Research Products International). Nagarse (Sigma type XXVII, lot 97F-0218) was reconsti- tuted prior to each isolation in a solution containing 225 mM mannitol and 75 mM sucrose (M/S). Rotenone and CCCP were dissolved in absolute ethanol (AAPER Alcohol and Chemical Co.). Phosphate solutions were prepared by titrating phosphoric acid with Tris to pH 7.4. Single comb white leghorn chicks were purchased from Michigan State University's Department of Animal Science. Chicks were fed Chick G0125 feed (Kent Feeds, Inc., Muscatine, IA) and were not starved prior to sacrifice. II. ISOLATION OF MITOCHONDRIA 202 Heart mitochondria were isolated from 14-21 day-old chicks using Nagarse, as previously described (Chance and Hagihara, 1961). Chick heart ventricular myocardium was minced into an ice-cold isotonic isolation medium that contained 225 mM mannitol, 75 mM sucrose, and 20 mM Tris, pH 7.4 (MST). The myocardium was then homogenized with five up-and-down strokes of a Potter-Elvejhem pestle (operated at 500 r.p.m.) in the presence of approximately 0.5 I.U. Nagarse/g muscle. The homogen- ized tissue was allowed to incubate on ice for 3 min, and was then sub- jected to two additional up-and-down strokes of the pestle. The crude homogenate was centrifuged for 5 min at 800g. The resulting supernatant liquid containing the mitochondria was filtered through two layers of cheesecloth and centrifuged at 8,000g for 10 min. The mitochondria were then rinsed twice by resuspending these organelles in 7-8 ml of MST and centrifuging at 8,000g for 10 min. The final washed mitochondrial pellet was resuspended in MST at 15-20 mg/ml. All experiments were completed within 4-5 hrs after isolation of mitochondria. EGTA was not added to the isolation or assay media, unless specified otherwise, because this chelating agent would be expected to interfere with divalent metal cation fluxes across the mitochondrial inner membrane. Bovine serum albumin was also excluded from these solutions because this protein is known to scatter light and has some capacity to bind cations. Preliminary studies show that Nagarse destroys a soluble protein factor that stimu- lates uncoupled respiration (i.e., oligomycin-insensitive) in chick heart mitochondria (Toth g£_gl., 1985). As described in greater detail in Chapter 3 of this dissertation, under optimal conditions (0.050 nmol cyt aa /ml, 20 mM Pi’ and 5 mM pyruvate/2.5 mM malate), these mitochon- 3 dria are highly coupled, as evidenced by respiratory control ratios 203 routinely a 100 and ADP:0 ratios of 3.42 i 0.12 (n=30). III. LIGHT SCATTERING MEASUREMENTS A computerized spectrometer/fluorometer was used to characterize the kinetics of mitochondrial swelling secondary to the addition of Tris-buffered Pi (pH 7.4). A Perkin-Elmer Model 512 spectrofluorometer was modified to the single beam mode. A quartz fluorescence cuvette with a 1 cm pathlength was mounted on an off-center cell rotation device (Adamsons gp_gl., 1982). Two photovoltaic cells (PVC; Hamamatsu, 81377- 101080) were added to the cell compartment in order to measure the intensity of light incident to and transmitted from the sample. The xenon lamp source and excitation and emission grating monochromators of the instrument were retained. The photomultiplier tube (PMT; fluorescence detector) output along with the two PVC outputs were amplified and then multiplexed into an analog/digital converter for processing by a PDP 8/e Digital computer. Rates of phosphate-induced mitochondrial swelling were continuously monitored by measuring changes in 500 nm light scattering intensities at 0° and/or 90°. No attenuation of the light intensity was observed at this wavelength. The intrinsic gain of the PMT, the intensity of the 150 W xenon lamp, and the cell geometry resulted in a higher signal-to-noise ratio and a greater dynamic range for 90° light scattering (yp 0°). After initial mixing of reactants (elapsed time of 2-3 sec), readings were collected. Each stored reading represents an average of 40 data points collected over a 200 msec interval in order to minimize 60 Hz line noise. Slit widths for the excitation and emission windows were set at 20 nm. All light scattering measurements were performed at room 204 temperature (22-23°C). First-order rate constants for P -induced i swelling were calculated with the following equation: L.S.max - L.S.min k = In t (2) L.S.t - L.S.min where L.S.max is the intensity of scattered light immediately after the addition of Pi’ L.S. is the intensity of scattered light at each time t t (measured in 4 sec intervals), and L.S.min is the intensity of scattered light once the maximal volume for a swelling phase has been attained. Experimental conditions are described in the legends to Figures 1 and 3. IV. DETERMINATION OF INTRAMITOCHONDRIAL 32Pi. Mitochondria (1.9 mg) were suspended in a 1.0 ml final volume con- taining 225 mM mannitol, 75 mM sucrose, 20 mM Tris (MST; pH 7.4), 5 mM pyruvate/2.5 mM malate, and increasing concentrations of Tris-buffered 32P Pi (pH 7.4) that was enriched with carrier-free i' Mitochondria were allowed to incubate with P1 for 10 min. The mitochondria were then pelleted by centrifuging the suspension at 15,600g for 4 min in an Eppendorf 5414 Microcentrifuge. Duplicate 100 pl aliquots of the super- natant liquid were withdrawn and added to 10 ml of a scintillation cocktail composed of toluene, Triton-X100, and 2.5-diphenyloxazole. The remaining supernatant liquid was discarded, and the wall of the centrifuge tube was swabbed and carefully wiped. The mitochondrial pellet was resuspended in 0.3 ml of MST. The mitochondria were then denatured so as to release internalized 32Pi by adding 0,7 ml of 2.0 M perchloric acid. The denatured mitochondria were pelleted by centri- 205 fuging the sample at 15,600g for 4 min. Duplicate 100 pl aliquots of the supernatant liquid were withdrawn and added to 10 ml of the scintil- lation cocktail described above. Liquid scintillation counting was 32 performed on a Packard TriCarb 300 CD counter. P1 was counted for 10 min with a window setting of 5-1700 KeV. 32p. The amount of 1 in the mitochondria was corrected for the amount of 32 P1 in the extramitochondrial water co-sedimenting with the mito- chondrial pellet. The amount of non-matrix water in the pellet was calculated from data obtained in an experiment in which 0.1 pCi (149 pmol) of [1uC(U)]-sucrose was incubated with mitochondria for 1 min before sedimenting. Non-matrix water associated with the mitochondrial pellets was calculated with the following equation (Jensen et al., 1986): 14 ( C )(v ) P S exvP = (3) 14 CS where exVP is the volume of extra-matrix water, V8 is the volume of the suspension, and 11‘CP and 1uCS are the dpm of [1uC(U)]-sucrose in the pellet and supernatant fractions, respectively. The insert to Fig. 6 shows that, subsequent to a 10 min pre-incubation with P the volume of i’ non-matrix water co-sedimenting with the mitochondria is independent of Pi concentration (mean: 9.31 i 1.25 pl H20/mg mitochondria). For that particular experiment, mitochondria were sedimented by centrifuging at 9,000g for 4 min in a Fisher 59A Microcentrifuge. By centrifuging these organelles at 15,600g for 4 min, the amount of extramitochondrial water was decreased over 2-fold to 4.10 i 0.31 pl H O/mg mitochondria for a 2 typical experiment run in triplicate. [1uC(U)]-sucrose samples were 206 counted for 10 min with a window setting of 0-156 KeV in the scintilla- tion cocktail noted earlier. Cpm were converted to dpm for each isotope by counting a series of ten acetone-quenched standards. The resulting quench curves (quench index yp counting efficiency) were stored in the scintillation counter's memory and dpm were automatically calculated. V. QUANTITATION OF TRANSMEMBRANE IONIC FLUXES The concentrations of Ca(II), Mg(II), and K+ were quantitated from absorbance measurements made at 422.7, 285.2, and 766.5 nm, respectively, with an Instrumentation Laboratory 951 atomic absorption spectrophotometer using an air-acetylene mixture as fuel and a spectral bandpass of 1.0 nm. Ca(II), Mg(II), and K+ standards were obtained from Varian Techtron (Palo Alto, CA). Incubation conditions are described in the legends to Figures 7-10. Reactions were terminated by pelleting mitochondria with a Fisher Model 59A Microcentrifuge Operated at 9,000g for 3 min. When assaying for Ca(II) or Mg(II), resulting supernatant liquids were acidified with one drop of 10 N HCl and made 0.1 M in Sr(III) in order to control for interfering absorption by P Acidification of these samples prevented 1' formation of a strontium phosphate precipitate. The addition of a releasing agent to K+-containing samples was unnecessary because P1 does not interfere with the measurement of this cation. Total mitochondrial content of these ions was determined by assaying the supernatant liquids of sedimented perchloric acid precipitates. Data were corrected for nonmitochondrial Ca(II), Mg(II), and K+ arising from impurities in incubation medium constituents. VI. OTHER ASSAYS 207 P1 concentrations were determined by the method of Bencini et al. (1983) using potassium phosphate as standard. Mitochondrial protein was determined by the modified Lowry method of Markwell et al. (1981) using BSA as the protein standard. 208 RESULTS I. VOLUME CHANGES INDUCED BY INORGANIC PHOSPHATE ARE COMPLEX Representative traces revealing changes in mitochondrial matrix volume secondary to Pi uptake under a variety of conditions are shown in Fig 1. The simultaneous addition of P and the substrate couple 1 pyruvate/malate (P/M) induces a volume change (Fig 1A) that obeys first-order kinetics through 90% of the reaction (Fig 2). It is clear that Pi uptake causes swelling since the intensity of scattered light is decreasing. Analysis of the data linearized according to equation I (Fig. 2) shows that swelling proceeds with a half-time of 40.8 i 4.0 sec and a first-order rate constant (k) equal to 1.08 i 0.1 min-1 (n=20). The final volume attained is stable for approximately 10 min. After this time the mitochondria appear to lose water. In agreement with Hackenbrock (1966), adding ADP triggers a conformational change in the mitochondrial matrix, the so-called "orthodox to condensed" transition. The ability of mitochondrial suspensions to undergo this transition was used as an index of structural intactness. Swelling induced by P1 in the presence of P/M leaves mitochondria morphologically intact, as evidenced by the immediate contraction of the mitochondrial matrix subsequent to the addition of ADP (Fig 1A). Once the ADP is nearly all phosphorylated, mitochondria reassume the volume, and presumably the morphology, observed prior to the initiation of state 3 respiration. In the presence of a large exogenous energy source (millimolar concentrations of P/M), P -induced swelling of mitochondria must be a highly controlled 1 process because the structure and function of the organelles is not Figure 1. (A). (B). (C). 209 Perturbations in 90° Light Scattering Subseqpent to the Addition of Pi to Suspensions of Chick Heart Mitochondria Mitochondria were suspended at 0.66 mg/ml in an osmotic support buffer that was 0.225 M mannitol, 0.075 M sucrose, and 20 mM Tris, pH 7.4. The suspension was then made 20 mM in Tris (pH 7.4) buffered Pi under a variety of conditions. Intensities of scattered light were measured as described under "Experimental Procedures." The sudden drop in light scatter intensity secondary to the addition of P or P1 1 and substrate is due to a small dilution of the sample and the loss of data during manual mixing (2-4 sec) of the mitochondrial suspension. The break in the time axis is used to indicate a variable (30-60 sec) lag in the time after which monitoring of the sample was reinitiated. Monitoring was briefly interrupted because the instrument was set up to collect data continuously for 11 min periods. 1 mM pyruvate/0.2 mM malate added simultaneously with P1. Once the perturbation induced by P1 went to completion, 2.1 pmoles of ADP were added. Addition of 20 mM Pi without exogenous substrate. Pi added to mitochondria suspended in M/S medium that contained 5 mM EGTA. No exogenous substrate was added. Intensity of Light Scattered at 90° (Arbitrary Units) 210 P. ADP jlllJ/lllllllllll I 8 10’12 14 IS 18 20 22 Time,min Figure 2. 211 Determination of the First-Order Kinetic Constants of Mitochondrial Swelling Changes Induced by Pi Data obtained from nephelometric measurements were linearized using equation 1. The first-order rate constant (k) is equal to the slope of the plot. The half-time for the swelling process was calculated with the relation t0 5 = ln 2/k. For the experiment shown, mitochondria were suspended in 2.0 ml of M/S/T at 0.90 mg/ml. Swelling was induced with the simultaneous addition of 18 mM Tris-P and 1 mM pyruvate/0.5 mM malate. The 1 line through the data was calculated using least-squares linear regression analysis (cc: 0.9990). k = 1.26 min-i and t0.5 = 33.2 sec. 212 I I I l l I l l l I I 0 I0 20 3040 5060 70 80 90 IOOIIO Time, sec 213 adversely affected. When Pi-induced swelling proceeds in the absence of exogenous substrate, a different sequence of structural alterations is initiated (Fig 1B). The initial swelling phase is again first-order described by the constants noted above. However, following a 3 min lag (measured from the time Pi is added), the mitochondria shrink and then undergo a large second phase of swelling. Subsequent to the second swelling phase, no conformational transition is observed upon the addition of P/M and ADP, indicating that these mitochondria have undergone extensive structural damage (data not shown). It has been suggested by others (Hunter and Haworth, 1979 g and p; Garlid and Beavis, 1987) that P -induced mitochondrial swelling is the 1 result of Ca(II)-dependent changes in inner membrane permeability. The trace shown in Fig 10 is not consistent with such a conclusion. In the presence of the Ca(II) chelating agent EGTA (5 mM), Pi still induces two phases of matrix swelling. Moreover, the shrinking phase is exacerbated. The kinetics of the first swelling phase are unaffected by EGTA. Considerable evidence indicates that the first swelling phase is coupled to P uptake. MalNEt is known to inhibit the mitochondrial P i 1 transport protein by alkylating thiol groups essential for activity (Fonyo and Ligeti, 1984; Klingenberg §£_gl., 1974). Preincubating mitochondria with MalNEt completely inhibits swelling subsequent to the administration of Pi (Fig. 3A). Uptake catalyzed by the P1 trans- port protein is believed to be a proton-coupled process (Hoek g£_§l., 1971; McGivan and Klingenberg, 1971; Coty and Pederson, 1974). Therefore, the addition of a protonophore should inhibit Pi-induced swelling if the volume change is associated with catalytic Pi inter- Figure 3. (A). (B). (C). 