: 3. .,_.5 \ NH: ”up.” flmfié c... ram. _ .5 er? I. Em. :2 mx, NERB , area ) m ' .Ef REED 73; A a n v r ~ I - in ‘ X. {D CHLORGP TH BF LAT: : .th 9N on S- Tunis ‘ 13 V _ ' ~ :C‘DNSERVAU . AW \l.. v.1.‘(.‘ , .4 I'v-v- W... . . .. 144:1... .yL 2: . jar. IIH,.E.»4U...E: .3521; .1. my; 1);? r“ L I B R A R Y Michigan Stats? U m'vchitY g” I ' :v/" This is to certify that the thesis entitled STUDIES ON THE MECHANISM OF CONSERVATION 0F REDOX ENERGY IN ATP BY ISOLATED CHLOROPLASTS presented by DONALD R. ORT has been accepted towards fulfillment of the requirements for Ph.D. Botany degree in //&07é90[& J01 / Z {cm Major professo/ Dateifllg- 2, 1974 ,JQ- '. .I‘ " ABSTRACT STUDIES ON THE MECHANISM OF CONSERVATION OF REDOX ENERGY IN ATP BY ISOLATED CHLOROPLASTS By Donald R. Ort This thesis deals with the mechanism.by which the free energy of certain chloroplast oxidation-reduction reactions is conserved in ATP. The results of the investigation are presented in three sections. I. The first section reports the existence of two sites of energy conservation in chloroplasts. Phosphorylation associated with the reduction of lipophilic oxidants and driven by Photosystem II is scarcely affected when plastocyanin is inactivated by poly-grlysine or KCN. In contrast phosphorylation associated with the oxidation of exo- genous electron donors and driven by Photosystem I is abolished by either inhibitor. Thus, there must be another site of ATP formation on the electron transport chain before plastoquinone in addition to the site known to occur after plastoquinone. The existence of a site of energy conservation on the Photosystem II side of plastoquinone was confirmed by experiments with dibromothymoquinone. This plastoquinone analogue appears to block electron flow between the photosystems by preventing the oxidation of reduced plastoquinone. Dibromothymoquinone thus abolishes the electron transport and phosphorylation of the Hill reac- tion but it does not affect electron flux or ATP formation in either of Donald R. Ort the abbreviated Photosystem I-dependent or Photosystem II-dependent transport chains. This virtually proves that there are two sites of energy conservation associated with non-cyclic electron transport in chloroplasts. Electron transport which is driven solely by Photosystem II is not stimulated by phosphorylation or by the addition of uncouplers. 0n the other hand, electron transport which is driven solely by Photo- system I is greatly enhanced by concomitant phosphorylation or by un- couplers. This difference may be related to differences in reversibil- ities of the redox reactions associated with the energy conservation steps. II. The second section of the thesis deals with the role of protons from water oxidation in conservation of the energy of Photosystem II reactions. This study involved the replacement of water as electron donor by a variety of exogenous donors. The use of exogenous electron donors required the complete and specific inhibition of water oxidation. Treatment of chloroplasts with NHZOH and EDTA under the proper condi- tions was found to obliterate water oxidation but not to harm either the phosphorylation mechanism or redox reactions not directly involved in 02 evolution. Moreover, it was discovered that accurate determination of phosphorylation efficiencies could be complicated by superoxide radicals. These radicals, generated via the aerobic oxidation of low potential electron acceptors, can react with the exogenous donor to inflate the 02 consumption and thereby make 02 uptake an unreliable measure of electron flux. This problem was solved by the inclusion of excess superoxide dismutase in the reaction mixture. Donald R. Ort Using chloroplasts preparations in which the water oxidation mechanism had been thus inhibited and the superoxide complication elimi- nated, a study was undertaken with exogenous donors in place of water as a source of electrons and methylviologen as the acceptor of electrons. Substances which had hydroxyl or amino groups as the oxidizable portion of the molecule were coupled to ATP synthesis with an efficiency char- acteristic of the overall Hill reaction. However, when non-proton- producing ions were oxidized by Photosystem II the efficiency declined to one half that of the Hill reaction and the remaining phosphorylation displayed many features characteristic of Photosystem I-dependent ATP formation. It is concluded that the transport of electrons from proton- producers such as catechol to methylviologen employs both of the coupling sites found in the Hill reaction whereas the transport of electrons from pure electron donors such as ferrocyanide or iodide employs only the Photosystem I—associated coupling site. Therefore, it is postulated that proton production is required for Photosystem II phosphorylation. These data are interpreted as evidence for a chemiosmotic type of mech- anism in the conservation of the energy of Photosystem II redox reactions. III. The third section of the thesis deals with Photosystem 1- dependent energy conservation. Photosystem I phosphorylation can be studied apart from Photosystem II phosphorylation when the flow of electrons from water is inhibited with diuron and a source of readily available electrons is supplied. Since a requirement for proton pro- duction in System II phosphorylation seems probable, the major purpose of this investigation was to determine if a similar correlation between protons and ATP formation existed in the System I reaction. To this Donald R. Ort end, the efficiency of phosphorylation (P/ez) and the efficiency of hydrogen ion accumulation (H+/e') were measured. However, Photosystem I reactions employing exogenous electron donors are plagued by a problem; once the exogenous donor has become oxidized, it may begin to act as an electron acceptor. Such a situation can result in a cryptic electron flow and thus a spuriously high apparent phosphorylation efficiency. When precautions were taken to avoid the above complication and super- oxide dismutase was added to the reaction mixture, a reliable measure of System I phosphorylation efficiency was obtained. The efficiency of the Photosystem I-driven proton pump was also measured. It was found that even though System I phosphorylation efficiency displays quite marked sensitivity to pH, the efficiency of hydrogen ion accumulation is essentially unaffected by pH. System II phosphorylation efficiency is insensitive to pH over the same range. Moreover, the two sites of energy conservation display vastly different sensitivities to the energy transfer inhibitor Hg++. System I phosphorylation is 50% sensitive and System II phosphorylation is insensitive. These observations on the different effects of pH and Hg++ are not consistent with the chemiosmotic notion of a common H+ pool shared by both sites of energy conservation. However, the observations can be accommodated with a modification of the chemiosmotic model in which the electrochemical gradient responsible for steady-state phosphorylation is localized within the lamellar membrane and no common high energy pool or coupling factor pool is directly involved. STUDIES ON THE MECHANISM OF CONSERVATION 0F REDOX ENERGY IN ATP BY ISOLATED CHLOROPLASTS By \ Donald RVfOrt A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Botany and Plant Pathology 1974 To Sara ii ACKNOWLEDGMENTS I would like to convey my appreciation to Dr. Seikichi Izawa for his advice and counsel. My collaboration with Dr. Izawa during the past several years was personally rewarding and exceedingly educational. Special thanks go to Dr. Norman E. Good for an inspiring education not confined to biochemistry. I would also like to thank Sandy Perry for her frequent and welcome distractions. This work was supported by the National Science Foundation of the United States through grants GB 22657 and GB 37959X to Drs. Izawa and Good. TABLE OF CONTENTS Page LIST OF TABLES ........................ vii LIST OF FIGURES ....................... ix LIST OF ABBREVIATIONS .................... xiii GENERAL INTRODUCTION ..................... 1 METHODS . . . . . . . . ...... . ............ 8 Chloroplast Isolation . . . .............. 9 Determination of ATP Formation ............... 10 Determination of Electron Transport Rates ......... 12 SECTION I DEMONSTRATION OF TWO SITES OF ENERGY CONSERVATION IN CHLOROPLASTS INTRODUCTION ....................... . . 15 RESULTS AND DISCUSSION .................... 25 Effects of Plastocyanin Inhibitors on Chloroplast Reactions ..................... . 25 Effects of a Plastoquinone Inhibitor on Chloroplast Reactions .................. . . . . 31 The Different Effects of Phosphorylating Conditions on Electron Transport through Sites I and II . ...... 37 SECTION II THE INVOLVEMENT OF THE PROTONS FROM WATER OXIDATION IN PS II-DEPENDENT ENERGY CONSERVATION INTRODUCTION ......................... 44 RESULTS AND DISCUSSION .................... 53 Treatment of Chloroplasts with NHZOH plus EDTA to Abolish Hater Oxidation while Maintaining Energy Conservation Efficiencies ......... . . . . . . 53 iv Page ATP Formation Associated with the Photosystem II- Oxidations of Electron Donors or Electron-Proton Donors ..... . .................... 57 SECTION III ENERGY CONSERVATION AT SITE I IN ISOLATED CHLOROPLASTS INTRODUCTION ....................... . . 78 RESULTS AND DISCUSSION . . . ..... . ........... 83 Site- -Specific Inhibition of Energy Conservation Reactions .................. 83 Quantitative Relationship between Photosystem I Electron Transport and ATP Formation ........... 85 Quantitative Relationship between Photosystem I ATP Formation Efficiency and H+ Accumulation Efficiency ........................ 95 CONCLUSIONS ......................... lOO LITERATURE CITED ....................... l06 APPENDICES ..... . . ........... . ....... l16 Appendix I: The Effects of the Plastocyanin Antagonists KCN and Poly-L- -lysine on Partial Reactions in Isolated Chloroplasts ....................... 116 Appendix II: Electron Transport and Photophosphorylation in Chloroplasts as a Function of the Electron Acceptor III. A Dibromothymoquinone-Insensitive Phosphorylation Reaction Associated with Photosystem II ..... . . 120 Appendix III: Studies on the Energy Coupling Sites of Photophosphorylation III. The Different Effects of Methylamine and ADP plus Phosphate on Electron Trans- port through Coupling Sites I and II in Isolated Chloroplasts ....................... l30 ApBendix IV: Studies on the Energy Coupling Sites of hotophosphorylation II. Treatment of Ch oroplasts with NHZOH plus Ethylenediaminetetraacetate to Inhibit Hater Oxidation while Maintaining Energy Coupling Efficiencies .............. . . . ..... . l40 Appendix V: Studies on the Energy Coupling Sites of Photophosphorylation V. Phosphorylation Efficiencies (P/e i Associated with Aerobic Photooxidation of Artig cial Electron Donors ................ l46 Appendix VI: Photooxidation of Ferrocyanide and Iodide Ions and Associated Phosphorylation in NHZOH-Treated Chloroplasts ....................... 153 Page Appendix VII: Site-Specific Inhibition of Photo- phosphorylation in Isolated Spinach Chloroplasts by Mercuric Chloride .................. l89 Appendix VIII: Quantitative Relationship between Photosystem I Electron Transport and ATP Formation . . . . l92 vi Table II. III. IV. II. LIST OF TABLES The effect of KCN or polylysine treatment on various pathways of electron transport (E.T.) and associated photophosphorylation (ATP) in isolated chloroplasts ................. The effect of DBMIB on electron transport and photophosphorylation in chloroplasts with different electron acceptors ............. The effect of DBMIB on electron transport and photophosphorylation in chloroplasts with different electron donors and methylviologen as electron acceptor ................. Electron transport and ATP formation in Photo- system II oxidation of exogenous donors by NHZOH-treated chloroplasts ......... . . . . . Stoichiometric relationship between ATP formation and electron transport associated with the PS 1- dependent oxidation of various electron donors . . . . APPENDIX I The effect of KCN or polylysine treatment on various pathways of electron transport (E.T.) and associated photophosphorylation (ATP) in isolated chloroplasts ................. APPENDIX II The effect of dibromothymoquinone on electron transport and photophosphorylation in chloroplasts with different electron acceptors . . ........ Inhibition of various reactions in chloroplasts by dibromothymoquinone and KCN . . ........ vii Page 28 33 35 68 93 118 123 127 Table Page APPENDIX IV I. Effect of NHfioH on postillumination ATP forma- tion (XE) an steady state photophosphorylation (Test for uncoupler action of NHZOH) ........ . . 141 II. Photooxidation of ascorbate in NHZOH-treated chloroplasts as assayed by 02 uptake and by titration with DCIP .................. l43 III. Effect of ascorbate on postillumination phos— phorylation (XE) .................... 143 APPENDIX V I. Effect of superoxide dismutase on O uptake and phosphorylation associated with varTous Photo- system I donor reactions ..... . .......... 150 II. Accumulation of H202 during the photooxidation of various Photosystem II and Photosystem I donors . . . 150 APPENDIX VI I. Electron transport from various artificial donors to methylviologen via Photosystem II and Photo- system I and the associated phosphorylation in NHZOH-washed chloroplasts . . . . . . . . . . . . . . . l74 APPENDIX VII 1. Effect of HgClz on photophosphorylation in spinach chloroplasts with various electron acceptors . ..... 190 II. Effect of HgClz on postillumination ATP formation (xE) .......................... 190 APPENDIX VIII I. Stoichiometric relationship between ATP formation and electron transport associated with the PS I- dependent oxidation of various electron donors . . . . . 209 11. Comparison of phosphorylation efficiencies (P/ez) in chloroplasts from different plants ......... 210 III. Effect of DBMIB on electron transport and ATP formation with various exogenous electron donors . . . . le viii LIST OF FIGURES Figure . Page l. Efficiency of ATP formation (P/e ) and effi- ciency of hydrogen ion accumulatgon (H+ /e) at Site II ..... . ........ . . . . . . . . . 97 2. Efficiency of ATP formation (P/e ) and+ effi- ciency of hydrogen ion accumulation (H+ /e) at Site I . ...................... 99 APPENDIX I 1. Effect of water-shock-time on inhibition of electron transport from water to ferricyanide by polylysine . . . ....... . ........ . . ll7 APPENDIX II 1. Effect of dibromothymoquinone on electron transport (E. T. ) and phosphorylation (ATP) with ferricyanide( (FeC ) or oxidized p- phenylenediamine (PDo x) as electron acceptor . . . . . . l24 2. Effect of dibromothymoquinone on electron transport and phosphor lation with ferricyanide and methylviologen (MV as electron acceptors ..... l24 3. Effect of higher concentrations of dibromothymo- quinone on electron transport and phosphorylation with ferricyanide or methylviologen as acceptors . . . . 125 4. Effects of dibromothymoquinone on digitonin- treated chloroplasts .................. l25 5. Simplified scheme of the electron transport path- ways, phosphorylation reactions and inhibition sites ..... . ................... l29 ix Figure 1A. 18. APPENDIX III Effect of the plastoquinone-antagonist dibromothymoquinone on electron transport and ATP formation associated with the photo- reduction of ferricyanide and oxidized p- phenylenediamine by isolated chloroplasts ....... Effect of the electron transport inhibitors dibromothymoquinone and KCN on electron trans- port (E.T.) and phosphorylation (ATP) when dimethquuinone is the electron acceptor . . ..... Effect of the uncoupler methylamine hydro- chloride on electron transport (E.T.) and ATP formation when ferricyanide (FeCy) and oxidized p—phenylenediamine (PDox) serve as electron acceptors .................... . . . Effect of methylamine on the rate of electron transport when ferricyanide (FeCy) and oxidized p-phenylenediamine (P ) are the electron D acceptors . . . . . . ox. APPENDIX IV Effect of time of hydroxylamine pretreatment of chloroplasts on the Hill reaction and Photosystem I-dependent donor reactions .......... . . . . Effect of varied concentrations of hydroxylamine on the Hill reaction and Photosystem I-mediated donor reactions ....... . ...... . . . . . . Effect of time of pretreatment of chloroplasts (at 21°C) with hydroxylamine and EDTA on the Hill reaction, Photosystem I-mediated donor reactions, and associated phosphorylation . . . . . ....... Photooxidation of ascorbate with MV as the elec- tron acceptor in chloroplasts pretreated with hydroxylamine plus EDTA ..... . . . . . . ..... Photooxidation of benzidine with MV as the elec- tron acceptor in chloroplasts pretreated with hydroxylamine plus EDTA ........ . ..... . . APPENDIX V Effect of SOD on ascorbate-stimulated 02 uptake in normal (untreated) chloroplasts . . . ....... Page 133 134 135 135 142 142 142 143 143 147 Figure 1A. 13. Page Effect of SOD on 02 uptake associated with photooxidation of ascorbate in hydroxylamine- washed chloroplasts .................. 148 Effect of SOD on 02 uptake and phosphorylation associated with ascorbate photooxidation in NHZOH-treated chloroplasts at various pH levels . . . . 148 Effect of SOD on O uptake associated with the PS II-mediated oxi ation of catechol in NHZOH- treated chloroplasts ................. l49 Effect of SOD on O uptake associated with the PS II-mediated ox15ation of dicyanohydroquinone in NHZOH-treated chloroplasts ........ . . . . . 149 Effect of SOD on 02 uptake associated with PS I- dependent oxidation of DAD and DAT in the presence of DCMU ..................... . . . 149 APPENDIX VI Electron transport from ferrocyanide to MV in NHZOH-washed chloroplasts as observed as 02 uptake . . l79 Lack of effect of ferrocyanide addition on MV Hill reaction in normal chloroplasts ........... . l79 Electron transport from ferrocyanide to MV in NHZOH-washed chloroplasts as observed by spectro- photometric determination of ferricyanide formation . . 180 Time courses of electron transport from ferro- cyanide to MV and of associated phosphorylation in NHZOH-washed chloroplasts ............... 181 Effect of increasing concentrations of exogenous ferricyanide on the ferrocyanide to MV reaction and associated phosphorylation in NHZOH-washed chloroplasts ............... . ..... I82 Electron transport from ferrocyanide to MV and associated phosphorylation in NHZOH-washed chloro- plasts as a function of the ferrocyanide concentra- 183 tion I O O O I 0000000000000 O O O O O O 0 Electron transport from ferrocyanide to NV and associated phosphorylation in NHZOH-washed chloro- plasts as a function of pH ............ . . l84 xi Figure Page 7. Time courses of electron transport from iodide to MV and associated phosphorylation in NHZOH- washed chloroplasts ............... . . . 185 8. Electron transport from iodide to NV and asso- ciated phosphorylation as a function of the KI concentration ..................... 186 9. Electron transport from iodide to NV and asso— ciated phosphorylation in NHZOH-washed chloro- plasts as a function of pH .............. 187 10. Changes in the pH of the medium associated with the transport of electrons from ferrocyanide, iodide and water to MV in NHZOH-washed chloro- plasts ........................ 188 APPENDIX VII 1. Effect of HgClz on electron transport (E.T.) and phosphorylation associated with various electron transport pathways . . . . . . . . . . . . . . . . . . 190 APPENDIX VIII l. Photooxidation of diaminodurene with MV as the electron acceptor in the presence of DCMU . . . . . . . 214 2. Time courses of 02 uptake and associated phos- phorylation for d1aminodurene photooxidation in the presence and absence of ascorbate . . . ...... 215 3. Photooxidation of 3,3.-diaminobenzidine with MV as the electron acceptor in the presence of DCMU . . . 216 4. Effect of superoxide dismutase on O uptake and phosphorylation associated with diaminobenzidine photooxidation ......... . .......... 217 5A. Time courses of O uptake and associated phos- phorylation for diaminobenzidine photooxidation in the absence of ascorbate .......... . . . . 218 58. Time courses of 02 uptake and associated phos- phorylation for diaminobenzidine photooxidation in the presence of ascorbate ....... . . . . . . 219 6. Time courses of DAB-supported phosphorylation (and 02 uptake) in the absence of ascorbate and 220 xii Cyt DAD DADox DAT DBMIB DCIP DCIPH DMQ E.T. FeCy HEPPS MV PC PD PD ox P/e2 PMS PQ PS I PS II LIST OF ABBREVIATIONS cytochrome 2,3,5,6-tetramethyl-p—phenylenediamine (diaminodurene) oxidized form of diaminodurene 2,5-diaminotoluene 2,5-dibromo-3-methyl-6-isopropyl-p-benzoquinone (dibromothymoquinone) 2,6-dichlorophenolindophenol reduced 2,6-dichlorophenolindophenol 2,5-dimethyl-p-benzoquinone electron transport potassium ferricyanide hydroxyethylpiperazinepropanesulfonic acid methylviologen plastocyanin p-phenylenediamine oxidized form of p-phenylenediamine the ratio of molecules of ATP formed per pairs of electrons transported phenazine methosulfate plastoquinone Photosystem I Photosystem II xiii Q quencher; primary electron acceptor for Photosystem II SOD superoxide dismutase X primary electron acceptor for Photosystem I xiv GENERAL INTRODUCTION GENERAL INTRODUCTION Photosynthesis is the only process known which can supply the energy all living systems ultimately rely on to do work and to maintain low entropy levels (organization). Without the ability of green plants to assimilate solar energy into chemical bonds, the living systems on this planet would soon equilibrate energetically with the environment and consequently cease to exist. The energy available from light absorbed by chlorophyll is utilized to excite electrons to higher energy levels. These electrons then "decay" to a less energetic state via discrete steps along a pre- scribed electron transport chain in the chloroplast lamellar membrane. In higher plants there are two separate photosystems. Their most dis- tinguishing feature is that each absorbs light of different energies. Photosystem 11 (PS II) is believed to sequentially absorb four quanta of light, convert these into four charge separations and store the energy as four holes. The energy available from these absorbed quanta are used to simultaneously oxidize two molecules of H20. The overall result is thus the elevation of four electrons to a higher energy level and the production of a molecule of 0 The excited electrons reduce the primary 2. electron acceptor of PS II, an unidentified compound known as Q, the first step in a series of energetically downhill reactions. Ultimately, the electrons reach the last carrier in this downhill chain (at present believed to be a copper containing protein known as plastocyanin). Photosystem I (PS I) now takes over the electron transport. A specialized chlorophyll which, when excited, via energy transfer from the antennae chlorophyll of PS 1, donates electrons to the primary acceptor of PS I (another unidentified compound thought to be a non-heme iron sulfur protein). This specialized chlorophyll is recognized by the spectral change associated with its oxidation (P700). P700 regains its reduced state by oxidizing plastocyanin. The electrons excited by the System I photoact are transferred via plant ferredoxin and a flavin enzyme until they reach NADP+. However, in isolated chloroplasts the ferredoxin-flavin-NADP+ can be replaced by a host of exogenous electron acceptors. As electrons are transferred between adjacent redox carriers, they continually flow toward the carriers which bind them more tena- ciously, that is to carriers having electrons of lower potential energy. (The only exception to this rule are those redox pairs which donate electrons to and accept electrons from light-excited chlorophyll.) In these instances, because of the input of light energy electrons actually travel from compounds of low electron potential to compounds of higher electron potential (e.g. H20 to Q). The free energy of the oxidation- reduction reactions becomes available to do work. If the energy avail- able between two adjacent carriers is sufficient, it is possible that the energy could be utilized to phosphorylate ADP. Indeed, this appears to be one fate of redox energy in chloroplasts. As originally demon- strated by Arnon gt_gl, (1954), illuminated chloroplast suspensions are capable of ATP synthesis. As technology improved and knowledge of chloroplast biochemistry accumulated, it became possible to examine the fundamental question of how ATP formation is coupled to electron transport reactions. Inherent in this investigation are the questions: what is a site of energy con- servation in mechanistic terms, how many of these sites exist in isolated chloroplasts, and how do they differ? Clearly, any place along the electron transport chain where redox energy is stored in the form of chemical bonds is a site of energy conservation. In this context, the reduction of NADP+ constitutes con- servation of redox energy. However, for the purposes of this thesis, I will restrict the term energy conservation to ATP formation only. To understand what such a site of energy conservation is, we must first understand the mechanism by which redox energy is coupled to ATP forma- tion. Currently there are three major hypotheses concerning the mechanism of ATP synthesis, each with numerous revisions, modifications and alleged refinements of the parent theory. 1. High Energy Chemical Intermediate Hypothesis. Based on the "group transfer" mechanism worked out for oxidative substrate level phosphorylation, Slater (1953) proposed an analogous mechanism for mem- brane bound phosphorylation. According to this hypothesis, the free energy of certain of the redox reactions in mitochondria (and presumably chloroplasts) is conserved in the following manner. The reduction of certain electron carriers result in the formation of covalently bound carrier-"coupling factor" complexes. Subsequent oxidation of the elec- tron carrier portion of this complex results in a series of changes in the covalent union generating a "high energy chemical intermediate." 1? Then, as in substrate level phosphorylation, the electron carrier group is "transferred" for a phosphate group available from the surrounding medium. Ultimately, the phosphate group is used to phosphorylate ADP and, thus, the redox energy originally stored in the oxidized carrier- coupling factor complex is used to make ATP. According to this mechan- ism, a site of energy conservation occurs wherever the oxidation of a redox carrier results in such a high energy complex. 2. Conformational Hypothesis. This theory does not provide a clearly discernible mechanism of ATP formation; rather, it concerns itself principally with the transformation of redox energy into mechani- cal energy. Boyer and associates (Mitchell gngg,, 1967) were the first to espouse this theory for the energy conservation of phosphorylation. They proposed that electron transport generates a conformational change in some protein within the non-aqueous matrix of the energy transducing membrane. The distorted configuration of the protein is postulated as being an unfavorable thermodynamic state, that is as possessing excess energy. Green gt_§l, (1968) pointed out that morphologic changes detected in the inner mitochondrial membrane during respiration might represent the conformational changes proposed by Boyer. The similarities of the conformational and chemical theories are many. Most notable of these similarities is the fact that, in both theories, the oxidation-reduction reactions of electron transport alter the environment of some molecule sufficiently to generate a molecule of high energy. Thus, a site of energy conservation according to the con- formational theory closely resembles a site of energy conservation in the chemical hypothesis, the chief difference being that the "conformational" theory specifies that the high energy complex is a macromolecule capable of storing redox energy in its configurational changes. 3. Chemiosmotic Hypothesis. This mechanism, proposed by Mitchell (1961, l966a, 1966b), is distinctly different from the preceding two in that the first high energy state is not a molecule but rather an electrochemical gradient across a membrane. This hypothesis (in its original form) requires a continuous, anisotropic membrane, enclosing a cisterna. Moreover, this membrane must be relatively impermeable to hydrogen ions as well as to other ions whose movement might dissipate the energy of the gradient. The role of electron transport in this theory is to utilize the free energy available from its redox reactions to translocate hydrogen ions against a concentration gradient. The potential energy inherent in such a gradient supplies the energy neces- sary for the removal of H20 from ADP and Pi in order to form ATP. It is the removal of water via an anisotropic ATPase which prompted Mitchell to include the term osmotic in the nomenclature of his new theory. Clearly, the chemiosmotic hypothesis does not preclude the participation of a high energy chemical intermediate at some point in the mechanism. Indeed, it would be difficult to formulate a mechanism of ATP synthesis which did not include such a compound. However, the salient point is that the initial storage of redox energy is postulated to be a proton and electric charge gradient across the membrane. Therefore, any car- rier which releases protons in an anisotropic manner (preponderantly inside or outside) would constitute a site of energy conservation according to the chemiosmotic hypothesis. (Mitchell, himself, defines a site of phosphorylation as a "redox loop.‘I This concept will be dealt with in detail in Section II.) The purpose of the research presented in this thesis was to investigate the mode of energy conservation in chloroplasts. The results of the investigation are divided into three sections. The first section deals with the number of sites of energy conservation associated with the linear electron transport chain of isolated chloroplasts. The second section is concerned with the involvement of protons from water-oxidation in the energy conservation associated with Photosystem II. The third portion of the thesis reports on findings pertaining to energy conserva- tion associated with Photosystem I-dependent electron transport. Included . is an extensive study on the coupling efficiency (P/ez) of this site and a comparison of this ATP formation efficiency with H+ accumulation effi- ciency (H+/e'). The thesis concludes with a brief section which consists largely of what I believe to be logical speculations concerning the mechanism of energy conservation in chloroplasts. Inasmuch as most of the data presented in this thesis have already appeared in press many references to this published material are used in the text. Therefore, for the convenience of the reader, the appropriate publications have been included as appendices at the end of the thesis. METHODS METHODS The methods section of this thesis intends to describe only those procedures which were used throughout the entire course of my research and which may not be adequately detailed in the appendices. Descriptions of all other techniques are located in the appendices to which the reader's attention will be directed at the appropriate time. Chloroplast Isolation Chloroplasts (unfragmented, naked lamellae) were isolated from commercial spinach (Spinacia oleracea L.) unless there are specific statements to the contrary. The entire isolation procedure was carried out in a cold room at approximately 4°C. Leaves selected for greenness and turgidity were washed with cold distilled water. About ten of the washed leaves were then ground in a Haring blender for five seconds in approximately 75 m1 of a medium consisting of 0.3 M NaCl, 30 mM N-tris (hydroxyethyl) methyglycine (Tricine)-Na0H buffer (pH 7.8), 3 mM M9012 and 0.5 mM EDTA. The homogenate was quickly filtered through eight layers of well-washed cheesecloth and the chloroplasts were sedimented at 2500 x g for 2 minutes. The chloroplast pellet was then resuspended with a paint brush in 50 ml of a medium containing 0.2 M sucrose, 5 mM N-2-hydroxyethylpeperazine-N'-2-ethanesulfonic acid (HEPES)-Na0H buffer (pH 7.5), 2 mM MgClz, and 0.05% bovine serum albumin. After a 45-second centrifugation at 2000 x g to remove cell debris, the chloroplasts were 10 spun down (2000 x g for four minutes) and then washed again in 25 ml of the same medium. The pellet was then resuspended in a few milliliters of the above suspending medium with a final chlorophyll concentration of about 1 mg/ml. The exact chlorophyll concentration of this chloroplast stock was determined as follows: A 0.05 ml sample of the stock was diluted in a volumetric flask to 10 ml with 80% acetone. This acetone extract was spun at 6500 x g for five minutes to remove particulate matter. The absorbance of the resulting supernatant was determined at 710, 663, 652, and 645 nm. Absorbance at 663, 652, and 645 nm was corrected for absorbance at 710 nm, assuming any "absorbance" at this wavelength was not due to chlorophyll and was probably a relatively wave-length- independent scattering. The concentration of chlorophyll of the acetone solution in ug chlorophyll/m1 was calculated according to the formula: (A652 x 28.9) + (A663 x 8.02) + (A645 x 20.2) Determination of ATP Formation Phosphorylation was measured as incorporation of 32P-labeled orthophosphate into ATP. One milliliter aliquots were taken from the 2 ml reaction mixture and immediately frozen in a darkened freezer (in 20 mm x 150 mm pyrex test tubes). The extraction of the remaining labeled orthophosphate as organic solvent-soluble phosphomolybdic acid from the water-soluble ATP was carried out at room temperature and was completed within fifteen minutes after the samples had been removed from the freezer. This extraction was done as follows: Ten ml of 10% per- chloric acid saturated with 1:1 (v/v) butanol-toluene was added to the 11 test tube containing the 1 ml frozen sample. Then 1.2 m1 of 100% acetone, 8 ml of 1:1 (v/v) butanol-toluene saturated with 10% perchloric acid and 1.0 ml of 10% ammonium molybdate were added in that order to the test tube. The phases were then mixed for sixty seconds with a glass plunger. The phases were allowed to separate and the phosphomolybdate- containing organic phase was carefully removed by suction. The ATP— containing aqueous phase was filtered through wet Whatman #4 to remove any last traces of the organic phase. More ammonium molybdate (0.1 ml) was added to the filtered aqueous phase and extracted as before (but without the addition of acetone). After the butanol-toluene was suc- tioned off the samples were ready to be counted. 