214 Characterization of some of the Factors Affecting the Pi-Induced Change in Mitochondrial Matrix Volume Inhibition of light scattering changes secondary to the addition of 20 mM Pi by preincubating chick heart mitochondria (M, 0.87 mg mlfii) with N-ethylmaleimide (NEM, 1.4 mM) for 6 min. Inhibition of mitochondrial swelling secondary to the addition of 20 mM P1 by preincubating chick heart mitochondria (0.22 mg ml-j) with 50 pM of the uncoupler CCCP. Perturbation in chick heart mitochondrial light scatter secondary to the addition of 1 mM pyruvate and 0.5 mM malate. Mitochondria were suspended in 2.0 ml of M/S media at 0.97 mg ml-i. 215 Ail. PI \ w D. C C C M MalNEt ./ b il‘ / l. Ea 008 8:28”. 8828 a 2:5 32:24 10 Time, min 216 nalization. Consistent with this rationale, CCCP (Fig 3B) inhibits the initial volume change. In addition, the first-order rate constant for the first swelling phase (1.08 i 0.1 min -1) is conspicuously close to the k determined for the purified P transporter of rat liver mito- i chondria (0.85 min-1; Kaplan §£_pl., 1986). Taken together, these data strongly suggest that the swelling observed immediately after Pi addition reflects properties of the Pi transport protein and, thus, is a measure of the rate of Pi uptake. In the absence of exogenous substrate, the matrix of these mito- chondria cannot undergo ADP-induced contraction after the second swelling phase. The most reasonable explanation for this lack of response is that the mitochondria become subject to such severe stress that the outer membrane is lysed, thereby allowing the inner membrane to completely unfold. As discussed by Beavis pp_§l. (1985), the latter process is irreversible and would obviate any possibility of ordered inner membrane transitions from taking place. II. LIMITATIONS ON THE RATE AND EXTENT OF SWELLING The above experimental evidence supports the assumption that Pi- induced swelling proceeds at a rate that is limited by the Pi trans- porter. Accordingly, light scatter measurements were used to estimate the Km of the Pi transporter for Pi' By adding increasing concentrations of P1 to a constant concentration of mitochondria, two types of measure- ments of the first swelling phase were made. The first was AL.S./At, the dependence of the rate of swelling on P1 concentration; and, second, the L.S. - L.S. ). The latter total change In light scatter (AL.S.Total = max min values provide information on the relative change in matrix volume as a 217 function of Pi concentration. Uptake was allowed to proceed in the presence of endogenous substrate only, because the uptake of pyruvate and malate is also accompanied by mitochondrial swelling (Fig 3C). The first-order rate constant of Pi-induced swelling is independent of Pi concentrations over a ninety-fold range (0.5-45 mM; Fig 4). These kinetic data indicate that the chick heart mitochondrial Pi transporter has a Km for P1 less than or equal to 50 pM. AL.S. varies linearly Total as a function of Pi up to approximately 18 mM (Fig 5). From 18-45 mM, no further increase in AL.S is observed, suggesting that the mitochon- 'Total drial matrix has attained its maximal volume while still subject to the morphological constraints exerted by the outer membrane. III. FACTORS REGULATING THE AMOUNT OF TRANSPORTED INTO THE MATRIX Since the results shown in Fig 5 imply that matrix volume could limit the amount of Pi that mitochondria transport in, the following experiment was designed. Mitochondria were coincubated with increasing 32 concentrations of Pi and 5 mM pyruvate/2.5 mM malate for 10 min. Exogenous substrate was provided so as to insure structural integrity of mitochondria for the entire course of an incubation (see Fig 1A). It is important to note that mitochondria coincubated with P1 and P/M also attain maximal volume at concentrations of P a 18 mM. The amount of P1 1 internalized under these conditions increases linearly from 3-100 mM Pi (Fig. 6). Clearly, net uptake continues above 18 mM P Once mitochondria 10 reach their maximal matrix volume and water can no longer flow in, some other matrix solute(s) must continuously flow out so as to compensate for osmotic imbalances that may develop as P continues to be transported i in. If this were not the case, the light scatter signal subsequent to 218 Figure 4. Kinetic Constants of the First Pi-Induced Swelling Phase Measured as a Function of Pi Concentration Changes in the intensity of scattered light were measured as described under "Experimental Procedures." First-order kinetic constants were calculated as described in the legend to Figure 2. All measurements were performed in the absence of exogenous substrate. (A). Half-time (tO in seconds). 5’ (B). First-order rate constant (k, min -1). *1/21 sec k, min‘' 219 220 Figure 5. Total Change in Chick Heart Mitochondrial Light Scattering Measured as a Function of Initial Extramitochondrial Pi Concentration Mitochondria (0.8 mg ml-1) were irradiated with green light (500 nm) and changes in 90° light scattering were monitored continuously from before the addition of P1 to the end of the first swelling phase. 221 F _ 0 0 O 0 8 6 1000 - _-_s . roe . .5858 28:4 45 40 I5 20 25 30 35 IO 222 32 Figure 6. Mitochondrial Uptake of Pi Measured as a Function of Pi Concentration 32 P1 as described under "Experimental Procedures." All points repre- Mitochondria were incubated with and assayed for sent the mean of two determinations. The inset shows the volume of extra-mitochondrial water that co-sediments with mitochondria that have been pre-incubated with the indicated concentrations of Pi' 223 728.88 .2228? 2:. .n. a 8% l00 708090 405060 I0 20 30 0 [Ft] . mM 224 the swelling phase could not remain stable (Fig. 1A). Considering the sheer number of inner membrane substrate transporters (Lanoue and Schoolwerth, 1986), as well as the complex interrelations possible between them, the variety of electrolytes that may participate to some degree in precise osmotic adjustments is large. In order to simplify the analysis, transmembrane fluxes of K+, Mg(II), and Ca(II) were examined because changes in the matrix concentrations of these cations are known to influence mitochondrial volume significantly (for review, see Garlid and Beavis, 1987). In the absence of exogenous substrate, both diluting mitochondria into a larger volume of isotonic M/S medium or adding the organelles to M/S made 20 mM in P1 (M/S/Pi) elicits the extrusion of K+ into the extramitochondrial space (Fig 7A). The response in the two cases is identical. K+ is released linearly as a function of time at a rate of 21.1 i 2.1 nmoles/min/mg mitochondria for 4 min. Between 4 and 10 min, no further release of K+ is observed. After reaching the plateau, these mitochondria release 61-65% of their endo- genous K+. With excess P/M added as an energy source, the kinetics of K+ extrusion in both the presence and absence of 20 mM P are dramatically i changed (Fig 78). With exogenous substrate only, K+ efflux proceeds after a 4 min lag. Subsequent to the lag, the mitochondria release approximately 25.6 nmoles K+/mg over an additional 4 min period. When mitochondria are simultaneously presented with P/M and P1, K+ efflux is biphasic with k1 = 14.1 i 1.6 and k2 = 4.4 i 0.34 nmoles/min/mg mitochondria. The difference in K+ efflux kinetics during Pi transport proceeding with and without exogenous P/M is striking. In the former case, mitochon- dria undergo a single swelling phase and remain structurally intact. The Figure 7. (A). (B). 225 The Effect of Pi and Exogenous Substrate on the Rate of K+ Efflux from Chick Heart Mitochondria K+ concentration in the extramitochondrial space measured as a function of time of incubation in either M/S ((:)) or M/S/Pi (.). Mitochondria at 0.56 mg ml“? were suspended in a final volume of 1.75 ml and incubated at 30.5°C. Time course for K+ efflux in M/S/T containing 5 mM pyruvate and 2.5 mM malate (P/M; Q ) or M/S/T enriched with both P/M and 20 mM Tris-Pi ( . ). Mitochondria at 0.65 mg ml-1 were suspended in a final volume of 1.76 ml and incubated at 30.5°C. 226 '_(Dupuoqoouw 6w). pascalag 1); salouuu m 1 0 £2 0. .. < l l l \___ 83.38988 0 I0 23456789 0 IO 23456789 0 Time, min Time, min 227 first, faster phase of K+ efflux proceeds until P induced swelling is i 90% complete. The rate of K+ efflux becomes three-fold slower as swelling nears and finally attains completion. At this point mitochondria main- tain a constant volume and take up Pi independent of water by closely matching the efflux of K+ and likely other ions with P1 influx. In the absence of added P/M, K+ loss is not coupled to P uptake. Consequently, i once swollen to their maximal volume, the mitochondria become stressed osmotically because K+ is no longer used to balance solute pressure on both sides of the inner membrane. If Pi uptake continues without osmotic control, the outer membrane ruptures and the inner membrane unfolds. The following major points can be drawn from these data: (a) K+ efflux occurs concomitantly with P1 influx; (b) Pi itself does not induce K+ extrusion; rather, K+ is used to relieve part of the osmotic stress resulting from the disruption of osmotic balance across the inner membrane; (a) exogenous substrate limits non-specific loss of K+ and appears to provide mitochondria with the energy necessary to continu- ously maintain osmotic balance as Pi and water are transported into the matrix. Fluxes of Mg(II) also participate in the volume changes induced by P1. The amount of Mg(II) released by chick heart mitochondria incubated in 20 mM Pi also depends on whether or not the incubation medium contains exogenous substrate (Fig 8). In the absence of exogenous substrate, these mitochondria release 36.9 1 0.63 nmoles Mg(II)/mg over a 20 min period. This decreases to 11.5 i 0.54 and 13.1 i 0.79 nmoles Mg(II)/mg when the mitochondria are co-incubated with pyruvate/ malate or succinate, respectively. These results suggest that in the presence of substrate the mitochondria either release less Mg(II) during 228 Figure 8. Mg(II) Concentration in the Extra-Mitochondrial Space Measured as a Function of Mitochondrial Concentration Mitochondria at the indicated concentration were allowed to incubate at 30.5°C for 20 min in 225 mM mannitol, 75 mM sucrose, and 20 mM Tris-Pi, pH 7.4, Suspensions were centrifuged and the supernatants were treated and assayed as described under "Experimental Procedures." Incubations conducted with: no substrate ( . ); 5 mM pyruvate/2.5 mM malate ( A ); or 5 mM succinate ( O ). 30 nmoles Mg(II) Released 229 l l I 0.25 0.50 0.75 mg Mitochondrial Protein-ml" 0.10 230 P1 uptake or that the released Mg(II) can be reinternalized in an energy- dependent manner. In order to distinguish between these possibilities, the following experiments were performed. Fig 9A shows the time course for Mg(II) release when mitochondria are incubated with or without P1 in the presence of P/M. In M/S medium enriched with P/M only, the amount of extramitochondrial Mg(II) remains constant at 6.4 i 0.47 nmoles/mg mito- chondria. This represents the amount of Mg(II) lost during isolation. As mitochondria take up Pi when suspended in M/S/P containing P/M, Mg(II) i is released at a steadily decreasing rate. This result suggests that Mg(II) efflux is also coupled to osmotic stabilization of the matrix as water and P1 are transported in. In the absence of exogenous P/M and Pi’ Mg(II) effluxes at a constant rate of 0.49 1 0.021 nmoles/min/mg mito- chondria (Fig 9B). Maintaining a constant matrix Mg(II) concentration thus requires energy. After an initial 3 min lag, the Pi-induced swell- ing process gives rise to a large increase in the rate of Mg(II) efflux from chick heart mitochondria metabolizing endogenous substrate only (Fig 9B). This results in the loss of 85% of endogenous Mg(II). The timing of this massive release of Mg(II) coincides with the transient matrix contraction which occurs subsequent to Pi-induced swelling (Fig 18). It is not possible to discern from these data whether the contraction detected by light scatter measurements represents water loss coupled to Mg(II) efflux or some as yet undefined change in inner membrane configuration triggered by the loss of Mg(II). One point favoring the former possibility is that both Mg(II) efflux and the mitochondrial contraction occur in synchrony, in that both begin after a 3 min lag and take 5 min to go to completion. The Mg(II) so released cannot be reinternalized under either standard state 3 or state 4 Figure 9. (A). (B). 231 The Effect of Pi and Exogenous Substrate on the Rate of Mg(II) Efflux from Chick Heart Mitochondria Mg(II) concentration in the extramitochondrial space measured as a function of time of incubation in either 225 mM mannitol, 75 mM sucrose, 5 mM pyruvate/2.5 mM malate, 20 mM Tris, pH 7.4 (O) or this assay medium enriched with 20 mM Tris-P1 (. ). Mitochondria (0.65 mg ml-i) were suspended in a final volume of 1.76 ml and incubated at 30.5°C. Mg(II) concentration in the extramitochondrial space measured as a function of time of incubation in either M/S (A) or M/S/Pi (.). The data represent two separate experiments, differentiated by open and closed symbols. The insert shows the concentration of Mg(II) in the extra-mitochondrial space as a function of time of incubation with 5 mM pyruvate/2.5 mM malate (state 4,(:)) or 5 mM pyruvate/2.5 mM malate and 4.2 1 pmoles of ADP (state 3, .) after mitochondria (0.61 mg ml_ ) were incubated for 20 min at 30.5°C in M/S/Pi. 232 I0 L 02468 Time, min 0 «189° but/(mow sa|outu 9L1 16.. v N g a) (0 <1- N O '_(Dppuoqoouw DLU) pas-08183 (mow saiowu 10 23456789 0 I0 3456789 Time, min Time, min 233 conditions (Insert, Fig 9B). During the course of Pi uptake, Ca(II) fluxes are not involved in preventing the development of osmotic stress in any specific way. Chick heart mitochondria release 92% of endogenous Ca(II) over a 10 min period when added to a larger volume of M/S or M/S/P media (Fig 10A). If i provided with exogenous P/M after the second swelling phase (Fig 1B), these mitochondria transport Ca(II) back into the matrix with an approximate half-time of 35 sec (Fig 108). In M/S or M/S/P1 media enriched with exogenous P/M, extra-matrix Ca(II) content remains essentially constant at 7.1 i 1.2 and 7.0 i 1.7 nmoles/mg mitochondria, respectively. These results show that: (a) P1 uptake does not induce Ca(II) efflux. Whether Ca(II) efflux occurs depends, under these con- ditions, solely upon the mitochondrial energy poise. (b) The mitochon- drial inner membrane is intact and respiration-competent subsequent to the second swelling phase that occurs when P uptake proceeds in the i absence of exogenous substrate. This conclusion also implies that the outer membrane is ruptured but not stripped away, because cytochrome 0 cannot be lost to the bulk medium if respiration is to occur. (a) When metabolizing endogenous substrate only, energy is preferentially comm- itted to the uptake of Pi rather than that of Ca(II). The functional and mechanistic basis for such preferential energy utilization is yet to be elucidated. Figure 10. (A). (B). 234 The Effect of Pi and Exogenous Substrate on Transmembrane Ca(II) Fluxes in Chick Heart Mitochondria Ca(II) concentration in the extra-mitochondrial space measured as a function of time of incubation in either M/S ( O ) or M/S/Pi ( . ). Mitochondria were suspended in a final volume of 1.75 ml at 0.73 mg ml-1. Internalization of Ca(II) by chick heart mitochondria. Mitochondria were incubated in M/S/Pi at 30.5°C for 20 min to allow for Ca(II) efflux. Ca(II) concentrations in the extra-mitochondrial space are shown as a function of time of incubation with 5 mM pyruvate/2.5 mM malate (state 4, . ) or 5 mM pyruvate/2.5 mM malate and 4.2 pmoles of ADP (state 3, O ). Mitochondria (0.61 mg/ml) were suspended in a final volume of 1.75 ml. 30.. I51. IO 0 N N 9 '_(Dppuoqoouw but) .(11)Do SO|OUJU IO I0 23456789 0 10 23456789 0 Time, min Time, min 236 DISCUSSION I. Ca(II) IS NOT REQUIRED FOR Pi-INDUCED SWELLING OF CHICK HEART MITOCHONDRIA A variety of studies have suggested that Pi-induced swelling is the result of an interaction between internalized Ca(II) and the mitochon- drial inner membrane. The "Ca(II)-induced membrane transition" (Hunter g£_gl., 1976, 1979 g and p; Haworth and Hunter, 1979) or the "Ca(II)- induced hole" (Garlid and Beavis, 1987) is believed to facilitate P1 permeation due to an increase in the permeability of the inner membrane to both charged and uncharged solutes. In this model the initiation of Pi-induced swelling depends absolutely on Ca(II) that is transported from the extramitochondrial space into the matrix. Consistent with this, Garlid and Beavis (1987) found that, in the absence of exogenous Ca(II), rat liver mitochondria oxidizing succinate do not swell secondary to the addition of Pi' The Ca(II)-dependent increase in membrane permeability is accompanied by increases in mitochondrial I-acyllysophospholipid content, indicating that once inside the matrix, Ca(II) activates phospholipase A2 (Beatrice pp_gl., 1980). Of interest is that in all of these studies, swelling in the presence of Pi proceeds after a lag time of approximately one to several min. For the swelling studies detailed here, no lag between the addition 0f Pi to mitochondrial suspensions and the initiation of swelling is seen. Moreover, when swelling was monitored in the absence of exogenous P/M, Ca(II) was lost to the extramitochondrial space (Fig 10A). Under 237 these conditions no Ca(II) internalization took place during the course of a series of volume changes (see Fig 18). That a generalized permea- bility change could not be facilitating P uptake is further evidenced i by the facts that the rate of Mg(II) efflux is unaffected during the initial swelling phase (Fig 9B) and this swelling occurs in the presence of a high concentration of EGTA (Fig 1C). When presented with exogenous P/M, chick heart mitochondria readily take up extramitochondrial Ca(II) (see Fig 108 of this Chapter, and Fig 19 of Chapter 3). Yet, as shown in Fig 1A, this did not lead to the type of large-scale, massive swelling to be expected subsequent to a Ca(II)-induced membrane transition. Rather, the mitochondria were able to maintain precise osmotic control over matrix volume after the initial swelling phase reached steady- state. Taken together, these results show that in chick heart mitochon- dria, the Ca(II)-dependent membrane transition is not requisite to P1- induced swelling. The swelling arises from P internalization catalyzed i by the Pi transport protein as well as concomitant cationic fluxes. The discrepancy between these studies and those of the authors cited above is difficult to account for. It is possible that chick heart mitochondria simply lack the biochemical apparatus underlying the Ca(II) induced membrane transition observed in rat liver (Garlid and Beavis, 1987) and bovine heart (Hunter and Haworth, 1979 p and p) mitochondria. II. K+ AND Mg(II) FLUXES SECONDARY TO Pi ADDITION K+ and Mg(II) appear to participate directly in osmotic stabiliza- tion of the mitochondrial matrix during the steady-state subsequent to Pi-induced swelling (see Fig's 7 and 9). It may be argued that K+ and Mg(II) are lost because: (a) they are passively diffusing down their 238 respective concentration gradients; or (b) the binding of Tris+ to mitochondrial membranes displaces these metal cations from negatively charged phospholipid head groups. Both of these possibilities are neg- ated by the data shown in Fig's 7B and 9A. Neither K+ nor Mg(II) is continuously lost when incubated in a Tris-buffered osmotic support medium containing exogenous P/M. In the case of Mg(II), for instance, matrix stores of this ion remain constant. It follows that when mito- chondria are suspended in a Tris-buffered medium containing P/M and Pi-induced swelling is initiated, changes in matrix Mg(II) must be coupled to the uptake of Pi' Consistent with the findings of others (Crompton et al., 1976), the control of Mg(II) efflux during P uptake 1 proceeding in the presence of excess substrate appears to be absolutely dependent on respiration. The degree of control over K+ extrusion and the apparent coupling of this process to mitochondrial volume changes (Fig. 7B) are striking. Although not discernible from these data, it is possible that the putative mitochondrial inner membrane K+/H+ antiporter (Nakashima and Garlid, 1982; Martin gp_§l., 1984) is responsible for controlling the rate of K+ efflux. In the steady-state (2-10 min after the addition of Pi)' these, and likely other additional (e.g., electro- phoretic Tris internalization and Pi export via malate exchange on the dicarboxylate carrier) processes operate to ensure that [solutes]in = [solutes]ou If this were not true the mitochondria would rapidly lyse t. subsequent to the continued uptake of P That the mitochondria remain i. functionally, as well as structurally, intact after P uptake in the 1 presence of exogenous P/M is further evidenced by the observation that respiratory control increases as a function of time of pre-incubation with these substrates prior to the initiation of state 3 respiration 239 with ADP (see Fig 5 of Chapter 3). It is apparent that if mitochondrial P uptake proceeds in the i absence of sufficient substrate, structural damage of the organelle occurs. Under these conditions, the contraction following Pi-induced swelling (Fig. 1B) is clearly coupled to the loss of Mg(II) (Fig. 9B). Depletion of matrix Mg(II) destabilizes the inner membrane (Binet and Volfin, 1974; Coelho and Vercesi, 1980) and increases the permeability of mitochondria to Ca(II) (Zoccarato pp pl., 1981) and K+ (Jung and Brierly, 1986). Mg(II) can only stabilize the inner membrane when membrane thiol groups are reduced; Mg(II) is released to the extra- mitochondrial space when these thiols become oxidized (Siliprandi g£_gl., 1975, 1979). In the absence of exogenous substrate, the loss of Mg(II) does not appear to be coupled to P internalization. Rather, as 1 matrix substrate is depleted subsequent to the uptake of P the energy 1’ requisite to the maintinence of matrix Mg(II) concentrations is no longer available. This loss of Mg(II) would thus be expected to proceed independent of respiration. Considering the distribution of Mg(II) between the outer membrane, inter-membrane space, inner membrane, and matrix (Bogucka and Wojtczak, 1971), the massive loss (86%) of this cation occurs in a surprisingly concerted fashion (i.e., it is mono- phasic). The efflux of Mg(II) is accompanied by a large loss of water. This phase cannot be due to the disruption of membranes because no additional K+ is lost (see Fig. 7A). As the process of Mg(II) efflux goes to completion, another swelling phase is initiated which leads to extensive disruption of mitochondrial morphology. The trigger for this transition appears to be the loss of Mg(II). The second swelling phase leads to unfolding of the inner membrane and, very likely, rupture of 240 the outer membrane. Although Mg(II) in the extramitochondrial space was not reinternalized after the second swelling phase, this does not imply that these mitochondria cannot transport Mg(II) in because net influx requires concentrations of Mg(II) in excess of 5 millimolar (Brierly §£_§l., 1963; Schuster and Olson, 1974). The second swelling phase does not lead to complete lysis of these organelles since Ca(II) was rapidly internalized when exogenous substrate was provided (Fig 108). In order for the mitochondria to be able to clear extra-matrix Ca(II) they must have had: (a) intact inner membranes; and (b) a capacity to transfer electrons to cytochrome oxidase (i.e., cytochrome a could not have been lost to the extramitochondrial space). It can be inferred from the data presented herein that substrate depletion leads to a breakdown in mito- chondrial morphology because Mg(II) is lost and the mitochondria are no longer able to match solute fluxes in and out of the matrix via energy dependent ion-exchange reactions. III. IMPLICATIONS FOR MYOCARDIAL ISCHEMIA The effects of myocardial ischemia on mitochondrial structure and function is an area currently under intensive investigation. P accumu- i lates in the cell during ischemia as the magnitude of the cytosolic phosphorylation potential decreases. Heart mitochondria swell in ische- mic myocardium (Jennings and Ganote, 1976). P is believed to participate i in the induction of this swelling (Matlib et al., 1983). Pi-induced swelling of heart mitochondria has been reported to inhibit oxidative phosphorylation in the presence of exogenous substrate (Vaghy et al., 1981; Lange et al., 1984). Both groups invoke Ca(II) as a participant in the inhibition process. The results of experiments reported here do 241 not support these findings. In the presence of exogenous substrate Pi induces swelling but compromises neither structure nor function. Moreover, Ca(II), which is actively taken-up under these conditions, exerts no detectable deleterious effects. During periods of severe ischemia distress, respiration will be inhibited due to a lack of oxygen. Protons will not be pumped, the flux through proton-coupled ion permeation pathways will be negligible, and thus the capability of mitochondria to make precise osmotic adjustments using K+, Mg(II), or other ions will be depressed. This situation is akin to that shown in Fig. 18, where subsequent to P internalization substrate is depleted. i The loss of Mg(II) and possibly other matrix components leads to the collapse of membrane organization and a loss in the capacity of mitochondria to phosphorylate ADP. The swelling induced by P1 does not per se lead to compromise; it is the energy deficiency that does 80. IV. USE OF LIGHT SCATTER METHODOLOGY T0 MEASURE Pi UPTAKE KINETICS By continuously monitoring changes in the intensity of light that is scattered by a suspension of chick heart mitochondria, it is possible to extract valuable information about the kinetic properties of the Pi transport protein and to ascertain the effect of Pi on mitochondrial structure and function. These measurements do not provide information about the specific transport rate of Pi (i.e., nmoles Pi/min/mg mito- chondria). However, by assuming that the rate of swelling is limited by the rate of Pi transport (Garlid and Beavis, 1985), it is possible to calculate an apparent rate constant for, and to identify the order of, the Pi translocation reaction. Pi-induced swelling revealed, in close agreement with the more direct studies of Kaplan et al. (1986), that Pi 242 transport is a first-order process with a k equal to approximately 1.0 . -1 min .. Error in k may arise from ionic fluxes that proceed concomitantly with P1 uptake. For instance, when Pi the presence of exogenous substrate, Ca(II) will be transported in while -induced swelling is monitored in K+ and Mg(II) are transported out. These fluxes will influence matrix volume. The extent of this effect depends on the degree to which these fluxes cancel one another. This cannot be estimated from the data reported in this communication. However, based on the inhibition of Pi- induced swelling changes with MalNEt and CCCP, it can be stated with confidence that Pi-uptake is the transport process limiting the rate of mitochondrial swelling under the conditions used in these experiments. Based on comparison with other published studies, 90° light scatter measurements are more sensitive to alterations in mitochondrial volume than monitoring changes in absorbance (0° scatter). To our knowledge, this is the first report showing that P induces a triphasic, not mono- i phasic, change in mitochondrial volume. That others have not detected this sequence of changes is probably due to a lack of sensitivity in the 0° light scatter instrumentation employed. V. ROLE OF THE Pi TRANSPORT PROTEIN IN RESPIRATORY CONTROL In recent years much attention has been directed toward identifying the rate limiting step or steps of oxidative phosphorylation (Lemasters and Sowers, 1979; Wilson, 1980; Groen g£_§l., 1982). The P1 transporter is believed to exert significant control over the rate of ATP synthesis in yeast mitochondria (Mazat §£_§l., 1986). Although swelling studies do not constitute direct kinetic assays of Pi transport, the fact that the 243 rate of Pi-induced swelling is independent of Pi concentration in the range of 0.5-45 mM suggests that the P transport protein is nearly 1 saturated at these concentrations. At pH 7.4, one would expect that the total P1 in solution will consist of an equilibrium between the mono- and divalent forms of this anion. Assuming that 90% saturation of an enzyme with its substrate is equal to IOKm, then it is estimated that the Km of the chick heart mitochondrial Pi transport protein for P1 is no higher than 25-50 pM. This value is in sharp contrast with that published for the rat liver Pi transport protein (4.21 mM; Coty and Pederson, 1974). The basis for the difference in the affinity of these mitochondria for P1 may in part be attributed to the specific energy requirements of heart muscle and liver cells. The metabolic focus of mitochondria in these two tissues is very different. Heart mitochondria may require greater efficiency in binding and transporting Pi into the matrix because of the constant beat-to-beat energy demands of myocardium. Under physiological conditions the P transporter in heart would thus be i expected to be very nearly saturated at all times. This would preclude a role for the Pi transporter in the control of oxidative phosphorylation. VI. CONCLUSIONS The studies described herein demonstrate that: (a) P1 induces res- piration-dependent swelling of chick heart mitochondria. Significantly, this process does not require a Ca(II) triggered alteration in mitochon- drial membrane permeability. (b) In the presence of excess oxidizable substrate, the volume change is monophasic and native mitochondrial mor- phology does not appear to be disrupted. (c) When mitochondria oxidize endogenous substrate only, Pi induces a triphasic change in mitochondrial 244 volume, resulting in extensive morphological damage. (d) Complex K+ and Mg(II) fluxes accompany the volume changes resulting from P1 uptake. (e) The rate of swelling observed secondary to the addition of Pi reflects the kinetic properties of the mitochondrial inner membrane Pi transport protein. 