0n the basis of the radioactivity remaining in the reaction blanks, they contained less than two parts per 100,000 of the remaining unreacted orthophosphate. The number of counts expected from incorporation of 32 32 Pi into 0.1 umole of ATP was determined as follows: 0.1 ml of the Pi stock (same stock as used for experiments) was diluted to 100 ml in a volu- metric flask with 0.1 M unlabeled phosphate. After thorough mixing a 1.0 m1 aliquot was taken and diluted to 13.8 ml with 10% perchloric acid. This was then counted as the experimental samples and represented the counts expected from the phosphorylation of 0.1 umole of ADP. Radioactivity in the final aqueous phase was determined by one of two procedures. The method of choice was to count scintillations resulting from Cerenkov radiation. Although this technique is consider- ably more sensitive than the available alternative, it is not reliable if the sample is colored. When colored samples were encountered, a Geiger-Mfiller immersion tube was used. 12 Determination of Electron Transport Rates The photoreduction of ferricyanide was continuously followed spectrophotometrically by loss of absorbance at 420 nm. These measure- ments were made using either a specially adapted Bausch and Lomb Spec- tronic 505 spectrophotometer or Beckman DU spectrophotometer with Gilford electronics. These instruments were fitted with the appropriate comple- mentary filters to prevent the actinic illumination from interfering with the 420 nm measurement. The temperatures in the reaction vessel compartments were thermostatically controlled. The amount of ferricyanide reduced was calculated using the millimolar extinction coefficient of 1.06. The other frequently used procedure for determination of elec- tron transport rates was 02 consumption resulting from the reoxidation of low potential electron acceptors such as methylviologen, flavin mononucleotide or dibromothymoquinone to form H202.~ Since the chloro- plast preparations were virtually free of catalase activity, no inhibitor of catalase was necessary. A Clark-type oxygen electrode (Yellow Springs Instruments Co.) covered with a teflon membrane was employed for these measurements. The signal from this electrode was fed into an amplifier and the amplified signal was then fed into a strip chart recorder. The number of electrons transported per 02 consumed varies according to the particular system used and will be clearly defined for each system as it is presented. Ferricyanide reduction can also be followed as oxygen evolu- tion. When this technique is employed, one atom of oxygen evolved is —~v———-——-——-—'~—’. wv fi___~.._, . 13 equivalent to the transport of two electrons. The reduction of ferricyanide by illuminated chloroplasts is virtually zero order with respect to ferricyanide concentration. Therefore, the oxygen measuring system can be accurately calibrated by adding a known amount of ferri- cyanide to a reaction mixture and illuminating until the electron accep- tor is exhausted, whereupon oxygen production ceases abruptly. SECTION I DEMONSTRATION OF TWO SITES OF ENERGY CONSERVATION IN CHLOROPLASTS SECTION I DEMONSTRATION OF TWO SITES OF ENERGY CONSERVATION IN CHLOROPLASTS INTRODUCTION The simultaneous discovery of photophosphorylation by Arnon §§;§l, (1954) in isolated chloroplasts from higher plants and by Frenkel (1954) in cell free preparations of photosynthetic bacteria represented the discovery of a phenomenon which is still largely an enigma. An interest in the number of sites of energy conservation in chloroplasts commenced with the demonstration by Arnon $3411. (1955) that rates of phosphorylation and rates of electron transfer could be measured con- currently, the efficiency of phosphorylation could therefore be computed. This efficiency has implications regarding the number of phosphoryla- tion reactions involved. Avron and Chance (1967) were successful in identifying and locating a site of non-cyclic ATP formation in chloroplasts. They demon- strated that, in the presence of an exogenous electron acceptor, illumi- nation with either 650 nm (PS II) or 730 nm (PS I) actinic light leads to the oxidation of cytochrome f_(the measurements were made with a dual wavelength spectrophotometer and cytochrome f_changes were followed as 554 - 540 nm). Furthermore, they showed that the addition of ADP + Pi ++ O O O O O + Mg or of an uncoupler under the same reaction condit1ons d1m1nishes 15 16 the rate-limitation and consequently cytochrome f_is largely reduced when PS II light impinges upon the chloroplast suspension. They con- cluded that a rate-limitation occurs before cytochrome f_and that this rate limiting step is due to a phosphorylation site. Later Kok gt_al. (1969) extended the work to show that the rate-limitation of electron transfer reactions occurs after the electrons from PS II are pooled. In a chain of dark reactions, electrons go from photoreduced Q' (primary electron acceptor of PS II) to photooxidized P700 so that the two traps of the two photosystems are restored to their photoactive state. In strong light, Q and P700 accumulate in their photoinactive state, which indicates that a rate-limitation occurs between the photoacts. Even under weak light the rate of electron transfer is regulated by this slow reaction. Reports by Witt's group (Stiehl and Witt, 1968; Schmidt-Mende and Rumberg, 1968; Schmidt-Mende and Witt, 1968) suggest that the kine- tics of light-induced absorbance changes at 254 nm (ascribed to plasto- quinone) are remarkably similar to those reported by Kok gtgal, (1969) for their carrier pool. These findings are consistent with the original scheme of Hill and Bendall (1960) postulating that a phosphorylation site exists between the two photosystems. Furthermore, we can conclude reasonably that this site of energy conservation is located between plastoquinone and cytochrome f_on the redox chain. The amount of rhetoric which has been dedicated to the argument for and against the existence of more than one site of energy conserva- tion in chloroplasts is formidable. Support of multiple sites of energy conservation can be divided into two categories. Many investigators have endorsed the premise that there exists a separate site of "cyclic 17 phosphorylation" not involved in electron flow from water to NADP+. This subject will be considered in Section III of the thesis. The sub- ject of this section of the thesis is the existence and probable location of multiple sites of non-cyclic ATP formation. The earliest assertions of more than one site at which light energy is conserved as ATP were based on calculations of efficiencies of ATP formation. Based on the precedent of substrate level phosphoryla- tion and organic reactions in general, it was assumed that a coupled redox reaction involves the transfer of two electrons. Therefore, the ratio of the number of molecules of ATP synthesized to the number of electrons transported through a site (P/ez) was taken to be a measure of the efficiency of that site. Moreover, each of the hypotheses for the mechanism of ATP formation presented earlier have an inherent theoretical maximum efficiency (P/ez) for each site of 1.0. Krogmann et_al, (1959) observed P/e2 values associated with the Hill reaction of approximately 0.9 using ferricyanide reduction to monitor electron transport. They also noted that a basal rate of ferri- cyanide reduction always occurred in their chloroplast preparations in the absence of phosphorylation or uncouplers. They pointed out that, if an identical rate of non-phosphorylating electron transport continues when the electron transport is enhanced by the addition of ADP, Pi and Mg++, "corrected" P/e2 ratios of 1.3 to 2.0 may be calculated. On this basis, Krogmann gngfl, suggested that two sites of energy conservation might accompany electron transfer from H20 +-FeCy. Izawa and Good (1968) greatly extended and clarified this approach. They demonstrated that under conditions in which electron transport is strictly limited by 18 phosphorylation the amount of ATP formed is proportional to the additional electron transport which occurs in the presence of Pi (ADP and Mg++ already included in the reaction mixture). That is to say, under the conditions prescribed above, the phosphorylating electron transport appears to be superimposed on a constant non-coupled electron flux. They showed that the efficiency (P/ez) of the phosphorylating component of the electron flux is 1.95 i 0.1 when phosphorylation reactions are rate-determining. As procedures for chloroplast preparation improved, it was no longer necessary to invoke such corrections in order to obtain P/e2 values which routinely exceeded 1.0 by a significant amount. Winget et_§1, (1965) as well as Izawa and Good (1968) reported "uncorrected" P/e2 ratios ranging from 1.1 to 1.3. Recently, efficiencies as high as 1.7 have been reported for Type D chloroplasts (e.g. Heathcote and Hall, 1974) which are chloroplasts that are morphologically intact as opposed to the naked lamella used in my studies. It is significant to note that the chloroplasts which yield higher P/e2 values have lower rates of basal electron transport on a percent basis than their less tightly coupled counterparts. These data are, indeed, suggestive of two sites of energy conservation in chloroplasts. However, they give no indica- tion as to where additional sites may be located along the redox chain. Neumann et_al. (1971) have presented evidence which has led them to postulate that there may be two sites of energy conservation associated with PS I-dependent electron transport. They found that the rate of phosphorylation attendant on the DCMU-insensitive photooxidation of DCIPHz/ascorbate (with methylviologen as an electron acceptor) could 19 actually be stimulated in PS I subchloroplast particles by low concentrations of either amine uncouplers or nigericin. However, no stimulation of ATP synthesis by these uncouplers was observed when phenazine methosulfate (PMS) was substituted for the reduced indophenol dye. Neumann gt_al, suggested that these data could be interpreted in terms of two sequential sites of energy conservation, only the first of which exerts control over the rate of electron transport. They reasoned that low concentrations of amines or nigericin enhanced the flow of electrons through the second site by relieving a restriction imposed by the first site. Assuming that the second site of energy conservation is less sensitive to certain uncouplers than the first, it is possible to visualize an increase in the rate of ATP formation even though the efficiency of the reaction has decreased. In their model the ATP syn- thesized by the PMS cyclic reaction was the product of the second site only. Another location where a site of energy conservation might occur in chloroplasts has been proposed by Kok and associates (1967). They suggested that energy conservation might occur as the result of the photoact itself. Primary charge separation at PS II, due to the excitation of chlorophyll by light, is thought to generate a high energy state which can be used for phosphorylation. Kok gt_al, reasoned that this high energy state may be in the form of a proton gradient or a thermodynamically unfavorable configuration of the chloroplast matrix. They were prompted to construct this picture of phosphorylation by their determination of a relatively high redox potential for Q (primary elec- tron acceptor for PS II). According to their value for the potential 20 of Q, there is not sufficient energy available between Q and P700 to support ATP formation. However, it now seems probable that their value for the redox potential for Q was in error (Cramer and Butler, 1969). The first actual demonstration of two distinct sites of energy conservation was accomplished in this laboratory by Saha gt_al, (1971). Oxidized prphenylenediamine (l,4-benzoquinonediimide) belongs to a class of relatively high potential, lipophilic oxidants. Saha et_al, observed that the addition of PDox (prphenylenediamine plus excess ferricyanide) stimulates electron transport to a level roughly equivalent to the amine- uncoupled rate. They also found that the presence of uncouplers only minimally enhanced the already high rates. These reaction characteristics were very reminiscent of uncoupled electron transport except for one salient difference; the rate of phosphorylation associated with PDox reduction was twice as high as the rate of ATP formation when ferri- cyanide was the electron acceptor even though the electron transport was only one half as efficient in supporting phosphorylation. The quantum efficiency for PDox reduction was the same as the quantum efficiency for the overall Hill reaction but, of course, the quantum efficiency for the phosphorylation with PDox decreased to one half. In addition, the pH optimum for phosphorylation efficiency during PDox reduction was rather broad (pH 7.0 to 8.5) whereas the optimum for phosphorylation efficiency of the Hill reaction showed a relative sharp maximum at about 8.5. A reasonable explanation for their data is depicted in the following scheme: 21 ATP (Site II) H20 ADP + S II p To explain these data, Saha §t_al. proposed a site of ATP formation closely associated with P5 II reactions. They felt that PDox intercepts the preponderant portion of the electrons before they can pass through the PS I coupling site. They reasoned that PS II electron transport could proceed very rapidly and for unknown reasons was regulated little, if any, by phosphorylation reactions. Although their model is, indeed, reasonable, their data do not exclude alternative explanations. One such explanation which comes to mind is this: PDox may be reduced within the chloroplast membrane by PS II in a fashion which is not coupled to ATP synthesis. Before PD can be reoxidized by the hydrophilic oxidant ferricyanide (FeCy), it may donate electrons to a portion of the 22 redox chain containing the site of energy conservation believed to be located between plastoquinone and cytochrome f, This alternative hypothesis was rendered untenable and the scheme largely confirmed by the subsequent investigation of Ouitrakul and Izawa (1973). They showed that incubation of chloroplasts for sixty to ninety minutes at pH 7.8 in the presence of 30 mM KCN suppresses electron transport from water to MV or FeCy by 95%. They also showed that the KCN treatment inactivates the redox carrier plastocyanin. Later, Izawa et_al. (1973) conclusively demonstrated that KCN inhibits the oxidation of cytochrome f_and the reduction of P700. Following the work of Izawa §t_§l,, Selman §t_al, (1974) verified the site of KCN inhibition to be plastocyanin. It is clear, therefore, that KCN inhibits MV and FeCy reduction by blocking the transfer of electrons through PS I. Nevertheless, the reduction of PDox is inhibited only about 20% by KCN treatment. Presumably the part of the reduction of PDox’ which is sen- sitive to this plastocyanin inactivation, represents a small component of the reaction which is still dependent upon PS I. Elimination of this small component by KCN-treatment lowers the phosphorylation efficiency of the electron transport (P/ez) a little more (to about 0.4) and removes any residual effect of uncouplers on the electron transport. 1 Although the work of Ouitrakul and Izawa (1973) added much to the tenability of Saha §t_al,'s claim that PDox intercepts electrons between two sites of non-cyclic energy conservation, the observations were still open to other interpretations. For instance, one could argue that PDox intercepts electrons from the redox chain on the PS I side of the plastoquinone-cytochrome f_coupling site but on the PS II site of evid 121‘. Fact plas tie :91: Stra (1)0 'T 1‘1 1 can 23 plastocyanin. However, Ouitrakul and Izawa did present circumstantial evidence which argues against such a claim: they demonstrated that DCMU-inhibition of PDox reduction was light-intensity independent, a fact which can be best understood if PDox accepts electrons before the plastoquinone electron pool. Another line of support for two sites of ATP formation involves the ability of energized chloroplast membranes to quench the fluores- cence of certain uncoupling acridine dyes. Kraayenhof (1970) demon- strated that the ability of chloroplast membranes to quench the fluorescence of these dyes, specifically atebrin, is strictly dependent upon a source of energy; light-induced electron transport, exogenous ATP or artificially induced pH gradients each under the proper conditions can serve as the membrane energizing agent. Moreover, the concentration of dye,whose fluorescence can be completely abolished by energized lamellar membranes,is linearly related to the chlorophyll concentration. This strict relationship between atebrin quenching and chlorophyll con- tent indicates that there are a fixed number of sites which can be titrated with atebrin on the transducing membrane. Collaborating with Izawa and Chance, Kraayenhof (1972) showed that approximately twice as many sites are available when electrons are transferred from water to FeCy or MV (via two phosphorylation sites?) than when the two abbreviated pathways H20-+ DAD + MV are traversed (each presumably utilizing only one site of phosphorylation). These findings suggest that there may be two separate regions of the transducing membrane, each associated with separate por- tions of the electron transport chain and each energized independently. 24 In this section of the thesis, and in the attendant appendices, data are presented which conclusively show that two sites of non-cyclic energy conservation exist in chloroplasts; one before plastoquinone, as predicted by Saha gt_al. (1971), and the one after plastoquinone identi- fied by Avron and Chance (1967) and by Kok et_al, (1969). The effects of the plastocyanin inhibitor poly-L:lysine on the various abbreviated electron transport reactions and associated phosphorylation, are compared to KCN-treatment. The new inhibitor 2,5-dibromo-3-methyl-6-isopropyl-p: benzoquinone (DBMIB), developed by Trebst et al. (1970), which is believed to inhibit the oxidation of plastoquinone, was tested. Utilizing DBMIB it was possible to unambiguously isolate both sites of energy conserva- tion. Finally, data are presented which deal with the influence of phos- phorylation and uncouplers on PS II-dependentelectron transport reactions. RESULTS AND DISCUSSION Effects of Plastocyanin Inhibitors on Chloroplast Reactions Brand §t_al. (1971) reported that when certain polycations are introduced into chloroplast suspensions low in salt content, PS I- dependent electron transport is inhibited. Brand §t_al, (1972) later demonstrated that polycation treatment prevents the reduction of P700 and the oxidation of cytochrome f, presumably by inactivating plastocyanin. Simultaneously and independently, Izawa and his associates (Izawa gt_al,, 1973a; Ouitrakul and Izawa, 1973) were showing that KCN-treatment of chloroplasts had very similar effects. The further studies reported here had several purposes: First, to examine the effect of poly-trlysine (M.W. 194,000) on phosphorylating electron transport which theretofore had not been explored; second, to compare polycation inhibition with KCN inhibition; third, to determine if the inhibitions are specific for PS I-requiring reactions as suggested by the alleged plastocyanin involvement and; fourth, to determine the relationship between the remaining plastocyanin-independent PS II reac- tions and phosphorylation. The procedure for polycation treatment of chloroplasts developed by Brand §t_al, (1971) involved the isolation of chloroplasts in media free of sucrose and salt. It did not seem that such a procedure would be satisfactory for studies directed at evaluating phosphorylation 25 26 parameters. However, we soon discovered that chloroplasts can be isolated in a conventional isotonic buffered media (see Methods) and then briefly suspended in greatly diluted buffer just before the addi- tion of the polylysine. Shortly after the polycation is added, the normal isotonic media can be restored. Using this procedure, the treated chloroplasts are effectively inhibited but are more tightly coupled and generally more intact than those employed in the earlier studies, mostly because the chloroplasts were not subjected to deleterious salt—free conditions for long. Optimal conditions for polylysine treatment are detailed in Appendix I. Treatment of chloroplasts with KCN was carried out according to the procedure of Ouitrakul and Izawa (1973). This tech- nique is reproduced in the Methods section of Appendix I. The data obtained from the investigation is summarized in Table I. With the described procedure for polycation treatment, nearly complete inhibition of ferricyanide reduction was observed. Previously, other investigators (Brand gngyL., 1971) described inhibitions for this reaction of only 50%. The explanation of the discrepancy probably involves the "leaky" nature of the lamellar membranes of chloroplasts isolated in a hypotonic, salt-free media. In these chloroplasts ferri- cyanide apparently has two sites of reduction, the normal hydrophilic PS I site and an additional site close to PS II which is accessible to only lipophilic oxidants in our chloroplast preparations. In the "leak- ier" chloroplasts this second, lipophilic reduction site, seems to be partially exposed to ferricyanide. In any event, the reduction of MV is in our chloroplasts a PS I-dependent reaction and such reduction involving PS I is completely prevented by KCN and polylysine treatment regardless of whether H20 or diaminodurene is the donor.' 27 Chloroplasts in which water oxidation has been abolished by NHZOH washing are capable of utilizing exogenous compounds such as cate- chol and ferrocyanide in the place of H20 as the source of electrons for the reduction of methylviologen via Photosystem II and I (Ort and Izawa, 1974). Table I shows that these reactions with MV are again almost completely abolished by polylysine treatment. (KCN inhibition of these photoreactions was not investigated because the NHZOH treatment and KCN treatment are not compatable as they were carried out in this study.) Much more detail pertaining to these types of donor reactions is provided in Section II of this thesis. In the introduction of this section, it was pointed out that Saha et_al. (1971) had discovered that lipophilic strong oxidants ("Class III acceptors") are reduced very rapidly by illuminated chloroplasts. They proposed that these oxidants intercept electrons close to PS II and between two sites of energy conservation. The insensitivities of the reductions of PD DAD and DMQ to plastocyanin inactivation confirms ox’ ox their contention that PS I is not necessary for the reduction. The reduction of oxidized prphenylenediamine (PDox) is affected the least by KCN and polylysine. Therefore, PDox reduction probably has a smaller PS I component than the reduction of the other lipophilic oxidants tested. The rate of reduction of 2,5—dimethyl-p;benzoquinone (DMQ) on the other hand is diminished more than 50% by KCN and polylysine. This indicated that DMQ is more efficient as a Class I acceptor than as a Class III acceptor. Dibromothymoquinone (DBMIB) is a unique Class III acceptor (Izawa et al., 1973a) in that it acts not only as a lipophilic electron 28 A~a\av ac< .E.m .cowu_p=ou Amm\av ap< .».m gawpwucoo «unmEummLu mcwm»_>_oa eucmsummgp zug Emumzm .Fpazaoaopsu me .c;\mh< mm—os: so mucmpm>waamn cw ummmmgaxm use mopmm .029 so xooa + cm: can m: cm >2 +.o2 + muwuow ucm wuwcmaoogsmw .Pocumpmu $04 a: oe mmHZma so how; +.o~: msmpm»m mgu so; prgaogopcu m: ow op pcmpm>wzam mm: manuxrs cowuumms FE o.~ mg» cm umucmmmam mummpaosopzu 4o acaosm use .muwvow 25 mp mmuwcmeoggme :2 om "mumngoomm :2 F msFa Pocomumu :5 m.o ”Ao2V :mmo—ow>_»cpms ze mo.o ”Azummv muwcmAUchme anmmmpoa ze ¢.o ”use: com: memumxm Lawnmoum :ogpumpm one .cmm: can: Empmam Locou coguowpm um moves? mg» use .Louamoum coaguapa umuao_u=E ago .mpmapaoao_su .ao< m.o ._amm :5 _ . pom: :2 N .amoeuam z _.o .Ao.m gov Lm$maa zomZImcwowgu :5 Om "co mcwpmwmcou waspst :o_powmc we FE o.~ uwcwmpcou muum>=o cowuomma one .mummpaoLo—so ummeOmw cw Aap co pamEpmmLH mcwmx_apoa co zox Lo pumywm one .H m—nmh 29 new AH> xwucmaq< mmmv mummpaoEo—gu =umsmmzuxomzz= .Ao.m lav aaLcsn Ioaz-maam: saw: cow: :3; mam: mgocou mmmcp mcwma mcowpummm ewe .umm: mm: m.~ In an cmwmza mmmm: gopamuom coguumpm mg» mm mHZmo mcwm: mcowpumma :H as .H xwucmaa< mom .mmeaumuoga any can * Am¢.ov Aoc.ov N_._ mN.o Mpm.ow P~.o Ao~.ov MA~.o mm.o Mm~.o _o.o “Nm.ov Aem._v mm mm cop mm Fe Fun NFP new mmp owe o¢¢ o nmm om FNN or om om m¢m mwm mum omm coop 0mm comp cwm omn— Pom omnm om com mcwmapz—oa Pogucou mcwmxpx_om Fogpcoo mcpmzpapom pocucou mcmm>P>Foa Fogucou m=_mapz_oa lpogpcou mcwmx—a—oa Fogacoo mcwmayapoa Pocpcoo mcwmxpapom Pogpcou marma_zpoa Fogpcoo r—OOOOOOOOOOO wwwwvwv \OOSSDNQLONQ'wa‘O NNmmmmmmmmmm MMMMAMA NV me e¢ mcm on emu mop o—m Fm owm mmN mmu «mm «am can mpc mpm mom ouop omm ommm mm pow zox Fogucoo zox Fogucou zox Pogucoo zux Pogucoo zux Pogpcou 28. Foapcoo eee>z + wuwuom rrr>z.+ onwcmxuoggmm *«¥>z + pocompmu N ramHzmo +_o I axe +.o~= xo N a2 +.owuw :mm ~m\m cowumELom pcoamcmgp Louamoo< -mHzmo Lo w\a ah< mo mpmm coguuwpm co mama cogpompm .szznoEo—go ms.;\mh< mmposa Lo >w=cwn cw uwmmwgaxm mam mmumm .z: m.o we: mHzmo vow: ems: .Axop:pmswvum.~ ze m.o “Axog _.O>:H Nw\& compo—biog. “Loam—has“. Locoo Ewumamopocm ak< we mung :ogpompm to mama coguom_m .H> xwucmaa< 4o H mpnmh cw cm>wm mam comumuwxoouoza NAmcwxovncz use muwuow .wuwcexuoggmm .pocmznocwsmum .Fogompmo Low unoppwccou composes och .ongume new mgocov cosuompw pcmgmmwwu sup: mummpaogopgu cw cowpmpxgoznmocaouoga use peoamcmgp cogpompm co mHzmn mo pomwem och .HHH mpnmh 36 presented. Bdhme et_al. (1971) noted that up to 50% of the transport of electrons to ferricyanide was insensitive to DBMIB in their chloro- plast preparations while in our chloroplasts the inhibitor resistant portion did not exceed 15%. When we treated chloroplasts mildly with digitonin the DBMIB-resistant FeCy reduction increased markedly, pre- sumably by "opening up" the lamellar membranes and giving FeCy access to the normally lipophilic site. It seems logical to conclude that the results obtained by 86hme gt_al, can be explained by such membrane "leakiness." A more serious discrepancy concerns the coupling of the residual FeCy reduction to ATP formation. Earlier in this thesis data were presented which showed that DBMIB-resistant FeCy reduction was firmly coupled, with a constant P/e2 of 0.3 to 0.4 even in chloroplasts treated mildly with digitonin (See Figures 3 and 4, Appendix II). Bmae SflLflls: on the other hand, observed a continual decline in the P/e2 ratio as the concentration of dibromothymoquinone was increased. They concluded from their experiments that the DBMIB-insensitive component of ferricyanide reduction is not coupled to ATP formation whereas we find that it is. The cause of the discrepancy is not clear. Inhibition of electron transport by dibromothymoquinone pro- vides evidence for two distinct sites of non-cyclic ATP formation which is even more convincing than the earlier work with KCN (Ouitrakul and Izawa, 1973) and polylysine (Ort et_al,, 1973). The unique feature of chloroplasts inhibited with DBMIB which makes them such useful tools is that they are able to support two completely separate coupled electron transport processes each of which seems to be a partial reaction of the overall Hill reaction. The DCMU-sensitive reduction of lipophilic 37 acceptors by PS II has already been discussed at length and is resistant to DBMIB as well as KCN and polylysine. The other partial reaction, i.e. the DCMU-insensitive transfer of electrons from exogenous donors such as DAD or DCIPHZ, is prevented by the plastocyanin inhibitors KCN and polylysine but is not affected by dibromothymoquinone (see Table III). Since both of these DBMIB-resistant abbreviated electron pathways are coupled to ATP synthesis the existence of two sites of non- cyclic energy conservation in chloroplasts seems a certainty. If dibromothymoquinone indeed blocks electron transport at the site of plastoquinone involvement, as the evidence suggests (see Appendix II for a more complete discussion of this evidence), there must be a site of energy conservation before and after plastoquinone. These data are completely consistent with the model, first proposed by Saha et_§l, (1971), which is presented in the introduction of this section of the thesis. The Different Effects of Phosphorylating Conditions on Electron Transport through Sites I and II Saha gt_al, (1971) were the first to notice that the reduction of PDox responded very little to uncouplers or to the addition of ADP plus Pi' Figure l of Appendix III depicts the effect of DBMIB on the reduction of ferricyanide and PDox in the presence and absence of ADP plus Pi' This figure shows that as the concentration of DBMIB reaches 0.5 uM the rate of reduction of PDox in the presence of phosphorylation falls to the level observed in the absence of phosphorylation. Never- theless, this DBMIB-resistant electron transport continues to support substantial phosphorylation, even though the stimulation of electron 38 transport by ADP plus Pi is no longer observable. Quite similar results are seen when FeCy is the electron acceptor. As pointed out previously DBMIB not only blocks electron transport between the photo- systems but can also function as a Class III acceptor, the reduced quinone being reoxidized by excess FeCy present in the reaction medium. Both of these reactions (in the presence of 0.5 uM DBMIB) are insensitive to CN' inactivation of plastocyanin and are coupled to ATP formation with the characteristic efficiency (P/ez) of Site II, approximately 0.4. While these observations clearly imply that Site 11 exerts no "control" over PS II electron transport, Trebst and Reimer (1973) have reported that the reduction of various substituted prbenzoquinones is stimulated by ADP plus Pi or by amine uncouplers even in the presence of DBMIB concentrations which completely block the transport of electrons from water to MV. In an attempt to resolve the discrepancy, we repeated the work of Trebst and Reimer using 2,S—dimethyl-E:benzoquinone (DMQ) as the Class III acceptor. Approximately 70% of DMQ photoreduction was sensitive to DBMIB (Table II). Presumably this sensitive electron transport represents transport which is dependent upon both light reac- tions. It is not surprising, therefore, that the rate of DMQ reduction shows a marked stimulation (approximately two-fold) when ADP and P1 are included in the reaction mixture (Figure 2A, Appendix III). Consistent with the findings of Trebst and Reimer, electron transport to DMQ was also stimulated by phosphorylation in the presence of 0.5 uM DBMIB. In this case, the rate of electron transport was only enhanced about 20%, but clearly significant stimulation was occurring (Figure 28, Appendix III). However, when chloroplasts were used in which plastocyanin had 39 been inactivated by CN', all stimulation by ADP plus Pi vanished (Figure 2C, Appendix III). Nevertheless, the remaining electron trans- port was coupled to ATP formation with an efficiency of 0.3 to 0.4. Thus, it is probable that, in the presence of DMQ plus DBMIB, there exists a small electron "leakage" around the DBMIB block which allows a slow rate of electron flux through Site I. This view is supported by the finding that 1.0 mM DMQ (plus 3 mM ascorbate and 100 pM MV) can support a rate of ATP formation of approximately 70 umoles/h/mg chl in chloroplasts inhibited with 1 uM DCMU and 0.5 uM DBMIB (unpublished data). Clearly reduced DMQ can serve as a donor of electrons to PS I. DMQ reduced within the lamellar membrane by electrons from PS II may be an even more efficient donor than the exogenous hydroquinone. Since Site I exerts "control" over electron transport, this would account for the stimulation of electron flow by ADP plus Pi which is observed in the DMQ reducing system. Thus when the alleged electron "leakage" is abolished by CN' inactivation of plastocyanin, the residual pure PS II reaction, utilizing only Site II, is not stimulated by phosphorylation. The complication introduced by electron "leakage" around the site of DBMIB inhibition appears to be unique to experiments involving the reduction of substituted benzoquinones. For instance PD0x reduc- tion and the reduction of any other substituted quinonediimides tested have no such DBMIB-resistant, KCN-sensitive component. The effect of methylamine on the reduction of PDox was also examined. As has long been known, methylamine greatly enhances the rate of electron transport when FeCy is the electron acceptor (Figure 3, Appendix III) by uncoupling electron flow from a rate-limiting energy 40 conservation reaction. Since we have shown that Coupling Site I rather than Coupling Site II is responsible for the rate-limitation imposed on the Hill reaction, we must conclude that methylamine's preponderant effect on electron transport is by releasing the rate-limitation at Site I. The amine uncoupler has an entirely different effect on PS II reactions catalyzed by lipophilic oxidants. When PDox is employed to intercept electrons, the addition of methylamine actually inhibits electron transport somewhat as Site II becomes uncoupled (Figure 3, Appendix III). Since this effect of methylamine on the rate of electron transport increases with increasing concentration of methylamine, even after ATP formation has been completely abolished, it seems likely that the inhibition is actually due to a secondary effect of the amine on PS II (Figure 4, Appendix III). High concentrations of methylamine are known to interfere with water-oxidation (Izawa et_§l,, 1969). The question of whether or not Site II can influence the rate of PS II electron transport is not a trivial consideration. The sugges- tion that phosphorylation does not control electron transport at Site II whereas it does at Site I implies some fundamental difference in the two sites. This difference now seems well established. It is therefore appropriate that we ponder the causes of the difference between Site I and Site II which allows one to regulate electron transport and one not. In many respects Site I resembles the sites of oxidative phoSphoryla- tion. By constructing a model based on what is known about oxidative phosphorylation in mitochondria, it becomes possible to view the electron transport between adjacent redox carriers (A and B) at Site I of chloro- plasts as an equilibrium system. Therefore when PS I is rapidly 41 removing electrons from carrier 8 during illumination, the reaction (A-+ B) will be pulled to the right. However, as the "high energy state" generated by the coupled reaction begins to build up, a back- pressure is exerted against the flow of electrons from A to B by the tendency of the energy conservation steps to be reversed. In the pre- sence of ADP plus Pi the "high energy state" is continually being drained to drive ATP synthesis. Hence, the back pressure exerted by this "high energy state" could be diminished and electron transport from A to B would be able to proceed more rapidly. Uncouplers, accord- ing to this model, would work in a similar fashion except the "high energy state" would be dissipated in their presence even more effectively. 