245 REFERENCES Adamsons, K., Sell, J.E., Holland, J.F., and Timnick, A. (1982) Anal. Biochem. 54: 2186-2190. Armston, A.E., Halestrap, A.P., and Scott, R.D. (1982) Biochim. Biophys. Acta 681: 429-439. Austin, J., and Aprille, J.R. (1984) J. Biol. Chem. 259: 154-160. Beatrice, M.C., Palmer, J.W., and Pfeiffer, D.R. (1980) J. Biol. Chem. 255: 8663-8671. " ‘ Beavis, A.D., Brannan, R.D., and Garlid, K.D. (1985) J. Biol. Chem. 260: 139u2u-139u33o . Bencini, D.A., Wild, J.R., and O'Donovan, G.A. (1983) Anal. Biochem. 132: 254-258; Binet, A., and Volfin, P. (1974) Arch. Biochem. Biophys. 164: 756-764. Brierly, G., Murer, E., Bachmann, E., and Green, D.E. (1963) J. Biol. Chem. 238: 3482-3489. Brierly, G.P., Jurkowitz, M., Scott, K.M., and Merola, A.J. (1971) Arch. Biochem. BiOphys. 147: 545-556. ‘ ' Chance, 8., and Hagihara, B. (1963) Proc. 5th int. Congr. Biochem. 5: 3‘37. Chappell, J.B., and Crofts, A.R. (1965) Biochem. J. 95: 378-386. Chavez, E., Jung, D.W., and Brierly, G.P. (1977) Arch. Biochem. Biophys. 183: 460-470. Coelhi, J.C., and Vercesi, A.E. (1980) Arch. Biochem. Biophys. 204: 141-147. Coty, W.A., and Pederson, P.L. (1974) J. Biol. Chem. 249: 2593-2598. Crofts, A.R., and Chappell, J.B. (1965) Biochem. J. 95: 387-392. Crompton, M., Capano, M., and Carafoli, E. (1976) Biochem. J. 154: 735-742. ' Fonyo, A., and Vignais, P.V. (1979) FEBS Lett. 102: 301-305. Garlid, K.D., and Beavis, A.D. (1985) J. Biol. Chem. 260: 13,434-13,441. 246 Garlid, K.D., and Beavis, A.D. (1987) Biochem. Biophys. Acta 853: 187-204. ’ Greenbaum, N.L., and Wilson, D.F. (1985) J. Biol. Chem. 260: 873-879. Groen, A.K., Wanders, R.J.A., Westerhoff, H.V., Van der Meer, R., and Tager, J.M. (1982) J. Biol. Chem. 257: 2754-2757. Hackenbrock, C.R. (1966) J. Cell Biol. 30: 269-297. Halestrap, A.P., Quinlan, P.T., Whipps, D.E., and Armston, A.E. (1986) Biochem. J. 236: 779-787. ' Haworth, R.A., Hunter, D.R. (1979) Arch. Biochem. Biophys. 195: 460-467. Hoek, J.B., Lofrumento, N.E., Meyer, A.J., and Tager, J.M. (1971) Biochem. Biophys. Acta 226: 297-308. ' Hunter, D.R., Haworth, R.A., and Southard, J.H. (1976) J. Biol. Chem. 251: 5069-5077. ‘ Hunter, D.R.. and Haworth, R.A. (1979) Arch. Biochem. Biophys. 195: 453-459. Hunter, D.R., and Haworth, R.A. (1979) Arch. Biochem. Biophys. 195: 468-477. Izzard, S., and Tedeschi, H. (1973) Arch. Biochem. Biophys. 154: 527-539. Jennings, R.B., and Ganote, C.E. (1976) Circ. Res. 38: 80-91. Jung, D.W., Chavez, E., and Brierly, G.P. (1977) Arch. Biochem. Biophys. 183: 452-459. Kaplan, R.S., and Pederson, P.L. (1983) Biochem. J. 212: 279-288. Kaplan, R.S., Pratt, R.D., and Pedersen, P.L. (1986) J. Biol. Chem. 261: 12,767-12y773. Klingenberg, M., Durand, R., and Guerin, B. (1974) Eur. J. Biochem. 42: 135-150. Lange, L.G., Hartman, M., and Sobel., B.E. (1984) J. Clin. Invest. 73: 1046-1052. LaNoue, K.F., and Schoolwerth, A.C. (1979) Ann. Rev. Biochem. 48: 871-922. Lehninger, A.L. (1962) Physiol. Rev. 42: 467-517. Lemasters, J.J., and Sowers, A.E. (1979) J. Biol. Chem. 254: 1248-1251. 247 Ligeti, E., and Fonyo, A. (1984) Eur. J. Biochem. 159: 279-285. Lotscher, H., Winterhalter, K.H., Carafoli, E., and Richter, C. (1980) Eur. J. Biochem. 110: 211-216. Markwell, M.A.K., Haas, S.M., Talbert, N.E., and Bieber, L.L. (1981) Methods Enzymol. 72: 296-303. Martin, W.H., Beavis, A.D., and Garlid, K.D. (1984) J. Biol. Chem. 259: 2062-2065 Matlib, M.A., Vaghy, P.L., Epps, D.E., and Schwartz, A. (1983) Biochem. Pharmacol. 32: 2622-2625. ' Mazat, J., Jean-Bart, E., Rigoulet, M., and Guerin, B. (1986) Biochem. J. 849: 7-15. ' McGivan, J.D., and Klingenberg, M. (1971) Eur. J. Biochem. 20: 292-399. McGivan, J.D., Grebe, K., and Klingenberg, M. (1971) Biochem. Biophys. Res. Comm. 45: 1533-1541. Nakashima, R.A., and Garlid, K.D. (1982) J. Biol. Chem. 9252-9254. Nicholls, D.C., and Crompton, M. (1980) FEBS Lett. 111: 261-268. Packer, L. (1961) J. Biol. Chem. 236: 214-220. Palmieri, F., Quagliariello, E., and Klingenberg, M. (1970) Eur. J. Biochem. 17: 230-238. Pozzan, T., Bragadin, M., and Azzone, G.F. (1977) Biochemistry 16: 5618-5625. ‘ Puskin, J.S., Gunter, T.E., Gunter, K.K., and Russell, P.R. (1976) Biochemistry 15: 3834-3842. Schuster, S.M., and Olson, M.S. (1974) J. Biol. Chem. 249: 7151-7158. Siliprandi, D., Toninello, A., Zoccarato, F., Rugolo, M., and Siliprandi, N. (1975) Biochem. Biophys. Res. Comm. 66: 956-961. Siliprandi, D., Rugolo, M., Zoccarato, F., Toninello, A., and Siliprandi, N. 1979) Biochem. Biophys. Res. Comm. 88: 388-394. Tedeschi, H. (1961) Biochim. Biophys. Acta 46: 159-169. Tyler, D.D. (1980) Biochem. J. 192: 821-828. Vaghy, P.L., Matlib, M.A., and Schwartz, A. (1981) Biochem. Biophys. Res. Comm. 100: 37-44. 248 Wilson, D.F. (1980) In Membrane Structure and Function (E.E. Bittar, ed.). Pp. 153-195. J. Wiley Press, New York. Wohlrab, H. (1986) Biochim. Biophys. Acta 853: 115-134. Zoccarato, F., Rugolo, M., Siliprandi D., and Siliprandi, N. (1981) Eur. J. Biochem 114: 195-199. Chapter 6 INTERACTION OF INORGANIC PHOSPHATE WITH CHICK HEART MITOCHONDRIA. II. SUBSTRATE-DEPENDENT MODULATION OF STATE 3 AND STATE 4 RATES OF RESPIRATION. 249 250 INTRODUCTION Inorganic phosphate (Pi) is a functionally versatile anion (Westheimer, 1987) which regulates the flux through numerous key meta- bolic pathways. Mitochondrial oxidative phosphorylation is one such pathway. However, the explicit kinetic and thermodynamic mechanisms by which Pi regulates the rate of mitochondrial respiration are unclear. A number of theories concerning this issue have been elaborated in recent years. Wilson and coworkers (Wilson §£_pl., 1974; Holian g£_§l., 1977) showed that the rate of mitochondrial respiration depends on the extra- mitochondrial phosphorylation potential, defined as: AGP = AGOP + RT ln([ATP]/[ADP][Pi]) (1) where all terms have their usual meaning. According to this theory, increased Pi concentrations activate respiration because the phosphory- lation potential (AGP) would no longer be at equilibrium with the redox state of the electron transfer chain components involved in energy con- servation. Lemasters and Sowers (1979) postulated that increasing Pi concentrations would stimulate the rate of oxidative phosphorylation by raising the magnitude of the Aw component of protonmotive force, thereby accelerating the rate at which ADP3- exchanges with ATP”- across the mitochondrial inner membrane. In contrast, other workers claim that the rate of respiration is independent of P and depends instead on the 1 ratio of ATP to ADP concentrations, primarily because ATP competes with the binding of ADP to the adenine nucleotide translocase (Slater et al., 251 1973; Davis and Lumeng, 1975; Kuster gp_§l., 1976). Early work by Chance and Williams (1955, 1956) with isolated mitochondria suggested that ADP is the primary regulator of respiratory rate. This regulation has been substantiated in intact skeletal muscle under steady-state conditions using phosphorous magnetic resonance spectroscopy (Chance p£_pl., 1985, 1986). In those models which attribute primary rate control to ADP, Pi may be rate-limiting if the Pi requirements for near maximal activity of the ATP synthase are not met. Thus, in kinetic models Pi apparently modulates the rate of respiration because it is a substrate of oxidative phosphorylation with a characteristic affinity for the F1FO-ATP synthetase. Integrative models describing the relative contribution of P1 to respiratory control must account for all possible ways that P regulates i the rates of mitochondrial substrate oxidation and ADP phosphorylation. The P1 carrier protein catalyzes the transport of Pi into the mitochon- drial matrix via Pi/H+ symport or Pi/OH— antiport (Hoek §£_gl., 1971; Greenbaum and Wilson, 1985). If the Km of this protein for P1 is in the millimolar range (Coty and Pedersen, 1974; Ligeti p£_pl., 1979), then it too could exert some control over the rate of oxidative phosphorylation. However, as shown in Chapter 5, in chick heart mitochondria the Km of this protein for P1 appears to be less than or equal to 50 pM. Since Pi concentrations in vivo approximate to 1-2 mM (Chance pp_pl., 1986), the Pi transport protein of these heart mitochondria is very likely nearly saturated under normal conditions and thus would not be expected to exert significant control over rates of respiration. Another possibility for regulation by P1 is that this anion may positively or negatively affect the kinetics of matrix dehydrogenases. In this case Pi would 252 limit respiration by modulating matrix concentrations of NADH. The following investigations were performed to characterize the kinetics of chick heart mitochondrial Pi utilization in an effort to contribute toward a more comprehensive model of how Pi regulates respiration. The data show that P regulates the rate of mitochondrial i respiration in several ways. It is, first, an effector of the dehydro- genases for a-ketoglutarate and B-hydroxybutyrate during state 3 respi- ration; second, a substrate for the ATP synthetase; and third, an inhibitor of state 4 respiration when the dehydrogenases for pyruvate and a-ketoglutarate are rate-limiting. The data support the arguments that Pi regulates respiration through kinetic rather than thermodynamic means and that mitochondrial matrix dehydrogenases participate in the phenomenon of respiratory control. 253 EXPERIMENTAL PROCEDURES I. MATERIALS The water used for these experiments was purified as previously described in Chapter 3 of this dissertation. The following substances were reagent grade or better, used without further purification, and obtained from the sources noted: ADP (grade X), GDP (type I), IDP, NAD (grade III-C), NADH (grade II), rotenone, phosphoenolpyruvate, thiamine pyrophosphate, INT, EGTA, fatty acid free BSA, Tris, sucrose (RNase- free), B-mercaptoethanol, lactate dehydrogenase (type II), pyruvate kinase (type III), collagenase (type VII, lot 47F-6829), lipoamide dehydrogenase (type III), and the sodium salts of a-ketoglutaric, pyruvic, B-hydroxybutyric, malic, and glutamic acids (Sigma Chemical Co., St. Louis, MO); HEPES (Boehringer-Mannheim Biochemicals, Indianapolis, IN); and mannitol (Fisher Scientific). Collagenase was reconstituted prior to each isolation in a solution containing 225 mM mannitol and 75 mM sucrose. Rotenone was dissolved in absolute ethanol (AAPER Alcohol and Chemical Co.) and stored at -20°C. Phosphate solu- tions were prepared by titrating a 1.0 M solution of phosphoric acid (Mallinckrodt) with either Tris or HEPES, as indicated, to pH 7.4. The lipoamide dehydrogenase was dialyzed at 5°C against two changes of a 20 mM HEPES, pH 7.4, buffer prior to use in order to remove the ammonium sulfate in which it was suspended. Single comb white leghorn chicks were obtained as described in Chapter 3. 254 II. ISOLATION AND PREPARATION OF MITOCHONDRIA For oxygen consumption experiments, highly coupled heart mitochon- dria were isolated from 14-21 day-old chicks using collagenase (Toth pp pl., 1986). The final mitochondrial pellet was resuspended in a medium that contained 225 mM mannitol, 75 mM sucrose, 1 mM EGTA, and 0.2% (w/v) BSA. Mitochondria used in matrix dehydrogenase assays were isolated as specified by Hinman and Blass (1981), with minor modifications. Chick ventricular myocardium was homogenized in a buffer that contained 220 mM mannitol, 70 mM sucrose, 1 mM B-mercaptoethanol, 1 mM EGTA, 0.05% (w/v) BSA, and 20 mM HEPES, pH 7.4. Mitochondria obtained after the first high speed centrifugation step (8,000xg) were washed six times (8,000xg for 20 min) in a hypotonic buffer comprised of 15 mM KCl, 1 mM EGTA, 1 mM B-mercaptoethanol, and 20 mM HEPES, pH 7.4. The resulting swollen mitochondria were resuspended at approximately 15 mg ml.1 and frozen in 1.5 ml Eppendorf centrifuge tubes at -105°C for 48 hrs. Prior to use mitochondria were broken by thawing for 15 min in a water bath adjusted to 37°C. III. ASSAYS Mitochondrial oxygen consumption was assayed at the critical mitochondrial concentration as detailed in Chapter 3. NADH production by mitochondrial matrix dehydrogenases was assayed according to the method of Hinman and Blass (1981). In this assay, NADH production is coupled to the reduction of INT using lipoamide dehydrogenase. Assays were performed at 500 nm and a molar absorptivity of 12,400 M-1 cm.1 for reduced INT was assumed (Owens and King, 1975). Broken mitochondria were suspended in 1.0 ml of an assay medium that contained 2.5 mM NAD, 2 mM 255 EGTA, 0.6 mM INT, 0.2 mM thiamin pyrophosphate, 0.3 mM DL-dithiothreitol, 1 mM MgCl 0.1% (w/v) BSA, 20 mM HEPES, pH 7.4, 10 pM rotenone, and an 2. excess of lipoamide dehydrogenase. In addition, the assay medium con- tained oxidizable substrate (5 mM), as well as Pi and a nucleotide diphosphate at concentrations indicated in Tables 2 and 3 and in the legend to Figure 5. In order to ensure that the activiation of dehydro- genase activity by P1 and nucleotide diphosphates was specific, EGTA was included in the assay medium because Ca(II) is known to activate the dehydrogenases for pyruvate and a-ketoglutarate (Denton and McCormack, 1986; McCormack and Denton, 1979). Rotenone was included in the assay medium so as to prevent the oxidation of NADH by site I of the electron transfer chain. ADP concentrations were assayed using the pyruvate kinase/lactate dehydrogenase assay detailed elsewhere (Toth §£_pl., 1988a). GDP and IDP concentrations were determined spectrOphotometrically in 1.0 ml of a 20 mM Tris, pH 7.4, buffer. The extinction coefficients used for GDP and IDP were 8(252 mm) = 13,700 M-1 cm“1 (Bock et al., 1956) and 8(265 nm) = 23,600 M.1 cm"1 (C.H. Suelter, personal communication), respectively. Other assays and statistical analyses were performed using the methods described in Chapter 3. 256 RESULTS I. THE INORGANIC PHOSPHATE REQUIREMENTS OF STATE 3 RESPIRATION The exogenous Pi requirements of chick heart mitochondrial state 3 respiration vary with the carboxylic acid substrates (Fig's 1 and 2). In these experiments, rates of state 3 respiration supported by endogenous Pi were substracted from rates measured at each concentration of added Pi' Therefore, the rates represent respiration stimulated by exogenous Pi' The kinetic parameters for P1 saturation of state 3 respiration are summarized in Table I. The Kso's for P1 saturation of state 3 respiration in the presence of pyruvate/malate (P/M) and glutamate/malate (G/M) (Fig 1) are very similar and are in close agreement with the K50 of bovine heart ATP synthase for P1 (Matsuno-Yagi and Hatefi, 1986). Therefore, the K50 obtained during the oxidation of these substrate pairs probably reflects directly the P requirements of the chick heart ATP synthase, 1 and no other site of interaction for P in the matrix is apparent. 1 Similar titrations with P1 when a-ketoglutarate (aKG; Fig 1) or B-hydroxybutyrate (BHB; Fig 2) are oxidized suggest, however, the exis- tence of more than one binding site for Pi' The kinetics of stimulating the rate of state 3 respiration with P during the oxidation of a-keto- i glutarate are sigmoidal. Consistent with sigmoid-type kinetics found with other enzymes (Monod et al., 1965; Koshland et al., 1966), it is possible that P binds cooperatively to the dehydrogenase for a-ketoglu- i tarate. It is reasonable to assume that the affinity of the ATP synthase for P1 will not vary as the source of reducing equivalents for electron 257 Figure 1. State 3 Rates of Respiration Measured as a Function of Pi Concentration When Mitochondria Oxidize Pyruvate/Malate, Glutamate/Malate, or a-Ketoglutarate Mitochondria phosphorylating 400 nmols of ADP oxidized either 5 mM pyruvate/2.5 mM malate (.). 5 mM glutamate/ 2.5 mM malate (A), or 5 mM a-ketoglutarate (I). The hyperbolic curves were calculated with the Michaelis-Menten equation by weighted non- linear regression. The sigmoid curve was calculated by fitting data to the following modified Michaelis-Menten equation using a non-linear least squares regression program: n vmax [Pi] v = . n n K50 + [Pi] Where "n" is equal to the Hill coefficient and all the other terms have their usual meaning. The insert shows a Klotz (1946) plot of the data normalized for n (equal to 1.0 for P1 titra- tions performed while mitochondria oxidize pyruvate/malate and glutamate/malate) and K50 values. Note that Pi concentrations vary from 10-90% saturation of the enzyme system. 258 1100 P Lnoo So .95: .758 . 0 E06- o: eozsamom m sea 259 A See . s m.ms A com 2.0 A m.wm w.o: A mmm 02.0 A ms.o voocoAm ooocsososxoeosm1m e38 . as m.zm A 2mm 82.0 A se.: AovoemAm oposoosAmoooxts w.mm A com m_.o A _:.F OHHOOLOQ>x mpmamz\mumsmpzao w.Fm A o_22 FF.o A :m._ oAAooeoosm oomfloz\osm>scse _ Ammm pzo Hoecv P CHE o sawmim: :8 xms > some ovoosAx oflmvvoA oAAsxootoo mme Hmsfixme on» ma 53m on» mpcwmogamg wsam> mfich .ucmwoflhmmoo HHHm on» ma m: .m.ex Lo umoumcfi m.omx uoaamo mam mozam> Oman» .coHmeHamOL m mumpm a“ noun mcfipHsHH cums on» mo zucfimpgoocs on» no Omsmoom .28 m.m um Snowman mm: OumHms .oumEmOsam cam mpm>scma Sufi: OOOmOSOCHoo can: “28 m was: mumszpspxxogumnim ocm .mmempsamouoxnm .mmempzam .mum>3L>q "mzoaaom mm was: mcofiumcpcmocoo mumsumosm H . m msocomoxo no cofiuflcum as» usonufiz cm>somno OH :oHmeHQmOL m mumum o: .mpmgumnsm mm wumgmusam uoumxio npfiz .>HO>HpomquL .Ammmofixo oeoscoou>o maoecv\cfle\o Eoumim: For new .oom .msp mum: Hm msocomonco now: cofipmgfiqmog m Oumum no meow; on» .mmpmnumnsm mm mumgausnmxocuxcim cam .Opmams\mpmemusam .mpmHms\OOm>:L>q mcfim: .Hm ho coHpHuom on» O» pcmscmmnsm uo>gmmno cums on» SOLO umuomcunsm mm: Hm msocmmoxm mo mocomnm 826 :H mHLBCOQOOpHE an omusoqqzm coflumgfiamon m mumpm no cums on» mpcmeflgoaxm omega :H =.mmcsnmoogm Hmpcmeflgmaxm: Loos: nonflsommo mm .nawLmOLmHoa cmwzxo cm 26H: cocouficos mm: Hm an cofiuqssmcoo cmmmxo no coHpmHzerm one 0 m 261 transfer is varied (i.e., using different exogenous substrates should have no effect on the Pi requirements of the ATP synthase). The concen- tration of Pi which gives half-maximal stimulation of state 3 rates of respiration during a-kg oxidation is 3-4 fold higher when compared with the KBO's observed during the oxidation of P/M and G/M (see Table I). Thus, Pi is rate-limiting for a-kg oxidation. However, it cannot be discerned from these data whether this rate-limitation is direct or indirect. The P1 requirement for state 3 respiration is biphasic when B-hydroxybutyrate is used as oxidizable substrate. Of interest with this substrate is that the KSO's for both phases are significantly different from the K50 of the ATP synthase for P genase has two mutually exclusive binding sites for P i' It is possible that BHB dehydro- 1' Alternatively, Pi may function as a positive effector of BHB dehydrogenase and some other matrix enzyme that is coupled to the further catabolism of aceto- acetate. The phosphorylation of ADP during the oxidation of BHB, then, appears to involve at least two and possibly three binding sites for Pi' These oxygen consumption studies cannot provide information on the precise identity of Pi binding sites in the mitochondrial matrix. II. THE INORGANIC PHOSPHATE REQUIREMENTS OF STATE 4 RESPIRATION Chick heart mitochondrial state 4 respiration also shows clear diff- erences in its response to increasing Pi concentrations as substrates are varied. With either P/M or aKG, increasing Pi concentrations inhibit uncoupled (i.e., oligomycin-insensitive) state 4 respiration in a hyper- bolic, saturable manner (Fig 3). The KI(50) values are 1.5 i 0.28 mM and 6.6 i 4.4 mM in the presence of P/M and aKG, respectively. On the other hand, Pi has little or no effect on state 4 respiration when G/M or BHB 262 Figure 2. State 3 Rates of Respiration Measured as a Function of Pi Concentration when Mitochondria Oxidize 5 mM B-Hydroxybutyrate The biphasic curve is theoretical, calculated with the kinetic constants obtained by a non-linear regression of the data fitted to the equation: 1 2 Vmax [Pi] Vmax [Pi] 1 2 Km, + [Pi] Km + [Pi] where the superscripts 1 and 2 distinguish kinetic parameters for the two phases of respiratory stimulation by P The insert 1. is an Eadie-Hofstee plot of the same data to emphasize their biphasic nature. The dotted lines are theoretical, based on the estimated K and V values for the two phases. 50 max 263 l00200300400500 o 800- . . 0 0 m m m Luca So _oES...:_E .0 £28 to: 5:238”; m 065 ng~atomO-min'i-(nmol cyt 0081715] 400- 300 200 l00 l 30 70 50 40 20 [P1]: mM 264 Figure 3. State 4 Rates of Respiration Measured as a Function of Pi Concentration When Mitochondria Oxidize Pyruvate/Malate, Glutamate/Malate, a-Ketoglutarate, or B-Hydroxybutyrate Mitochondria were titrated with increasing Pi concentrations in the presence of either 5 mM pyruvate/2.5 mM malate (‘), 5 mM glutamate/2.5 mM malate (.), 5 mM a-ketoglutarate (I), or 5 mM B-hydroxybutyrate (CD). The theoretical curves for data obtained with pyruvate/malate and a-ketoglutarate were calculated by a least squares non-linear regression of the data fitted to the following equation assuming a constant absolute error: vmaXEPi:I K150 + [Pi] where b is the y-intercept (respiration at 0 Pi) and KISO is the concentration of Pi which gives half maximal inhibition of state 4 respiration. 265 \ 40 .v 9...?! L - . 0 0 0 8 6 2 140 120 IOO , 7188 So .855 .758 .0 606.9. 8:282 8 saw 0 70 80 90 I00 60 50 30 20 [P,] , mM 266 are used as substrates (Fig 3). These data indicate that in the absence of ADP certain enzymes in, or coupled to, the tricarboxylic acid cycle can be rate-limiting for respiration. III. ACTIVATION OF SUBSTRATE DEHYDROGENASE ACTIVITIES BY P1 In an effort to probe possible sites of interaction for P more 1 directly, disrupted mitochondria were titrated with P1 in the presence of a variety of substrates. Rates of NADH production were inferred from the rate at which lipoamide dehydrogenase transferred reducing equiva- lents from pyridine nucleotides to the redox-sensitive dye INT. HEPES was used as the buffer in this assay system because Tris interferes with the transfer efficiency (Fig 4A). The efficiency of transferring reducing equivalents from NADH to INT is 100% (i.e., the slope of a plot of nmoles INT reduced measured as a function of nmoles of NADH added is 1.0) when the assay medium is buffered with 20 mM HEPES. In contrast, the effici- ency drops to only 67% when the system is buffered with Tris. As shown in Fig 4B, the transfer efficiency remains uncompromised regardless of HEPES-P1 concentration. Thus, for the following experiments, mitochon- drial matrix substrate dehydrogenases were titrated with HEPES-Pi. The kinetic constants obtained for the substrate-dependent activa- tion of dehydrogenase activity by P1 are summarized in Table 2. ADP (1 mM) was included in the assay medium in order to simulate state 3 con- ditions. When glutamate or a-ketoglutarate are used as substrates (Fig 5), Pi stimulates NADH production in a hyperbolic manner. With a much lower ADP concentration (40 pM) the activation by P1 remains hyperbolic. In agreement with the data of McCormack and Denton (1986) on other mito- chondria, Ca(II) also stimulates the a-ketoglutarate dehydrogenase acti- 267 Figure 4. Efficacy of HEPES in Buffering the INT Reduction Assanyystem (A) Effect of buffers on the efficiency with which lipoamide dehdyrogenase transfers reducing equivalents from NADH to INT. Calibrated amounts of NADH were added to a solution of 0.6 mM INT buffered with either 20 mM HEPES ((3) or 20 mM Tris (O). The final absorbance was measured and the amount of reduced INT formed was calculated. The lines through the data were calculated by least-squares linear regression. The lepes of these lines were 1.02 and 0.67 for the assays performed in HEPES and Tris, respectively. (B) Effect of HEPES-Pi on the efficiency with which lipoamide dehydrogenase transfers reducing equivalents from a constant concentration of NADH to INT. 68 nmoles of NADH were added to 1.0 ml of a 0.6 mM solution of INT which contained the indicated concentrations of HEPES-P1. The final absorbance was measured and the amount of reduced INT formed was calcula- ted. The line was drawn through the mean value of the data. 268 .28 ._-1.m 183:“— Om O¢ Om ON m 1cm ox. 828 :22 was: CON OO. Om. OON 0mm slowu P930991 .LNI 269 Figure 5. Titration of Mitochondrial Dehydrogenase Activity with Pi using either a-Ketoglutarate or Glutamate as Substrates The rate of NADH production was inferred from the rate at which INT was reduced, as described under "Experimental Procedures." The assays were performed in the presence of either 5 mM a-keto- glutarate (.) or 5 mM glutamate (O) using broken chick heart mitochondria suspended at 0.15 mg/ml. For each point, the rate of NADH production measured in the absence of P1 was subtracted from the rate measured after the addition of Pi' The curves drawn through the data are theoretical, calculated with the Km and Vmax values estimated by a non-linear regression of the data fitted to the Michaelis-Menten equation. 270 13:95:83.2 9A.-15E .uoosuom ._.2_ SEE: 271 TABLE II APPARENT KINETIC CONSTANTS FOR THE STIMULATION OF CHICK HEART MITOCHONDRIAL MATRIX DEHYDROGENASES BY INORGANIC PHOSPHATEa CARBOXYLIC ACID(S) NDP V m max mM mM Pyruvate/2.5 mM Malate ADP 75.9 t 24.4 468 t 110 mM Glutamate ADP 2.0 t 0.29 22.6 i 0.86 mM a-Ketoglutarate ADP 2.6 1 0.37 39.0 t 1.5 mM a-Ketoglutarate GDP 4.7 t 0.68 37.6 i 1.9 mM a-Ketoglutarate IDP 7.4 i 1.4 34.7 i 2.4 The stimulation of the reduction of INT by P1 was monitored using the method of Hinman and Blass (1981), as described under "Experimental Procedures." The rate of reduced INT production in the presence of substrate and nucleo- tide diphosphate was subtracted from all rates observed after the addition of Pi“ NDP, nucleotide diphosphate. The designated NDPs were present at 1 mM during titration of the enzyme system with inorganic phosphate. Maximal rates are expressed as nmoles of INT reduced/min/(mg mitochondria). 272 vity in chick heart mitochondria (data not shown). A further question arises: Does Ca(II) alter the interaction of P with the dehydrogenase i for a-ketoglutarate? In the absence of EGTA, Pi stimulates a-ketoglutar- ate oxidation in a hyperbolic manner with a K = 1.12 i 0.16 mM (data 50 not shown). These data show that P does not bind cooperatively to a- i ketoglutarate dehydrogenase. The K for the stimulation of dehydrogenase 50 activity by P1 in the presence of either glutamate or a-ketoglutarate is approximately 2 mM. This is a physiologically attainable concentration of P When Pi titrations are performed using a-ketoglutarate as sub- 1. strate in the presence of IDP or GDP, the apparent kinetic parameters for activation are identical, within error, to those obtained in the presence of ADP (Table 2). The apparent K for P1 activation of NADH 50 production when pyruvate and malate are oxidized is 76 mM (Table 2). In addition, the Vmax for NADH production from the substrate couple pyru- vate/malate is approximately 10-fold higher than that observed with either a-ketoglutarate or glutamate. The following experiments were performed to localize the site at which Pi induces an increase in the rate of a-ketoglutarate oxidation. It was of interest to determine whether a-ketoglutarate was catabolized beyond succinyl 00A in these assays. In a reaction mixture containing all the standard components except for INT, formation of the CoA thioester was monitored at 230 nm (Bridger, Ramaley, and Boyer, 1969). While the absorbance of the sample increased at 340 nm (i.e., NADH was being produced), no change in absorbance at 230 nm was observed (data not shown). These results indicate that although a-ketoglutarate was being oxidized, succinyl COA was not accumulating. Because the concen- tration of succinyl CoA appears to remain constant, the CoA available to Figure 6. 273 Titration of Mitochondrial Dehydrogenase Activity with Ca(II) using B-Hydroxybutyrate as Substrate Rates of NADH production were inferred from the rate at which INT was reduced in the assay medium described under "Experimental Procedures." Mitochondria were suspended at 0.16 mg/ml of the assay medium which contained, in addition to the usual constituents, 1 mM ADP, 5 mM B-hydroxybutyrate, and 20 mM Hepes-Pi, and were titrated with the indicated concentrations of Ca(II). The assay medium did not contain EGTA. The rates shown represent rates of NADH production stimulated by Ca(II). 