0n the other hand, as has been pointed out, Site II does not appear to exert any control over associated electron transport. This difference can be readily explained, however, if the oxidation-reduction reaction which gives rise to the energy conservation step at Site II is essentially an irreversible reaction. That is, if the forward reaction (A:+ B) is highly favored for thermodynamic reasons, then the back reaction (8 + A) would occur at negligible rates. Under these condi- tions it is easy to imagine how the back-pressure exerted by the "high energy state" would have no effect on the rate of electron flux through the site. Therefore, dissipation of this "high energy state" by ATP formation or uncouplers would not influence electron transport through Site II. Site II is known to be closely associated with PS II and to occur before electrons are pooled (at plastoquinone?). There are two known reactions in this portion of the photosynthetic redox chain which 42 are essentially irreversible. One of these reactions is the System II photoact. It has been suggested that the energy of absorbed light is utilized to generate a reduced acceptor Q' and an oxidized donor Z+ on opposite sides of the photosynthetic membrane (e.g. Witt, 1971). The energy available in this electrical gradient could supply the energy necessary to phosphorylate ADP. The second of the essentially irreversible reactions associated with PS II is the water oxidation reaction. Protons resulting from the oxidation of water may be released anisotropically and accumulate "inside" the membrane, generating a H+ gradient across the membrane capable of driving ATP formation. This model is the subject of Section II of the thesis. SECTION II THE INVOLVEMENT OF THE PROTONS FROM WATER OXIDATION IN PS II-DEPENDENT ENERGY CONSERVATION SECTION II THE INVOLVEMENT OF THE PROTONS FROM WATER OXIDATION IN PS II-DEPENDENT ENERGY CONSERVATION INTRODUCTION The notion that hydrogen ions might be intimately inVolved in the photosynthetic and oxidative phosphorylation came into being when Mitchell (1961) formulated his "Chemiosmotic Hypothesis." This proposed mechanism of membrane bound ATP synthesis was briefly discussed in the introduction of the thesis and will be elaborated here. My purpose is to examine the major tenets of the chemiosmotic hypothesis and the experimental support for the theory coming from research on chloroplasts. Mitchell's theory (e.g. 1961, 1966a, 1966b) requires that the energy transducing organelle consist of an ion-impermeable membrane sur- rounding an enclosed space. The lamellar membrane of chloroplasts, which contains the photosynthetic redox carriers, is freely permeable to water but seems relatively impermeable to protons and many other ions. It may be, however, that other ions can be translocated through these membranes via specialized carriers as in mitochondria. The second requirement of Mitchell's theory is for a proton- translocating photosynthetic redox chain. The oxidation-reduction reactions of chloroplasts responsible for the transport of electrons are thought to be responsible for the translocation of H+ to the inner 44 45 space of the lamellar membrane. Mitchell postulates (cf. refs. 1961, 1966a, l966b, 1970) that each site of energy conservation is associated with a "redox loop." A typical "loop" consists of a carrier whose reduction involves protonation with two hydrogen atoms and a carrier whose reduction involves only electrons. Thus, in the operation of a photosynthetic redox loop, a carrier with two hydrogen atoms is first translocated inward and then two electrons are moved across the membrane in the opposite direction. Since a proton is left behind, this creates a pH difference as well as an electrical potential (negative outside) across the membrane. The difference in hydrogen ion concentration on either side of the membrane and the membrane potential jointly exert a "driving force" on the protons. Thus, the "driving force" or energized state postulated by the chemiosmotic mechanism consists of a chemical ("osmotic") component, due to the disequilibrium of protons, and an electrical component, the membrane potential. Finally, the chemiosmotic theory requires a reversible proton- translocating ATPase. Since the ATPase equilibrium ATP + Hzo=—,ADP + P1. (1) greatly favors ATP hydrolysis, an energy input mechanism is required to drive this reaction in the reverse direction to phosphorylate ADP. Mitchell contends this is accomplished by coupling the ATPase reaction with a translocation of protons across the membrane (inside to outside in chloroplasts). Hence, this theory places another restriction on the reversible ATPase, it must be directional in its enzymatic function, which implies that its orientation in the membrane is critical. 46 Photosynthetic phosphorylation can be visualized as resulting from a cyclic flow of protons, the redox chain acting to translocate the hydrogen ions in the inward direction. In escaping again they drive equation 1 to the left. However, so far nothing has been said as to how the translocation of protons from a region of high concentration to a region of low concentration can be coupled to the synthesis of ATP at the level of the ATPase. Mitchell reasoned (e.g. 1970, 1972) that accu- mulation of protons is not detectably different from the depletion of OH', and if the active site of the membrane-bound chloroplast ATPase is inacclgsible to water, then the dehydration reaction (ADP + P1) could occur so that protons are released to the outside of the membrane and OH' to the inside. This mechanism implies a stoichiometry of one proton released for one ATP synthesized. However, on the basis of experimental evidence, the stoichiometry of 2 seems more likely. + ¢ + ATP+H0+2Hout———,ADP+P1.+2Hin (2) 2 The value of two protons per ATP conforms with data of Mitchell and Mcyle (1965, 1968) from mitochondria as well as the value suggested by other workers for chloroplasts (Schwartz, 1968; Izawa, 1970). Clearly, the dehydration reaction may then have an equilibrium constant favorable for ATP formation with more modest proton pressures. ATP synthesis is favored over ATP hydrolysis when the activity of water is lowered in the vicinity of the ATPase. According to calcu- lations by Mitchell (1966b), in order to obtain an ATP/ADP concentration ratio of unity in the presence of 10 mM P1, the activity of water in a model system would have to be lowered to 5 uM: and probably much lower yet at the active site in the actual situation where these concentrations 47 of ADP, ATP and Pi are not realistic. The activity of water in the vicinity of the ATPase can be theoretically diminished by vectoral movement of protons. The equilibrium constant, K, for equation 2 is [ADP] [P.] [H+ )2 K = 1 in (3) + 2 [ATP] [H20] [H out] Therefore, Mitchell argues that the effective activity of water in the microenvironment surrounding the active site of the ATPase is described by the left hand side of equation 4. H+ 2 ADP P. ”20* [ out] = 1/K [ 1 [ 11 (4) LH+1n12 [ATP] where H20* is the activity of water in the aqueous phase. From these equations the in/out proton concentration ratio requisite to drive ATP synthesis can be computed. With realistic ATP/ADP ratios and P1 con- centrations the pH difference would have to be 4 to 5 units. However, these computations neglect the membrane potential which could act as an added driving force. Jagendorf and Hind (1963) were the first to show that there is indeed a reversible light-induced hydrogen ion uptake associated with electron transport. Furthermore, they demonstrated that preillumi- nated chloroplasts, which had accumulated hydrogen ions, were capable of ATP synthesis when ADP and Pi were added immediately after the light was turned off. Shen and Shen (1962) independently described similar observations. Clearly, electron transport in chloroplasts results in H+ accumulation and a build up of the capacity of chloroplasts to syn- thesize ATP (this capacity has been termed XE). Moreover the 48 time-courses of the proton accumulation (and subsequent decay in the dark) are closely parallel to the time course of build-up and dissipation of XE (Izawa, 1970). Since the light-induced H+ accumulation and XE have nearly identical characteristics, it came to be thought that they were one in the same. This belief was greatly strengthed by Jagendorf and Uribe (1966) who demonstrated that an artificially induced pH gradient can support ATP synthesis in chloroplasts without any light stage at all (acid-bath phosphorylation). If the pH gradient is not XE’ then surely it must be in free equilibrium with X This pH rise in the suspending E' media of illuminated chloroplasts has been the subject of extensive research since it seems one of the most direct means of examining the predictions of the chemiosmotic model of energy transduction in chloro- plasts. It has been suggested that the difference in pH across the lamellar membrane of chloroplasts may be as high as 3.0 pH units (Rottenburg and Grunwald, 1972; Schuldiner et_al,, 1972; Rumberg and Siegel, 1969; Portis and McCarty, 1973), and this has been presented as establishing the thermodynamic feasibility of chemiosmotic coupling in chloroplasts. However, as we have seen a pH difference of 3 units is really not quite enough to be the sole driving force for ATP synthesis. Such massive inward translocation of hydrogen ions as is implied by observed pH changes would not be possible without compensatory ion movements to preserve electrical neutrality. The chloride ion has long been a popular candidate as the major mobile counter ion and indeed Cl' uptake does occur (Schroder et al., 1971: Rottenburg et al., 1972: Deamer and Packer, 1969). Dilley and Vernon (1965) found that the 49 combined K+ and Mg++ efflux nearly compensated for H+ influx which suggested that the movement of C1" did not occur under their conditions. The significance of K+ and Mg++ efflux observed by Dilley and Vernon is obscure because the suspending media they employed was devoid of these cations. Perhaps the most definitive work on compensatory ion movement to date was done by Hind, Nakatani and Izawa (1974). They reported that fully reversible Cl' influx and Mg++ efflux accompany the H+ uptake and together compensate almost equally for the charge transferred in H+ translocation. A small efflux of K+ appears to complete the electrical balance. The ratio of the number of protons taken up to the number of electrons transported (H+/e') is a pertinent measurement since, accord- ing to the chemiosmotic hypothesis, the value of this ratio will impose an upper limit on the efficiency of ATP formation (P/ez). Various investigators have reported quite high values for H+/e' ratios in iso- lated chloroplasts from slightly over 3 to nearly 6 (Lynn and Brown, 1967; Karlish and Avron, 1967; Dilley and Vernon, 1967; Crofts, 1968); however, it is generally felt that these values are erroneous. Izawa and Hind (1967) were able to resolve the initial kinetics of the pH rise, by employing a "flash yield" method to measure pH change and thereby compensate for sluggish instrument response. Using FeCy and DCIP as electron acceptors and correcting for outward proton leakage, they obtained H+/e' values of 1.6 to 1.7 measuring proton uptake imme- diately after completion of the "pH gush." (The pH gush is a very rapid phase of hydrogen ion consumption which occurs at the very beginning of the illumination period. This rapid hydrogen ion uptake may be due to 50 photoreduction of the plastoquinone pool.) This value was reproduced by Gould and Izawa (1974) using a flash yield technique and MV as the electron acceptor. Similarly, Rumberg et_al, (1969), employing the method of Izawa and Hind (1967), determined that the value of the H+/e' ratio was 2.1 using FeCy as the electron acceptor and extrapolating to zero time. Schliephake giggl, (1968), illuminating with short flashes and using a pH indicating dye, obtained an H+/e' of 2.0 with NADPI. The fact that there seems to be two sites of proton trans- location in chloroplasts agrees well with this observed H+/e' ratio of 2.0. One site appears to be associated with PS I-dependent electron transport (Witt gt_al,, 1968; Schwartz, 1968; Strotmann and von Gosseln, 1971; Ort gt_al., 1974). For instance, Strotmann gt_al. (1971) measured the quantitative relationship between H+-uptake and the transport of electrons from DCIPH2 to MV: the H+/e' ratio associated with this system was reported to be 1.0. The other site of proton translocation appears to be PS II-dependent (Gould and Izawa, 1974; Schliephake et_al,, 1968). Gould and Izawa (1974) demonstrated that there is a reversible light- induced proton pump associated with the PS II-driven transport of elec- trons from H20 to DBMIB. This proton pump operated with an observed efficiency (H+/e') of about 0.5. If we assume that the equivalent of the efflux of two protons from inside of the lamellae is required for the synthesis of one molecule of ATP (in accord with the value of Mitchell and Moyle for mitochondria-- refs. 1965, 1968) then, according to the chemiosmotic theory, the P/e2 ratio should not exceed the H+/e' ratio. In the first section of this thesis I discussed in detail energy conservation efficiencies (P/ez) 51 of various chloroplast electron transport reactions. Clearly, Gould and Izawa's (1974) value for the H+/e- ratio of 1.7 for the H20 to MV reaction agrees rather well with the observed P/e2 of this reaction, 1.2-1.3. Furthermore, the P/e2 ratio of the PS II partial reaction H201+ DBMIB of approximately 0.4 (Izawa gt_al,, 1973b) closely approaches the value of H+/e' for this system, 0.5 (Gould and Izawa, 1974). At pH 8.0 the observed value for the H+/e' ratio of the Site I utilizing reactions DCIPHZ +»MV (Strotmann and von Gosslen, 1971) and ferrocyanide to MV (Ort et_al,, 1974) is about 1.0. This value is also in the range which would be predicted by the chemiosmotic hypothesis for these reac- tions which have P/e2 values of 0.6 to 0.7. Above, I chose to assume that the H+/ATP ratio in chloroplasts is two as Mitchell and Moyle (1965, 1968) claim for mitochondrial oxida- tive phosphorylation. In agreement, Izawa (1970) has suggested that the ratio of H+ accumulation to potential post-illumination ATP formation slightly exceeds 2.0 on the grounds that only 50% of the XE is actually trapped as ATP. Swartz (1968) also reported H+/ATP ratios of 2.0, using steady-state rates of ATP synthesis and H+ translocation deduced from the slopes of the pH tracing immediately before and after the light was turned off, but this method is questionable on theoretical grounds. However, Witt §t_al, (1971) suggest the ratio may be 3.0 rather than 2.0 in chloroplasts. If the higher value is correct, this would, of course, have the effect of diminishing the maximum P/e2 values predicted by Mitchell's theory given the observed H+/e' ratios. The chemiosmotic theory predicts that the steady-state pH gradient at high light intensity under phosphorylating conditions should 52 be significantly less than that reached under non-phosphorylating conditions and this is indeed the case (Schwartz, 1968; Gould and Izawa, 1974). Furthermore, Gould and Izawa (1974) demonstrated that the steady-state level of the pH gradient is also diminished by arsenylation of ADP. As can be seen from the above literature survey, the assertion that energy conservation in chloroplasts occurs via a "chemiosmotic" mechanism is not without foundation. Certainly electron transport is associated with a rather massive translocation of protons in chloroplasts, a translocation which is capable of storing a good deal of the redox energy. Furthermore, the energy which is stored in the proton gradient can be used for ATP synthesis (Jagendorf and Uribe, 1966). It is diffi- cult, therefore, to escape the conclusion that the proton gradient is somehow on the mainstream of redox energy conservation. Nevertheless evidence will be presented later in this thesis which probably necessi- tates at least some modifications of the theory as it now stands. I will present data in this section of the thesis which strongly suggest that hydrogen ion production is an obligatory require- ment for ATP formation at Site II. These data provide us with the first direct evidence of a causal relationship between proton production and ATP synthesis. Heretofore the evidence has been circumstantial. Pre- sented also are several new techniques which were developed to make the experiments mentioned above possible, methods which should prove useful in future investigations. RESULTS AND DISCUSSION Treatment of Chloroplasts with NHOOnglus EDTA to Abolish Water Oxidation While‘Maintaining Energy Cbnservation Efficiendies In the first section of this thesis, it was concluded that there are two sites where the redox energy of the non-cyclic electron transport of chloroplasts is made available to phosphorylate ADP. The data presented strongly suggest that one of these sites (Site 11) is close to PS 11 (probably before plastoquinone). The other, Site I, is the previously recognized site of phosphorylation between plastoquinone and cytochrome f_(Avron and Chance, 1967; Bdhme and Cramer, 1969). The purpose of the studies reported below was to locate Site II more pre- cisely and, in so doing, to gain insight into the nature of energy con- servation. According to existing data (Cramer and Butler, 1969), it does not appear thermodynamically feasible for a site of energy conserva- tion to lie between PS II and plastoquinone. Hence, it seemed plausible that the energy conservation reaction might be coupled to the process of water oxidation or the System II photoact. If water oxidation (or the proton-production from water oxida- tion) is in fact intimately involved in the mechanism of phosphorylation at Site II, an examination of the relationship between electron transport and ATP formation where exogenous donors of electrons replace water should be of great interest. Such a study requires a complete and 53 54 specific inhibition of water oxidation. Although chloroplasts in which oxygen evolution has been inhibited by "tris-washing" (Yamashita and Butler, 1968; 1969) or mild heat treatment (Bdhme and Trebst, 1969; Katoh and San Pietro, 1967) are known to be capable of coupled Photo- system II-mediated reactions using exogenous electron donors, these techniques were not satisfactory for our purposes. The chloroplasts used in this study were somewhat resistant to "tris-washing," thus it was not possible to obtain complete inhibition of water oxidation without severely impairing the phosphorylation mechanisms. The effect of heat- treatment on the phosphorylation mechanism of these membranes was even harsher. Treatment of chloroplasts with hydroxylamine is known to result in the specific and irreversible inhibition of water oxidation due to release of manganese from the lamellar membranes (Cheniae and Martin, 1966). However, heretofore the use of NH20H as an electron transport inhibitor for investigations aimed at evaluating photophos- phorylation efficiencies has been avoided presumably because of undesirable effects of this compound, primarily caused by the fact that as an amine it uncouples and as a reductant it donates electrons to PS II. Fortunately, the undesirable effects of hydroxylamine can be completely eliminated if the hydroxylamine is washed out of the chloro- plasts. Presumably because of the removal of manganese, the inhibition of water oxidation is irreversible and unaffected by the removal of the amine. Since it is impossible to remove absolutely all of the NHZOH by washing, the feasibility of the washing technique depended somewhat 55 on the potency of hydroxylamine as an uncoupler. We examined the effect of NHZOH added in the reaction mixture on postillumination ATP formation (Hind and Jagendorf, 1965) in the presence of pyocyanine and on the steady- state phosphorylation supported by the transport of electrons from DAD to MV, both of which reactions are sensitive to uncouplers but not to inhibition of water oxidation. These data are presented in Table I of Appendix IV. In the concentration range of NHZOH employed to inhibit 02 evolution, the uncoupling action is weak. The fact that hydroxylamine is a weak uncoupler is predictable from the low bascity (pKa = 6) of this amine (Good, 1960; Hind and Whittingham, 1963). Therefore, on removal the greater part of the amine uncoupling side effect became negligible. Hydroxylamine inhibition of oxygen evolution is quite tempera- ture dependent. The Q10 for this inhibition has been reported to be 2.43 (Cheniae and Martin, 1966). A time-course of NHZOH inhibition of water oxidation at 0°C and 21°C is presented in Figure 1A, Appendix IV. The optimum concentration of NHZOH (Figure 18, Appendix IV) was approxi- mately three times greater for treatments carried out at 0°C (about 15 mM) than for those carried out at room temperature (about 5 mM). It is notable that there was no discernable effect of NHZOH treatment on the electron transport rate or phosphorylation efficiency associated with the PS I reaction transferring electrons from DAD to MV if the amine had been removed by washing. See Methods of Appendix IV for exact details of the washing procedure. Unfortunately, these treatments were not wholly satisfactory for another reason. Significant rates of electron transport and ATP 56 formation remained (typically 40-60 acquiv. or 20-30 umoles ATP/h/mg Chl) when water was the electron donor. However, the residual electron flow and the accompanying phosphorylation were obliterated when the treatment medium included 1 mM EDTA. Since EDTA alone has no effect on the water-splitting machinery and is not able to bring about the release of manganese from the membrane (Cheniae and Martin, 1971), the observed effect of EDTA on water oxidation in the presence of NHZOH is probably indirect. It is possible that a portion of the manganese extraction by NHZOH is reversible, and the binding of released manganese by EDTA does not allow the reversal to occur. Since the difference between 0°C and 21°C treatments appeared to be only in the rate of the development of inhibition, the higher temp- erature was chosen for a closer look at pretreatment of chloroplasts with combined NHZOH and EDTA (Figure 2, Appendix IV). In these studies, it was found that the P/e2 ratio of non-cyclic photophosphorylation (H20 + FeCy) remained the same until the rates of electron transport and ATP forma- tion became so low that they could not be measured accurately. The PS I-dependent electron transport (DAD to MV or DCIPH to MV) and the coupled 2 phosphorylation were totally unaffected. Therefore it appears that a brief treatment of chloroplasts at 21°C in the dark with NHZOH, EDTA and Mg++ at pH 7.5, followed by washing at 0°C with amine-free medium, pro- vides a reliable method of abolishing the water-oxidizing ability of chloroplasts without impairing the phosphorylation mechanism. Further- more, there were no detectable adverse effects of the treatment on any of the redox reactions other than those directly involved in water oxida- tion. ATP Formation Associated with the Photosystem II-Oxidations of Electron Donors or EleCtran-ProtonTDonors In 1969, Bfihme and Trebst provided an argument for two sites of ATP synthesis associated with non-cyclic electron flow in chloroplasts. Their suggestion was based on the observation that the DCMU-sensitive photooxidation of ascorbate (to form stoichiometric amounts of dehydro- ascorbate) supported ATP formation in heat-treated chloroplasts with an apparent P/ez of only 0.5. Since this value was only half that of the P/e2 associated with the normal Hill reaction (H20 + MV) they concluded that the electrons removed by PS II from ascorbate were by-passing a site of ATP synthesis. In NHZOH-treated chloroplasts very similar results were ob- tained (Figure 3, Appendix IV). An apparent P/e2 of approximately 0.6 was observed for this highly DCMU-sensitive reaction. However, we have shown that these diminished P/e2 ratios are not due to electrons from ascorbate detouring around a site of ATP formation, but rather to an over-estimation of electron flux. Exogenous low potential electron acceptors, such as the viologens and anthroquinones, are in wide use for the study of chloroplast electron transport and phosphorylation reac- tions. The reduction of these acceptors is conveniently followed as 02 consumption resulting from the autoxidation of the reduced low potential acceptors. The widely accepted formulae (see ref. Trebst gt;gl,, 1963) for the transfer of electrons from a donor (AHZ) to an acceptor such as methylviologen and the subsequent reoxidation of the reduced acceptor with the production of H202 must be rewritten. A new term for the 58 intermediate in H202 formation, the superoxide radical '02- (Misra and Fridovich, 1972) must be introduced: ‘7 AH2 + 214v2+ c“'°”°P'aSts A + 2H+ + 2-Mv+ (1) light z-Mv+ + 202 59°"ta”e°”5 > 2MV2+ + 2.02” (2) . - + spontaneous ,_ 2 02 + 2” endogenous SOD" 02 + H202 (3) AHZ + 02 2e- trangfigrta A + H202 (4) and therefore the uptake of a molecule of 02 represents the biological transfer of two electrons. In the normal Hill reaction where AH2 = H20 (i.e. A = EOZ), the over-all reaction becomes what is known as the Mehler reaction: H20 + 50 2e' transport H20 . 19 and the uptake of a single atom of oxygen represents the transfer of two 2 2 (5) electrons. Until now it has been assumed that calculations of electron fluxes could be accurately made according to equation 4 (or 5 if H20 is the electron source). The validity of equation 5 was established long ago (Brown and Good, 1955) by mass spectroscopic studies of the 02 exchange reactions involved in the Mehler reaction. However, the valid- ity of the formulation for exogenous electron donor systems has never been rigorously established. Indeed, it now appears a certainty that the involvement of ~02' in the aerobic photooxidation of exogenous reductants considerably complicates the straight-forward formulation presented in equations 1 through 4 (Elstner et al., 1970). For example, 59 if a portion of the -02' generated by the aerobic reoxidation of -MV+ (eq 2) directly oxidized the exogenous donor AHZ’ the rate of 0 con- 2 consumption would exceed the rate predicted by equation 4 and would not be an accurate measure of the electron transport which was occurring: that is, the uptake of 02 would not represent the transport of a pair of electrons. This enhancement of 02 consumption results because -02', which normally dismutates and thereby regenerates half of the consumed O2 (eq 3), is now merely reduced to H202 with no regeneration of 02: AH2 + 2.021 ————+ A + zHoo' (6) However, such reactions and the resulting enhancement of 02 consumption should be abolished when the dismutation of :02“ (eq 3) is accelerated by superoxide dismutase (SOD) (McCord and Fridovich, 1969). That is to say, if sufficient superoxide dismutase is present, the aerobic photo- oxidation of exogenous donors should follow exactly equations 1 through 4 and the relation 02 = 2e' should be valid. These considerations prompted us to re-examine the photooxida- tion of ascorbate both by normal chloroplasts and chloroplasts treated with hydroxylamine. Figure l of Appendix.‘V shows the effect of ascorbate addition on 02 consumption and phosphorylation in normal chloroplasts which are capable of oxidizing water. The addition of 5 mM ascorbate (Figure 1, Appendix V) nearly doubles the rate of O2 consumption observed for the H20 + MV reaction. Nevertheless, the rate of ATP formation remains virtually unchanged, but due to the enhanced uptake of 02, the apparent efficiency of phosphorylation (P/O) drops sharply from 1.2, a value typical of the Hill reaction, to 0.5. However, the ascorbate- stimulated portion of the 02 consumption is abolished by the addition 60 of sufficient SOD (approximately 180 units/ml). Thus, with SOD the observed P/O ratio is restored to nearly the original value. It must be concluded from these observations that, in normal 02-producing chloroplasts, ascorbate oxidation is due almost entirely to the super- oxide radical which is generated by the aerobic reoxidation of reduced MV. Thus, it is unlikely that ascorbate replaces water as the primary electron donor to any significant extent, Bmae and Trebst (1969) not- withstanding. Epel and Neumann (1972) and Allen and Hall (1973) have suggested also that such a mechanism involving -02' might cause ascorbate oxidation in 02-producing chloroplasts. . On the other hand, ascorbate can serve as an alternative electron donor (Bmae and Trebst, 1969; Yamashita and Butler, 1969; Ben- Hayyim and Avron, 1970; Cheniae and Martin, 1970; Ort and Izawa, 1973). Apparently, the inability of chloroplasts to oxidize water creates favor- able conditions for the strong oxidant produced by PS II to oxidize exogenous reductants. As reported above, the P/O2 value observed for the oxidation of ascorbate by NHZOH-treated chloroplasts is approximately 0.6; similar results were reported by Bbhme and Trebst (1969) using heat- treated chloroplasts. However, it seemed certain that the uptake of a 02 molecule is not equivalent to two electrons in the systems reported above because the superoxide radical, produced via the univalent reduc- tion of 02 by MV, must react with ascorbate. As suspected, the aerobic photooxidation of ascorbate in NHZOH-treated chloroplasts (Figure 2;. Appendix V) contains a large portion which is sensitive to $00 which- shows that -02' is involved. It is the SOD-insensitive portion of the 02 consumption--somewhat more than half--which is an accurate measure 61 of electron flux according to equations 1 through 4. As a consequence of the removal of the irrelevant 02 uptake by inclusion of $00, the P/O2 is elevated to a value approaching 0.9; a value not greatly different from the value observed during the normal Hill reaction. Thus, in the presence of $00, the P/O2 ratio is virtually equivalent to the P/e2 ratio. Further evidence that ascorbate is substituting for water as the primary donor of electrons to PS 11 can be gleaned from the pH pro- file of ascorbate oxidation (Figure 3, Appendix V). Predictably, the absence or presence of SOD affects only the height of the 02 uptake vs. pH curves with little other effect on their shapes. Phosphorylation remains virtually unchanged by $00 at all pH's. The shapes of these activity-pH curves and the marked stimulation of electron transport by phosphorylation are remarkably similar to the shapes observed with the H20 +1MV reaction (Gould and Izawa, 1973b). This suggests that the photo- oxidation of both water and ascorbate may be controlled by the same rate-limiting phosphorylation site. The value of the phosphorylation efficiency ratio in ascorbate photooxidation is no longer low enough (P/e2 = 0.9) to support the con- tention of Bohme and Trebst that one site of energy conservation is being by-passed. It is significantly lower, however, than the P/ez ratio characteristic of the Hill reaction (1.1-1.3). An explanation for this will be developed later in the thesis. (A more detailed account of other published data on ascorbate oxidation can be found in the Dis- cussion of Appendix V.) 62 Using several other exogenous donor systems, it was possible to restore the energy conservation efficiency characteristic of the Hill reaction almost completely. In all cases except one it was necessary to include 500 to attain these high P/O2 ratios. Catechol (with 0.5 mM ascorbate as a reductant pool) was the best exogenous donor of electrons to PS II of all compounds tested. The rates of electron transport and phosphorylation (in the presence of excess SOD) were approximately one half the rate of the Hill reaction and were highly DCMU-sensitive (Figure 4, Appendix V). The observed P/O2 was 0.6 in the absence of $00, but when the dismutation of the superoxide radical was enhanced by the inclusion of SOD, in the reaction mixture the P/O2 (where P/O2 = P/ez) exceeded 1.1, which is virtually identical with the ratio char- acteristic of the Hill reaction. A similar efficiency was observed with a new donor, 2,3-dicyanohydroquinone (again with 0.5 mM ascorbate), although the rates of reaction with this reductant were not as rapid as those observed with catechol (Figure 5, Appendix V). The photooxidation of praminophenol (with 0.5 mM ascorbate) in the presence of 300 units SOD/ml was also found to be coupled to ATP formation with an efficiency which exceeded 1.0. Yamashita and Butler (1969) had reported a P/e2 value of 0.97 in their tris-washed chloroplasts for the anaerobic photo- oxidation of this compound with NADP+ as the electron acceptor. It is clear that complications introduced by superoxide radical occur in practically all reactions employing exogenous donors if electron transport is monitored as 0 consumption. Very often the electron donor 2 system includes ascorbate which we know reacts with superoxide radicals. In addition many of the commonly used exogenous donors themselves must 63 be expected, on grounds of redox potential, to be susceptible to oxidation by -02', for instance the prhydroquinones (Rao and Hayon, 1973) and catechols. Nevertheless, it was possible to obtain accurate measurements of electron fluxes associated with these donor/ascorbate systems by abolishing donor-oOZ' interactions with $00. Since the curves of SOD inhibition of 02 uptake always reached a well-defined plateau and the P/O2 ratios measured in these plateau regions were remarkably uniform with a variety of donors (Table I, Appendix V), it seems likely that 02 uptake in the presence of saturating amounts of SOD provides an accurate measure of electron flux. See Table IV. Benzidine is a unique donor of electrons to PS II in that the photooxidation of this compound can be followed for a period of time (90 seconds) in the absence of ascorbate without inducing a cycle (i.e. unmeasured electron transport from the donor to the product of the donor oxidation). The P/O2 ratio obtained was 1.05 (Table I, Appendix V). Even in the presence of very low concentrations of ascorbate (0.2 mM) the observed P/O2 was 1.0 (Figure 4, Appendix IV). Predictably, the presence of SOD in the reaction mixture, under these conditions, had virtually no effect on the O2 consumption associated with the photo- oxidation of benzidine. Yamashita and Butler (1969) also tested benzi- dine using tris-washed chloroplasts, NADP+ and anaerobic conditions; they observed P/e2 ratios similar to those reported above. In view of the fact that the P/e2 ratios observed during the oxidation of all of these exogenous donors are almost identical to the P/e2 ratio observed in the overall Hill reaction, it seems quite certain that the transfer of electrons from the donors to MV utilizes both sites 64 of energy conservation. Since the true P/e2 ratio associated with the transport of electrons through both photosystems is near or just over 1 regardless of the donor employed, it is clear that Site II does not have a specific requirement for the oxidation of water. This suggests the following model of the chemiosmotic mechanism of energy conservation in chloroplasts. Exogenous electron donors, with either amino or hydroxyl groups as the oxidizable portion of the molecule, might mimic water by also serving as hydrogen donors. That is, if the machinery of water oxidation operates in such a way that the protons released by water oxidation are discharged to the inside of the lamellar membrane, the same directional discharge of protons might occur when the exogenous hydrogen donors are oxidized by PS 11. Such a chemiosmotic view is supported by the discovery of Gould and Izawa (1973b, 1974) of the involvement of a light-induced reversible proton pump in the electron transport reaction H204+ PS II + DBMIB. As mentioned earlier in the thesis, this reaction is believed to involve only that portion of the natural redox chain up to and including the plastoquinone pool. With the availability of the newly developed techniques of using exogenous electron donors discussed above, a crucial test of the feasibility of this chemiosmotic model of Site II seemed possible. There are a number of substances whose oxidation does not involve pro- tons, substances which may be called "pure electron donors.“ Since the model predicts that the inward discharge of H+ from hydrogen donors (including water) represents the initial event in energy conservation at Site II, it follows that no energy conservation should take place at this site when PS II oxidizes non-proton-producing substances. Thus, 65 when electrons are transferred from these pure electron donors through both photosystems to MV, the ATP formed should be the product of Site I only and operate with an efficiency (P/ez) characteristic of this site alone, that is to say, about 0.55. (The value found to be associated with a large number of DCMU-insensitive reactions which are believed to include only one of the two sites of ATP formation (Site I) is 0.55 t 0.1.) In the course of the studies described in this thesis, it was ‘ discovered that PS II of NH OH-treated chloroplasts can oxidize ferro- 2 cyanide and iodide ions. The 02 electrode tracings presented in Figure 1A of Appendix VI depict some of the reaction characteristics of ferrocyanide ' oxidation. When high concentrations (>10 mM) of ferrocyanide were included in the reaction mixture, a rapid consumption of 02 occurred. When catalase was added to the reaction vessel after completion of illu- mination one half of the 02 consumed during the light was released. This is consistent with the assumption that all of the 02 consumed was reduced to the level of H202. The last tracing (Figure 1, Appendix VI) demon- strates the high DCMU sensitivity of this reaction, thus implicating PS II involvement in the reaction. The addition of 300 units SOD/m1 had no effect on the ferrocyanide-supported 02 consumption. This insen- sitivity to $00 is not surprising in light of the relatively high E; of ferrocyanide/ferricyanide (0.43 V) and the fact that no ascorbate was employed as a reductant pool: presumably, there was nothing present which the superoxide radicals could oxidize. Therefore, we conclude that in this system two electrons generated by the light-dependent bio- logical oxidation of ferrocyanide are utilized exclusively to reduce 66 02 to H202. Since the uptake of one molecule of molecular oxygen corresponds to the transport of a pair of electrons from ferrocyanide to MV, we can again conclude that P/O2 = P/ez. The rate of reduction of MV by electrons from water is com- pletely unaffected by high concentrations (30 mM) of ferrocyanide ions (Figure 18, Appendix VI). Moreover, the ATP formation associated with the H20 to MV transport proceeded unhindered by the high concentration of the weak reductant. From these data, it seems certain that ferro- cyanide cannot replace water as the electron source as long as the water- oxidation mechanism is unimpaired. Furthermore, we may conclude that ferrocyanide, even at quite high concentrations, has no uncoupling or other deliterious effects on the ATP synthesizing system. We were able to demonstrate that ferricyanide accumulates in the light in approximately the amounts expected from the extent of ferro- cyanide-supported 02 uptake. Illumination of the reaction mixture con- taining ferrocyanide resulted in irreversible absorbance changes in 400-450 nm region, and the wavelength dependence of these spectral changes, corrected for reversible scattering changes, corresponded well with the absorbance spectrum of ferricyanide (Figure 2A and B, Appendix VI). The absorbance change at 420 nm resulting from 60 second illumina- tion was equivalent to the formation of 54 nmoles of ferricyanide in the 2 ml reaction mixture. A duplicate reaction mixture consumed 24 nmoles of 02 (48 nequiv) during the same illumination period. The tracing of ferrocyanide-supported 02 consumption presented in Figure 1A of Appendix VI shows a slight deviation from linearity as the reaction proceeds. To more clearly resolve the apparent non-linear 67 kinetics of this reaction a time course of 02 uptake and ATP formation were obtained by "yield determinations," using a series of identical reaction mixtures illuminated for different periods of time. The very fast initial phase of O2 consumption apparent from the reconstructed curve (Figure 3, Appendix VI) had been masked in the tracings presented in Figure 1A (Appendix VI) by the slowly responding membrane covered oxygen electrode used for these measurements. The rate of electron transport in this experiment, computed on the basis of total 02 uptake resulting from five seconds of illumination, was 220 uequiv/h/mg Chl. Thus, at t = 0 the rate of this reaction may closely approach the rate of the Hill reaction. The reason for the non-linear kinetics of this reaction will be dealt with below. The ratio of the non-linear ferrocyanide-supported electron transport and phosphorylation reactions (P/Oz or P/ez) is, however, quite uniform for about sixty seconds, the value of the ratio being about 0.6. The value of the P/O2 ratio rises slowly from this point on (Inset Figure 3, Appendix VI). The explanation of the rise in P/O2 probably involves the accumulation of the oxidation product of ferro- cyanide, ferricyanide. As ferricyanide accumulates, it begins to accept electrons and thereby partially replaces MV as the electron acceptor. This transfer of electrons from ferrocyanide to ferricyanide is, of course, not measured as 02 consumption yet it supports ATP formation. The obvious consequence is an inflation in the P/Oz value so that P/OZ > P/ez. Figure 4 of Appendix VI shows that such a conversion from methylviologen-mediated to ferricyanide-mediated electron transport can indeed by induced by exogenous ferricyanide. A concentration of added 68 .m.e eaeaaexa HHeeeoee Owen; age e n H H< «.4 .Lepmew epew megs» e» e3» xpee23m loge use: o u e we menus FewpweH .meueg mewgmwewewe xweweeg we Amomuov meewe> eemeem>eumeww ea .woe awepeswxeceee we w>z + omzwpeewu newxgesemece mueum weeeum eweaeecs ea eceew we: meeseesee ewes» we coweecucweee eeueewe one a - mN.e e - eH e - e4 e.eH 4H xweceee< .H epeew we ecmmew cw =e>wm age meewpwecee :ewueeom .mumepeegewcu emuemgpuzomzz we egocee meeeemexe we :ewueewxe HH Empmzmeuege cw cewuesgew ew< ece uceemceee :ecpeeHm .>H oweew 69 ferricyanide of 30 uM was sufficient to reduce O2 consumption by 50% without any decline in ATP formation rate. Careful scrutiny of the data will show that the rise in P/O2 seen in Figure 3 (Appendix VI) is about the extent predicted from the concentration of ferricyanide deduced from the data of Figure 4 (Appendix VI). The concentration of ferrocyanide ions required to saturate the electron transport and phosphorylation reactions is astonishingly high, about 30 mM (Figure 5, Appendix VI). Nevertheless, once the reaction becomes measurable (5 mM) the efficiency (about 0.55) with which the electron transport is coupled to ATP formation is quite constant up to a concentration of 60 mM. The consistency of this phosphorylation efficiency over such a wide range of donor concentrations further sub- stantiates our contention that the low P/e2 values associated with this reaction are not due to uncoupling by the ferrocyanide ion. Nor is it likely that any kind of product inhibition can be responsible for the low phosphorylation efficiencies, since the P/e2 value is essentially independent of the reaction time (60 seconds). It appears that the high concentration requirement for ferrocyanide is primarily a function of the low lipid solubility of this highly charged ion. Surely, ferro- cyanide must have a rather extensive sphere of hydration which should tend to delocalize the charges; this may be responsible for the limited access to the lipid laden membranes which the ion apparently does enjoy. The implication is clearly that the site of oxidation of these exogenous donors, and presumably water, is "buried" in a hydrophobic region of the lamellar membrane. The relative inaccessibility of this site to ferro- cyanide ions could explain the biphasic kinetics of the ferrocyanide- 70 supported reactions. The initial phase of the reaction when the rate of oxidation is rapid and approaches that of the Hill reaction may repre- sent the oxidation of that portion of the ferrocyanide which has already permeated the membrane (possibly into the internal space of the thylakoid) while the subsequent much slower phase may represent a diffusion limited process. We have found that concentrations of digitonin which do not uncouple enhance the rate of oxidation of sub-saturating concentrations of ferrocyanide; this observation is consistent with the existence of a strong permeability barrier limiting the accessibility of the PS II oxidation site to ferrocyanide. The iodide ions seem to permeate the lamellar lipid phase somewhat more easily than do ferrocyanide ions, as suggested by the lower concentration requirement (Figure 8, Appendix VI) and by the more linear kinetics (Figure 7, Appendix VI). However, even the 15 mM iodide required to saturate the electron transport and phos- phorylation reactions is in striking contrast to the much lower concent- rations required for lipid soluble donors (e.g. catechol, praminophenol, etc.). See Table IV for complete listing of concentration requirements. The concentration of ascorbate needed to optimize its donation to PS II (5 mM) is roughly intermediate between the ions (ferrocyanide and iodide) and the lipid soluble donors. The implication is that the PS II oxidizing equivalents are not available to the external medium and may well be located near the inside of the lamellar membrane. This is of crucial importance to the argument that Site II represents internal production of hydrogen ions from the internal oxidation of hydrogen donors: if water and other hydrogen donors are to release protons into an internal space 71 there should be a requirement, although perhaps not absolute, that the site of oxidation be at or near the inner surface of the thylakoid mem- brane. Since it seemed likely that the diminished P/e2 ratios accom- panying ferrocyanide oxidation might result from only one of the two sites of ATP formation being operative, we were interested in another diagnostic test (that is something other than P/e2 values one half of those of the overall reaction). Gould and Izawa (1973b) have presented evidence that the portion of the redox chain employed in the transfer of electrons from DCIPH2 to MV via PS I includes only Site I. Phosphor- ylation and its efficiency when associated with this pathway peak at pH 8 or slightly above (where P/ez = 0.65) and sharply decline toward zero between pH 6.5 and 7.0. This very marked sensitivity of the phos- phorylation efficiency in PS I-mediated reactions differs very much from the virtual pH insensitivity of phosphorylation efficiency in PS II reactions such as the transport of electrons from H20 to DBMIB. The pH profile of phosphorylation efficiency in the Hill reaction (H20 + MV) can be closely reproduced by summing the profiles of the isolated sites. Both phosphorylation and its efficiency peak at pH 8 and quickly approach zero below pH 7 when associated with ferrocyanide oxidation (Figure 6, Appendix VI). Clearly, these data suggest that only Site I is operative in ferrocyanide-mediated chloroplast reactions. The slight distortion of the pH curves toward the acid side is discussed in Appendix VI. Electron transport and phosphorylation supported by iodide oxidation proceed more linearily than do the ferrocyanide-supported reactions. The efficiency with which the electron transport is coupled 72 to ATP formation (P/ez) initially exceeds 0.5 (15 seconds) but falls slightly with reaction time (Figure 7, Appendix VI). The P/e2 ratio was virtually independent of the concentration of KI (Figure 8, Appendix VI) and, therefore, the possibility of the low P/e2 resulting from uncoupling by iodide is unlikely. Furthermore, concentrations of KI up to 30 mM were tested and found to have essentially no uncoupling effect on the H201+ MV system (in normal 02-producing chloroplasts). As in the ferrocyanide-oxidizing system, the pH sensitivity displayed by the P/e2 ratio computed from the iodide-supported reactions has the character- istics expected for an isolated Site I (Figure 9, Appendix VI). Electron transport and phosphorylation of this exogenous electron donor reaction share a common pH optimum at 8.0 and a marked stimulation of electron flux by phosphorylation was observed. The iodide-oxidizing system was not as fully characterized as the ferrocyanide-oxidizing system because attempts to detect the oxida- tion product of iodide, free iodine (or 13') were unsuccessful. Any 12 formed would likely have been consumed by the reaction medium, since it was found that both the chemicals of the reaction mixture, and the chloroplasts themselves reduced added 12. In the MV-reducing system, there was no indication of H202 formed being consumed by a reaction with I2 produced during a reaction of 30 seconds. Neither of the more weakly reducing halogen ions 01' or Br" was able to substitute for I" as the electron donor for PS II. However, several metal complexes were oxidized by PS II in NHZOH-treated chloro- plasts and the electron transport supported phosphorylation with an efficiency of 0.5-0.6 (Table IV). The lower P/e2 ratios associated 73 with the oxidation of two of the complexes (Mn(dipyridyl)2 and Mn (EDTA) ) are apparently due to partial uncoupling. Nevertheless, the full efficiency in the absence of this uncoupling probably would not exceed 0.6. Characterization of the reactions supported by the metal complexes has not been completed. Table IV presents a summary listing most of the exogenous donors tested and found to donate electrons pri- marily to PS II. Note that, with the exception of ascorbate and the two metal complexes mentioned above, these reactions fall into two dis- tinct categories; those which support ATP formation with an efficiency (P/ez) of about 1 and those with a P/e2 of about 0.5. In the above discussion it has been postulated that the lower phosphorylation efficiencies result from the less efficient formation of hydrogen ion gradients. There is direct experimental evidence that this is indeed so. The pH of a weakly-buffered suspending medium should rise irreversibly during the methylviologen-mediated aerobic photo- oxidation of ferrocyanide according to the formula: 4- + 2e' transport 3- 2Fe CN + 2H + 20 >- 2Fe CN + H 0 ( )5 2 methylviologen ( )5 2 2 Moreover, according to our model there should also exist a reversible light-induced proton translocation due to the energy conservation reac- tions at Site I. As predicted, proton uptake did occur upon illumina- tion, and it had two components: a gramicidin-insensitive irreversible component (Figure 108, Appendix VI) and a gramicidin-sensitive reversible component (Figure 10A, Appendix VI). However, as will be shown later, the efficiency of the reversible proton pump H+/e' was only about 1.0, which is significantly less than the 1.7 measured for the Hill reaction 74 but agrees well with the H+/e' ratio observed with isolated PS I reactions (Section III of thesis). Because of the reactivity of the oxidation product of iodide, the methylviologen mediated photooxidation of I' did not involve a gram- icidin-insensitive irreversible proton consumption (Figure 10C and 0, Appendix VI). Apparently 12 is produced and then quickly reduced again by the reaction mixture with a stoichiometric production of hydrogen ions (AH2 + I2 i A + 2H+ + 21') which compensates for the proton con- sumption due to electron transport (21' + 2H+ + 02-————-¢) 12 + 2H202)' Consequently, the only detectable proton change is the gramicidin- sensitive reversible proton pump (Figure 10 C and 0, Appendix VI). The production of acid by I2 reduction in the reaction mixture is confirmed in trace E (Figure 10, Appendix VI). Clearly, the oxidation of ferrocyanide to yield ferricyanide involves only a valence change (at physiological pH's) due to the very stable natures of both the ferro- and ferri- forms of the complexes. No detectable H+ or 0H' changes take place as indicated by the pH insen- sitive E; value. Moreover, the oxidation of iodide to iodine does not induce proton changes except above pH 9.0 where IOH begins to occur. In contrast to these low P/e2 yielding ions, all of the PS II donors which gave P/e2 values above 1.0 (e.g. water, catechol, praminophenol, etc.--see Table IV) are hydrogen donors whose oxidation involves nearly stoichiometric production of protons and electrons at physiological pH's (AEO'lApH = -o.osv or AH+/ e' = 1) (Clark, 1950). Therefore, if we ignore for the moment the complication introduced by the H+ production in 12 reduction, our findings provide a strong experimental basis for 75 believing in a chemiosmotic type of energy conservation at Site II or at very least an energy conservation reaction which requires H+ produc- tion. This line of reasoning may be extended to encompass the inter- mediate efficiency reported for ascorbate oxidation. At pH's above 4.5 the oxidation of the ascorbic acid anion to the neutral dehydroascorbate molecule (Bdhme and Trebst, 1969) liberates only one proton per pair of electrons (AH' —+ A + H+ + 2e'). Thus, according to our model, it might be that Site II operates at half efficiency when ascorbate is the electron donor. It may be worthy of mention that ferrocyanide and iodide differ from the hydrogen donors in another respect; these ions are one-electron donors whereas the oxidation of the hydrogen donors results in the release of two electrons per molecule. We feel this distinction is nominal because oxidation of several hydrogen donors by PS 11 has been demonstrated to occur in one quantum steps requiring no charge accumula- tion (Bennoun and Joliot, 1969; Babock and Sauer, 1971). Indeed, it is believed that the oxidations of organic substances, although ultimately requiring the removal of two electrons, almost always occur in one electron steps. Since the reduction of 12 is accompanied by an acid production, iodide may be considered as a proton-producing electron donor. Never- theless, the data seem to indicate that Site II in NHZOH-treated chloro- plasts is not operative when iodide is the electron source. If Site II does indeed consist of a "redox loop" producing H+ inside as predicted by the chemiosmotic mechanism, the implication is that the protons produced by reduction of I2 are not available for phosphorylation and 76 hence presumably not inside. Such a situation seems plausible. The region of the membrane where the strong oxidant of PS II is produced may be so devoid of oxidizable substances as to afford I2 a sufficient life- time to diffuse away. The reduction of I2 could then occur where the H+ produced could not be used in energy conservation reactions, perhaps on the outer surface of the membrane where abundant oxidizable substances are known to exist. This is consistent with the observation that no membrane components essential for electron transport and phosphorylation seem to be destroyed by the highly reactive 12. The iodide oxidation and associated phosphorylation proceed linearly for at least 60 seconds, during which time the amount of I2 produced and consumed is roughly equivalent to the chlorophyll present. SECTION III ENERGY CONSERVATION AT SITE I IN ISOLATED CHLOROPLASTS . SECTION III ENERGY CONSERVATION AT SITE I IN ISOLATED CHLOROPLASTS INTRODUCTION A good deal of information has accumulated suggesting that there may be some rather fundamental differences in the mode of energy transduction at Site I and Site II. In the first section of the thesis, data was presented which demonstrated a difference in the ability of these sites to influence the rate of electron flux (Gould and Ort, 1973). Segments of the electron transport chain containing Site I can support PS I dependent-electron transfer when an appropriate electron donor system (e.g. ascorbate/DCIP) and electron acceptor (e.g. MV) are supplied. The rate of electron flux through isolated Site I is greatly enhanced when phosphorylation occurs concurrently or when uncouplers are added. On the other hand, when Site II is isolated from Site I as in the PS II-dependent reaction H20 +PDox in KCN-treated chloroplasts, the already very high rates of electron transport are not further enhanced by concomitant ATP synthesis or by uncouplers. The difference in the activities of these two sites at low pH's was first noticed by Saha gt_al, (1971). This observation was greatly extended by Gould and Izawa (1973). They demonstrated that the difference in the response of the phosphorylation efficiency of these 78 79 sites to pH was exceedingly pronounced. The efficiency of Site II was virtually unchanged through the pH range 6.0 to 9.0 (with a P/e2 value of approximately 0.35). In contrast, the P/e2 ratio at Site I peaked around 8.0 with a value of approximately 0.6 but declined sharply with pH, being near zero by pH 7.0. Data will be presented in this section of the thesis showing that Site I and Site II also differ in their sensitivity to the energy transfer inhibitor Hg++. The sensitivity of phosphorylation in the overall Hill reaction to Hg++ (Izawa and Good, 1969) can probably be accounted for completely by the sensitivity of Site I. When Site II is isolated, phosphorylation proceeds unhindered by Hg++. Prompted by the knowledge of these differences in the sites of energy conservation, we set out to attempt to learn something of the mechanism of energy conservation at Site I. Because of the very strong evidence for the intimate involvement of protons in energy conservation at Site II, we felt it unlikely that nature would devise a completely different mechanism to make ATP elsewhere in the chloroplasts. Thus, the major purpose of the investigation was to examine the role of pro- tons and proton gradients in the conservation of redox energy at Site 1. One approach to this problem was to examine the correlation between the efficiency of ATP formation (P/ez) at Site I with the efficiency of proton accumulation (H+/e') at the site. To measure proton accumulation, the light-induced reversible loss of protons from the surrounding weakly buffered medium was monitored. There is some experimental evidence which indicates these hydrogen ions lost from the medium are translocated to the inner thylakoid space (e.g. Rumberg and 8O Siegel, 1969; Rottenburg and Grunwald, 1972; Portis and McCarty, 1974). Gould and Izawa (1973b) discovered that there is a light- dependent, reversible proton translocation associated with the PS II- dependent transfer of electrons from water to DBMIB. This partial reaction turned out to be an extremely convenient system in which to examine the PS II-dependent proton pump (Gould and Izawa, 1974). The initial phase of both the DBMIB-mediated H+ uptake and 02 evolution were linear for more than 3 seconds as analyzed by a "flash yield tech— nique." (In contrast, H+ uptake has a preliminary fast phase during the Hill reaction.) Using initial rates of 02 evolution and H+ uptake a H+/e' ratio close to 0.5 was determined for the PS II-dependent reaction. Furthermore, this value did not change significantly when the pH of the suspending medium was varied between 6 and 8. As predicted by the chemiosmotic mechanism, the H+/e’ value is quantitatively similar to the P/ez value of this reaction (0.3-0.4) and shows the same lack of pH sensitivity. The extent of the DBMIB-mediated pH rise was markedly reduced by concomitant phosphorylation or arsenylation, again in con- formity with the postulates of the chemiosmotic theory. A similar analysis of the correlation between the efficiency of ATP synthesis and the efficiency of proton accumulation at Site I could help to explain some of the differences between the two sites which have been observed. The initial task in this investigation was to obtain a reliable measurement of the efficiency at Site I. Hereto- fore, the accurate measurement of the energy conservation efficiency of Site I has been bedeviled by three factors. First, the rate of electron 81 flux of photosystem I-dependent reactions can be obscured by the involvement of the superoxide radical generated by the aerobic oxidation of the low potential electron acceptors used (Ort and Izawa, 1974: Epel and Neumann, 1973) since the rate of reduction of these compounds is routinely followed as 02 consumption. When the superoxide radical is allowed to react with the exogenous donors the rate of 02 uptake is enhanced and consequently the rate of electron transport is overestimated. The end result is an underestimation of the efficiency of Site I. A second factor which can prevent the accurate determination of the effi- ciency at Site I is the condition of the lamellar membranes. In certain chloroplast preparations ascorbate/DCIP has at least two sites of dona- tion of electrons to the PS I, one before and one after Site I (Arntzen §t_al,, 1971 and Gould, 1974). The more "intact" the lamellar membranes are, the smaller the non-phosphorylation component of electron donation from DCIPH2 becomes. Clearly, situations similar to the one described above would result in an overestimation of the electron flux through the energy conservation site and, thus, an underestimation of the effi- ciency of the site. The third problem encountered is the possibility of cyclic electron transport. That is, after the donor has been photo- oxidized by System I it may compete with the exogenous electron acceptor. The portion of the electrons intercepted by the oxidized form of the donor would be unmeasured and a spuriously high P/e2 would result. With these complications in mind, a careful characterization of Site I phos- phorylation efficiency was undertaken and is presented in this section of the thesis. 82 The chemiosmotic model of energy transduction predicts that the efficiency of phosphorylation should be quantitatively related to the efficiency of the light-induced reversible H+ uptake (H+/e') as seems to be the case for Site II. The observed H+/e’ ratio at Site I is reported here. RESULTS AND DISCUSSION Site-SEecific Inhibition of Energy onservation Reactions It has been well established in this thesis that electron flux- through Site I can be greatly influenced by phosphorylation. The point has been made that the control imposed by Site I can be best explained' in terms of a build up of "high energy state." Thus, anything which aids in the dissipation of the "high energy state" (ADP + P1 or uncoup- lers) will permit electron flux through Site I to proceed more rapidly. Conversely, anything which prevents the utilization of this “high energy state" should reduce the rate of electron transport through Site I to the lower level which is characteristic of non-phosphorylating condi- tions. Compounds which prevent ATP formation in chloroplasts and reduce the rate of electron transport to the level of the non-phosphorylati rate (e.g. phlorizin--Winget gtgal,, 1969) are known as energy transfer inhibitors. They do not affect uncoupled or basallelectron transport and are, therefore, thought to prevent some reaction directly involved in ATP synthesis. Izawa and Good (1969) discovered thgt'prchloromercuri- benzoate and mercuric ion both behave as energy transfer inhibitors, but with one difference. These compounds only inhibit ATP formation and 83 84 phosphorylating electron transport by approximately 50%. This 50% inhibition level, once attained, is very resistant to any further inhibi- tion by increased concentrations of the inhibitor. Only when the concentration exceeds by many times the amount necessary to cause the initial 50% inhibition level does further decline in rates of electron transport and ATP formation occur. At the higher concentrations the effect is probably on some redox reaction because uncoupled electron transport is also affected. These observations are confirmed in this thesis. At concentrations of less than 0.1 uM HgClZ/mg Chl the inhibi- tion of the H20 to FeCy-supported electron transport and phosphorylation had reached its maximum level. There was no further inhibition when the HgCl2 concentration was increased five times and the rate of uncoupled electron transport was not attenuated by this high concentration of inhibitor (Figure 1A, Appendix VII). The fact that only one half of the ATP formation is abolished by these unique energy transfer inhibitors suggested that perhaps only one site of ATP formation is mercury sensitive. As serendipity would have it, Hg++ does appear to be a site-specific energy transfer inhibitor. However, one site (Site II) is completely insensitive to Hg++ while the other (Site I) displays the 50% levels of inhibition observed for the Hill reaction. Thus, we know no more now than we did before we began the investigation about the nature and cause of the 50% sensitivity. (However, see Bradeen and Winget, 1974.) The rate of reduction of PDox and the associated ATP formation (in the presence of 0.5 uM DBMIB) were not diminished by HgCl2 (Figure 1C, Appendix VII). Similar insensitivity to HgCl2 was observed with 85 other PS 11 reactions mediated by DBMIB, DMQ or DADox (Table I, Appendix VII). The absence of inhibition by HgCl2 with Class III acceptors almost certainly does not result from the acceptors themselves preventing the inhibition. In all cases, chloroplasts were incubated with HgCl2 for 30 seconds before the addition of the acceptor system. Since -SH compounds can reverse HgC12 inhibition (Izawa and Good, 1969), It 15 probable that H9++ is reacting with a membrane (perhaps coupling factor) sulfhydryl group. Thus, it seems unlikely that binding between Hg++ and the quinonediimides and substituted benzoquinones used as Class III acceptors would be strong enough to reverse the inhibition. This prediction becomes almost a certainty when we consider that the UV spectra of DMQ, DADox’ PDox and DBMIB remained virtually unchanged in the presence of high concentrations of HgCl2 (33 uM). Clearly, the acceptor did not bind Hg++ to a significant extent. The sensitivity of Site I to Hg++ is shown in Figure 18, Appendix VII. Quantitative Relationship Between Photosystem I Electron Transport and ATP Formation As pointed out in the introduction to this section of the thesis, a crucial step in the evaluation of energy conservation at Site I is an accurate measure of the efficiency of ATP formation (P/ez) at that site. Herein it will be demonstrated that under proper condi- tions Site I can be studied apart from Site II. The P/e2 at Site I can be established if the transport of electrons from water is inhibited by DCMU and an exogenous source of readily available electrons is supplied. 