274 - 5 2 Lotncocoozfi oEv ...-c_E. noosuom ._.ZH 86E: 0 2 IO.- 0 [Ca(II)],mM 275 the system will cycle (see Fig 8 for schematic depiction). This is confirmed by the observation that even when NADH production from a- ketoglutarate is monitored in the absence of exogenous COA, the rate of INT reduction remains constant for the entire time course over which the assay is monitored (> 5 min). Thus, even in disrupted mitochondria, a- ketoglutarate dehydrogenase and succinyl thiokinase are still highly coupled. Therefore, one Obvious explanation for the Pi requirement during a-ketoglutarate oxidation is that succinate thiokinase must have a source of this anion if substrate-level phosphorylation is to proceed. In support of this hypothesis are the apparent affinities of the various nucleotide diphosphates for the reaction activated by P1 (Table 3). The K of the reaction for GDP is 4 pM, in excellent agreement with the K 50 of pig heart succinyl thiokinase (3 uM; Cha and Parks, 1964). It is 50 known that this enzyme has a K for IDP that is very similar to that 50 for GDP (Mazumder et al., 1960). The data summarized in Table 3 are con- sistent with this. The apparent affinity of ADP for the Pi activated reaction is 10-fold lower. It is possible that this K represents the 50 affinity of nucleoside diphosphokinase for ADP. In support of this suggestion, most of the nucleoside diphosphokinases yet studied have Michaelis constants for ADP ranging from 40-50 pM (review: Parks and Agarwal, 1973). Activation by ADP is to be expected since the trans- phosphorylation reaction will insure that GDP is rapidly recycled within the system (see Fig. 8). Pi has no effect on the rate of NADH production when B-hydroxybuty- rate is oxidized (data not shown). This result is not consistent with the hypothesis presented above which assumes that P interacts with B- i hydroxybutyrate dehydrogenase. In order to examine this problem in a b 276 TABLE III APPARENT Km's FOR THE STIMULATION 0F CHICK HEART MITOCHONDRIAL NADH PRODUCTION BY NUCLEOTIDE DIPHOSPHATES WHEN a-KETOGLUTARATE IS OXIDIZED NDPb K m uM ADP 41.5 i 3.4 GDP 4.0 i 1.8 IDP 1.9 i 0.9 The stimulation of the reduction of INT by the various NDP's was monitored using the method of Hinman and Blass (1981), as described under "Experimental Procedures." The rate of reduced INT production in the presence of 5 mM a-ketoglutarate and 20 mM HEPES-phosphate was subtracted from all rates observed after the addition of the respective NDP. NDP denotes the nucleoside diphosphate with which the system was titrated. 277 greater detail, it was reasoned that Pi may not stimulate B-hydroxybuty- rate dehydrogenase directly. Rather, P may stimulate this enzyme by i activating another enzyme that is closely coupled to it. One candidate for this enzyme would be a-ketoglutarate dehydrogenase. Succinyl CoA is a necessary co-substrate in the formation of acetoacetyl CoA via B-oxo- acid CoA transferase (Fig 8). As shown in Fig 5, Pi limits the oxidation of a-ketoglutarate. At low Pi concentrations, only small amounts of succinyl CoA can be produced. It is possible that this would limit the rate of B-hydroxybutyrate dehydrogenase due to the accumulation of acetoacetate. Therefore, another Pi titration was performed in the presence of 5 mM B-hydroxybutyrate and 5 mM malate. By including malate in the assay medium, the acetyl units produced from acetoacetyl COA could be used to produce citrate, and this in turn could be oxidized to succinyl CoA. In the foregoing it was assumed that some succinyl CoA endogenous to the mitochondria would be available to initiate the transferase reaction. Pi once again had no effect on the rate of NADH production (data not shown). In the course of these experiments, it was found that B-hydroxybutyrate could be oxidized at a faster rate when EGTA was excluded from the assay medium. A titration with Ca(II) shows that this cation stimulates the rate at which NADH is produced from B- hydroxybutyrate (Fig 6). The possibility that activation by P1 is a Ca(II)-dependent process was explored. Broken chick heart mitochondria oxidizing B-hydroxybutyrate in the presence of 1 mM Ca(II) were titrated once again with Pi' No effect by P1 could be discerned (data not shown). IV. INHIBITION OF SUBSTRATE DEHYDROGENASE ACTIVITIES WITH Pi 278 The method of Hinman and Blass (1981) was also used to monitor NADH production during the oxidation of pyruvate and a-ketoglutarate in the absence of ADP (i.e., state 4). In agreement with the respiration experi- ments (Fig 3), increasing concentrations of Pi inhibit the the rate at which the dehydrogenases for these substrates produce NADH (Fig 7). Much higher concentrations of Pi must be used in order to obtain half-maximal inhibition with the dehydrogenases contained in the preparation of dis- rupted mitochondria. This probably reflects (a) changes in the chemical and physical environment of the enzymes subsequent to lysis of the mito- chondria, or (b) that the concentration of Pi inside the matrix of intact mitochondria is greater than that in the extramitochondrial space. 279 Figure 7. Titration of Mitochondrial Dehydrogenase Activity with Pi in the Absence of ADP using Pyruvate/Malate and a-Ketoglutarate as Substrates Rates of NADH production were measured as described in the legend to Figure 6. The disrupted mitochondria were incubated in the assay medium described under "Experi- mental Procedures," except it did not contain ADP. The mitochondria were suspended at 0.35 and 0.41 mg/ml when oxidizing 5 mM pyruvate/2.5 mM malate (.) and 5 mM a-ketoglutarate ((:)), respectively. 280 L 258525 oEV . TEE . noosnom 60 50 3O 40 [i],mM 20 1 l0 ./ O .nlu 0 -_.2H 86E: Figure 8. 10. 11. 12. 13. 14. 281 Integrative Scheme Depicting the Reactions in and Coupled to the Tricarboxylic Acid Cycle The numbers between intermediates are used to designate the identity of the following enzymes: Pyruvate Dehydrogenase Complex Citrate Synthase Aconitate Hydratase Isocitrate Dehydrogenase Glutamate Dehydrogenase a-Ketoglutarate Dehydrogenase Complex Succinyl-GOA Synthetase Nucleoside Diphosphokinase Succinate Dehydrogenase Fumarate Hydratase Malate Dehydrogenase B-Hydroxybutyrate Dehydrogenase B-Oxoacid CoA-Transferase Acetoacetyl-COA Thiolase 282 %© NADHoH‘ TRICARBOXYLIC H 1 ACID 283 DISCUSSION I. PHOSPHATE CONTROL OF DEHYDROGENASES DURING STATE 3 RESPIRATION The oxygen consumption experiments reported in this study indicate that the regulation of state 3 rates of respiration by P is complex in i the sense that a number of binding sites for this anion are implicated. The sigmoidal and biphasic saturation kinetics observed during the oxi- dation of a-ketoglutarate and B-hydroxybutyrate, respectively, suggested that Pi is a direct effector of the dehydrogenases for these substrates. Measurement of the rates of NADH production by these dehydrogenases do not, however, support this hypothesis. In broken mitochondria, P control over rates of a-ketoglutarate 1 oxidation is due to the Pi requirement of the substrate-level phosphory- lation catalyzed by succinyl CoA synthetase. The rate of glutamate oxi- dation is stimulated by P1 because a-ketoglutarate is the product of the glutamate dehydrogenase reaction. Thus, for this substrate, P1 simply stimulates catabolism beyond a-ketoglutarate. The sigmoidal saturation kinetics of state 3 respiration during a-ketoglutarate oxidation does not arise from cooperative binding of P to the dehydrogenase for this i substrate. Instead, it is likely that this sigmoidicity arises from the substrate requirements for a-ketoglutarate transport across the inner membrane. The dicarboxylate carrier can catalyze the antiport of such dicarboxylic acids as succinate, malate, and a-ketoglutarate (Palmieri et al., 1971). In addition, this carrier can also catalyze the import of dicarboxylates via exchange with matrix Pi' Since there would not be 284 sufficient concentrations of endogenous dicarboxylic acids to drive the import of large amounts of exogenous a-ketoglutarate, the sigmoidal saturation of state 3 respiration is a probable result of P limiting i the rate at which a-ketoglutarate can be transported into the mito- chondrial matrix via the dicarboxylate carrier. Seen in this light, the discrepancy between the Kms for P1 activation of a-ketoglutarate oxida- tion (1.2 mM) and state 3 respiration when a-ketoglutarate is used as substrate (4.7 mM) reflects differences in the affinity of succinyl COA synthetase and the dicarboxylate carrier for Pi' The manner by which P stimulates state 3 respiration during B-hyd- i roxybutyrate oxidation is not apparent. B-Hydroxybutyrate dehydrogenase exists as a dimer and is associated with the matrix side of the mitochon- drial inner membrane (McIntyre §£_pl., 1978). The oxygen consumption data suggest that this enzyme experiences a complex biphasic interaction with P1 during state 3 respiration. Direct measurement of rates of NADH production by matrix enzymes in the presence of B-hydroxybutyrate revealed no stimulation by P It is possible that when the mitochondria 1. were disrupted, B-hydroxybutyrate dehydrogenase was irreversibly damaged. It is also possible that by removing the enzyme from its native environ- ment (i.e., disrupting mitochondrial membranes and destroying the envir- onment unique to the matrix), all sensitivity to P1 is lost. Because this substrate is transported into the matrix by the pyruvate carrier (Parvin and Pande, 1979), it is unlikely that the biphasic activation of respiration by P1 is a transport phenomenon since (a) this carrier does not transport Pi and (b) it does not diplay biphasic saturation kinetics. At the least, these data show that state 3 respiration driven by B-hyd- roxybutyrate displays two separate affinities for Pi that are significan- 285 tly different from the Pi affinities of the F1FO-ATP synthetase and succ- inyl COA synthetase. Whether the two binding sites for P1 exist on the dehydrogenase for B-hydroxybutyrate cannot be discerned from these data. The P1 saturation kinetics of state 3 respiration when pyruvate/malate and glutamate/malate are used as substrates indicate that this anion is largely committed to ATP synthesis. Pi stimulates NADH production in the presence of pyruvate and malate with a high Km (76 mM). This is physiolog- ically untenable and may reflect an ionic strength effect. Pi does stimu- late NADH production when glutamate and malate are used as substrates. However, one molecule of NADH is produced during the glutamate dehydrogen- ase reaction. Thus, even if a-ketoglutarate oxidation is rate-limited by Pi’ it is apparent that NADH concentrations sufficient to drive oxidative phosphorylation are produced by glutamate dehydrogenase within intact mitochondria. Finally, the nearly identical Vmax values for oxygen consumption at infinite P and high substrate concentrations suggest i that the rate of oxidative phosphorylation is limited by a reaction common to all of the substrates. II. PHOSPHATE CONTROL OF DEHYDROGENASES DURING STATE 4 RESPIRATION Based on the data shown in Fig's 3 and 7, the dehydrogenases for pyruvate and a-ketoglutarate are inhibited by P1 during state 4. As evidenced by its insensitivity to inhibition by oligomycin, all of the state 4 respiration in these experiments was due to uncoupled respiration. It is unlikely that Pi inhibits enzymes of the electron transfer chain under these conditions. All of the substrates tested produce NADH. If Pi acted at the level of NADH dehydrogenase (complex I) or the cytochromes, one would expect uniform inhibition of respiration independent of the 286 source of reducing equivalents. This is not observed. These data indicate that in the absence of ADP the enzymes catalyzing two of the rate-limiting steps in or coupled to the tricarboxylic acid cycle can regulate the rate of respiration. As Pi concentrations are increased during the oxidation of either pyruvate or a-ketoglutarate, the rate of oxygen consumption during state 4 approaches zero. This is in sharp contrast to what one would expect if the near-equilibrium hypothesis (Wilson g£_§l., 1974; Erecinska p£_pl., 1977) for the control of respira- tion were true for chick heart mitochondria. As Pi concentrations are increased without compensatory alterations in extramitochondrial ATP concentrations, this theory suggests that the rate of respiration should increase since the magnitude of the cytosolic phosphorylation potential will decrease. The data reported herein support the notion advanced in recent years (Hansford, 1977, 1980, 1985; Hansford and Castro, 1981) that under some conditions the matrix dehydrogenases catalyzing non- equilibrium reactions in the matrix control the rate of respiration. The inhibition of a significant percentage of uncoupled respiration by P1 during the oxidation of pyruvate/malate and a-ketoglutarate is attainable by physiological concentrations of P . Consequently, as cytosolic ADP i concentrations change during variations in cardiac contraction rates, Pi may play a significant role in the transient regulation of dehydrogenase activity in the mitochondrial matrix. The actual mechanism by which this inhibition of the dehydrogenases for pyruvate and a-ketoglutarate is effected is yet to be elucidated. However, it may be stated that these data support a model in which Pi exerts kinetic rather than thermodynamic control over the rate of respiration. In addition, these data show that Pi can be used as a non-toxic inhibitor of uncoupled 287 respiration. Thus, at the cmc, when Pi concentrations are optimized, the mitochondria become highly coupled and respire with respiratory control ratios that approach infinity. III. CONCLUSIONS The experiments reported in this study show that: (a) The P1 satura- tion kinetics of state 3 respiration vary with the nature of the carbox- ylic acid(s) used as oxidizable substrate; (b) The control of state 3 respiration by P1 is distributed among a number of enzymes: the F1FO-ATP synthetase, succinyl COA synthetase, and possibly B-hydroxybutyrate dehy- drogenase as well as the dicarboxylate carrier during a-ketoglutarate oxidation; (c) The cytosolic phosphorylation potential does not regulate the rate of respiration; (a) P1 inhibits the oxidation of pyruvate and a-ketoglutarate in the absence of ADP; and (e) The data are consistent with the notion that particular mitochondrial matrix dehydrogenases participate in the phenomenon of respiratory control via a mechanism that is independent of changes in the ratios of NADH/NAD, ATP/ADP, and succinyl (or acetyl) COA/COA. 288 REFERENCES Bencini, D.A., Wild, J.R., and O'Donovan, G.A. (1983) Anal. Biochem. 