86 Most substances used to supply electrons to PS I are of relatively low redox potentials (i.e. tend to release electrons readily). Very often excess ascorbate, which is not directly oxidized by PS I (Ort and Izawa, 1974), is included in the reaction to maintain the primary donor in its reduced state. Because these PS I reactions are commonly mediated by low potential oxidants such as MV, we must suspect that the measurement of electron flux by 02 consumption (to form H202) may be complicated by the involvement of the superoxide radical. As demonstrated earlier in the thesis, .02' is produced via the aerobic oxidation of photoreduced MV and this radical can then react with either or both constituents of the exogenous donors. In such cases, 02 consump- tion is not an accurate measure of electron transport rate (see equa- tions 1 through 6 in Results, Section II). The rate of oxygen consumption supported by the ascorbate + diaminodurene (DAD) +'MV reaction decreased by over 30% when 600 units SOD/m1 were included in the reaction mixture without any reduction in the rate of ATP formation (Figure 6, Appendix V). When ascorbate/ diaminotoluene were the exogenous donors employed, the rate of 02 con- sumption was reduced more than 40% by the inclusion of SOD. Oxygen uptake and phosphorylation in these two PS I reactions are faster, by almost an order of magnitude, than the PS II-mediated reactions described in Section II. Partly for this reason, much higher levels of SOD are required to suppress the donor -02' interactions. However, the chemical nature of the donor seems to determine, to a large degree, the extent it reacts with the superoxide radical. The reaction of diaminodurene or diaminotoluene with ~02" competes more effectively with the dismutation 87 reaction than does the reaction with ascorbate or with any of the higher potential PS II donors. The effects of SOD on the 0 consumption sup- 2 ported by a number of other exogenous electron donors to PS I were tested. In all cases the 02 consumption was markedly suppressed by $00 but rates of ATP formation remained unaltered (Table I, Appendix V). Clearly, unless precautions are taken to abolish donor- superoxide radical interactions electron transport cannot be reliably measured as 02 consumption. When such precautions are taken, there is a remarkable uniformity in the P/e2 ratios with these PS I-dependent reactions, 0.58 f 0.1 (see Table V). Unfortunately, the superoxide complication is not the only source of error which can befall the determination of Site I phosphoryla- tion efficiencies. Cyclic electron transport is a complication which is unique to these exogenous donor reactions since once the donor has been oxidized it may begin to function as an electron acceptor. The inclu- sion of ascorbate is designed to avert such cyclic flow by maintaining the primary reductant in its reduced form. However, even this precaution in some instances does not appear to be sufficient to prevent cyclic electron flow. Indeed, the ATP formation accompanying the oxidation of reduced 2,6-dichlorophenolindophenol by PS I has been reported to be largely independent of any added electron acceptor, even when excess ascorbate is present (Trebst and Eck, 1961: Wessels, 1964; Avron, 1964). This total independence of the rate of DCIPHz-supported ATP formation from added acceptor does not appear to be true of the more intact chloro- plasts which can be prepared today. But a small amount of cyclic electron flow may account for the fact that DCIPH2 oxidation seems to 88 be coupled to phosphorylation with a slightly higher efficiency than the oxidations of the other donors tested (Table V). If so, the higher efficiency is spurious. The use of 3,3'-diaminobenzidine (DAB) should have the major advantage of eliminating unmeasured cyclic electron flow because the primary oxidation product of DAB should not be available as a competing electron acceptor. This is because oxidized DAB is known to form an insoluble polymer (Seligman gt;al,, 1968; Nir and Seligman, 1970). Goffer and Neumann (1973) examined the System I-dependent photooxidation of DAB and the accompanying ATP formation. They found that this reaction supported phosphorylation with an apparent efficiency approaching 1.0. Judging by the method of chloroplast isolation employed by these investi- gators one would predict that the overall non-cyclic (H20 + MV) P/e2 would not exceed 1.0. Therefore, such a high efficiency of ATP formation for a PS I reaction is difficult to reconcile with the emerging picture of chloroplast energy conservation and is not consistent with the P/e2 values reported in Table V for other PS I reactions. These considerations prompted us to reexamine the phosphoryla- tion efficiencies of PS I-dependent reactions using DAB and other exo- genous donors. It was discovered that the high P/e2 value associated with DAB oxidation reported by Goffer and Neumann can be explained on the basis of a bizarre short-lived cyclic electron flow. The efficiency of phosphorylation during DAB oxidation was found to be actually about 0.5 and is, thus, no different from the efficiency observed with any other PS I donor. 89 As previously stated, reactionsinvolving the donation of electrons to PS I are usually carried out in the presence of excess ascorbate to prevent the accumulation of the oxidized form of the donor since the newly formed oxidant can accept electrons and, in so doing, contribute an unmeasured component to the electron flow. The importance of supplying excess ascorbate is illustrated in Figures 1 and 2 of Appendix VIII. When ascorbate was omitted from the DAD +~MV system the apparent value of the P/ez ratio (actually P/Oz) was high and increased with illumination time presumably due to the accumulation of DADox (Figure 2, Appendix VIII). Indeed 10 second illumination produces an amount of DADox which intercepts more than half of the total electron flux even in the presence of 0.1 mM MV. Apparently, oxidized diamino- durene formed in situ is an exceedingly efficient electron acceptor because the DADox concentration at the end of 10 seconds illumination should be no more than 20 uM. The change in P/O2 in the absence of ascorbate is dramatized by the upper inset of Figure 2, Appendix VIII where the change in ATP concentration OAATP) and 02 concentration @302) are presented for each illumination period. However, when ascorbate is included in the reaction mixture and the requisite SOD has been added, the P/O2 ratio associated with the MV-mediated oxidation of DAD is time- independent and has a value of approximately 0.55. It is quite clear that the increase in the P/O2 ratio is not due to an increase in phos- phorylation rate but rather to a diminution in the rate of 02 uptake (Figures 1 and 2, Appendix VIII). It seems certain, therefore, that in the absence of ascorbate oxidized DAD is competing for electrons with 90 NV and, as a consequence, 02 uptake is not a full measure of the electron flux occurring (i.e. P/02>P/e2). Diaminobenzidine behaves quite differently from diaminodurene as a donor of electrons to PS I: The maximum rate of electron transport with DAB is only about one fourth the rate with DAD, and even this rate is not sustained (compare Figures 1 and 2 with Figures 3-5, Appendix VIII). The apparent efficiency of ATP formation (P/OZ) is dependent on the concentration of DAB both in the presence and absence of ascorbate (Figure 3, Appendix VIII). This concentration-dependence seems to be the result of a short-lived cycle which is more significant at low donor concentrations, perhaps for reasons to be discussed later. When the conditions used by Goffer and Neumann (1973) were selected (these conditions included suboptimal 0.14 mM of DAB, no ascorb- ate and 10 second illumination), we observed a P/O2 ratio almost exactly the same as they reported (0.83). However, Goffer and Neumann apparently were not aware of the peculiar effect of the DAB concentration on the apparent phosphorylation efficiency and were not aware of the influence of ascorbate thereon (Figure 3, Appendix VIII). The cyclic nature of the electron transport which occurred at the onset of illumination when DAB was the donor is demonstrated in Figure 5, Appendix VIII. In the absence of ascorbate (Figure 5a, Appen- dix VIII) this cycle is particularly evident and, hence, unrealistically high P/O2 values were observed (>2.0) for short illumination periods. The high P/O2 clearly resulted from a temporary suppression of oxygen uptake, indicated by the S-shaped uptake curve. The suppression of oxygen uptake illustrated in Figure 5a (Appendix VIII) seems to result 91 from oxidized but not yet polymerized DAB partially replacing MV as the electron acceptor, since the effect is largely removed if ascorbate is present to reduce the photooxidized DAB (Figure 5b, Appendix VIII). The lowering of the P/O2 by added ascorbate is not simply due to an enhancement of 02 uptake resulting from ~02'lascorbate interactions because $00 was included in all reactions (see Figure 4, Appendix VIII). This interpretation of the high P/O2 ratios with DAB is born out by experiments conducted in the absence of MV (Figure 6, Appendix VIII). During the initial phase of illumination (<25) DAB-supported phosphorylation is practically independent of the presence or absence of MV which suggests that unpolymerized DABox is accepting most of the electrons near the beginning of the reaction. Cyclic transport in the DAB-oxidizing system is not as readily interpretable as is cyclic transport in the DAD-oxidizing system. In the case of DAD-mediated reactions the accumulating oxidized DAD accepts an increasing proportion of the electrons, competing very effectively with MV. The accumulation of DADox can be easily arrested by the inclu- sion of ascorbate, in which case the oxidized donor is reduced almost as soon as it is formed. Thus, in the presence of ascorbate there seems to be absolutely no cyclic electron flow associated with the DAD-oxidizing system and, with the addition of SOD, electron transport can be accurately determined from 02 uptake data. In the case of the DAB-oxidizing system, cyclic transport seems to be maximal near the inception of the reaction when the accumu- lation of oxidation products must be minimal. This curious behavior of DAB is probably related to the chemical nature of the primary oxidation 92 product which is both a strong oxidant and a highly reactive substance. Being a strong oxidant and formed in situ it should compete especially favorably with MV for electrons from PS I, much as does DADox' However, as the oxidation product of DAB accumulates it, unlike DAD, begins to polymerize (Seligman §t_al,, 1968). This polymerization, like many polymerizations, is probably autocatalytic and, therefore, once polymeri- zation has started, the level of oxidized DAB should fall and the cycle diminish. For this reason, factors which limit the accumulation of oxidized DAB, such as suboptimal concentrations of DAB and, above all, short reaction times might be expected to increase the proportion of cyclic transport by delaying or preventing the onset of polymerization (Figure 5a, Appendix VIII). However, in the presence of excess ascorb- ate, at no time would the concentration of the primary oxidation product of DAB reach levels capable of supporting high rates of cyclic trans- port (Figure 5b, Appendix VIII). When the contribution of such hidden cyclic transport systems has been eliminated the observed phosphorylation efficiencies of these two reactions are actually quite similar. The same is true of the photooxidation of a variety of other PS I donors (Table V) and with chloroplasts from a variety of plants (Table I, Appendix VIII). All of the electrons donors presented in Table V were photooxidized entirely by System I since the oxidation occurred in the presence of DCMU. Plasto- quinone, which presumably acts as a carrier of electrons between the two photosystems, seems also to be unnecessary since the oxidations of the donors and the associated phosphorylation reactions are completely insensitive to DBMIB (Table III, Appendix VIII). 93 Table V. Stoichiometric relationship between ATP formation and electron transport associated with the PS I-dependent oxidation of various electron donors. Reaction conditions are similar to those in Table I for the DAD to MV reaction. The concentration of donors used were: diaminodurene (1.0 mM), 3,3'-diminobenzidine (1.0 mM), 2,5-diaminotoluene (0.5 mM), tetrachlorohydroquinone (1.25 mM), 4,5-dimethyl-o-pheny1enediamine (1.25 mM), 2,6-dichloro henolindophenol (0.4 mM), ortoTidine (0.5 mM), diphenyl- hydrazine (0.5 mM), p:pheneylenediamine (0.5 mM). The concentration of ascorbate present in each system was 2.0 mM. The chlorophyll concentra- tion of the chloroplasts was 40 ug/Zml for all systems except diamino- durene, 3,3'-diaminobenzidine 810 ug/Zml) and 2,5-diaminotoluene (20 ug/ 2ml). $00 (1200 units) was ad ed to each reaction mixture. Illumination time was 20 seconds except for diaminodurene and diaminobenzidine which was 10 seconds. Rates for the exogenous donor reactions are given in nmoles O of ATP per h per mg Chl. For the reaction involving H O as the elect- r n donor, chloroplasts containing 40 pg Chl were used anO ascorbate, DCMU and S00 were omitted. Rates of the Hill reaction are reported in uatoms O or nmoles ATP per h per mg Chl. Note that when water is the donor 0 = e2 but when exogenous donors are employed 02 = e2. 02 uptake ATP P/e2 Diaminodurene 1435 790 0.55 3,3'-Diaminobenzidine 510 245 0.48 2,5-Diaminotoluene 760 445 0.58 Tetrachlorohydroquinone 72 41 0.54 4,5-Dimethyl-9:phenylene- 180 91 0.51 diamine 2,6-Dichlorophenolindophenol 138 91 0.66 (reduced) Q7Tolidine 50 26 0.53 Diphenylhydrazine 40 24 0.59 pyPhenylenediamine 135 84 0.62 (H20) (310) (385) (1.24) 94 It is important to establish beyond reasonable doubt that the site of phosphorylation associated with the oxidation of these exogenous donors is actually one of the sites associated with the Hill reaction. There are several strong reasons for equating phosphorylation at Site I of the overall Hill reaction to the PS I-dependent, exogenous electron donor-supported phosphorylation described herein. Perhaps the most obvious reason is that the efficiency of the Hill reaction can be adequately accounted for by summing the efficiency of Site II phosphory- lation and the efficiency of donor-supported phosphorylation (Gould and Izawa, 1973b). Moreover, Gould (1974) has demonstrated that the rates of photooxidation of diaminodurene and diaminotoluene are nearly doubled when phosphorylation occurs concurrently. We have observed similar stimulation of electron transport by phosphorylation when DAB or dimethyl- orphenylenediamine are the donors employed. Such stimulations, although smaller, bear a conspicuous resemblance to the stimulation of the Hill reaction (H20 +-MV) by phosphorylation which has been attributed to the relief of a rate limitation imposed by Site I, the coupling site between plastoquinone and cytochrome f_(Avron and Chance, 1967). As suggested above, these electron donors appear to interact with the redox chain after plastoquinone. In addition, McCarty (1974) has published evidence that the photooxidation of DAD involves cytochrome j, Also, as pointed out above, the sensitivity which Hill reaction phosphorylation displays to the energy transfer inhibitor Hg++ and the sensitivity of the phos- phorylation efficiency of the Hill reaction to pH is strikingly similar to the sensitivities of these reactions involving exogenous electron donors. When these observations are viewed in concert they present a 95 convincing case for the contention that the exogenous donor reactions discussed above utilize one of the energy conservation sites (Site I) of the Hill reaction. Quantitative Relationship between Photosystem I ATP Formation Efficiency and H+ Accumulation Efficiency In the introduction to this section, it was pointed out that coupling Sites I and II have several notable differences. One of the most significant of these differences is the effect of pH on the effi- ciency of phosphorylation at each site. The insensitivity of Site II phosphorylation efficiency to pH (~0.4) over the range 6.0 to 9.0 is in striking contrast to the sensitivity of Site I phosphorylation efficiency which is optimal (~0.6) at just over pH 8.0 and falls very rapidly on either side of this optimum. According to the chemiosmotic model, the initial and obligatory event in energy conservation in chloroplasts is the electron transport- dependent translocation of hydrogen ions to the inside of the lamellar membrane. If one assumes that internal protons can be used equally well to generate ATP regardless of which of the two energy conservation sites is responsible for formation of the proton gradient, it follows that the efficiency of hydrogen ion accumulation (H+/6') at each site ShOUId display the same pH profile as the efficiency of ATP formation (P/ez) at that site. This does not seem to be the case and, therefore, one must question the assumption. The work of Gould and Izawa (1974) on the correlation between H+/e‘ ratios and P/e2 ratios of Site II over the pH range 6.0 to 9.0 is 96 confirmed in Figure l. The proton uptake efficiency (H+/e') and ATP formation efficiency (P/ez) supported by the PS II-dependent transfer of electrons from H20 + DBMIB exhibits almost complete insensitivity to the medium pH (over the range 6 to 9). Note also that the quantitative relationship of these two ratios is within the range predicted by the chemiosmotic hypothesis (for H+/ATP = 2 to 3). Thus, to reiterate a point made previously in the thesis, when the close correlation between H+/e' and P/e2 ratios at Site II is viewed in conjunction with the apparent absolute requirement for proton production at that site (Izawa and Ort, 1974) the suggestion is quite strong that a chemiosmotic type of coupling occurs at Site II. f The accurate determination of H+/e' ratios at Site I is not as easily accomplished as for Site II. When Site I is isolated from Site II by the use of exogenous donors such as described in Table V, the biological proton pump is obscured by purely chemical proton changes which accompany the oxidations of the donors. This problem was averted by employing the ferrocyanide +~MV system in NHZOH-treated chloroplasts. It was established earlier in the thesis that even though this reaction requires PS II, it does not include Site II. It was also shown that no proton changes occur which are directly due to the oxidation of ferro- cyanide to ferricyanide. As shown in Figure 10 of Appendix VI, there are two components to the pH change associated with ferrocyanide photo- oxidation: an irreversible, gramicidin-insensitive proton consumption due to the formation of H202 (product of aerobic oxidation of reduced MV) and a reversible, gramicidin-sensitive Site I proton pump. The pH change supported by this system was determined by a "flash-yield" H+/e' or P/e2 1.0 0.5 97 l I 1 H20<+ PS II + DBMIB H+/e’ H. c) C) .__ 0" o P/e2 l L I 6 7 8 , PH Figure 1. Efficiency of ATP formation (P/e3) and efficiency of hydrogen ion accumulation (H /e‘ at Site II. The 2 ml reaction mixture consisted of 0.1 M sucrose, 2 mM MQCI , 50 mM NaCl 0.5 mM buff r SMES-NaOH from 6 to 7 anfi HEPES-NaOH from 7 to 8.4 , 0 uM DBMIB and chloroplasts containing 100 ug chlorophyll. 98 technique to resolve the true kinetics of‘the change. (Fast pH changes could be masked by slow instrument response if continuous illumination measurements only were made.) It was possible to determine the portion of the total pH change which was due to the proton pump simply by sub- tracting the gramicidin-sensitive change from the gramicidin-insensitive change. This was possible because the rate of electron transport in the ferrocyanide to MV reaction is not stimulated by uncouplers such as gramicidin, which abolish the reversible proton pump. However, the sub- traction method is only valid after the gramicidin-insensitive "pH gush,“ believed to represent the reduction of the plastoquinone pool (Izawa and Hind, 1967) is over (about 1 second under our conditions). Using the above method a ratio (H+/e') of about 1.0 was determined for Site I. Contrary to the predictions of the chemiosmotic hypothesis, the efficiency of proton uptake at Site I is independent of the pH of the surrounding medium over the range 6 to 8.5 even though the P/e2 ratio is close to zero below pH 7 (Figure 2). These data do not seem to be consistent with premise that the reversible pH rise of the sus- pending media is to any significant extent related to steady-state phosphorylation at Site 1. Therefore, the utilization of protons in the formation of ATP at Site I must be very pH sensitive. On the other hand, the insensitivity of the efficiency of Site II supported- phosphorylation to pH suggests that the enzymatic machinery (i.e. coup- ling factor1) of ATP formation can operate with unimpaired efficiency at pH's below 7.0. Clearly, this discrepancy cannot be reconciled if we adhere to the chemiosmotic notion of a common H+ pool, but the or P/e2 H-i-le- 1.0 0.5 99 I i I C) Fe ro a ide + MV C) C) r cy n ‘7‘ (D H. (3 o o 7 (D H+/e' ,zz”’TTT".-___TTT" —— . __ PISETTT‘w‘dilhw ii 6 7 8 Figure 2. Efficiency of ATP formation (P/e ) and efficiency of hydrogen ion accumulation (HTFe') at Site I. The 2 m1 reaction mixture consisted of 0.1 M sucrose, 2 mM MgCl , 40 mM KCl, 0.5 mM buffer (MES-NaOH from 6 t3 7 and HEPPS-NaOH from 7.5 to 8.5), 30 mM ferrocyanide, 100 uM methylviologen and chloroplasts containing 100 ug ch orophy 1 100 discrepancy and many of the differences between the sites can be accounted for if the two sites exist in non-identical "microenvironments" and each site has its own reversible ATPase. The possible existence of such physically isolated sites in the lamellar membrane will be dealt with in the concluding portion of this thesis. CONCLUSIONS CONCLUSIONS There is little doubt that, under the proper conditions, a hydrogen ion gradient across a lamellar membrane can supply the requisite energy for the phosphorylation of ADP. Jagendorf and Uribe (1966) showed that ATP synthesis could be induced in chloroplasts by artificially creating a pH difference across the membrane (by loading the internal thylakoid space with an organic acid) without any electron transport whatsoever. Quite recently, Racker and Stoechenius (1974) have demon- strated that artificial vesicles impregnated with coupling factor and bacteriorhodopsin are able to synthesize ATP whenhydrogen ions were pumped inside the vesicles by the light-excited rhodopsin. It appears, therefore, that coupling factor by itself can utilize the energy of a proton gradient to produce ATP. Thus, the conclusion is inescapable that a proton gradient across a vesicular membrane can fulfill the energy requirements of ATP synthesis. The case for the chemiosmotic mechanism of phosphorylation becomes even stronger when we consider that phosphorylation at Site II seems to have a requirement for hydrogen ion production. While observations like those presented above must be con- sidered as very strong evidence in support of the chemiosmotic hypothesis, the issue is by no means settled. It has been implied several times in this thesis that some facts are not readily explicable in terms of a model of energy conservation in which the driving force of steady-state 101 102 phosphorylation is an "inside-outside" electrochemical gradient. An inevitable requirement of such an "inside-outside" model would be the formation of a critical proton gradient and/or charge separation agrg§§_ the thylakoid membrane before ATP synthesis could occur. According to Mitchell's own calculations (1966b), the minimum size of such a gradient which would be capable of driving ATP synthesis is quite large, equivalent to a difference of at least 3 pH units between the inside and the outside. Since the gradient is thought to be formed via light-induced electron transport, it follows that there should be some amount of light below which not enough energy could be stored in the gradient to synthesize ATP. Several investigators (Turner gt;al,, 1962; Sakurai gt_al,, 1965; Schwartz, 1968) reported that photophosphorylation did not begin until the light intensity reached a critical level and they attributed the "dead space" to a need for a threshold proton gradient. However, Saha gt_al, (1970) found no such well defined phosphorylation "dead space" and reported that ATP formation was linear with light intensity until the intensity became extremely low, at which time only a slight non- linearity in the phosphOrylation rate was observed. They suggested that the "dead space" observed by earlier investigators might be expli- cable on the basis of membrane integrity (or lack of it); the “critical" light intensity might not exist in very good chloroplasts in which little leakage occurs. Another and better way to investigate the "threshold" is to limit the duration rather than the rate of the reac- tion. Witt's group (1971) has reported that they were able to observe ATP formation supported by flashes of light much too brief to cause any detectable pH change in the weakly buffered suspending medium. 103 According to Mitchell, for a given amount of proton translocation without compensatory ion movements, there will be a proton gradient and an electrical potential gradient formed, the ability of these grad- ients to drive ATP synthesis will be determined in part by the buffering capacity of the inner thylakoid space and in part by the electrical capacitance of the thylakoid membrane. The data of Witt's group (1971) suggest that if Mitchell is correct the electrical capacitance of the membrane must be very small and the consequent large electrical potential across the membrane must be capable of driving phosphorylation in the absence of any significant increase in internal pH. Much more damaging to the chemiosmotic theory than the apparent absence of "thresholds," is the evidence for site specificity in phos- phorylation reactions. As has been pointed out, electron transport through either Site I or Site II results in an inward movement of hydro- gen ions. Furthermore, the efficiencies of these “proton pumps" are almost independent of pH. However, only the protons transported by Site II can be used efficiently to make ATP at low pH. Other evidence for site specificity in ATP synthesis lies in the very different sen- sitivities of Site I and Site II phosphorylations to Hg++. Finally, the studies of Kraayenhof algal, (1972) on atebrin binding strongly suggest that there are two different kinds of energizable sites on the chloro- plasts membrane. The implication that the phosphorylation of ADP can take place at two different sites depending on which redox reaction is occurring does not fit the chemiosmotic model. This is because the chemiosmotic model has an intermediate energy conservation step which is common to both sites--protons on the inside of the thylakoid. 104 Williams (1961, 1969) has proposed a modification on the chemiosmotic mechanism which may help reconcile the observations dis- cussed above with Mitchell's picture of electrochemical gradient- dependent phosphorylation. The suggestion Williams has made is that the electrochemical gradient directly involved in steady-state phosphory- lation is formed and utilized with1n_the phosphorylating membrane. Thus, it seems plausible that the accumulation of only a few protons in the membrane could be coupled to an ATPase and drive ATP synthesis. In fact, the number of hydrogen ions required at a given "site" might be as low as 2 (assuming H+/ATP = 2). Clearly, one would not expect to find threshold light intensities or threshold pH gradients if such minute accumulations of protons and electrical charge are sufficient to drive ATP formation. An interesting consequence of Williams' version of chemiosmotic coupling is that it does not necessarily require an enclosed vesicle. (Mitchell (1972) has discussed a similar model which he calls the "micro" chemiosmotic mechanism of energy transduction.) Another consequence of the modification, not visualized by Williams, is that it readily accommodates the site specifity described in the preced- ing paragraph if one supposes that each energy conservation site in the membrane is supplied with its own reversible ATPase. If Site II actually consists of proton production due to water oxidation as I have suggested, it may be that this entire process is "buried" in the lipid laden lamellar membrane and is, thus, protected from changes in the medium pH (and also from Hg++). Perhaps as protons are released from water, they are consumed by a reversible ATPase also buried in the membrane. On the other hand, Site I may be much more 105 accessible to constituents of the aqueous phase. As the pH is lowered we can imagine how this might affect the energy which could be stored in the intra-membrane electrochemical gradient. It is interesting that Saha et_al, (1970) noticed that, at low light intensities, the quantum efficiency of phosphorylation supported by the Hill reaction is not diminished as the medium pH is varied from the optimum (about 8.0). The interpretation of this datum is very uncertain. However, it seems to suggest that at very low rates of proton uptake even Site I phosphoryla- tion efficiency is not affected by medium pH. For these reasons, I would like to suggest that the association between the site of a particular proton-producing redox reaction and a particular proton-utilizing coupling factor may be very intimate. Such a notion fits very well with Williams' model. LITERATURE CITED LITERATURE CITED Allen, J. F. and D. 0. Hall. 1973. Superoxide reduction as a mechanism of ascorbate-stimulated oxygen uptake by isolated chloroplasts. Biochem. Biophys. Res. Commun. 52: 856-862. Arnon, D. I., M. 8. Allen and F. R. Whatle . 1954. Photosynthesis by isolated chloroplasts. Nature 1 4: 394-396. Arnon, D. I., F. R. Whatley and M. B. Allen. 1955. Vitamin K as a cofactor of photosynthetic phosphorylation. Biochim. Biophys. Acta 16: 607-608. Arntzen, C. J., J. Neumann and R. A. Dilley. 1971. Inhibition of elec- tron transport in chloroplasts by a quinone analogue: Evidence for two sites of DCIPHZ oxidation. J. Bioenergetics 2: 73-83. Avron, M. 1964. On the site of ATP formation in photophosphorylation. Biochem. Biophys. Res. Commun. 17: 430-432. Avron, M. and B. Chance. 1967. Relation of phosphorylation to electron tgansport in isolated chloroplasts. Brookhaven Symp. 19: 149- O. Babcock, G. T. and Sauer, K. 1971. Photos stem II oxidation of artifi- cial donors using 20 usecond flas es in tris-washed chloro- ‘plasts. Abst. Congress on Primary Photochemistry of Photo- synthesis, Argon National Lab., p. 29. Ben-Hayyim, G. and M. Avron. 1970. Involvement of Photosystem Two in non-oxygen evolving non-cyclic and in cyclic electron flow processes in chloroplasts. Eur. J. Biochem. 15: 155-160. Bennoun, P. and A. Joliot. 1969. Studies on hydroxylamine photooxida- tion by spinach chloroplasts. Biochim. Biophys. Acta 189: 85- 94. Bdhme, H. and W. A. Cramer. 1971. Plastoquinone mediates electron transport between cytochrome b559 and cytochrome f.in spinach chloroplasts. FEBS Letts. 15: 349-351. Bfihme, H. and W. A. Cramer. 1972. Localization of a site of energy coupling between plastoquinone and cytochrome f_in the electron transport chain of spinach chloroplasts. Biochemistry 11: 1155-1160. 106 107 Bmae, H., S. Reimer and A. Trebst. 1971. The effect of dibromothymo- quinone, an antagonist of plastoquinone, on non-cyclic and cyclic electron transport. Z. Naturforsch. 26b: 341-352. B3hme, H. and A. Trebst. 1969. On the properties of ascorbate photo- oxidation in isolated chloroplasts. Evidence for two ATP sites in non-cyclic photophosphorylation. Biochim. Biophys. Acta 180: 137-148. Bradeen, D. A. and G. D. Winget. 1974. Site- specific inhibition of photophosphorylation in isolated spinach chloroplasts by HgCl II. Evidence for three sites of energy conservation associatea gggh ggn- gyglic electron transport. Biochim. Biophys. Acta Brand, J., T. Baszynski, F. Crane and D. Krogmann. 1971. Photosystem i inhibition by polycations. Biochem. Biophys. Res. Commun. 5: 538-543. Brand, J., A. San Pietro and B. Mayne. 1972. Site of polylysine inhibi- tion of Photosystem I in spinach chloroplasts. Arch. Biochem. Biophys. 152: 426-428. Brown, A. H. and N. E. Good. 1955. Photochemical reduction of oxygen in chloroplasts preparations and in green plant cells. I. The study of oxygen exchanges in vitro and in vivo. Arch. Biochem. Biophys. 57: 340-354. Cheniae, G. M. and I. F. Martin. 1967. Studies on the function of manganese in photosynthesis. Brookhaven Symp. Biol. 19: 406- 417. Clark, W. M. 1960. In Oxidation-reduction potentials of organic systems. Williams and—Wilkins. Baltimore, p. 397- 402. Cramer, W. A. and W. L. Butler. 1969. Potentiometric titration of the fluorescence yield of spinach chloroplasts. Biochim. Biophys. Acta 172: 503-510. Crofts, A. R. 1968. In Regulatory function of biological membranes. J. Jarnefelt, ed. Elsevier. Amsterdam, 47-263. Deamer, D. and L. Packer. 1969. Light-dependent anion transport in isolated spinach chloroplasts. Biochim. Biophys. Acta 172: 539-545. Dilley, R. A. 1968. Effect of poly- -L- lysine on energy-linked chloro-"P plast reactions. Biochemistry 7: 338- 346. - 108 Dilley, R. A. and L. P. Vernon. 1965. Ion and water transport processes related to the light-dependent shrinkage of chloroplasts. Arch. Biochem. Biophys. 111: 365-375. Dilley, R. A. and L. P. Vernon. 1967. Quantum requirement of the light- induced proton uptake by spinach chloroplasts. Proc. Natl. Acad. Sci. U. S. 57: 395-400. Elstner, E. F., A. Heupel, and S. Vaklinova. 1970. Uber die Oxidation von Hydroxylamin durch isolierte Choroplasten und die mfigliche Funktion einer Peroxidase aus Spinatblattern bei der Oxidation v32 gagorbinsaure und Glykolsfiure. Z. Pflanzenphysiol. 62: Epel, B. L. and J. Neumann. 1972. In VI. Intern. Congr. Photobiol. Biochem. Germany Abstract N61 237. Epel, B. L. and J. Neumann;+ 1973. The mechanism of the oxidation of ascorbate and Mn by chloroplasts. The role of the radical superoxide. Biochim. Biophys. Acta 520-529. Frenkel, A. W. 1954. Light induced phosphorylation by cell-free prepar- ations of photosynthetic bacteria. J. Am. Chem. Soc. 76:5568- 5569. Goffer, J. and J. Neumann. 1973. Evidence for system I mediated non— gycéZc photophosphorylation in chloroplasts. FEBS Letts. 36: Good, N. E. 1960. Activation of the Hill reaction by amines. Biochim. Biophys. Acta 40: 502-517. Gould, J. M. 1974. The phosphorylation site associated with the oxida- tion of exogenous electron donors by Photosystem I. (Submitted to Biochim. Biophys. Acta) Gould, J. M. and S. Izawa. 1973a. Photosystem II electron transport and phosphorylation with dibromothymoguinone as the electron acceptor. Eur. J. Biochem. 37: 18 -1 2. Gould, J. M. and S. Izawa. 1973b. Studies on the energy coupling sites of photophosphorylation 1. Separation of Site I and Site II by partial reactions of the chloroplast electron transport chain. Biochim. Biophys. Acta 314: 211-223. Gould, J. M. and S. Izawa. 1974. Studies on the energy coupling sites of photophosphorylation IV. The relation of proton fluxes to electron transport and ATP formation associated with Photo- system II. Biochim. Biophys. Acta 333: 509-524. 109 Gould, J. M. and D. R. Ort. 1973. Studies on the energy coupling sites of photophosphorylation III. The different effects of methyl- amine and ADP plus P on electron transport through coupling Site I and II in isoIated chloroplasts. Biochim. Biophys. Acta 325: 157-166. Green, D. E., J. Asai, R. A. Harris and J. T. Penniston. 1968. Conform- ational basis of energy transformation in membrane systems. III. Configurational changes in the mitochondrial inner mem- brane induced by changes in functional state. Arch. Biochem. Biophys. 125: 684-705. Heathcote, P. and D. 0. Hall. 1974. Photosynthetic control and photo- phosphorylation in Photosystem II of isolated spinach chloro- plasts. Biochem. Biophys. Res. Commun. 56: 767-774. Hill, R. and F. Bendall. 1960. Function of the cytochrome components in chloroplasts: A working hypothesis. Nature 186: 136-137. Hind, G. and A. T. Jagendorf. 1965. Effect of uncouplers on the conform- ational and high energy states in chloroplasts. J. Biol. Chem. 240: 3202-3209. Hind, G., H. Nakatani and S. Izawa. 1974. Light-dependent redistribution of ions in suspensions of chloro last thylakoid membranes. Proc. Nat. Acad. Sci. U. S. 71: 484-1488. Hind, G. and C. P. Whittingham. 1963. Reduction of ferricyanide by chloroplasts in presence of nitrogenous bases. Biochim. Bio- phys. Acta 75: 194-201. Izawa, S. 1970. The relation of post-illumination ATP formation capacity (X ) to H+ accumulation in chloroplasts. Biochim. Biophys. Ac a 223: 165-173. Izawa, S. and N. E. Good. 1968. The stoichiometric relation of phos- phorylation to electron transport in isolated chloroplasts. Biochim. Biophys. Acta 162: 380-391. Izawa, S. and N. E. Good. 1969. Effect of p-chloromercuribenzoate and mercuric ion on chloroplast photophosphorylation. In Progress in photosynthesis research. Vol. III. H. Metzner, EEK Internat. Union Biol. Sci. Tubingen., pp. 1288-1298. Izawa, S., J. M. Gould, D. R. Ort, P. Felker and N. E. Good. 1973a. Electron transport and photophosphorylation in chloroplasts as a function of the electron acceptor. III. A dibromothymo- quinone-insensitive phosphorylation reaction associated with Photosystem II. Biochim. Biophys. Acta 305: 119-128. 