132: 254-258. Bock, R.M., Ling, W.S., Morell, S.A., and Lipton, S.R. (1956) Arch. Biochem. Bigphys. 62: 253- Bridger, W.A., Ramaley, R.F., and Boyer, P.D. (1969) Methods Enzymol. 13: 70-75. Bridger, W.A. (1974) The Enzymes (3rd ed.) 10: 581-606. Brooks, S.P.J., and Suelter, C.H. (1987) Int. J. Bio-Med. Camp. 19: 89-99. Bulos, B.A., Thomas, B.J., Shukla, S.P., and Sacktor, B. (1984) Arch. Biochem. Biophys. 234: 382-393. Cha, 8., and Parks, B.E. (1964) J. Biol. Chem. 239: 1961-1967. Cha, S. (1969) Methods Enzymol. 13: 62-69. Chance, B., and Williams, J.R. (1955) J. Biol. Chem. 217: 383-393. Chance, B., Leigh, J.S., Clark, B.J., Maris, J., Kent, J., Nioka, S., and Smith, D. (1985) Proc. Nat. Acad. Sci. (U.S.A.) 82: 8384-8388. Chance, 8., Leigh, J.S., Kent, J., McCully, K., Nioka, S., Clark, B.J., Maris, J.M., and Graham, T. (1986) Proc. Nat. Acad. Sci. (U.S.A.) 83: 9458-9462. Coty, W.A., and Pedersen. P.L. (1974) J. Biol. Chem. 250: 2593-2598. Davis, E.J., and Lumeng, L. (1975) J. Biol. Chem. 250: 2275-2282. Denton, R.M., and McCormack, J.G. (1986) In Hormones and Cell Regulation (Nunez, J., et al., eds) Vol. 139, pp. 249-259, John Libbey Eurotext Ltd. Erecinska, M., Stubbs, M., Miyata, Y., Ditre, C.M., and Wilson, D.F. (1977) Biochim. Biophys. Acta 462: 20-35. Greenbaum, N.L., and Wilson, D.F. (1985) J. Biol. Chem. 260: 873-879. Hansford, R.G. (1976) J. Biol. Chem. 251: 5483-5489. Hansford, R.G. (1980) Current Topics in Bioenergetics 10: 217-278. 289 Hansford, R.G., and Castro, F. (1981) Biochem. J. 198: 525-533. Hansford, R.G. (1985) Rev. Physiol. Biochem. Pharmacol. 102: 1-72. Hinman, L.M., and Blass, J.P. (1981) J. Biol. Chem. 256: 6583-6586. Holian, A., Owen, 0.3., and Wilson, D.F. (1977) Arch. Biochem. Biophys. 181: 164-171. Klotz, I.M. (1946) Arch. Biochem. Biophys. 9: 109-117. Koshland, D.E., Nemethy, G., and Filmer, D. (1966) Biochemistry 5: 365-385. Kuster, U., Bohnensack, R., and Kunz, W. (1976) Biochim. Biophys. Acta 440: 391-402. Lemasters, J.J., and Sowers, A.E. (1979) J. Biol. Chem. 254: 1248-1251. Mannervik, B. (1982) Methods Enzymol. 87: 370-390. Matsuno-Yagi, A., and Hatefi, Y. (1986) J. Biol. Chem. 261: 14,031-14,038. Mazumder, R., Sanadi, D.R., and Rodwell, V.W. (1960) J. Biol. Chem. 235: 2546-2550. McIntyre, J.D., Bock, H.O., and Fleischer, S. (1978) Biochim. Biophys. Acta 513: 255-267. McCormack, J.G., and Denton, R.M. (1979) Biochem. J. 180: 533-544. Monod, J., Wyman, J., and Changeux, J.P. (1965) J. Mol. Biol. 12: 88-118. Nishimura, J.S., and Grinnell, F. (1972) Adv. Enzymol. 36: 183-202. Owens, T.G., and King, F.D. (1975) Mar. Biol. 30: 27-36. Palmieri, F., Prezioso, G., Quagliarello, E., and Klingenberg, M. (1971) Eur. J. Biochem. 22: 66-74. Parks, B.E., and Agarwal, R.P. (1973) The Enzymes (3rd ed) 8: 307-333. Parvin, R., and Pande, S.V. (1979) J. Biol. Chem. 254: 5423-5429. Rigoulet, M., Velours, J., and Guerin, B. (1985) Eur. J. Biochem. 153: 601-607. Slater, E.C., Rosing, J., and M01, A. (1973) Biochim. Biophys. Acta 292: 534-553. Toth, P.P., Ferguson-Miller, S.M., and Suelter, C.H. (1986) Methods Enzymol. 125: 16-25. 290 Westheimer, F.H. (1987) Science (Washington, D.C.) 235:1173-1178. Wilkinson, G.N. (1961) Biochem. J. 80: 324-332. Wilson, D.F., Stubbs, M., Veech, R.L., Erecinska, M., and Krebs, H.A. (1974) Biochem. J. 140: 57-64. Chapter 7 IDENTIFICATION OF A SOLUBLE FACTOR THAT UNCOUPLES THE OXIDATION OF SUBSTRATES FROM THE PHOSPHORYLATION OF ADP IN CHICK MYOCARDIUM 291 292 INTRODUCTION The mitochondria of most tissues have evolved both structurally and functionally in ways which maximize the potential of these organelles to commit the free energy of substrate oxidation toward the synthesis of ATP. An exception to this are the mitochondria of brown fat adipocytes. These mitochondria are able to release the free energy of fatty acid oxidation in the form of heat (Nicholls pp_§l., 1972; Cannon gp_§l., 1984). This nonshivering thermogenesis is induced by the activity of a 32’OOO-Mr uncoupling protein located in the mitochondrial inner membrane (review: Nicholls, 1979). The function of the uncoupler protein is to provide protons with a means of reentering the matrix without flowing through the F moiety of the ATP synthetase. The conductance of the 0 proton channel formed by this uncoupler protein is subject to regulation by purine nucleotide di- and triphosphates and by fatty acids (Nicholls, 1984; Klingenberg, 1984). The uncoupler protein from rat (Bouillaud pp gl., 1986) and hamster (Aquila gp_§l., 1985) brown fat has been isolated, purified, and its amino acid sequence determined. The uncoupling of substrate oxidation from the phosphorylation of ADP can also be achieved by subjecting mitochondria to lipOphilic protonOphores (Kell and Westerhoff, 1985), low oxygen tensions (Kramer and Pearlstein, 1983), and large Ca(II) loads (Lehninger g£_pl., 1967). Significantly, however, no protein has yet been found to uncouple mitochondria from tissues other than mammalian brown fat. The purpose of this brief report is to present prliminary evidence 293 for the existence of a soluble uncoupler protein in chick myocardium. The protein is small (515 kD), binds with high apparent affinity to chick heart mitochondria, and its activity is insensitive to oligomycin and EGTA, agents that would inhibit respiration attributable to the phosphorylation of ADP or the uptake of Ca(II), respectively. Phenom- enologically, the protein exists; however, the teleological basis for its existence is not clear. As shown in this dissertation, isolated chick heart mitochondria are extraordinarily highly coupled. It is apparent from the work to be described herein that, due to the presence of this uncoupling protein, it is unlikely this degree of coupling is actually available to myocardium in vivo. One possible explanation for the existence of this protein is that its controlled expression may regulate rates of cardiac thermogenesis. In contrast, its function could be related to a range of novel kinetic and/or thermodynamic mechanisms regulating the complex processes responsible for cardiac energy transduction. I 294 EXPERIMENTAL PROCEDURES I. MATERIALS Water was purified as described in Chapter 3. The following sub- stances were reagent grade or better, used without further purification, and obtained from the sources noted: essentially fatty acid free BSA (fraction V), EGTA, glycine, mannitol, sucrose, DL-dithiothreitol, ammonium sulfate, bromophenol blue, rotenone, oligomycin A, Dowex MR-3, Tris, collagenase (type VII), Nagarse, (type VII), and the sodium salts of ADP and pyruvic and malic acids (Sigma, St. Louis, MO); urea, glycerol, glacial acetic acid, silver nitrate, formaldehyde, and phosphoric acid (Mallinckrodt, Paris, KY); acrylamide (Serva, West Germany); N,N'-methyl- enebisacrylamide (Miles Laboratories); TEMED (Merck Laboratories); sodium dodecyl sulfate (Matheson Coleman and Belle); absolute methanol and ammonium hydroxide (J.T. Baker Chemical). Single comb white leghorn chicks were obtained from the Department of Animal Science at Michigan State University. Ammonium sulfate (enzyme grade, Mallinckrodt) was recrystallyzed from an aqueous 1 mM EGTA (pH 7.4) solution. Dialysis tubing (3500 molecular weight cutoff, Spectrapore) was soaked for 1 hr at 60° in a solution that contained 1% (w/v) ammonium carbonate and 0.1% (w/v) NaDodSOu. The soaked dialysis tubing was subsequently rinsed exhaustively with ultra-pure water. II. ISOLATION OF MITOCHONDRIA Mitochondria were isolated from chick heart muscle using either collagenase (Toth et al., 1986) or Nagarse (Chance and Hagihara, 1963). 295 These methods are described in detail in Chapters 2 and 4, respectively. III. COLLECTION OF CRUDE FRACTION CONTAINING THE UNCOUPLER PROTEIN A crude fraction of the uncoupler protein is prepared from the supernatant liquid overlying the first mitochondrial pellet obtained after centrifuging these organelles at 8,000g for 10 min. Henceforth, this supernatant liquid will be referred to as "S8000." To remove particulate matter, 88000 is filtered through a double layer of cheesecloth and centrifuged at 40,000g for 20 min. The S8000 is once again filtered through a double layer of cheesecloth and loaded into dialysis tubing. The S8000 is then concentrated by placing the dialysis tubing over cheesecloth and covering it with crystalline sucrose for 12 hrs at 4°C. The resulting concentrate is centrifuged at 10,000g for 20min. The clear supernatant liquid is collected and stored at 4°C. The uncoupling activity of this concentrated S8000 (088000) is stable for at least two weeks. IV. STABILIZATION OF CONCENTRATED 88000 IN GLYCEROL Because sucrose promotes rapid bacterial growth in the stored S8000, another agent that stabilizes the uncoupling activity was sought. In order to test the stability of the coupling activity in glycerol, aliquots of cS8000 were dialyzed against 1.5 L of the following buffers for 21 hrs at 4°C: (a) 10% glycerol, 20 mM Tris, 2 mM EGTA, pH 7.4; (b) 25% glycerol, 20 mM Tris, 2 mM EGTA, pH 7.4; (c) 50% glycerol, 20 Mm Tris, 2 mM EGTA, pH 7.4. After the dialysis step, these c88000 aliquots were assayed for uncoupling activity. 296 V. AMMONIUM SULFATE PRECIPITATION OF THE UNCOUPLING ACTIVITY IN S8000 In an attempt to partially purify the uncoupling activity, the cS8000 was subjected to a series of ammonium sulfate precipitation steps. 5 ml of the 088000 was dialyzed at 4°C against 1.5 liters of a buffer that contained 50% glycerol (v/v), 20 mM Tris, pH 7.4, and 1 mM EGTA (GTE). This buffer was adjusted to either 50, 60, 70, 80, or 90% saturation with ammonium sulfate. After the various ammonium sulfate cuts, samples were centrifuged at 40,000g for 15 min. The supernatant liquids were withdrawn, and precipitates were redissolved in minimal volumes of GTE. In order to remove the ammonium sulfate, the precipi- tates and the supernatant liquids were dialyzed for 12 hrs at 4°C against 1.5 liters of a buffer that contained 50% glycerol (v/v), 20 mM Tris, pH 7.4, and 2 mM EGTA. The dialyzed samples were then concentrated against solid crystalline sucrose as described above. VI. ASSAY FOR UNCOUPLING ACTIVITY The activity of this protein was measured by determining its effect on the state 4 respiration of intact chick heart mitochondria isolated with collagenase. One unit of uncoupling activity is that amount of crude protein/mg mitochondria that stimulates mitochondrial oxygen con- sumption by 100 ng atom O (min)-i (mg mitochondria)-1. Oxygen consumption experiments were performed as described in Chapter 3. The assay medium was comprised of 0.225 M mannitol, 0.075 M sucrose, 20 mM Tris-P1, pH 7.4, and 5 mM pyruvate/2.5 mM malate. Other experimental conditions were as described in the legend to Fig 1. VII. POLYACRYLAMIDE GEL ELECTROPHORESIS 297 The protein banding profiles of 088000 and of mitochondria isolated with either collagenase or Nagarse were compared using NaDodSOu-polyacry- lamide gel electrophoresis. Electrophoresis was performed using the dis- continuous buffer system of Laemmli (1970) in slab gels (16cm x 18cm x 0.75 mm). The resolving gel contained 15% (w/v) acrylamide/0.40% bisacry- lamide and 3.75 M urea. The stacking gel (typically 1-1.5 cm) contained 3% acrylamide/0.08% bisacrylamide and 0.95 M urea. Prior to mixing, acrylamide solutions were swirled over a bed of Dowex MR-3 ion exchange resin (Sigma) in order to remove contaminating acrylic acid. Samples were solubilized in boiling sample buffer (0.125 M Tris, pH 6.8, 4% (w/v) NaDodSOu (w/v), 20% (v/v) glycerol, 0.1 M DL-dithiothreitol, 3.75 M urea, and 0.002% (w/v) bromophenol blue) and incubated in a boiling water bath for 5 min. It was important when working with mitochondria isolated with Nagarse that boiling sample buffer be used; if not, residual Nagarse that could not be removed from mitochondria by rinsing would completely hydrolyze the solubilized protein. The electrophoresis apparatus was not cooled with tap water during the course of a run because the urea crystallizes out of the gel. Gels were run for 1.5 hrs at 100 V, and then for 10-12 hrs at 300 V. The protein standards were those obtained in the Sigma Dalton Mark VI kit: bovine serum albumin (66,000); ovalbumin (45,000); Porcine pepsin (34,700); bovine trypsinogen (24,000); bovine B-lactoglobulin (18,400); and lysozyme (14,300). Gels were fixed and then stained with silver as described by Giulian et al. (1983). VII. OTHER ASSAYS Protein concentrations were determined by the modified Lowry method 298 of Markwell p£_pl. (1981) using BSA as standard. Pi concentrations were assayed using the method of Bencini §£_pl. (1983) using potassium orthOphosphate as standard. Cytochrome oxidase was quantitated as described in Chapter 2. 