110 Izawa, S., R. L. Heath and G. Hind. 1969. The role of chloride ion in photos nthesis III. The effect of artificial electron donors upon e ectron transport. Biochim. Biophys. Acta 180: 388-398. Izawa, S. and G. Hind. 1967. The kinetics of the pH rise in illuminated chloroplast suspensions. Biochim. Biophys. Acta 143: 377-390. Izawa, S., R. Kraayenhof, E. D. Ruuge and D. Devault. 1973b. The site of KCN inhibition in the photosynthetic electron transport pathway. Biochim. Biophys. Acta 314: 328-339. Izawa, S. and D. R. Ort. 1974. Photooxidation of ferrocyanide and iodide ions and associated phosphorylation in NHZOH-treated chloroplasts. Biochim. Biophys. Acta In Press. Jagendorf, A. T. and M. Smith. 1962. Uncoupling phosphorylation in spinach chloroplasts by absence of cations. Plant Physiol. 37: 135-141. Jagendorf, A. T. and E. Uribe. 1966. ATP formation caused by acid-base transition of spinach chloroplasts. Proc. Nat. Acad. Sci. U. S. 55: 170-177. Karlish, S. and M. Avron. 1967. Relevance of proton uptake induced by light to the mechanism of energy coupling in photophosphoryla- tion. Nature 216: 1107-1109. Katoh, S. and A. San Pietro. 1967. Ascorbate supported NADP photo- reduction by heated Euglena chloroplasts. Arch. Biochem. Biophys. 122: 144-152. Kok, B., P. Joliot and M. P. McGloin. 1969. Electron transport between the photoacts. In Progress in photosynthesis research. Vol. II. H. Metzner,-Ed. Internat. Union Biol. Sci. Tubingen. pp. 1042-1056. Kok, B., S. Malkin, O. Owens and B. Forbush. 1967. Observations on the reducing side of the 02-evolving photoact. Brookhaven Symp. Biol. 19: 446-458. Kraayenhof, R. 1970. Quenching of uncoupler fluorescence in relation to the energized state in chloroplasts. FEBS Letts. 6: 161- 165. Kraayenhof, R., S. Izawa and B. Chance. 1972. Use of uncoupling acri- dine dyes as stoichiometric energy probes in chloroplasts. Plant Physiol. 50: 713-718. Krogmann, D. W., A. T. Jagendorf and M. Avron. 1959. Uncouplers of spinach chloroplast photosynthetic phosphorylation. Plant Physiol. 34: 272-277. 111 Lozier, R. H. and W. L. Butler. 1972. The effects of dibromothymo- quinone on fluorescence and electron transport of spinach chloroplasts. FEBS Letts. 26: 161-164. Lynn, W. S. and Brown, R. 1967. Competition between phosphate and protons on phosphorylation and cation exchange in chloroplasts. J. Biol. Chem. 242: 426-431. McCarty, R. 1974. Inhibition of electron transport in chloroplasts between the two photosystems by a water-soluble carbodiimide. Arch. Biochem. Biophys. 161: 93-99. McCarty, R. and E. Racker. 1966. Effects of an antiserum to the chloro- plast coupling factor on photophosphorylation and related processes. Fed. Proc. 25: 226. McCord, J. M. and I. Fridovich. 1969. Superoxide dismutase. An enzymic function for erythrocuprein (hemocuprein). J. Biol. Chem. 244: 6049-6055. Mehler, A. H. 1951. Studies on reactions of illuminated chloroplasts. II. Stimulation and inhibition of the reaction with molecular oxygen. Arch. Biochem. Biophys. 34: 339-351. Misra, H. P. and I. Fridovich. 1972. The univalent reduction of oxygen by reduced flavins and quinones. J. Biol. Chem. 247: 188- 2 Mitchell, P. 1961. Coupling of phosphorylation to electron transport and hydrogen transfer by a chemiosmotic type of mechanism. Nature 191: 144-148. Mitchell, P. l966a. Ig_Chemiosmotic coupling in oxidative and photo- .synthetic phosphorylation. Glynn Research LTD., Bodwin, Cornwall, England. Mitchell, P. l966b. The intrinsic anisotropy of the cytochrome oxidase region of the mitochondrial respiratory chain and the conse- quent vectorial property of respiration. In Regulation of metabolic processes in mitochondria. J. MTTTager, S. Papa, E. Quagliariello and E. C. Slater, eds. Elsevier. Amsterdam, pp. 65-85. Mitchell, P. 1970. Reversible coupling between transport and chemical reactions. In_Membranes and 1on transport and chemical reac- tions, I. E. E. Bittar, ed. Wiley. 'London, pp. 192-256. Mitchell, p. 1972. Chemiosmotic coupling in energy transduction: A logical development of biochemical knowledge. J. of Bioener- getics. 3: 5-24. 112 Mitchell, R. A., R. D. Hill and P. D. Boyer. 1967. Mechanistic implications of Mg++ adenine nucleotide and inhibitor effects on energy linked reactions of submitochondrial particles. J. Biol. Chem. 242: 1793-1801. Mitchell, P. and J. Moyle. 1965. Stoichiometry of proton translocation through the respiratory chain and adenosine triphosphatase systems of rat liver mitochondria. Nature 208: 147-151. Mitchell, P. and J. Moyle. 1968. Proton translocation coupled to ATP hydrolysis in rat liver mitochondria. Eur. J. Biochem. 4: 530-539. Neumann, J., C. Arntzen and R. A. Dilley. 1971. Two sites of adenosine triphosphate formation in photosynthetic electron transport mediated by Photosystem I. Evidence from digitonin subchloro- plasts particles. Biochemistry 10: 866-873. Nir, I. and A. Seligman. 1970. Photooxidation of diaminobenzidine (DAB) by chloroplast lamellae. J. Cell Biol. 46: 617-620. Ort, D. R., J. M. Gould. S. Izawa and N. E. Good. 1974. The relation between the efficiencies of proton uptake (H+/e') and phos- phorylation (P/ez) in chloroplasts. Manuscript in Preparation. Ort, D. R. and S. Izawa. 1973. Studies on the energy coupling sites of photophos horylation. II. Treatment of chloroplasts with NH OH plus E TA to inhibit water oxidation while maintaining en rgy coupling efficiencies. Plant Physiol. 52: 595-600. Ort, D. R. and S. Izawa. 1974. Studies on the energy coupling sites of photophosphorylation. V. Phosphorylation efficiencies (P/ez) associated with aerobic photooxidation of artificial electron donors. Plant Physiol. 53: 370-376. Ort, D. R., S. Izawa, N. E. Good and D. W. Krogmann. 1973. Effects of the plastocyanin antagonists KCN and poly-L:lysine on partial reactions in isolated chloroplasts. FEBS Letts. 31: 119-122. Ouitrakul, R. and S. Izawa. 1973. Electron transport and photophos- phorylation in chloroplasts as a function of the electron acceptor. II. Acceptor-specific inhibition by KCN. Biochim. Biophys. Acta 305: 105-118. Portis, A. and R. McCarty. 1973. On the pH-dependence of the light- induced hydrogen ion gradient in spinach chloroplasts. Arch. Biochem. Biophys. 156: 621-625. Racker, E. and W. Stoeckenius. 1974. Reconstitution of purple membrane vesicles catalyzing light-driven proton uptake and adenosine triphosphate formation. J. Biol. Chem. 249: 662-663. 113 Rao, P. S. and E. Hayon. 1973. Experimental determination of the redox otential of the su eroxide radical :02". Biochem. Biophys. es. Commun. 51: 46 -473. Rideal, E. K. 1925. A note on the reduction potential of dicyan quin- hydrone. Trans. Faraday Soc. 21: 143-144. Rottenburg, H. and T. Grunwald. 1972. Determination ofA pH in chloro- plasts. 3. Ammonium uptake as a measure onSpH in chloroplasts and sub-chloroplast particles. Eur. J. Biochem. 25: 71-74. Rottenburg, H., T. Grunwald and M. Avron. 19721 Determination OftSpH in chloroplasts. 1. Distribution of ( 4C) methylamine. Eur. J. Biochem. 25: 54-63. Rumberg, B., E. Reinwald, H. Schrfider and U. Siggel. 1969. Correlation between electron flow, proton translocation and phosphorylation in chloroplasts. Die Naturwissensch. 2: 77-79. Rumberg, F. and U. Siggel. 1969. pH changes in the inner phase of the thylakoids during photosynthesis. Die Naturwissensch. 56: 130- 132. Saha, S., R. Ouitrakul, S. Izawa and N. E. Good. 1971. Electron trans- port and photophosphorylation in chloroplasts as a function of the electron acceptor. J. Biol. Chem. 246: 3204-3209. Saha, S., S. Izawa and N. E. Good. 1970. Photophosphorylation as a {gzction of light intensity. Biochim. Biophys. Acta 223: 158- Sakurai, H., M. Nishimura and A. Takamiya. 1965. Studies on photophos- phorylation. I. Two-step excitation kinetics of photophosphor- ylation. Plant and Cell Physiol. 6: 309-324. Schliephake, W., W. Junge and H. T. Witt. 1968. Correlation between field formation, proton translocation, and the light reactions in photosynthesis. Z. Naturforsch. 23b: 1571-1578. Schmidt-Mende, P. and B. Rumberg. 1968. Zur Plastochinonreduktion bei der Photosynthese. Z. Naturforsch. 23b: 225-227. Schmidt-Mende, P. and H. T. Witt. 1968. Zur Plastochinonoxydation bei der Photosynthese. Z. Naturforsch. 23b: 228-235. Schrfider, H., H. Muhle and B. Rumberg. 1971. Relationship between ion transport phenomena and phosphorylation in chloroplasts. In_ Proceedings 2nd international congress of photosynthesis. G. Forti, M. Avron, and A. Melandri, eds. Dr. W. Junk, N. V. Publishers. The Hague, pp. 919-930. 114 Schuldiner, S., H. Rottenburg and M. Avron. 1972. Determination of ZSpH in chloroplasts. Eur. J. Biochem. 25: 64-70. Schwartz, M. 1968. Light-induced proton gradient links electron trans- port and phosphorylation. Nature 219: 915-917. Seligman, A., M. Karnovsky, H. Wasserkurg and J. Hander. 1968. Non- droplet ultrastructural demonstration of cytochrome oxidase activity with a polymerizing osmiophilic reagent, diamino- benzidine (DAB). J. Cell Biol. 38: 1-14. Selman, 8., G. Johnson, R. Giaquinta and R. A. Dilley. 1974. Evidence for cyanide and mercury inactivation of endogenous plasto- cyanin. Biochim. Biophys. Acta In Press. Shen, Y. K. and G. M. Shen. 1962. Studies on photophosphorylation. II. The "light intensity effect" and intermediate steps in photophosphorylation. Scienta Sinica 11: 1097-1106. Slater, E. C. 1953. Mechanism of phosphorylation in the respiratory chain. Nature. 172: 975-978. Stiehl, H. H. and H. T. Witt. 1968. Die kurzzeitigen ultravioletten differenzspektren bei der Photosynthese. Z. Naturforsch. 23b: 220-224. Strotmann, H. and C. van Gosslen. 1971. Photosystem I-dependent phos- phorylation of isolated chloroplasts with ascorbate/2,6- dichlorophenolindophenol as electron donor and methylviologen as electron acceptor. Z. Naturforsch. 27b: 445-455. Trebst, A. and H. Eck. 1961. Photosystem I-dependent oxidation of 2,6-dichlorophenolindophenol by isolated chloroplasts. Z. Naturforsch. 16b: 455-461. Trebst, A., H. Eck and S. Wagner. 1963. Effects of quinones and oxygen in the electron transport system of chloroplasts. Ig_Photo- synthetic mechanisms of green plants. Nat. Acad. Sci. Nat. Res. Council Publication 1145, pp. 174-194. Trebst, A., E. Harth and W. Draber. 1970. On a new inhibitor of photo- synthetic electron-transport in isolated chloroplasts. Z. Naturforsch. 25b: 1157-1159. Trebst, A. and S. Reimer. 1973. Properties of photoreduction by Photo- system II in isolated chloroplasts. An energy-conserving step in the photoreduction of benzoquinone by Photosystem II in the presence of DBMIB. Biochim. Biophys. Acta 305: 129-139. 115 Turner, J., C. Black and M. Gibbs. 1962. Studies on photosynthetic processes. I. The effect of light intensity on NADPT reduction, ATP formation and CO assimilation in spinach chloroplasts. J. Biol. Chem. 237: 77-579. Vambutas, V. and E. Racker. 1965. Partial resolution of the enzymes catalyzing photophosphorylation. I. Stimulation of photophos- phorylation by a preparation of a latent, Ca+ +-dependent adenosine triphosphatase from chloroplasts. J. Biol. Chem. 240: 2660-2667. Vaklinova, S., N. Tomova, E. Nikolova and M. SeEenska. 1965. The effect of certain factors in the photooxidation of hydroxylamine in isolated chloroplasts. Compt. rend. Acad. Bu 9. Sci. 18: 659- 662. Wessels, J. S. C. 1964. ATP formation accompanying photoreduction of NADP+ by ascorbate-indophenol in chloroplast fragments. Bio- chim. Biophys. Acta 79: 640- 642. Williams, R. J. P. 1961. Possible function of chains of catalysts. J. Theoret. Biol. 1: 1-17. Williams, R. J. P. 1969. Electron transfer and energy conservation. In Current topics in bioenergetics. 3. D. R. Sanadi, ed. AEademic Press. New York. pp. 79-156. Winget, G., S. Izawa and N. E. Good. 1965. The stoichiometry of photo- phosphorylation. Biochem. Biophys. Res. Commun. 21: 438-443. Winget, G. 0., S. Izawa and N. E. Good. 1969. The inhibition of photo- phosphorylation by phlorizin and closely related compounds. Biochemistry 8: 2067-2074. Witt, H. T. 1971. Coupling of quanta, electrons, fields and phosphoryla- tion in the functional membrane of photosynthesis. Q. Rev. Biophys. 4: 365-477. Witt, H. T, B. Rumberg and W. Junge. 1968. Electron transfer, field charges, proton translocation and phosghorylation in photo- synthesis. l9. In Colloq. Ges. Biol em (Mosbach) Springer- Verlag. New York. pp. 262- 284. Yamashita, T. and W. L. Butler. 1968. Photoreduction and photophosphor- ylation with tris-washed chloroplasts. Plant Physiol. 43: 1978-1986. Yamashita, T. and W. L. Butler. 1969. Inhibition of the Hill reaction by tris and restoration by electron donation to Photosystem 11. Plant Physiol. 44: 435-438. Volume 31, number 1 FEBS LETTERS April 1973 EFFECTS OF THE PLASTOCYANIN ANTAGONISTS KCN AND POLY-L-LYSINE ON PARTIAL REACTIONS IN ISOLATED CH LOROPLASTS Donald R. ORT, S. IZAWA, N.E. GOOD and David W. KROGMANN“ Department of Botany and Plant Pathology, Michigan State University, East Lansing, Mich. 48823, USA Received 12 February 1973 I. Introduction A very large number of compounds are known to inhibit chloroplast electron transport at or near photosystem [1. However, compounds which have been shown to block photosystem I are very few de- spite the great importance of such inhibitors in ana- lyzing the pathway of photosynthetic electron trans- port and the sites of phosphorylation associated with it. Hauska et a1. [1] demonstrated that an antibody to plastocyanin can be a useful inhibitor but only when applied to finely fragmented chloroplasts. Kimimura and Katoh [2] have recently reported that incubation of chloroplasts in HgClz can inhibit elec- tron flow at plastocyanin. However, prolonged expo- sure to HgClz has a number of deleterious effects on the chloroplasts; moreover it strongly inhibits phos- phorylation [3] . This leaves us only two specific photosystem I inhibitors which can be applied to un- fragmented chloroplasts, i.e., polylysine (and certain other polycations) of Brand et a]. [4] and KCN of Ouitrakul and Izawa [5] .Brand et a1. [6] have located the site of polycation inhibition between cytochrome f and P700. Izawa and associates (manuscript in prep- aration) have shown spectrophotometrically that KCN inhibits cytochrome f oxidation. Moreover KCN has been shown to react readily with isolated plastocyanin under the conditions required for effecting electron transport inhibition. Thus both polylysine and KCN appear to block electron transport at the level of * Permanent address: Department of Biochemistry, Purdue University, West Lafayette, Ind. 47907, USA. North-Holland Publishing Company — Amsterdam plastocyanin involvement. However, the effect of polylysine on phosphorylating electron transport has not yet been explored. Furthermore, there are some discrepancies between the effects of polylysine re- ported by Brand et al. [7] and that of KCN reported by Ouitrakul and Izawa [5] . (For instance polylysine inhibited ferricyanide reduction only by 50% while KCN inhibited it almost completely). The purpose of the present study was to compare these two inhibitions in well-coupled chloroplasts and under similar experimental conditions, to determine if the inhibitions were specific for photosystem l-re- quiring reactions as suggested by the plastocyanin in- volvement, and to determine the relation of the resid- ual photosystem II reaction to phosphorylation. Us- ing photosystem II and photosystem I partial reac- tions, we were able to show the photosystem I speci- ficity of poly-L-lysine (M.W. 194,000) and to con- firm the photosystem I specificity of the KCN treat- ment. Chloroplasts blocked with either inhibitor con- tinue to phosphorylate during the reduction oflipo- philic Class III acceptors. This electron transport which is entirely dependent on photosystem II is somewhat less efficient in phOSphorylation when poly- lysine is used. Moreover, we found that exact condi- tions for polylysine inhibition were critical and there- fore it was difficult to reproduce precisely various levels of polylysine inhibition. 2. Materials and methods Chloroplasts were isolated from commercial Spin- ach (Spinacia oleracea L.) as described elsewhere [8] . 119 117 Volume 31, number 1 The resulting stock suspension contained 800 pg chlorophyll/ml, 0.2 M sucrose, 0.005 M HEPES, and 2 mM MgClz. The procedure for measuring the reduc- tion of the lipophilic oxidants, e.g. oxidized p-phenyl- enediamine (PDOX), oxidized diaminodurene (DADOX), 2,5-dimethyl-p-benzoquinone (DMQ), is detailed in a paper by Saha et al. [8] . Photoreduction of dibromo- thymoquinone (DBMIB) was measured in a similar manner [9] . Electron flow from diaminodurene (DAD) to methylviologen (MV) was measured as de- scribed by Izawa et al. [10]. Cyanide-treated chloroplasts were prepared as de- scribed in a previous paper [5] by incubating chloro- plasts for 90 min at 0° in a 30 mM solution of KCN buffered at pH 7.8. Control chloroplasts were sus- pended for the same time in a similar medium con- taining KOH instead of KCN. In all experiments designed to test the effect of polycations on chloroplasts, poly-L-lysine of molecu- lar weight 194,000 was used. The polylysine was dis- solved in glass distilled water at 1 mg/ml. The appro- priate amount of chloroplast suspension (see legend of table 1) was added to 0.4 to 0.6 ml of water at room temp. in the reaction cuvette prior to the addi- tion of the polylysine. This room temp. water-shock- ing (see fig. 1) was essential to obtain inhibition of electron transport by polylysine. The addition of poly- lysine was followed immediately by addition of the sucrose and buffer, and then by the remainder of the reaction components. Control chloroplasts were wa- ter-shocked in a similar manner. Photophosphorylation was measured using a modi- fied version of the method of Avron [l l] . 3. Results and discussion When Brand and his associates [4, 7] first studied the polycation inhibition of electron transport, they isolated the chloroplasts in media free of sucrose and salts. In contrast, we have found that the chloroplasts may be isolated in conventional isotonic, buffered media then suspended briefly in greatly diluted buffer just before addition of polylysine. Once the polyly- sine has been added the normal media can be restored. Thus the chloroplasts need be subjected to harmful hypotonic, salt-free conditions for only a few seconds. Therefore, our chloroplasts are more active, more 120 FEBS KETTE RS April 1973 800 #- ,_ T CONTHUL o O U) . Z 600 -l <1 m i... Z 0 50 0| I sine/ ml _ 400 '- PO 9 Y V g \oxo— ’— 0 El LI.) 200- -4 0 I00 9 polylysine/ml O 15 30 45 60 SECONDS Fig. 1. Effect of water-shock-time on inhibition of electron transport from water to ferricyanide by polylysine. Rates are given in pequivalents/hr- mg chlorophyll. The reaction cuvette contained a 2 ml reaction mixture consisting of the following 0.1 M sucrose, 50 mM tricine-NaOH buffer (pH 8.0), 2 mM MgC12, 0.5 mM ADP, 5 mM Pi, 0.4 mM FeCy, 40 pg chloro- phyll and polylysine as indicated. 0.5 ml of water at room temp. were placed in the cuvette. Then 0.05 ml of chloro- plasts containing 40 pg chlorophyll from the stock suspen- sion (see Methods) was added. After incubation for the indi- cated number of seconds the polylysine was added in 0.1 ml and this was followed immediately by sucrose and buffer and then by the other components of the reaction mixture. tightly coupled and generally more intact than those employed by earlier investigators [4,7] . Fig. l dem- onstrates the relationship between time of water shock and degree of inhibition of electron transport from water to ferricyanide. The time of water shock required to achieve 100% inhibition with 100 pg poly- lysine/ml varied with the different chloroplast prep- arations. To insure complete inhibition of electron flow through plastocyanin the chloroplasts were usu- ally water shocked for 90 sec. Fig. 1 also shows that 50 pg polylysine/ml is sufficient to cause only 50% inhibition of the Hill reaction. This suggests that the polycation may be titrating negative charges on the site of inhibition and elsewhere and that 50 pg poly- lysine/ ml does not supply enOUgh positive charges to bind all available negative sites. Polylysine levels above 100 pg/ml resulted in excessive clumping of the chloroplasts and therefore made spectrophotometric assays very difficult. Table 1 presents the data obtained from various partial reactions using KCN and polylysine-treated 118 Volume 31, number 1 FEBS LETTERS April 1973 Table 1 The effect of KCN or polylysine treatment on various pathways of electron transport (E.T.) and associated photophosphorylation (ATP) in isolated chloroplasts. :1: , a: KCN treatment Polylysme treatment System Condition E.T. ATP (P/ez) Condition E.T. ATP (P/ez) H20 —~ FeCy Control 401 253 (1.26) Control 560 357 (1.24) KCN 35 5.2 (0.29) Polylysine 20 0 DAD -> MV Control 3250 580 (0.36) Control 2750 440 (0.32) KCN 230 31 (0.27) Polylysine 291 0 H20 -* PDox Control 1070 310 (0.58) Control 1390 429 (0.61) KCN 605 108 (0.36) Polylysine 980 128 (0.26) H20 -* DADox Control 915 256 (0.57) Control 1360 407 (0.59) KCN 415 70.5 (0.34) Polylysine 830 112 (0.27) H20 -’ DMQ Control 790 245 (0.62) Control 1060 371 (0.70) H KCN 292 44.2 (0.33) Polylysine 590 62 (0.21) H20 -> DBMIB Control 262 49.7 (0.38) Control 272 41.5 (0.31) KCN 237 42.9 (0.36) Polylysine 285 35.7 (0.25) The reaction cuvette 3contained 2. 0 m1 of reaction mixture consisting of: 50 mM tricine- NaOH buffer (pH 8. 0), 0. l M sucrose, 2 mM MgCl;, 1 mM3 21’!) 0. 5 ADP, chloroplasts, the indicated electron acceptor, and the indicated electron donor system when used. The electron acceptor systems used were: 0. 4 mM potassium ferricyanide (FeCy); 0. 5 mM methylviologen (MV); 0. 5 mM p-phenylenediamine plus 1.5 mM ferricyanide (PD0 x); 0. 5 mM diaminodurene plus 1.5 mM ferricyanide (DAD0 x); 0. 5 mM 2 ,5- dimethquuinone plus 0. 4 mM ferricyanide (DMQ); l X 10 5M dibromothymoquinone plus 0. 4 mM ferricyanide (DBMIB). The electron donor system used was 1 mM ascorbate plus 0. 5 mM diaminodurene (DAD). The amount of chloroplasts suspended in the 2.0 m1 reaction mixture was equivalent to 40 pg chlorophyll for the systems H20 —* FeCy, DBMIB; 8 pg for DAD -' MV; 20 pg for H20 -> PDox, DMQ. Rates are expressed in pequivalents or pmoles ATP/hr- mg chlorophyll. * For the procedures, see Methods. In reactions using DBMIB as the electron acceptor HEPES buffer at pH 7.5 was used. chloroplasts. We were able to obtain nearly 100% in- hibition of electron tranSport from water to ferricya- nide by polylysine whereas previous investigators [7] saw only 50% inhibition of this system. This discrep- ancy is probably due to the “leaky” nature of those chloroplasts which had been exposed to prolonged osmotic shock. In leaky chloroplasts ferricyanide seems capable of accepting electrons not only at its normal hydrophilic photosystem I site, but also at a site close to photosystem II which is normally acces- sible only to lipophilic oxidants. Saha et a1. [8] have shown that lipophilic strong oxidants (e.g. oxidized p-phenylenediamine) are re- duced very rapidly by illuminated chloroplasts and that the reaction is not stimulated by uncouplers or by the addition of ADP and Pi- Electron flow to such lipophilic oxidants is only about 50% as efficient in phosphorylation as is the flow of electrons to hydro- philic Hill acceptors such as ferricyanide. This sug- gests to us that lipophilic oxidants can accept electrons between two sites of phosphorylation. Table 1 shows that both KCN and polylysine prevent electron transport to Class I acceptors almost com- pletely but have much less effect on electron trans- port to Class III acceptors. Oxidized p-phenylenedia- mine (PDOX) is the best of the Class III acceptors, that is it has the smallest photosystem I-dependent component. By the same criterion dimethquuinone (DMQ) is the worst Class III acceptor tested. It is our contention that KCN and polylysine inhibit that por- tion of the electron flow through plastocyanin but hardly affects the transport of electrons from a donor close to photosystem II to Class III acceptors. Dibromothymoquinone (DBMIB) is a unique Class III acceptor in that it acts not only as a lipophil- ic electron acceptor but also as an inhibitor of elec- tron transport, probably by blocking at the site of plastoquinone involvement [12] . Therefore the re- duction of this Class 111 acceptor has no photosystem [dependent component and should not be effected 121 119 Volume 31, number 1 by KCN or polylysine blocks. Table 1 shows this to be the case. These findings confirm the photosystem I specificity of polylysine inhibition and point out the similarity to KCN inhibition. To complete the comparison of KCN and polyly- sine inhibitions it is necessary to consider phosphoryla- tion. Dilley [13] has reported that polylysine is an un- coupler of phosphorylation at levels similar to those needed for inhibition of the photoreduction of P700. We found that when polylysine was added to the 2 m1 reaction mixture after the addition of 2 pmoles of MgC12 no phosphorylation occurred and electron transport from water to ferricyanide was stimulated. However, when the polycation was handled as de- scribed in Methods phosphorylation did occur when electrons flowed from water to a lipophilic oxidant. Table 1 shows that higher phosphorylation efficiencies were obtained when Class 111 acceptors were used with KCN-treated rather than with polylysineotreated chloroplasts. Furthermore, in some chloroplast prep- arations the polylysine treatment was found to un- couple even more severely than indicated in table 1. Since water shocking itself had little if any effect on phosphorylation efficiencies, we must conclude the partial uncoupling is due to the polycation. On the basis of these experiments and similar ex- periments published elsewhere [4, 5, 7, 8] we conclude that polylysine and KCN treatments described provide convenient and specific inhibitors of photosystem I reactions, almost certainly by blocking the electron transport through plastocyanin. 122 FE BS LETTERS April 1973 Acknowledgements We thank J.M. Gould for valuable consultation. This work was supported by a grant (GB 22657) from the National Science Foundation, USA. References [l] G. Hauska, R. McMarty, R. Berzborn and F. Racker, .1. Biol. Chem. 246 (1971) 3524. [2] M. Kimimura and S. Katoh, Biochem. Biophys. Acta 183 (1972) 279. [3] S. Izawa and NE. Good, in: Progress in photosynthesis research, ed. H. Metzner (International Union of Biological Sciences, Tubingen, 1969) p. 1288. [4] .1. Brand, T. Baszynski, F. Crane and D. Krogmann, J. Biol. Chem. 247 (1972) 2814. [5] R. Ouitrakul and S. Izawa, Biochim. Biophys. Acta, in press. [6] .1. Brand. A. San Pietro and B. Mayne, Arch. Biochem. Biochem. Biophys. 152 (1972) 426. [7] J. Brand, T. Baszynski, F. Crane and D. Krogmann, Biochem. Biophys. Res. Commun. 45 (1971) 538. [8] S. Saha. R. Ouitrakul, S. Izawa and N. Good, J. Biol. Chem. 246 (1971) 3204. [9] S. Izawa, I. Could, D. Ort, P. Felker and N. Good. Biochim. Biophys. Acta, submitted Jan., 1973. [10] S. Izawa, T. Connolly, G. Winget and N. Good, Brookhaven Symposia in Biology 19 (1966) 169. [11] M. Avron, Biochim. Biophys. Acta 40 (1960) 257. [12] H. Bohme, S. Reiner and A. Trebst, Z. Natruforsch. 26b (1971) 341. [13] RA. Dilley, Biochemistry 7 (1968) 338. APPENDIX II ELECTRON TRANSPORT AND PHOTOPHOSPHORYLATION IN CHLOROPLASTS AS A FUNCTION OF THE ELECTRON ACCEPTOR III. A DIBROMOTYMOQUINONE- INSENSITIVE PHOSPHORYLATION REACTION ASSOCIATED WITH PHOTOSYSTEM II 120 Reprinted from Biochimica et Biophysica Acta, 305 (1973) 119-128 © Elsevier Scientific Publishing Company, Amsterdam — Printed in The Netherlands BBA 46 544 ELECTRON TRANSPORT AND PHOTOPHOSPHORYLATION IN CHLORO- PLASTS AS A FUNCTION OF THE ELECTRON ACCEPT OR III. A DIBROMOTHYMOQUINONE-INSENSITIVE PHOSPHORYLATION REACTION ASSOCIATED WITH PHOTOSYSTEM II" S. IZAWA, J. MICHAEL GOULD, DONALD R. ORT, P. FELKER and N. E. GOOD Department of Botany and Plant Pathology, Michigan State University, East Lansing, Mich. 48823 (U. S .A.) (Received January 8th, 1973) SUMMARY Dibromothymoquinone (2,5-dibromo—3-methyl-6-isopropyl-p—benzoquinone) is reputed to be a plastoquinone antagonist which prevents the photoreduction of hydrophilic oxidants such as ferredoxin-NADP”. However, we have found that dibromothymoquinone inhibits only a small part of the photoreduction of lipo- philic oxidants such as oxidized p-phenylenediamine. Dibromothymoquinone-resistant photoreduction reactions are coupled to phosphorylation, about 0.4 molecules of ATP consistently being formed for every pair of electrons transported. Dibromo- thymoquinone itself is a lipophilic oxidant which can be photoreduced by chloro- plasts, then reoxidized by ferricyanide or oxygen. The electron transport thus catalysed also supports phosphorylation and the P/e2 ratio is again 0.4. It is concluded that there is a site of phosphorylation before the dibromothymoquinone block and another site of phosphorylation after the block. The former site must be associated with electron transfer reactions near Photosystem II, while the latter site is presum- ably associated with the transfer of electrons from plastoquinone to cytochrome f. INTRODUCTION Two quite different arguments lead us to the conclusion that non-cyclic photo- phosphorylation involves more than one site of energy conservation. Our reasons for believing that there are at least two phosphorylation sites are as follows: (1) The overall efficiency of photophosphorylation (P/ez) is considerably higher than one ATP molecule formed for every pair of electrons transported‘. Furthermore, the P/ez ratio approaches 2.0 if one subtracts that part of the electron transport which can occur in the absence of phosphorylationz. (2) Lipophilic strong oxidants (Class III acceptors), such as the oxidized form of p-phenylenediamine, intercept electrons by reacting with some intermediate Abbreviations: DCMU, 3-(3,4wdichloropheny1)-l,l-dimethylurea; P/ea, ratio of the mole- cules of ATP formed to the pairs of electrons transported. ' Journal Article No. 6219 of the Michigan Agricultural Experiment Station. 121 120 S. IZAWA et al. carrier which normally transfers electrons from Photosystem II to Photosystem I (refs 3, 4). This interception of electrons does not abolish phosphorylation but instead decreases the efiiciency to about half of the value observed when hydrophilic Class I acceptors are reduced. It therefore seems that the intermediate carrier re- sponsible for the reduction of oxidized p-phenylenediamine is situated between two sites of phosphorylation in the electron transport chain. The work described in this paper was undertaken in an attempt to define the location of the two phosphorylation sites in terms of the sites of action of known electron carriers. It has long been thought that there must be a rate-determining, phosphorylation- dependent reaction transferring electrons between the two photosystemss“7. This rate-determining step presumably lies between plastoquinone and cytochrome f since the rate of reduction of cytochrome f by Photosystem II and the rate of oxidation of plastoquinone by Photosystem I are accelerated during phosphorylation or when uncouplers are added. However, there is good reason for doubting that this rate- determining phosphorylating process is involved in the reduction of oxidized p- phenylenediamine. The reduction of oxidized p-phenylenediamine is very fast and independent of phosphorylation’. Moreover, the fact that 3-(3,4-dichlorophenyl)- 1,1-dimethylurea (DCMU) inhibition of oxidized p-phenylenediamine reduction is independent of light intensity suggests that oxidized p-phenylenediamine accepts electrons from a carrier situated close to Photosystem 114. We wish now to present evidence which lends further support to the concept of a site of phosphorylation close to Photosystem II and at the same time virtually precludes the participation of a plastoquinone—cytochrome f phosphorylating re- action in oxidized p-phenylenediamine reduction. The new evidence has been provided by studies of the effects of dibromothymoquinone (2,5-dibromo-3-methyl-6-isopropyl- p-benzoquinone) on electron transport and phosphorylation. This inhibitor was first introduced by Trebst and his associates8 as a plastoquinone antagonist. At very low concentrations it blocks all of the transfer of electrons from water to Class I acceptors such as ferredoxin-NADP+ or methylviologen and a large part of the transfer of electrons from water to ferricyanide9. It also seems to block the transfer of electrons from cytochrome [7559 to cytochrome f and it certainly prevents the reduction of cytochrome f by electrons from Photosystem 1110'”. These observations do indeed suggest that the inhibitor acts at the level of plastoquinone involvement. In any event, it is clear that the inhibitor prevents electron transport at some point after Photosystem II but before cytochrome f. Yet dibromothymoquinone does not greatly inhibit either oxidized p-phenylenediamine reduction or the associated phosphorylation reaction. It follows that there must be a site of phosphorylation before the site of dibromothymoquinone inhibition and probably therefore before the site of involvement of plastoquinone. MATERIALS AND M ETHODS The procedures employed in this study were similar to those employed in the earlier papers of the series3'4. Chloroplasts were isolated from commercial spinach (Spinacia oleracea L.) as already described’. Cyanide-treated chloroplasts were pre- pared by incubating chloroplasts at 0 °C for 90 min in a 30 mM KCN solution 122 PHOTOSYSTEM II PHOSPHORYLATION 121 buffered at pH 7.8 as described in the previous paper“. Control chloroplasts were suspended for the same time in a similar medium containing KOH instead of KCN. The inhibitor dibromothymoquinone was prepared by bromination of thymo- quinone in water and was recrystallized several times from alcohol. Stock solutions were prepared by dissolving dibromothymoquinone in ethanol-ethylene glycol (1:1, v/v). The concentration of the stock was such that the organic solvent in the reaction mixture never exceeded 1%. The reduction of ferricyanide was measured as the decrease in absorbance of the reaction mixture at 420 nm. In experiments with oxidized p-phenylenediamine, recrystallized colorless p-phenylenediamine dihydrochloride was added to the buffered reaction mixture, then oxidizedimmediately before the reaction with excess ferri- cyanide. Electron transport was measured as reduction of the excess ferricyanide since the oxidized p-phenylenediamine reduced during the reaction is immediately re- oxidized by ferricyanide. Reactions involving other aromatic diamines and quinones were measured in the same indirect way. The reduction of methylviologen was measured as oxygen uptake since reduced methylviologen reacts rapidly with oxygen to form H 202”. For these measurements a Clark-type, membrane-covered electrode was used. Phosphorylation was measured by a modification of the method of Avron” as the residual radioactivity after extraction of the 32P-labeled orthophosphate from the reaction mixture as phosphomolybdic acid. In all experiments the tempera- ture was 19 °C. RESULTS (1) The sensitivity of electron transport and phosphorylation to dibromothymoquinone with dtflerent electron acceptors As we have reported elsewhere3, lipophilic oxidants tend to increase the rate of electron transport in illuminated chloroplasts, decrease the dependence of electron transport on phosphorylation and reduce the efficiency of phosphorylation (P/ez) toward one-half. The extent to which acceptors are able to intercept electrons between two phosphorylation sites can be roughly judged by the increase in the rate of electron transport and the decline in the P/e2 ratios. (These criteria only apply, of course, if the P/ez ratios fall to a plateau rather than to zero and it can be shown that the acceptor is not an uncoupler.) Among the lipophilic acceptors listed in Table l, oxidized p-phenylenediamine is most nearly a typical Class III acceptor while 2,5- dimethyl-p-benzoquinone is the least typical. Ferricyanide ion is not lipophilic at all and, in our chloroplasts, seems to intercept very few of the electrons generated by Photosystem II. As can be seen in Table I, the transport of electrons to lipophilic acceptors has a large component which is resistant to dibromothymoquinone. This component is largest with the best Class III acceptor, oxidized p-phenylenediamine and smallest with the worst Class III acceptor, 2,6—dimethyl-p-benzoquinone. Clearly, that part of the electron transport which results from the interception of electrons between the phosphorylation sites is largely insensitive to the inhibitor. This is even more obvious when one notes the effect of dibromothymoquinone on the P/ez ratios. Regardless of the ratio in the absence of the inhibitor, that is regardless of what proportion of the electrons are intercepted between the phosphorylation sites. the P/e2 ratio always falls to about 0.4 invthe presence of the inhibitor. This is true even 123 122 S. IZAWA et al. TABLE I THE EFFECT OF DIBROMOTHYMOQUINONE ON ELECTRON TRANSPORT AND PHOTOPHOSPHORYLATION IN CHLOROPLASTS WITH DIFFERENT ELECTRON ACCEPTORS The 2.0-m1 reaction mixture consisted of the following: 0.1 M sucrose, 50 mM Tricine buffer (pH 8.2), 2 mM MgC12. 