299 RESULTS I. STABILITY OF UNCOUPLER ACTIVITY The uncoupler activity of the supernatant liquid overlying the first mitochondrial pellet is unstable in the 0.225 M mannitol/0.075 M sucrose medium used to isolate mitochondria. Within 24 hrs little if any activity remains. In the 10 and 25% glycerol buffers, most Of the uncoupling activity is lost within 2-3 days. However, if the 088000 is dissolved in either the 50% glycerol buffer or in concentrated sucrose, the uncoupling activity is stable for up to two weeks. Longer storage periods were not attempted. Consequently, the maximal storage time could well be in excess of two weeks. II. STIMULATION OF RESPIRATION The addition of an aliquot of 888000 stimulates uncoupled respira- tion in chick heart mitochondria (Fig 1). A number of observations strongly suggest that the stimulation of oxygen consumption is indeed attributable to an uncoupler protein. First, the 088000 stimulates mitochondrial respiration in the presence of a concentration of oligo- mycin sufficient to completely inhibit the stimulation of respiration by ADP. Therefore, the stimulation of respiration is not due to the activity of an ATPase. Second, respiration is not being stimulated by the addition of contaminating Ca(II), since the rate of uncoupled respiration stimu- lated by 888000 is unaffected by the presence of 10 mM EGTA. Third, the uncoupling activity is completely abolished by treating S8000 with the nonspecific protease Nagarse or by incubating this crude protein mixture 300 Figure 1. The Stimulation of Chick Heart Mitochondrial Oxygen Consumption by 088000 Mitochondria isolated with collagenase were suspended in 1.75 ml of an assay medium that contained 0.225 M mannitol, 0.075 M sucrose, 10 mM EGTA, and 20 mM Tris- buffered Pi' pH 7.4. Oxygen consumption was measured with a Clark electrode at 30.5°C. The following addit- ions to the reaction mixture were made: M, mitochondria (0.091 nmoles cyt aa /ml); S, substrate (5mM pyruvate/ 3 2.5 mM malate); ADP, 400 nmoles per addition; 0, olig- omycin A, 66 pg/nmole of cyt aa ; cS8000, crude mixture 3 of protein containing the uncoupling factor, 3.4 mg/ml; and rotenone, 10 pM. 301 Mito. Olig. _j 1 088000 f 64 nmol 02 Roten one A T2 min 302 in a boiling water bath for 5 min. Titrating chick heart mitochondria with increasing amounts of 088000 suggests that the uncoupler protein has an apparent high affinity for these organelles (Fig 2). From this plot it is discerned that one unit of uncoupling activity corresponds to 0.35 mg crude protein/mg mitochon- dria. Maximal uncoupling is obtained subsequent to the addition of 9.1 units of crude protein/mg mitochondria. The fact that the stimulation of oxygen consumption is linear with rapid saturation suggests that this protein binds to mitochondria stoichiometrically. In the experiment shown, cS8000 at saturating concentrations stimulates oxygen consumption at a rate equal to 70% of the Vmax for state 3 respiration. III. AMMONIUM SULFATE PRECIPITATION OF THE UNCOUPLER PROTEIN Additional evidence suggesting that the uncoupling factor is a protein is that it can be salted-out of solution with ammonium sulfate (Fig 3). The uncoupler begins to precipitate above 60% saturation with ammonium sulfate, as indicated by a steadily decreasing uncoupling activity in the GTE buffer. Importantly, the uncoupling activity lost during precipitation can be recovered by removing ammonium sulfate and resuspending the precipitate in GTE buffer. IV. MOLECULAR WEIGHT ESTIMATE FOR THE UNCOUPLER PROTEIN Experiments in which dialysis tubing of different pore sizes were used indicated that the uncoupler protein has an apparent molecular weight or 5'15 x 103 (i.e., if dialysis tubing with a molecular weight cutoff greater than 15,000 was used, the uncoupling activity was lost to the dialyzing buffer). Treatment of chick heart mitochondria with 303 Figure 2. Effect of Increasing Concentrations of 088000 on the Rate of Oxygen Consumption by Chick Heart Mitochondria Assay conditions were as described in the legend to Figure 1. 088000 was prepared as described under "Experimental Procedures." % Vmax Of State 3 Respiration 80 70 60 50 4O 30 304 l l l l l L I 0 LC 2.0 3.0 4.0 5.0 6.0 7. mg Crude Protein/mg Mitochondria 305 Figure 3. Activity of the Uncoupler Protein in the Ammonium Sulfate Cuts of 088000 Aliquots of the supernatant liquids (.) and the precipi- tates ((:)) resulting from treatment of 088000 with the indicated concentrations of ammonium sulfate were added to 0.38 mg mitochondria/ml. Oxygen consumption assays were performed as described in the legend to Figure 1. 306 80 90 7O °/o Ammonium Sulfate Saturation 60 50 . _._ r 73:2 oEEEEi O O O O 3 2 I 0020 oEv. ...EE .0 E2019. 307 Nagarse produces a substantial increase in their respiratory control ratio (see Chapter 3). It was of interest to determine whether Nagarse destoys a protein with a molecular weight in the range estimated by the dialysis tubing experiments. Mitochondria isolated with either Nagarse or collagenase and 088000 were loaded into separate lanes and electro- phoresed through a denaturing polyacrylamide gel. Densitometric mea- surements reveal that over 50 protein bands are obtained in the lane that contained mitochondria isolated with collagenase (data not shown). After comparing the banding patterns in the three lanes, a band with an apparent Mr of tu,300 found in the lanes for cSBOOO and collagenase mitochondria is missing from the lane containing mitochondria isolated with Nagarse (data not shown). The two experiments are thus consistent. However, much work must still be done in order to establish the identity of the 1M.3 kD band that is missing. It is possible that it is the uncoupling protein. At the least, these experiments provide a reasonably good candidate with which to begin the search for the uncoupling protein. 308 DISCUSSION The data reported herein provide strong suggestive evidence for the existence of a relatively small soluble protein that uncouples mitochon- drial electron transport from the synthesis of ATP in chick heart muscle. Because it is soluble and is found in a tissue other than mammalian brown fat, it is the first such protein to be discovered. These data establish that the uncoupling factor is a protein since it can be inactivated by treatment with heat and protease digestion and it can be precipitated by ammonium sulfate. The activity of the protein can be stabilized in highly viscous solvents containing either sucrose or glycerol. It is also shown that the stimulation of respiration is not attributable to Ca(II) or ADP contaminants in the 088000. Future work will be aimed at purifying the uncoupler protein and characterizing its structure, function, and the mechanism by which it stimulates electron transport. The sensitivity of this protein to Nagarse indicates that it binds to mitochondria on the outer membrane. Questions concerning whether the protein binds to porin or some other receptor, and the nature of the regulatory mechanisms (if any) regulating its activity and the strength of its binding to mitochondria will all be of consi- derable interest to elucidate. At this point one can only speculate about the function of this protein in vivo. Chick myocardium has evolved a cytosolic factor that appears to compromise the intrinsic high degree of coupling of the mitochondria which it contains. This is somewhat astonishing for two 309 reasons. First, a fundamental struggle in nature has always seemed to center around the develOpment of more and more efficient means of procuring and distributing energy within an organism. Second, heart muscle mitochondria have long been assumed to be particularly specia- lized for ATP synthesis (i.e., other mitochondrial functions such as urea metabolism and erythropoietic heme biosynthesis are performed in non-cardiac tissues). The uncoupler protein provides myocardium with a molecule capable of inducing mitochondria to increase the rate of respiration independent of changes in cytosolic ADP or Ca(II) concentrations. Instead of ATP, heat presumably will be produced. At this point it is not possible to say why or under what conditions heart muscle would uncouple its mitochondria, but it is important to know that it can and, based on findings with isolated heart mitochon- dria, does. Further work on this protein will provide novel insight into the means by which heart muscle regulates mitochondrial coupling and respiratory control. It will also be of interest to determine whether this protein plays a role in any forms of cardiac dysfunction. For instance, if a patient suffers from cardiac ischemia, an uncoupling protein, if it exists in the human heart, will only exacerbate the oxygen deficit. On the other hand, the controlled clinical use of an organic uncoupler could be beneficial, since it may well promote rapid weight loss if it can be used to uncouple the mitochondria of adipose tissue in obese persons. 310 REFERENCES Aquila, H., Link, T.A., and Klingenberg, M. (1985) EMBO J. 4: 2369-2376. Bencini, D.A., Wild, J.R., and O'Donovan, G.A. (1983) Anal. Biochem. 132: 25N-258. ' Bouillaud, F., Weissenbach, J., and Ricquier, D. (1986) J. Biol. Chem. 261: 1N87‘1H90. ‘ Cannon, E., Bernson, V.S., and Nedergaard, J. (198M) Biochim. Biophys. Acta 766: N83-A91. ' Guilian, G.G., Moss, R.L., and Greaser, M. (1983) Anal. Biochem. 129: 277-287. ' ‘ Kell, D.B., and Westerhoff, H.V. (1985) In Oragnized Mutlienzyme Complexes: Catalytic Properties (Welch, R., ed.), pp. 63-139. Academic Press, New York. ' ‘ Klingenberg, M. (198A) Biochem. Soc. Trans. 12: 390-393. Kramer, R.S., and Pearlstein, R.D. (1983) Proc. Nat. Acad. Sci. (U.S.A.) 80: 5807-5811. Laemmli, U.K. (1970) Nature (London) 227: 680-685. Markwell, M.A.K., Haas, S.M., Tolbert, N.E., and Bieber, L.L. (1981) Methods Enzymol. 72: 296-303. ‘ Nicholls, D.C., Grav, H.J., and Lindberg, O. (1972) Eur. J. Biochem. Nicholls, D.C. (1979) Biochim. Biophys. Acta 5N9: 1-29. Nicholls, D.C., Snelling, R., and Rial, E. (198A) Biochem. Soc. Trans. 12: 388-390. ' APPENDIX 311 312 BIBLIOGRAPHY Published Papers 1. 10. Toth, P.P. (1981) Transfer ribonucleic acid tertiary structure stabilization by cation binding measured as a function of terbium fluorescence behavior. Undergraduate Thesis. 119 pp. Toth, P.P., Ferguson-Miller, S.M., and Suelter, C.H. (1986) The isolation of highly coupled heart mitochondria in high yield using a bacterial collagenase. Methods Enzymol. 125: 16-27. Lysiak, W., Toth, P.P., Suelter, C.H., and Bieber, L.L. (1986) Quantitation of the efflux of acylcarnitines from rat heart, brain, and liver mitochondria. J. Biol. Chem. 261: 13698-13703. Lysiak, W., Lilly, K., DiLisa, F., Toth, P.P., and Bieber, L.L. (1988) Quantitation of the effect of L-carnitine on the levels of acid soluble short chain acyl-COA and CoASH in rat heart and liver mitochondria. J. Biol. Chem. 263: 1151-1156. Lysiak, W., Lilly, K., Toth, P.P., and Bieber, L.L. (1988) Effect of the concentration of carnitine on acetylcarnitine production by rat heart mitochondria oxidizing pyruvate. Nutrition A: 214-219. Toth, P.P., Chance, 8., Sell, J.E., Holland, J.F., Ferguson-Miller, 8., and Suelter, C.H. (1988) The kinetics of inorganic phosphate uptake and utilization by chick heart mitochondria. In Integration of Mitochondrial Function (Lemasters, J.J., Hackenbrock, C.R., Thurman, R.G., and Westerhoff, H.V., eds). Plenum Press, New York. In press. ‘ ' Toth, P.P., Sell, J.E., Holland, J.F., Ferguson-Miller, S., and Suelter, C.H. (1988) Interaction of inorganic phosphate with chick heart mitochondria. I. Changes in volume and ion content of the matrix. To be submitted to J. Biol. Chem. Toth, P.P., Ferguson-Miller, S., and Suelter, C.H. (1988) Interaction of inorganic phosphate with chick heart mitochondria. II. Substrate-dependent modulation of state 3 and state A rates of respiration. To be submitted to J. Biol. Chem. Toth, P.P., Sumerix, K., Ferguson-Miller, S., and Suelter, C.H. (1988) Characterization of the factors affecting the respiratory cantrol and ADP:O coupling ratios of isolated chick heart mitochondria. To be submitted to J. Biol. Chem. Toth, P.P., Sell, J.E., Ferguson-Miller, S., Suelter, C.H., and Holland, J.F. (1988) Simultaneous measurement of absorbance, fluorescence, and light scattering intensities to determine the kinetic parameters of carrier-mediated transport processes in heart mitochondria. To be submitted to Anal. Biochem. 313 Abstracts 1. 10. 11. Toth, P.P., Ferguson-Miller, S.M., and Suelter, C.H. (198“) Heart mitochondria with high respiratory control isolated in high yield using collagenase. Fed Proc. “3(7): 2001. Toth, P.P., Ferguson-Miller, S.M., and Suelter, C.H. (1985) Characteristics of highly coupled heart mitochondria. Proc. 13th Intn'l Cong. Biochem., vol 5, p. 767. Lilly, K., Lysiak, W., Toth, P.P., and Bieber, L.L. (1986) Effect of carnitine concentration on the efflux of specific short-chain acylcarnitines from rat heart mitochondria. 16th Annual Michigan Cardiovascular Research Forum, Detroit, Michigan. Lysiak, W., Toth, P.P., Suelter, C.H., and Bieber, L.L. (1986) Quantitation of the efflux of short-chain acylcarnitines from rat mitochondria. In Clinical Aspects of Human Carnitine Deficiency (Borum, P., ed.), Pergamon Press, New York, p. “1. Toth, P.P., Ferguson-Miller, S.M., and Suelter, C.H. (1986) Coupling and respiratory control of chick heart mitochondria. Fed. Proc. “5(6): 1922. Lilly, K., Lysiak, W., Toth, P.P., and Bieber, L.L. (1987) Regulation of the acyl CoA/CoAsh ratio in rat heart and liver mitochondria by carnitine. Fed. Proc. “6(3): 1013. Toth, P.P., Ferguson-Miller, S.M., and Suelter, C.H. (1987) The kinetic parameters for phosphate saturation of chick heart mitochondrial state 3 respiration vary with the carboxylic acid substrates. Fed. Proc. “6(6): 1931. Toth, P.P., Chance, 8., Sell, J.E., Holland, J.F., Ferguson-Miller, S.M., and Suelter, C.H. (1987) Kinetics of phosphate uptake and utilization by chick heart mitochondria. International Conference on the Integration of Mitochondrial Function, “'7 June 1987. Chapel Hill, North Carolina. ‘ Toth, P.P., Chance, 8., Sell, J.E., Ferguson-Miller, S.M., Suelter, C.H., and Holland, J.F. (1987) Kinetic studies of carrier-mediated uptake processes in heart mitochondria by light scattering measurements. Federation of Analytical Chemistry and Spectroscopic Societies, “-9 Oct 1987, Detroit, Michigan. p. 116. Toth, P.P., Sumerix, K., Ferguson-Miller, S.M., and Suelter, C.H. (1987) The role of Ca(II) and Mg(II) in the uncoupled and oligomycin- sensitive components of state “ respiration in isolated chick heart mitochondria. 7th Annual Michigan Cardiovascular Research Forum, East Lansing, Michigan. Toth, P.P., Ferguson-Miller, S.M., and Suelter, C.H. (1987) ADP:0 12. 13. 1“. 15. 16. 314 stoichiometries of chick heart mitochondria. 17th Annual Michigan Cardiovascular Research Forum, East Lansing, Michigan. Toth, P.P., Ferguson-Miller, S.M., and Suelter, C.H. (1987) Kinetics of phosphate utilization during chick heart mitochondrial state 3 and state “ respiration. 17th Annual Michigan Cardiovascular Research Forum, East Lansing, Michigan. Toth, P.P., Sell, J.E., Suelter, C.H., and Holland, J.F. (1987) Characterization and use of chick heart mitochondrial light scattering properties to measure phosphate uptake kinetics. 17th Annual Michigan Cardiovascular Research Forum, East Lansing, Michigan. Lilly, K., Lysiak, W., DiLisa, F., Toth, P.P., and Bieber, L.L. (1987) The effect of L-carnitine on the short-chain acyl-CoA/CoASH ratio in rat heart mitochondria. 17th Annual Michigan Cardiovascular Research Forum, East Lansing, Michigan. Suelter, C.H., Toth, P.P., Bertsch, R.A., and Ferguson-Miller, S.M. (1988) Inorganic phosphate is a complex activator of chick heart mitochondrial a-ketoglutarate dehydrogenase. FASEB J. 2(“): A55“. Toth, P.P., Sumerix, K., Ferguson-Miller, S., and Suelter, C.H. (1988) ADP:O stoichiometries of chick heart mitochondria. FASEB J. 2(5): A1122. SEMINARS PRESENTED 1. Department of Biochemistry and Biophysics, School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania. 10 September 1986. Department of Physiology and Biophysics, School of Medicine, Case Western Reserve University, Cleveland, Ohio. 7 October 1986. Department of Pharmacology and Therapeutics, Medical College of Ohio, Toledo, Ohio. 21 August 1987. 1“th Annual Meeting of the Federation of Analytical Chemistry and Spectroscopy Societies, Detroit, Michigan. 8 October 1987.