1 mM ADP, 5 mM 32P1, chloroplasts containing 30 pg chlorophyll, and the indicated acceptor system. These acceptor systems were: 0.5 mM potassium ferricyanide (Fecy); 0.5 mM p-phenylenediamine plus 1.5 mM ferricyanide (PDox); 0.5 mM diaminodurene plus 1.5 mM ferricyanide (DADox); 0.5 mM 2,5-dimethyl-p-benzoquinone plus 0.5 mM ferri- cyanide (DMQ); 0.5 mM 2,5-diaminotoluene plus 1.5 mM ferricyanide (DATox). When used dibromothymoquinone was 0.5pM. Rates are expressed in pequiv or pmoles ATP/h per mg chlorophyll. Electron Rate of electron transport Rate of AT P formation P/ee ac ‘e tor ( p Control + dibromo- Control + dibromo- Control + dibromo- thymoquinone thymoquinone thymoquinone Fecy 430 58 228 13 1.06 0.45 P130): 1260 695 292 149 0.46 0.43 DADox 735 383 244 56 0.66 0.39 DMQ 902 294 325 52 0.72 0.36 DATox 791 396 280 95 0.71 0.48 for the tiny residue of electron transport with ferricyanide as acceptor. We have also found that, regardless of the acceptor, the dibromothymoquinone-resistant component of the electron transport is always independent of the presence or absence of ADP and phosphate. Further effects of dibromothymoquinone are illustrated in Figs 1—4. Again, the transport of electrons from water to oxidized p-phenylenediamine has two components: one large, insensitive to dibromothymoquinone and supporting phos- phorylation with a P/e2 ratio of about 0.4; the other smaller, sensitive to dibromo- thymoquinone with a computed P/ez about 1.0 (Fig. 1). In contrast, the transport of electrons to ferricyanide is mostly sensitive to the inhibitor while the transport to methylviologen is almost all sensitive (Fig. 2). Once again the small residue of dibromothymoquinone-insensitive ferricyanide reduction supports phosphorylation with a P/e2 ratio of 0.4—0.5. This residual dibromothymoquinone—resistant ferricyanide reduction deserves attention since we have here the unusual situation of an inhibitor seeming to catalyze the reaction it inhibits; increasing concentrations of dibromothymoquinone actually increase the rate of ferricyanide reduction (Figs 3 and 4). The inhibitor, in addition to blocking electron transport, is itself a lipophilic oxidant which accepts electrons before or at its own site ofinhibition. Apparently the reduced dibromothymoquinone is quickly reoxidized by ferricyanide and thus the inhibited ferricyanide reduction is in part restored. However, this dibromothymoquinone-mediated ferricyanide reduction is quite different from the usual ferricyanide Hill reaction, having instead many of the characteristics of oxidized p-phenylenediamine reduction; the rate is independent of the presence or absence of ADP and phosphate or of uncouplers such as methylamine, and the efficiency of phosphorylation (We) is only 0.3—0.4. Although 124 PHOTOSYSTEM II PHOSPHORYLATION 123 f F4 I I I 1000— Lom ~ “K o I L l l L (I I 2 3 4 '&m ELECTRON TRANSPORT (in '41 or ATP °j\ Fecy at (P0,...) 0- .‘E b 5004K — § 1 X are (900.) 1- 2°04 ‘ i i] g .— ATP(Fecy) ‘3 25° 1 " '00 0 ATP °\ E.T(Fecy) é 6% '— / - d ~0— O 1 1 1 L o l l 4:?- 4 1 u..A_ 0 1 2 3 4 5 0 I 2 ” 5 0 I 2 " 5' Dibromothymoquinone x107(M) Dibrvomott'ryrr'ioquinonex107 (M) Fig. 1. Effect of dibromothymoquinone 011 electron transport (E.T.) and phosphorylation (ATP) with ferricyanide (Fecy) or oxidized p-phenylenediamine (PDox) as electron acceptor. Rates are expressed in pequiv or pmoles ATP/h per mg chlorophyll. The 2.0-ml reaction mixture consisted of the following: sucrose, 0.1 M; Tricine-NaOH buffer (pH 8.2), 50 mM; M302, 2 mM; ADP, 1 mM; ”P1, 10 mM; chloroplasts containing 40 pg (ferricyanide) or 30 pg (PDox) chlorophyll; and either 0.5 mM potassium ferricyanide or a combination of 1.5 mM ferricyanide and 0.5 mM p-phenylenediamine dihydrochloride. Note the great sensitivity of ferricyanide reduction to the inhibitor, the relative insensitivity of oxidized p-phenylenediamine reduction and the high rate of ATP formation associated with the resistant oxidized p-phenylenediamine reduction. Fig.2. Effect of dibromothymoquinone on electron transport and phosphorylation with ferri- cyanide and methylviologen (MV) as electron acceptors. Reaction conditions, units and ab- breviations as in Fig. 1 except that ”P1 was 5 mM. potassium ferricyanide was 0.4 mM and methylviologen was 50 pM. Electron transport was measured as oxygen production with ferri- cyanide and as oxygen consumption with methylviologen. The dotted curve for oxidized p. phenylenediamine (PDox) in the inset figure is taken from Fig. l for comparison. Note from the inset figure that the small amount of residual ferricyanide reduction supports phosphorylation with the efficiency characteristic of the oxidized p-phenylenediamine-reducing system. Presumably ferricyanide can intercept electrons, either directly or indirectly, between two sites of phospho- rylation as can oxidized p-phenylenediamine whereas methylviologen cannot. this catalysis of ferricyanide reduction is most conspicuous at high dibromothymo- quinone concentrations, there is no reason to doubt that it is already taking place at the point of apparent maximum inhibition of ferricyanide reduction and even there constitutes a large fraction of the residual reaction. This is clearly indicated by the fact that the P/ez ratio associated with ferricyanide reduction declines to 0.4—0.5 as the dibromothymoquinone inhibition approaches its maximum (Figs 1 and 2). No such decline is observed when the low potential acceptor methylviologen is being reduced; for thermodynamic reasons reduced dibromothymoquinone cannot donate electrons to methylviologen and therefore the inhibitor cannot catalyze methyl- viologen reduction. 125 124 S. IZAWA et al fl // , e T I l 1 // 1 l T 1 F 7 t .300 - " 300 _ 1.04 o. 1 Oil Fecy : ' Fecy Q 2 'No 5 T N b . 2 g X I- Rib—Ohm o— N 0.5 " o *200 - x o _ — A 200 - _ 3': 0"0 0 _\~ 0 1 1 1 1 1 1- ,, o 2 4 6 8 10 m 1 1 1 1 1 V Dibromothymoquinon 106 (M) E o o 2 4 6 8 IO E e x <2 1 Dibromothyrr'ioqulnoneat106 (M) 8 V ’0‘ m er (F ) é / z /———<>—~ z *- 100 o .4 31: 100 \ 0 0‘ E.T.(F ) _ Z 1 ecy / .— [ O/ 8 2 S l o\o/O ATPtFecy) 8 2\ ATP\(Fecy) .J \ / ATPiMV) w 1 E r 0. Ac”, 2 L 1 41 LJJ ‘ ' ‘_____/ O I 1 - 1 I . o a L 1 1% O l 2 3 4 IO 0 l 2 3 4 , O Dibromothymoquinone x106 (M) Dibriomothymoqumone x106 (M ) Fig. 3. Effect of higher concentrations of dibromothymoquinone on electron transport and phos- phorylation with ferricyanide or methylviologen as acceptors. Reaction conditions, units and abbreviations as in Fig. 2, except that methylviologen was 100 pM. Note the much greater residual electron transport with ferricyanide and the increasing rate of ferricyanide reduction with in- creasing dibromothymoquinone concentration. Note also from the inset figure that the inhibitor- insensitive electron transport again supports phosphorylation with the efficiency characteristic of the oxidized p-phenylenediaminc-reducing system. Apparently the inhibitory lipid soluble quinone dibromothymoquinone is reduced by chloroplasts, in a reaction which does not include the site of dibromothymoquinone inhibition and is then rapidly reoxidized by the excess ferricyanide. Fig. 4. Effects of dibromothymoquinone on digitonin-treated chloroplasts. Conditions, units and abbreviations as in previous figures. Chloroplasts were treated with 0.05% digitonin in buffer (pH 7.4) at 0 “C for 15 min, spun down at 4000> methylviologen, 40pg chlorophyll; water» PDox, 30 pg chlorophyll; diaminodurene» methylviologen and diaminodurene (cyclic), 10 pg chlorophyll. In the diaminodurene» methylviologen system ascorbate (1 mM), DCMU (1 pM), diaminodurene (0.5 mM) and methylviologen (0.1 mM) were added and the pH was lowered to 7.7 to eliminate much of the non-biological ascorbate oxidation. The diaminodurene (cyclic) phosphorylation system was similar except that methylviologen and ascorbate were omitted and 0.1 mM ferricyanide was added to establish an appropriate diaminodurene/oxidized diaminodurene ratio. System Condition Electron ATP Plea transport formation Water» methylviologen Control 646 395 1 . 1 3 + dibromothymoquinone 44 6 — KCN treated 0 0 — Water» PDox Control 1720 386 0.45 + dibromothymoquinone 1300 257 0.40 KCN treated 1200 198 0.33 Diaminodurene» methylviologen Control 4180 733 0.35 + dibromothymoquinone 4580 705 0.31 KCN treated 440 33 (0.15) Diaminodurene (cyclic) Control — 702 —- + dibromothymoquinone — 605 — KCN treated — 12 —— conclude that the site of dibromothymoquinone inhibition falls between the DCMU inhibition site and the KCN inhibition site. This is consistent with the view that di- bromothymoquinone acts as a plastoquinone antagonist. DISCUSSION In the first paper of this series3 we noted that lipophilic strong oxidants (e.g. oxidized p-phenylenediamine) can be reduced very rapidly by illuminated chloro- plasts whether or not phosphorylation occurs. Nevertheless, in the presence of ADP and phosphate, the high rate of electron transport is associated with a great deal of phosphorylation. We.have called such oxidants Class III electron acceptors. Conventional hydrophilic oxidants such as methylviologen, ferredoxin-NADP+ and ferricyanide we have called Class I acceptors. Since the reduction of Class III acceptors supports only half as much phosphorylation as the reduction of an equiva- lent amount of Class I acceptor, we suggested that lipophilic oxidants have access to 128 PHOTOSYSTEM II PHOSPHORYLATION 127 and accept electrons from some electron carrier which lies between two sites of phosphorylation. Furthermore, we suggested that the second phosphorylation site, the one not employed in the reduction of Class III acceptors, is responsible for limiting the rate of the Hill reaction. Hence the high rate of electron transport in the presence of Class III acceptors. In the second paper“ we showed that KCN treatment of chloroplasts prevents the reduction of Class I acceptors but not the reduction of Class III acceptors. This virtually proves that Class III acceptors do react directly with some inter- mediate carrier in the electron transport chain, a carrier operating before the KCN block. Moreover we postulated that this intermediate carrier is close to Photo- system 11 on the basis ofthe kinetics of DCMU inhibition of the reduction of Class III acceptors. This in turn implies the existence of a phosphorylation site closely asso- ciated with Photosystem 11, since the reduction of Class III acceptors via the KC N- insensitive shortened pathway is still coupled with a P/e2 ratio of 0.3—0.4. In the present paper, we have shown that dibromothymoquinone also inhi‘ .its the reduction of Class I acceptors without severely inhibiting the reduction of Clas III acceptors. Regardless of the Class III acceptor used (oxidized p-phenylenediamme, oxidized diaminodurene, oxidized diaminotoluene or 2,6-dimethyl-p-benzoquinone) and therefore, regardless of the rate of electron transport and the P/e2 ratio in the absence of the inhibitor, the dibromothymoquinone-resistant portion of electron transport is coupled to phosphorylation with a P/e2 ratio of 0.35—0.45. A very similar P/e2 ratio (0.3—0.4) is associated with the residual ferricyanide reduction in the presence of low concentrations of dibromothymoquinone. It is quite clear that all these reactions involve only Photosystem II and that segment of the electron transport chain which ends in the dibromothymoquinone block. We must therefore conclude that there is a site of phosphorylation associated with Photosystem II and located before the dibromothymoquinone inhibition site. If dibromothymo- quinone indeed blocks electron transport at the site of plastoquinone involvement. as the evidence suggests, there must be a site of phosphorylation both before and after plastoquinone. Thus the rate-limiting phosphorylation reaction presumed to occur between plastoquinone and cytochrome f6'7 may be equated to the slow step postulated in our first paper3. On the basis of cross-over point determinations, Bohme and Cramer7 concluded that only one phosphorylation site in the electron transport chain exerted a control over the rate of electron transport. However, our observations are in no way in- consistent with their conclusion. The transport of electrons to Class III acceptors proceeds at high rates whether or not phosphorylation occurs and it is axiomatic that cross-over data cannot yield information on sites of phosphorylation unless the electron transport through the site is phosphorylation dependent. We have presented the bare bones of our conclusions in Fig. 5. No doubt alternative interpretations of the data could be devised but none has occurred to us. The precise location of Site 11, the site close to Photosystem II which we have pro- posed in this paper, remains a matter for conjecture. Neumann et al.15 have provided a model of non-cyclic photophosphorylation in which two sites of phosphorylation are assumed to be involved in Photosystem I reactions. We find it difficult to re- concile their model with our data unless Site I in Fig. 5 is further divided into two sites. It is, however, possible that there is another site close to Photosystem I which 129 128 S. IZAWA et al. 02 PDoxve’c' / Fecy 081118/ E: /l ' Fecy W—v -———->POI l?)—-——>cyt. f—vP.C—>m MV 1 + 0cm Darius KCN NADP W—d SITE II SITE I Fig. 5. Simplified scheme of the electron transport pathways, phosphorylation reactions and inhi- bition sites discussed in this paper. PDox, oxidized p—phenylenediamine; Fecy, ferricyanide; DBMIB, 2,6-dibromo-3-methyl-6-isopropyI-p-benzoquinone (dibromothymoquinone); PQ, plasto- quinone; PC, plastocyanin; DCMU, 3-(3,4-dichlorophenyl)-1,l-dimethylurea; MV, methyl- viologen; PS I and PS 11, Photosystems I and II, respectively. It should be noted that dibromo- thymoquinone and KCN both block reduction of the hydrophilic electron acceptors but not the reduction of the lipophilic acceptors. Moreover, the residual electron transport with either in- hibitor present supports phosphorylation with an efficiency (P/e2 ratio) of 0.4. is responsible for some DCMU-insensitive cyclic photophosphorylation reactions, but this possibility is outside the scope of our present investigation. POSTSCRIPT After we had completed the manuscript of this paper we received a communica- tion from Dr Achim Trebst describing similar experiments conducted in his laboratory which have led him also to conclude that there is an energy conservation step associat- ed with Photosystem 11 reactions. ACKNOWLEDGEMENT This work was supported by a grant, GB 22657, from the National Science Foundation, U.S.A. REFERENCES Winget, G. D., Izawa, S. and Good, N. E. (1965) Biochem. Biophys. Res. Commun. 21, 438—443 Izawa, S. and Good, N. E. (1968) Biochim. Biophys. Acta 162, 380-391 Saha, S., Ouitrakul, R., Izawa, S. and Good, N. E. (1971) J. Biol. Chem. 246, 3204—3209 Ouitrakul, R. and Izawa, S. (1973) Biochim. Biophys. Acta 305, 105—1 18 Avron, M. and Chance, B. (1966) Brookhaven Symp. Biol. 19, 149-160 Kok, B., Joliot, P. and McGloin, M. P. (1969) in Progress in Photosynthesis Research, pp. 1042—1056, International Union of Biological Sciences, Tiibingen 7 Bohme, H. and Cramer, W. A. (1972) Biochemistry 11, 1155—1160 8 Trebst, A., Harth, E. and Draber, W. (1970) Z. Naturforsch. 25b, 1157—1159 9 B'dhme, H., Reiner, S. and Trebst, A. (1971) Z. Naturforsch. 26b, 341-352 1;) Bdhme, H. and Cramer, W. A. (1971) FEBS Lett. 15, 349—351 11 Knaff, D. B. (1972) FEBS Lett. 23, 142-144 12 Good, N. E. and Hill, R. (1955) Arch. Biochem. Biophys. 57, 355—366 13 Avron, M. (1960) Biochim. Biophys. Acta 40, 257—272 14 Lozier, R. H. and Butler, W. L. (1972) FEBS Lett. 26, 161—164 15 Neumann, J., Arntzen, C. J. and Dilley, R. A. (1971) Biochemistry 10, 866—873 OKLA-DUN— APPENDIX III STUDIES ON THE ENERGY COUPLING SITES OF PHOTOPHOSPHORYLATION III. THE DIFFERENT EFFECTS OF METHYLAMINE AND ADP PLUS PHOSPHATE ON ELECTRON TRANSPORT THROUGH COUPLING SITES I AND II IN ISOLATED CHLOROPLASTS 130 Reprinted from Biochimica et Biophysica Acta, 325 (1973) 157—166 © Elsevier Scientific Publishing Company, Amsterdam - Printed in The Netherlands BBA 46 624 STUDIES ON THE ENERGY COUPLING SITES OF PHOTOPHOSPHORY- LATION III. THE DIFFERENT EFFECTS OF METHYLAMINE AND ADP PLUS PHOSPHATE ON ELECTRON TRANSPORT THROUGH COUPLING SITES I AND II IN ISOLATED CHLOROPLASTS J. MICHAEL GOULD and DONALD R. ORT Department of Botany and Plant Pathology, Michigan State University, East Lansing, Mich. 48824 (U.S.A.) (Received June 4th, 1973) SUMMARY 1. The reduction of lipophilic (Class III) oxidants such as oxidized p-phenyl- enediamine consists of two components. One component requires both Photosystem II and Photosystem I and includes both sites of energy coupling associated with non- cyclic electron transport. The second component requires only Photosystem II and includes only the site of energy coupling located before plastoquinone (Site 11). When oxidized p-phenylenediamine is being reduced by both pathways, the overall rate of electron transport is stimulated by the addition of ADP plus phosphate or the un- coupler methylamine. However, if the Photosystem 1 component of oxidized p- phenylenediamine reduction is eliminated by a low concentration of the plasto- quinone-antagonist dibromothymoquinone, the stimulation of electron transport by ADP plus phosphate or methylamine is also abolished, although the remaining Photo- system II-dependent electron transport remains firmly coupled to phosphorylation (via coupling Site 11). These results indicate that coupling Site II, unlike the well- known rate-limiting coupling site between plastoquinone and cytochrome f (Site I), does not exert any control over the rate of associated electron transport. 2. When substituted p—benzoquinones (e.g. 2,5-dimethyl-p-benzoquinone) or quinonediimides (e.g. p-phenylenediimine) are used as Class III acceptors in con- junction with dibromothymoquinone, a small but significant stimultation of electron transport by ADP plus phosphate is observed. However, it can be shown that this stimulation does not arise from coupling Site 11 but rather is due to a low rate of electron flux through coupling Site I even in the presence of dibromothymoquinone. Apparently the p—benzoquinones can catalyze an electron “bypass” around the dibro- mothymoquinone-induced block at plastoquinone, possibly by substituting partially for the natural electron carrier. If this bypass electron flow is blocked at plastocyanin by KCN treatment, the stimulation of electron transport by ADP plus phosphate is eliminated, although a high rate of phosphorylation (from Site 11 only) remains. Abbreviations: P/Cz ratio, the ratio of the number of molecules of ATP formed to the number of pairs of electrons transported; dimethquuinonc (DMQ), 2,5-dimethyl-p-benzoquinone. 131 U8 J.M.GOULD,D.R.ORT 3. These results provide strong evidence that a profound difierence exists between the two sites of energy coupling associated with non-cyclic electron trans- port in isolated chloroplasts. That is, the rate of electron flow through coupling Site I, which is the rate-determining step in the Hill reaction, is strictly regulated by phosphorylating conditions, whereas the rate of electron flux through coupling Site 11 is independent of phosphorylating conditions. 4. A model is presented which accounts for the lack of control over electron transport exhibited by coupling Site 11. It is postulated that Site II is coupled to an essentially irreversible electron transport step, so that conditions which affect the phosphorylation reaction would have no effect on the rate of electron transport through the coupling site. Two essentially irreversible reactions, closely associated with Photosystem II ——the water-splitting reaction and the System 11 photoact itself— are discussed as possible locations for coupling Site 11. INTRODUCUON Saha et al.1 were the first to point out that lipophilic strong oxidants such as oxidized p-phenylenediamines could intercept electrons from the chloroplast electron transport chain primarily at a point between the two photosystems. Subsequent work has shown that the preponderant portion of the electron transport to these “Class III” oxidants is insensitive to the plastocyanin inhibitors KCN (ref. 2) and poly(L)-lysine (ref. 3) and the plastoquinone antagonist dibromothymoquinone”. Thus, when electron flow to Photosystem I is blocked by one of these inhibitors, Class III acceptors are reduced by an electron pathway which includes only Photo- system 11. Recently our laboratory“5 and Trebst’s laboratory6 have shown that there is a site of energy conservation closely associated with Photosystem II-driven photo- reduction of Class III acceptors. Evidence is now accumulating that this newly discovered coupling site close to Photosystem II differs in several fundamental aspects from the well—known coupling site located after plastoquinone and before cytochrome f 7’8. When the two coup- ling sites are functionally isolated by partial reactions of the electron transport chain9, the coupling site between plastoquinone and cytochrome f (Site I) exhibits a pH-dependent phosphorylation efifciency (P/ez ratio) (optimal P/e2=0.6 at pH 8.0—8.5) whereas the coupling site located before plastoquinone (Site 11) is less efficient, with a pH-independent P/e2 ratio of 0.3—0.4. In addition, we have notedg'lo that coupling Site 11 apparently exerts no control over the rate of coupled electron trans- port. That is, the rate of electron flux through coupling Site 11 is not stimulated by the presence of uncouplers or ADP plus phosphate (Pi). Conversely, Site 1 exerts tight control over the associated electron flow, responding sharply to the presence of phosphorylating or uncoupling conditions. From these results we have concluded that coupling Site I alone constitutes the rate-determining step in the reduction of conventional Hill oxidants such as ferricyanide or methylviologen". However, Trebst and Reimer6 have reported data which indicates that elec- tron transport through coupling Site 11 is regulated by phosphorylating conditions. They reported that the reduction of substituted p-benzoquinones in the presence of dibromothymoquinone was stimulated by the addition of ADP plus Pi or amine 132 DIFFERENCES IN CHLOROPLAST COUPLING SITES 159 uncouplers. In an effort to resolve the apparent discrepancy between their data and our own, we have re-examined this problem in considerable detail. In this paper we report conclusive evidence that substituted p-benzoquinones such as 2,5-dimethyl- quinone not only accept electrons at a point before the site ofdibromothymoquinone inhibition5 (i.e. before plastoquinone”) but also catalyze a “bypass” around the dibromothynoquinone block, allowing electrons to pass through coupling Site I to Photosystem 1. When KCN2 is used to block the bypass electron flow by inacti- vating plastocyanin”, no stimulation of electron transport by ADP plus Pi is ob- served, indicating that Site 11 in fact does not exert control over coupled electron transport. This important difference in the properties of coupling Sites land 11 may reflect a fundamental difference in the mode of energy transduction at the two sites. EXPERIMENTAL METHODS The techniques employed in this study were similar to those described in previous papers. Chloroplasts were isolated from fresh market spinach (Spinacia oleracea L.) as” described earlierg. The photoreductions of 2,5-dimethquuinone and oxidized p-phenylenediamine were measured spectrophotometrically as the decrease in absorbance of the reaction mixture at 420 nm due to the reduction of excess ferricyanide'. Reactions (in a final volume of 2.0 ml) were run in thermo- stated cuvettes at 19 °C. Actinic light (>600 nm; 400 kergs-s’1'cm‘2) was supplied by a 500-W slide projector and the appropiate colored glass filters. ATP formation was determined for an aliquot of the reaction mixture as the residual radioactivity in the aqueous phase after extracting unreacted ortho- phosphate as phosphomolybdic acid into a butanol—toluene mixture (1:1, v/v) as described by Saha and Good”. Radioactivity in the final aqueous phase was deter- mined using the Cerenkov technique of Gould et al.”. KCN-treated chloroplasts were prepared by incubating chloroplasts in 30 mM KCN (buffered at pH 7.8) at 0 "C for 90 min as described by Ouitrakul and Izawaz. Stock solutions of 2,5-dimethquuinone and dibromothymoquinone were pre- pared in ethanol-ethylene glycol (I :1, v/v) and diluted so that the final concentration of organic solvent in the reaction mixture never exceeded 2%. RESULTS We noted previously that dibromothymoquinone, which blocks electron flow at plastoquinone, strongly inhibits electron transport and ATP formation when ferricyanide is the electron acceptor, but only partially inhibits electron transport and phosphorylation when the lipophilic Class III oxidant p-phenylenediimine (oxidized p-phenylenediamine) serves as the electron acceptors. Since the reduction ofClass III acceptors is known to contain two components, one solely dependent on Photosystem II and one requiring both Photosystem II and Photosystem 1 (ref. 2), it was concluded that dibromothymoquinone was blocking the Photosystem I com- ponent of oxidized p-phenylenediamine reduction. This conclusion was confirmed by the observation that the reduction of Class III acceptors in the presence of dibro- mothymoquinone is completely insensitive to plastocyanin inhibition by KCN"). Fig. I shows the effect of dibromothymoquinone on the reduction of ferri- 133 160 J. M. GOULD, D. R. ORT cyanide and oxidized p-phenylenediamine in the presence and absence of a complete phosphorylating system. As the dibromothymoquinone concentration approaches 510‘7 M the rate of electron transport to oxidized p-phenylenediamine in the 1000 r 1 at r r r 1 r v I LofF-o-o F 1. ELECTRON TRANSPORT \"11 F’Io2 N \ . , - . coo» ' "20*PDO‘isl/2) - E 05 ‘ko .P\-s: (I 0' - . o J I l Doégfi/epi 0 600 \- o " 500 r r 1 r r 1. \g. are FORMATION E. .c a 0 5‘2 1. 5 8 fig 400» HzO—c-Fccy '1 '1'? 400 '1 .. o ‘4: . -p. j: \ Poo: . , 1 .2 \/ OF E \o 8 200 \{o /' ‘ ’5 200» \.: \ \O—OI g o 1 1 1 1 3 o 1 1 9 1 - " O I 2 3 4 5 O I 2 3 4 5 DIBROMOTHYMOOUINONE (M11 107) Fig. 1. Effect of the plastoquinone-antagonist dibromothymoquinone on electron transport and ATP formation associated with the photoreduction of ferricyanide and oxidized p-phenylene- diamine by isolated chloroplasts. The reaction mixture contained 0.1 M sucrose, 2 mM MgC12, 50 mM Tricine—NaOI—I buffer (pH 8.0), 0.75 mM ADP, 5 mM Na2H32P04 (when added), chloro- plasts equivalent to 15 ug chlorophyll, and the indicated acceptor system. The acceptor systems were: Fecy, 0.4 mM potassium ferricyanide; PDox, 0.5 mM p-phenylenediamine plus 1.5 mM potassium ferricyanide. Note that as the Photosystem I component of PDox reduction is elimi- nated, the stimulation of electron transport by P1 is also eliminated, even though a high rate of ATP formation remains. Also note that at dibromothymoquinone concentrations ; 2.5- 10—7 M, where dibromothymoquinone itself functions as a Class III acceptor”, the stimulation of electron transport by P. is not observed, although phosphorylation, with the characteristic Site 11 P/Cz ratio of 0.3—0.4, does occur. phosphorylating (+P,) system falls to the level of the nonphosphorylating (—Pi) system. Nevertheless, this dibromothymoquinone-insensitive electron transport remains firmly coupled to ATP formation, even though the stimulation of electron transport by ADP plus Pi is no longer observed. Similar results can be seen when ferricyanide acts as the electron acceptor. This is because dibromothymoquinone, at concentrations greater than 2.5- 10‘7 M, not only blocks at plastoquinone but also functions as a Class III electron acceptor”, the reduced dibromothymoquinone being rapidly reoxidized by the excess ferricyanide present in the reaction mixture. In both systems the P/ez ratio falls to around 0.4, the characteristic efficiency of coupling Site [12'4‘5'9'19 Since both systems (in the presence of 5- 10’7 M dibromo- thymoquinone) are insensitive to plastocyanin inhibition by KCN, we can conclude that only coupling Site 11 is involved. While the data in Fig. 1 clearly show that coupling Site 11 exerts no control over Photosystem II electron transport, Trebst and Reimeré, using substituted p-benzoquinones as Class III acceptors, have shown that electron transport to these compounds is stimulated by ADP plus Pi and by amine uncoupling, even in the pres- ence of dibromothymoquinone concentrations which completely block electron transport to Photosystem I. From this data they concluded that coupling Site II does exert control over the rate of electron transport. To resolve this apparent discrepancy 134 DIFFERENCES IN CHLOROPLAST COUPLING SITES 161 with our own results, we performed similar experiments using 2,5-dimethyl-p- benzoquinone as the Class III acceptor. Fig. 2A shows that in the mixed system (i.e. dimethquuinone reduction by both Photosystem II alone and Photosystem 11 plus Photosystem I), considerable stimulation of electron transport by the complete phosphorylation system (+P,) is observed. These data confirm the earlier results of Saha et a1}. If 5- 10'7 M dibro- mothymoquinone is added (Fig. 28), however, the rate of electron transport is inhibited, indicating that the Photosystem I component of dimethquuinone reduc- tionz'S has been largely eliminated. When no dimethquuinone is present in the reaction mixture (dimethquuinone=0, Fig. 2B), the residual rate of electron trans— port, which is due to the Class III-type reduction of ferricyanide via dibromothymo- quinone“), shows no stimulation by ADP plus P,. Nevertheless, when dimethyl- quinone is added, a small but significant stimulation of electron transport by ADP plus P, is observed. This confirms the findings of Trebst and Reimer6 that control r v 1 r r T‘ n r r I I —' 1 to’\ A '01- a '0” C 'l N . °\,, 1 .3 N .. N _‘ {1 05'- " " Q 051' Q 05.. Q a <-o—-—0-————-o——————°—— a e e c .. O I I 4 1 O l 1 1 1 o L L I L 2000 r T r 1 1000 I T r f 10°C I r r r E“ 3 ET OP E 2 1.1.1.» /°——-—*°—— a aoo~ -- . ° . a 1.01.. . 5 ° \ a a - OP — _ C / ' C o/ .C U o 57 -—° // a” 5'1" 3P. To '0' . o 0’ \ 0.. E .2001 /-P. — E 6001— /‘..\_p 4 E 600 o « h /0 o— C '1 I I: O/ \ o \ \ c1 .‘1 / °- 2 / >3 .1... / 2 11.0.1/ 800 O ‘1 u, '- 3 " W -4 g 0 are 23’ % E \ E e 3. A 6_ 3- 3- 4 h A—5 . .1 5 0 400A —1 B 200 ‘/ATP 0 200 .ATP 3 O E >- I ______._—o— 2 g l/FERRICYANIOE ALON 3 , /¢/O 3 a/0 o b_ 1 ol 1 1 L 1 L ‘1/ 1 1 1 1 1 o / 1 1 1 1 0 0.25 0.5 0.75 1.0 O 0.25 0.5 O 75 I O O 0.25 0.5 0.75 1.0 DIMETHYLOUINONE (mM) DIMETHYLOUINONE (mM) DIMETHYLOUINONE (mM) Fig. 2. Effect of the electron transport inhibitors dibromothymoquinone and KCN on electron transport (E.T.) and phosphorylation (ATP) when dimethquuinone is the electron acceptor. The basic reaction mixture is as described in Fig. 1 when ferricyanide was the electron acceptor. Specific conditions are as follows: (A) No inhibitor added. Note that when no dimethquuinone is present (i.e. ferricyanide is being reduced via the normal Hill reaction) there is a large stimula- tion of electron transport by the complete phosphorylating system (+ P1). When dimethquuinone is added, a large increase in the rates of electron transport is observed, although the absolute amount of stimulation by P1 remains about the same. (B) Dibromothymoquinone (510—7 M) was added. Here rates of electron transport are lower since dibromothymoquinone blocks the Photosystem I component of dimethquuinone reduction. Note that in the absence of dimethyl- quinone (i.e. when dibromothymoquinone serves as the electron acceptor) no stimulation by ADP plus P1 is observed. When increasing concentrations of dimethquuinone are added, however, a small but significant stimulation of electron transport by the complete (+ P1) system is observed. (C) Chloroplasts blocked at plastocyanin by KCN treatment (see Methods); 5-10‘7 M dibromo- thymoquinone added. Note that the stimulation of electron transport by the complete phosphory- lating system (see B) is abolished by plastocyanin inhibition, indicating that electrons are by- passing the dibromothymoquinone-induced block. Nevertheless, a substantial rate of ATP for- mation remains even when this bypass is blocked by KCN. .135 162 J. M. GOULD, D. R. ORT is present in this system, even in the presence of a dibromothymoquinone block. However, by using the plastocyanin inhibitor KCN it is possible to show that this control of electron transport by phosphorylating conditions is not due to coupling Site II. If electron flow is completely blocked at plastoquinone by dibromo- thymoquinone, then inactivation of plastocyanin by KCN should have no effect on dimethquuinone reduction. However, as Fig. 2C shows, treatment of chloroplasts with KCN completely eliminates the stimulation of electron transport by ADP plus Pi seen in Fig. 2B. Nevertheless, the remaining electron transport is coupled to ATP formation with an efficiency of 0.3—0.4. Thus, it can be concluded that in the presence of dibromothymoquinone plus dimethquuinone, there is a small amount of electron leakage around the dibromothymoquinone block which allows a slow rate of electron flux through coupling Site I. Since Site I exerts tight control over electron transport9, this would account for the stimulation of electron flow by ADP plus Pi which is observed in this system. Indeed, when this electron leakage is blocked at plastocyanin, the remaining pure Photosystem ll reaction, utilizing only coupling Site II, is not stimulated by ADP plus Pi. It should be pointed out that we have not observed this leakage phenomenon when oxidized p-phenylenediamine (Fig. l) (g I I 1 IO 5,) ~ 2000 l l I \,.,Fccy A N : Q 0.5CM 0 /p0011 (9 DBMIB) 4 E gaze—”pom ‘1 \I\ 8 0 A b i \O\ .1 i, E U \. E \o \,Po°, to 06MB) 1: \ : iooo— \ o ‘5 o 7‘ P\ 0’ \ a . / \ ° 3 g / U. P . .x _. 0 / 6 800L- / ‘ -l g 'OOOF ./ .\i Q. (E) o \‘\ g / f \FCC’ q . a. soo~/ a E H20——>-Fecy ’— ‘1 z ‘3 f o '5 400 E 500‘ _ s 1 a 5 .1 _>_ DJ 3 U 200 _ g /ATP A 0 J\ A A o l i l 1 o 5 T6 fit, 0 5 IO I5 20 METHYLAMINE (mM) ME THYLAMINE (mM) Fig. 3. Effect of the uncoupler methylamine hydrochloride on electron transport (E.T.) and ATP formation when ferricyanide (Fecy) and oxidized p-phenylenediamine (PDrx) serve as electron ac- ceptors. Reaction conditions are as in Fig. I. When PDQx served as the electron acceptor, 5' 10‘7 M dibromothymoquinone (DBMIB) was added to block the Photosystem I component of PDQ): reduction. Note that the ferricyanide system shows a big stimulation of electron transport as the rate limitation at coupling Site I is relieved, while electron transport in the PD”. system is in- hibited as Site II becomes uncoupled. Fig. 4. Elect of methylamine on the rate of electron transport when ferricyanide (Fecy) and oxidized p-phenylenediamine (PDox) are the electron acceptors. Reaction conditions are as in Fig. 3 except that dibromothymoquinone was omitted from the Pan system to eliminate a secon- dary effect of the inhibitor on the quantum efficiency of Photosystem [I (ref. 10). I36 DIFFERENCES IN CHLOROPLAST COUPLING SITES 163 or other, substituted quinonediimides are used as Class III acceptors, suggesting that the p-benzoquinones may be able to substitute partially for the natural electron carrier plastoquinone. The effect of the uncoupler methylamine on electron transport and phos- phorylation associated with the reduction of ferricyanide and oxidized p-phenyl- enediamine (in the presence of 5- 10‘7 M dibromothymoquinone) is shown in Fig. 3. As is already widely known, methylamine stimulates the rate of electron transport when ferricyanide is the electron acceptor by uncoupling electron flow from a rate- limiting energy conservation reaction”. Since we have shown previously that coup- ling Site I is the rate-determining step for the Hill reaction9, we can conclude that methylamine’s preponderant effect on electron transport is by releasing the rate- limitation at Site I. Methylamine has an entirely different effect on a Photosystem II partial reaction (utilizing only Site II), however. Instead of stimulating the rate of electron transport as Site II becomes uncoupled, methylamine inhibits electron flow. Since this inhibition increases with increasing concentrations of methylamine, even after ATP formation has been completely abolished, it seems likely that the inhibi- tion is actually due to a secondary effect of the amine on Photosystem 11. Indeed, high concentrations of methylamine have been shown to inhibit the water-splitting reaction”. In this experiment (Fig. 3) the rate of electron transport in the presence of )5 mM methylamine is somewhat lower when oxidized p-phenylenediamine is the electron acceptor than when ferricyanide is the electron acceptor. This is due to the dibromothymoquinone present in the p-phenylenediamine system, since dibromo- thymoquinone has been shown to have a secondary effect on the quantum efficiency of Photosystem II“). In the absence of dibromothymoquinone (Fig. 4) similar results are obtained for the effect of methylamine on ferricyanide and oxidized p-phenylene- diamine reduction. It is clear that, as the rate limitation at coupling Site I is released, the rate of ferricyanide reduction increases until the secondary inhibition of Photo- system II by methylamine becomes the rate-limiting factor. DISCUSSION There is an impressive amount of evidence accumulating which indicates that the mechanisms of energy conservation at Sites I and II are not identical. When these sites are isolated by partial reactions of the electron transport chain9, it can be demonstrated that they differ in their coupling efficiencies (P/e2 ratios) and in the effect of pH on these efficiencies. Site II exhibits a characteristic P/ez ratio of 0.3—0.4 which is practically pH-independent from pH 6 to 9, whereas Site I, which is the rate-limiting step for the Hill reaction, exhibits a strongly pH-dependent P/ez ratio having an optimal value of about 0.6 at pH 8.0—8.5. Furthermore, it has recently been shown that HgClZ, which is an energy transfer inhibitor in chloroplasts”, can preferentially inhibit the coupling reaction at Site I without affecting ATP for- mation supported by Site II”. In this paper we have reconfirmed that, unlike Site I, coupling Site [I does not exert control over the rate of associated electron flux. That is, the rate of electron transport through coupling Site II is independent of the presence of ADP plus Pi or uncouplers. Nevertheless, under phosphorylating conditions ATP formation supported by Site II can occur at very high rates. 137 164 J. M. GOULD, D. R. ORT Furthermore, we have presented data which allows a reinterpretation of the findings of Trebst and Reimer", who concluded from experiments with dibromo- thymoquinone and 2,6-dimethyl-p-benzoquinone that coupling Site [I does exert control over electron transport. Apparently certain p-benzoquinones, in the presence of dibromothymoquinone, can catalyze an electron bypass around the dibromo- thymoquinone-induced plastoquinone block, perhaps by substituting for plastoqui- none. This does not seem unlikely in view of the structural similarities between these p-benzoquinones and plastoquinone. When KCN blocks this bypass, however, no control over electron transport by Site II is observed. Indeed, Trebst and Reimer6 noted that when they used higher concentrations of dibromothymoquinone in the presence of ferricyanide (so that dibromothymoquinone itself functioned as aClass I II acceptor) no effect of uncouplers on electron transport was observed. This obser- vation is in agreement with our own results obtained in a more extensive study of dibromothymoquinone as an electron acceptor”. It is also important to note that oxidized p-phenylenediamines, when used as Class III acceptors, do not catalyze this bypass reaction around the dibromothymoquinone block. It is possible to construct a model which explains the observed differences between Site II and Site I in their ability to regulate electron transport. In many respects coupling Site I resembles sites associated with the mitochondrial respiratory chaing. By analogy with mitochondria”, the electron transport between adjacent electron carriers (A and B) at Site I can be viewed as an equilibrium system. Thus, in the light, when Photosystem I is rapidly draining electrons from B, the reaction (A—> B) would be pulled to the right. However, as the high energy state (~) generated in the coupled energy conservation reaction begins to accumulate, a back pressure is created against the flow of electrons from A to B by reversal of the energy conser- vation steps. When a complete phosphorylating system (i.e. ADP plus Pi) is present, however, the pool of high energy intermediate is much smaller since it is continually being utilized to drive ATP formation. Thus the back pressure exerted by this pool is diminished and electron transport from A to B is stimulated. Similarly, in the pres- ence of uncouplers, the high energy state is rapidly dissipated and no significant back pressure is present, allowing very high rates of electron transport. Coupling Site II does not exhibit the tight control over electron transport seen at Site I. This can be readily explained, however, if the oxidation—reduction reaction which gives rise to the energy c0nservation step at Site II is essentially an irreversible step. That is, the nature of the forward reaction (A—> B) is such that the reverse reaction is thermodynamically prevented. In this case, the accumulation of the high energy intermediate or state (~) would still exert a back pressure on the forward reaction, but this back pressure would have no effect on the rate of electron transport from A to B. This would account for the fact that conditions which drain or dissipate the high energy intermediate pool do not effeCt the rate of electron flow at Site II. Several observations lead us to believe that this model does indeed provide a reasonable explanation for the differences between Site II and Site I discussed in this paper. We have shown previously that coupling Site II occurs at a point in the electron transport chain before the electrons from the independent Photosystem 11 units are pooled'o. This indicates that coupling Site II is located prior to plasto- quinone in the electron transport chain. The insensitivity of the phosphorylation I38 DIFFERENCES IN CHLOROPLAST COUPLING SITES 165 associated with the reduction of Class III acceptors to the plastoquinone antagonist dibromothymoquinone lends strong support to this argument. Indeed, the existence of a Photosystem II-driven “proton pump” before plastoquinone has been demon- strated9. There are at least two reactions in this portion of the electron transport chain which could be considered essentially irreversible. One of these is the System II photo- act itself. It has been suggested that a photochemical quantum conversiOn results in the formation of a reduced acceptor Q‘ and an oxidized donor Z+ on opposite sides of the thylakoid membrane (e.g. ref. 20). According to this model the resulting electrical field or membrane potential could serve as an energy reservoir to drive ATP formation as elaborated by Mitchell“. A second essentially irreversible reaction closely associated with Photosystem II electron transport is the water-splitting reaction. The existence of a coupling site on the water-oxidizing side of Photosystem II has already been considered by several author56'22‘24. It has been suggested that the protons from water oxidation are released to the inside of the thylakoid membrane, generating a transmembrane H+ gradient which is capable of driving ATP formation”. Experiments are currently in progress in an attempt to further define the exact location of coupling Site II in chloroplasts. ACKNOWLEDGEMENTS The authors would like to thank Drs S. Izawa and N. B. Good for many helpful comments on this work. We would also like to acknowledge valuable dis- cussions with Dr Bessel Kok concerning the interpretation of the control mechanisms discussed above. This work was supported by grants (GB 22657 and GB 37959X) from the National Science Foundation, U.S.A. REFERENCES l Saha, S., Ouitrakul, R., Izawa, S. and Good, N. E. (1971) J. Biol. Chem. 246, 3204—3209 2 Ouitrakul, R. and Izawa, S. (1973) Biochim. Biophys. Acta 305, 105—118 3 Ort, D. R., Izawa, 8., Good, N. E. and Krogmann, D. W. (1973) FEBS Lett. 31, 119—122 4 Gould, J. M., Izawa, S. and Good, N. E. (1973) Fed. Proc. 32, 632 5 Izawa, S., Gould, J. M., Ort, D. R., Felker, P. and Good, N. E. (1973) Biochim. Biophys. Acta 305, ll9-128 6 Trebst, A. and Reimer, S. (1973) Biochim. Biophys. Acta 305, 129—139 7 Avron, M. and Chance, B, (1966) Brookhaven Symp. Biol. 19, 149—160 8 Bohme, H. and Cramer, W. A. (1972) Biochemistry 11, 1155—1160 9 Gould, J. M. and Izawa, S. (1973) Biochim. Biophys. Acta 314, 211—223 10 Gould, J. M. and Izawa, S. (1973) Eur. J. Biochem. 37, 185—192 11 Bohme, H., Reimer, S. and Trebst, A. (1972) Z. Naturforsch. 26b, 341 -352 12 Izawa, S., Kraayenhof, R., Ruuge, E. K. and DeVault, D. (1973) Biochim. Biophys. Acta 314, 328-339 13 Saha, S. and Good, N. E. (1970) J. Biol. Chem. 245, 5017—5021 14 Gould, J. M., Cather, R. and Winget, G. D. (1972) Anal. Biochem. 50, 540—548 15 Good, N. E. (1960) Biochim. Biophys. Acta 40, 502—517 16 Izawa, 8., Heath, R. L. and Hind, G. (1969) Biochim. Biophys. Acta 180, 388—398 17 Izawa, S. and Good, N. E. (1969) Prog. Photosynlh. Res. 3, 1288-1298 18 Bradeen, D. A., Gould, J. M., Ort, D. R. and Winget, G. D. (1973) Plant Physiol., in the press I39 166 J. M. GOULD, D. R. ORT l9 Chance, B. (1972) FEBS Lett. 23, 2—30 20 Witt, H. T. (1971) Q. Rev. Biophys. 4, 365—477 21 Mitchell, P. (1966) Biol. Rev. 41, 445—602 22 Schwartz, M. (1968) Nature 219, 915—919 23 Yamashita, T. and Butler, W. L. (1968) Plant Physiol. 43, 1978-1986 24 Bohme, H. and Trebst, A. (1969) Biochim. Biophvs Acta 180, l37~l48 25 Junge, W., Rumberg, B. and Schroder, H. (1970) Eur. J. Biochem. 14, 575—581 APPENDIX IV STUDIES ON THE ENERGY COUPLING SITES 0F PHOTOPHOSPHORYLATION II. TREATMENT OF CHLOROPLASTS WITH NHZOH PLUS ETHYLENEDIAMINE- TETRAACETATE TO INHIBIT WATER OXIDATION WHILE MAINTAINING ENERGY COUPLING EFFICIENCIES 14;) Studies on the Energy-coupling Sites of Photophosphorylation II. TREATMENT OF CHLOROPLASTS WITH NH20H PLUS'ETHYLENEDIAMINETETRAACETATE TO INHIBIT WATER OXIDATION WHILE MAINTAINING ENERGY-COUPLING EFFICIENCIESl DONALD R. ORT AND SEIKICHI IZAWA Received for publication May 11, 1973 Department of Botany and Plant Pathology, Michigan State University, East Lansing, Michigan 48824 ABSTRACT Artificial electron donors to photosystem II provide an im- portant means for characterizing the newly discovered site of energy coupling near photosystem II. However, water oxidation must be completely abolished, without harming the phosphoryl- ation mechanism, for these donor reactions and the associated phosphorylation to withstand rigorous quantitative analysis. In this paper we have demonstrated that treatment of chloroplasts with hydroxylamine plus EDTA at pH 7.5 in the presence of Mg“ followed by washing to remove the amine is a highly re- liable technique for this purpose. The decline of the Hill reaction and the coupled phosphorylation during the treatment were carefully followed. No change in the efficiency of phos- phorylation (P/e: 1.0—1.1) was observed until the reactions became immeasurable. Photosystem l-dependent reactions, such as the transfer of electrons from diaminodurene or re- duced 2,6-dichlorophenolindophenol to methylviologen, and the associated phosphorylation were totally unaffected. It is clear that the hydroxylamine treatment is highly specific, with no adverse efl'ect on the mechanism of phosphorylation itself. Benzidine photooxidation via both photosystems II and I in hydroxylamine-treated chloroplasts (electron acceptor, methyl- viologen; assayed as 02 uptake) supports phosphorylation with the same efficiency as that observed for the normal Hill reaction (P/e2 ‘2 1.1). An apparent P/e: ratio of 0.6 was computed for the photooxidation of ascorbate. The recent papers from this and other laboratories (15, 21, 25—28), which dealt with partial reactions of chloroplast elec- tron transport, strongly indicated the existence of a site of energy coupling in the vicinity of photosystem II (most prob- ably before plastoquinone), in addition to the well recognized site of phosphorylation between plastoquinone and cytochrome f (4, 5). Consequently, the question has arisen as to the exact location of this second site of phosphorylation. We must seriously consider here the possibility that an energy-con- servation reaction is coupled to the process of water oxidation or to photoact 11 itself, since available thermodynamic data (10) seem somewhat unfavorable to the existence of an energy- conservation reaction between photosystem II and plasto- quinone. ‘ This work was supported by Grant GB 22657 from the National Science Foundation. As an approach to the problem of mapping the location of this site, investigations of the quantitative relationships between electron transport and phosphorylation supported by artificial donors to photosystem II become quite important. This ap- proach calls for a specific and complete inhibition of water oxidation. Yamashita and Butler (31, 32), using their “tris- washed” chloroplasts, and Bohme and Trebst (6), using mildly heat-treated chloroplasts, have already shown that the donor reactions mediated by photosystem II can support phosphoryla- tion with various efficiencies (P/ e,2 ratios) depending upon the electron donor used. It is clear, however, that more extensive studies are required, if one is to draw decisive conclusions as to the site of phosphorylation, paying careful attention to possible adverse effects of these treatments or of the electron donors used or both on the machinery of phosphorylation itself. For instance. we have noticed that our chloroplasts are somewhat resistant to tris treatment, and our attempts to totally abolish water oxidation without appreciably impairing the phosphoryla- tion mechanism have not been successful. Even greater diffi- culties in terms of the inhibition specificity were encountered with the heat treatment. Hydroxylamine is a potent inhibitor of water oxidation. Cheniae and Martin (8) have shown that its effect on isolated chloroplasts is specific and irreversible, involving a release of Mn from the chloroplast membranes. No inhibition of photo- system I-mediated electron transport was found. The use of hydroxylamine as an electron transport inhibitor for photo- phosphorylation studies have been shunned because of the ambiguous results one would expect from its possible un- coupling effect as an amine or its ability to serve as an elec- tron donor to photosystem II (22, 29). Wessels (30) did employ NH20H in his very early studies on photophosphorylation. The effect of hydroxylamine-O-sulfonate on photophosphorylation has recently been studied by Elstner et al. (11). We have ex- amined the effect of hydroxylamine on chloroplast reactions under various conditions and found a simple method of treating chloroplasts with this amine which allows total inhibition of water oxidation without any detectable damage to the mech- anism of phosphorylation. This paper describes details of the method and some preliminary results of photophosphorylation experiments with the hydroxylamine-treated chloroplasts. MATERIALS AND METHODS Chloroplast Isolation. Chloroplasts were isolated from commercial spinach (Spinacia oleracea L.). Leaves were washed ” Abbreviations: P/e2: the ratio of the number of ATP molecules formed to the number of pairs of electrons transported; DAD: di- aminodurene (2,3 ,5 , 6-tetramethyl-p-phenylenediamine); DC IP: 2 , 6-dichlorophenolindophenolz MV: methylviologen. 595 141 with cold-distilled water and ground in a Waring Blendor for 5 sec in a medium consisting of 0.3 M NaCl, 30 mM Tricine- NaOH buffer (pH 7.8), 3 mM MgC12, and 0.5 mM EDTA. The homogenate was filtered through eight layers of cheesecloth, and the chloroplasts were sedimented at 2500g for 2 min. The chloroplast pellet was then resuspended in a medium con- taining 0.2 M sucrose, 5 mM HEPES-NaOH buffer (pH 7.5), 2 mM MgC12, and 0.05% bovine serum albumin. After a 45-sec centrifugation at 2000g to remove cell debris, the chloroplasts were spun down again (2000g 4 min) and finally suspended in a few milliliters of the above suspending medium. Chemicals. A stock solution of 0.1 M NH,OH was made by dissolving the hydrochloride salt in 0.05 N HCl and stored at 0 C. Fresh solutions were prepared every 3 to 4 days. When necessary, the pH of the NH20H solution was adjusted to de- sired pH values immediately before use. D-Ascorbate solution (0.1 M; pH adjusted to 6.5 with NaOH) was stored at —20 C in small, tightly sealed vials. Diaminodurene dihydrochloride and benzidine dihydrochloride were recrystalized from char- coal-treated aqueous alcoholic and aqueous solutions, respec- tively, by adding excess HCI at 0 C. The completely colorless crystals thus obtained were stored at —20 C. Fresh aqueous solutions of these compounds were made up daily and kept at 0 C during the experiment. Hydroxylamine Treatment. NH,OH and EDTA (when used) were added to the suspending medium described above, pH adjusted with NaOH to 7.5, and used immediately. The treat- ment was done in the dark at either room temperature (21 C) or at 0 C as indicated. The Chl concentration during the NH,OH treatment was approximately 100 pg/ ml. Upon the completion of the prescribed treatment period, the chloroplasts were spun down (2000g, 4 min) and washed twice with sus- pending medium to remove the NH20H. The Chl concentra- tions of final stock suspensions were determined by the method of Arnon (2). FJectron Transport and Phosphorylation Assays. The fer- ricyanide Hill reaction was assayed as 0, evolution, and the MV Hill reaction as 02 uptake resulting from aerobic re- oxidation of reduced MV. Electron transport from artificial donors to MV was assayed as 02 uptake (20). A membrane- covered Clark-type oxygen electrode was used for these 02 assays. When artificial donors were used, the observed rate of electron transport was corrected for the slow rate of dark autooxidation of the donors which ranged from 5 to 20% of the rate in the light. In no case was it necessary to add a H202 trap to the reaction mixture, since the chloroplast prepar- ations used were free from catalase activity. The intensity of actinic light (600—700 nm) was approximately 600 kergs-sec‘1 cm‘”. The reaction temperature was 19 C. Phosphorylation was measured as the residual radioactivity after the extraction of the 32P-labeled orthophosphate as phosphomolybdicv acid in butanol-toluene (3). Radioactivity was determined by Cerenkov radiation as described by Gould et al. (14). RESULTS Effect of NILOH Added in the Reaction Mixture of Phos- phorylation. To test the potency of NHgOH as an uncoupler, we have examined the effect of NH30H added in the reaction mixture on postillumination phosphorylation (X3) (17) and on the steady state phosphorylation supported by the transfer of electrons from DAD to MV (Table 1). In the XE experiments, hydroxylamine was present only during the dark phosphoryla- tion stage. The inhibition of X5 thus observed has been shown to be a sensitive indicator of uncoupling (16). The concentra- tions of NH20H used in these experiments are those commonly used for inhibition of 0. evolution. Clearly, the uncoupler Table I. Eflect of NH30H on Postillumination ATP Formation (X3) and Steady State Photophosphorylation (Test for Uncoupler Action of NH20H) The XE experiments were carried out as described before (19). NH30H was present only in the dark phosphorylation stage. The reaction mixture for the steady state photophosphorylation (2 ml) contained 0.1 M sucrose, 50 mM Tricine-NaOH buffer (pH 8.0), 2 mM MgCl;, 0.75 mM ADP, 5 mM NagH’2P04, 0.5 mM DAD, 1 mM ascorbate, 50 pM MV, 1 pM DCMU, chloroplasts equivalent to 20 pg of Chl, and indicated concentrations for NHaOH. l Steady State Phosphorylation Postillumination (DAD _' MY) NBA)” ATP Formation - a - (XE) Electron . i transport ATP P" a” my nmoles/[00 at: (It! utqilthrl-mg nmolcCsI/‘ftr-mg ratio 0 8.3 2840 586 0.41 0. l 8. l 2.0 6.0 3.0 2660 442 0.35 5.0 4.0 i 2600 395 0.30 action of NH.OH is weak, as one would predict from the low basicity of this amine (pK. = 6) (13, 18). Weak as it is, this side effect of NH,OH is definitely undesirable when a rather precise assessment of phosphorylation efficiency is required. As described below, the uncoupling effect of NH20H can be completely eliminated by washing the NHZOH-treated chloro- plasts, without relieving the desired inhibition of water oxida- tion. Pretreatment of Chloroplasts with NH20H and EDTA. In all of the following experiments, chloroplasts were treated with NH20H under a variety of conditions and then washed twice with a large volume of amine-free suspending medium (see “Materials and Methods”) at 0 to 4 C. The data are for these washed chloroplasts. Figure 1A shows that the NH,OH inhibition of water oxida- tion proceeds much more slowly at 0 C than at 21 C. Cheniae and Martin (8) reported a Q... of 2.43 for the development of NH,OH inhibition. The effect of increasing concentrations of NHZOH is shown in Figure 1B for both 0 C and 21 C treat- ments. Notably, in both figures, there are significant rates of electron transport and phosphorylation remaining (typically 40—60 peq or 20—30 nmoles ATP hr" -mg Chl") when NH20H alone is relied upon to abolish water oxidation. This residual electron flow and accompanying phosphorylation are ob- literated when the treatment includes 1 mM EDTA. The above residual rates of electron transport (ferricyanide reduction) and phosphorylation are still in a ratio of approximately 2 (i.e. P/e2 : l). The rate of photosystem I-dependent electron trans- port from DAD to MV and the efi‘iciency of the associated phosphorylation are totally unaffected (inset, Fig. IB). These results clearly indicate that the weak uncoupling effect of NH20H is completely eliminated when NH20H-treated chloro- plasts are washed as described. Since the inclusion of EDTA seemed to be highly effective in achieving the complete inhibition of water oxidation, and the difference between 0 and 21 C treatments seems to be only in the rate of the development of inhibition. we have examined more closely the effect of pretreatment of chloroplasts with NH2OH plus EDTA at 21 C (Fig. 2). The P/e, ratio of non- cylic photophosphorylation (H20 -+ MV) remains the same until the rates of electron transport and phosphorylation be- come immeasurable. The photosystem I-dependent electron 142 3200 A I I f I E. . , g (D 0 __ >‘ E 3IOO- / 1 g 0 o 3 5 E a ,L 2 ° ’ DAD—+Mv 4, o a- - o o: E ’i EDTAIO ) E f SOOh ‘ 5 .z' _=_.- 3 0 g 0 a) 3 400 - _. 3 +— E s o —. 35 300 _ H20 MV (0) - 3) 2 -EDTA (2l°) E g / E f— __‘F 2005 ° H20 ecy _ 2 g kO/\ E DTA (21°) - E cratoot 8 a: /o *— 5 + E DTA (2l°) + EDTA (0°) 0 m IOOA\ .. u.) .1 1%?) _J Lu °-——o— m o “1u\“‘j“-.---i-—-—---.--- 0 I0 20 30 40 TREATMENT TIME (min) FIG. 1. A: Effect of time of hydroxylamine pretreatment of chloroplasts on the Hill reaction and photosystem I-dependent do- nor reactions. Chloroplasts were pretreated for the indicated periods of time with NH2OH dissolved in the suspending medium (see “Ma- terials and Methods"). The Chl concentration at this stage was ap- proximately 100 rig/ml. The concentrations of NHaOH were 5 mM for 0 C and 3 mM for 21 C treatment. EDTA was 1 mM when in- cluded. After the treatment, the chloroplasts were spun down, washed twice with amine-free suspending medium, and finally sus- pended in a few ml of the same medium. These procedures were carried out at 0 to 4 C. The basic ingredients of the reaction mix- ture (2 ml) were: 0.1 M sucrose, 50 mM Tricine-NaOH buffer (pH 2400- ELECTRON TRANSPORT _+_ ATP FORMATION J 1.; I X l/2 I C', HzO-O Fecy a lo .‘O—O—O—------ S 2300 " "r' d 7% DCIPHz-OMV N 0—0 0—0— _ . DAD—ow _L ‘b 05 b ... g 2200 a : E th—O—o—o— ; 9 ° DAD—ow o. .c ZIOO [ ‘- O —« \ O 2 4 6 8 IO (1 , / minutes treatment 2 4? 1r on O J 32’ 800 "it 0 o 0" :1. if DAD—UNIV :E a T . _\~ 2005;, . =3 - ~ 3': DCIPH2-.MV 8 I00 \= ‘ - — m H O —-Fec =- 2 Y H o —.Fecy O L $9flm 1 O‘NPXOE O 2 4 6 8 0 2 4 6 8 IO TREATMEN T TIME (minutes) FIG. 2. Effect of time of pretreatment of chloroplasts (at 21 C) with hydroxylamine plus EDTA on the Hill reaction, photosystem I-mediated donor reactions, and associated phosphorylation. The reaction mixture for the DCIPHa -> MV reaction contained 0.4 mM DCIP, 2.5 mM ascorbate, 50 1M MV, 1 11M DCMU. For other con- ditions see Fig. 1A. 26008 I r I r DAD—.MV 0 0 25000_ ° '-EDTA(O°) _ I °-o—-o\_o__ I.Ot- O / H20 —>Fecy (0°) 1’ 500 a N Q osr o 0- tit-o-——0——o —-—o __ 400— DADo—o-MVIO°) _ O l I L 300 - ° 20 3 4o 60 _ NHZOH (x IO M) in treatment (- EDTA, 0° ) HZO—bF_—ecy 200 - —-4 /-EDTA (2| °) -EDTA (0°) //+EDTA(21°) +EDTA (0°) IOO // a L / ._ A—A-_ O "-1. l “““““ O---:—------i ------ o—- 0 IO 20 30 4O 50 NHZOH (x 103M) in treatment media 8.0), 2 mM MgC12, 0.75 mM ADP, 5 mM NagHmPOn and chloro- plasts containing 40 pg of Chl (or 10 pg of Chl for the DAD -’ MV system). The concentrations of MV and Fecy (ferricyanide) were 50 1M and 0.4 mM, respectively. The DAD -> MV reaction was run in the presence of 1 MM DCMU and 2.5 mM ascorbate. The concen- tration of DAD was 0.5 mM. For assay conditions, see “Materials and Methods.” The temperatures indicated in the figure are for pre- treatment periods. B: Effect of varied concentrations of hydroxyl- amine (in pretreatment medium) on the Hill reaction and photo- system I-mediated donor reactions. The treatment time was 9 min in both 0 and 21 C treatment. For other conditions, see A. transport (DAD -> MV or DCIPH2 -’ MV) and the coupled phosphorylation are totally unaffected. Therefore it seems quite safe to conclude that brief 21 C (room temperature) treatment of chloroplasts with NHQOH plus EDTA at pH 7.5 in the presence of Mg“ and subsequent washing at 0 to 4 C with NHgOH-free media, provide a highly reliable method of abolishing the water-oxidizing ability of chloroplasts without impairing any other functions of chloroplasts including phos- phorylation. Photosystem II-mediated Donor Reactions in NH,0H- treated Chloroplasts. Figure 3 shows the photooxidation of D- ascorbate and the associated phosphorylation in chloroplasts which have been pretreated with NHQOH plus EDTA as de- scribed above. The electron acceptor used was methylviologen. The rate of electron transport was calculated according to the formula: ascorbate + 02 -> dehydroascorbate + H202 (for each pair of electrons transported). The stoichiometric disappear- ance of ascorbate was confirmed by titration with DCIP at acidic pH (Table II). The stoichiometric formation of dehy- droascorbate has also been reported by Bohme and Trebst (6), who found a P/e2 ratio of 0.5 for this donor reaction in heat-treated chloroplasts. Our data show a P/e2 ratio of 0.6 at an optimal pH of 8.0 to 8.5. The remote possibility that these low P/e2 ratios may be due to weak uncoupling by ascorbate has been ruled out by X, experiments (Table III) which failed to detect any uncoupling action of ascorbate up to 30 mM. l43 I j l l j (.0- ’3: o E. a? oso—o—o-o-o—o—o— z a ' Q 8 O E U o o 1 L E 300— 0 IO 20 .. 5 ASCORBATE I (03 (M) Q E.T. 4 200i- 0 -* U) 2 / E o 3. 8 ET (+DCMU) o '00 ATP ATP(7DCMU) 4 ——-— --o — g ./.’. . / .TC— ’6 1 o 9 ; i4 T: 0 5 IO I5 20 25 ASCORBATE x (03 (M) F IG. 3. Photooxidation of ascorbate with MV as the electron ac- ceptor in chloroplasts pretreated with hydroxylamine plus EDTA. Chloroplasts were pretereated for 9 min at 21 C with 3 mM NH=OH plus 1 mM EDTA as described in Fig. 1A. The basic ingredients of the reaction mixture are also given in Fig. 1A. The concentration of DCMU (when used) was 1 pM. 0.4 mM KCN was included so as to minimize the autoxidation of ascorbate. Note that the reactions contain practically no DCMU-insensitive components. Table II. Photooxidation of Ascorbate in NHgOH-lrealed Chloroplasts as Assayed by 0, Uptake and by Titration with DCIP Basic reaction conditions were as in Figure 3. The concentration of ascorbate used was 1.0 mM. After 3 to 4 min illumination, during which 02 uptake was continuously monitored, the reaction mix- ture was acidified to pH 6 with MES buffer (containing DCMU to prevent further oxidation in room light), centrifuged, and the resultant supernatant was titrated quickly using a microburette with a standardized 1 mM DCIP solution. The DCIP solution, freshly prepared, was standardized spectrophotometrically at pH 8.0 using the molecular extinction coefficient (Eeoo = 21,800 M”- cm“) of Armstrong (1), and also by titration with a fresh solution of ascorbate according to the stoichiometry—DCIP + ascorbate a DCIPH. + dehydroascorbate. Both methods gave the same value for the DCIP concentration within an error of 5%. A good agreement between the values obtained by O, assay and by titra- tion indicates that under the conditions employed ascorbate was photooxidized only to the level of dehydroascorbate. Ascorbate Consumption Experiment 0': uptake DCIP titration No. neg/'2 ml reaction m ixlurc l 235 251 2 235 229 3 223 285 Thus it seems that the apparent phosphorylation efficiency of ascorbate photooxidation via Photosystems II and I is close to half of the efficiency when water is photooxidized (see “Dis- cussion”). Also worthy of note here is the fact that with these NH,OH-treated chloroplasts the ascorbate oxidation contains practically no DCMU-resistant component (Fig. 3). A signifi- cant rate of DCMU-insensitive ascorbate photooxidation was observed for tris-washed chloroplasts (32), suggesting that as- corbate could be an electron donor for photosystem I, depend- ing on the integrity of the chloroplast membranes (24). Figure 4 shows that benzidine, unlike ascorbate, supports phosphorylation with a P/e, ratio of 1.1 which is almost the same as the value for normal noncyclic photophosphoryla- tion involving water oxidation (P/e2 = 1.0 to 1.2). In this experiment a low level of ascorbate (0.2 mM) was present in the reaction mixture to eliminate the possibility of a cyclic reaction with oxidized benzidine. Nearly identical data (not shown) were obtained without ascorbate, indicating that no significant cyclic electron flow occurred during the short illumination period employed (1—2 min). Table III. Eflect of Ascorbate on Postillumr’nation Phosphorylation (XE) The procedure and conditions for the experiments were as de- scribed before (19). Ascorbate was present only in the dark phos- phorylation stage. Note that ascorbate had no effect on the dark phosphorylation while the known uncoupler methylamine (also present at the dark stage only) greatly diminished the yield of ATP. Addition at Dark Stage ATP formed my nmoles/100 pg Chl None 6.8 Ascorbate (3) 8.0 Ascorbate (30) 7.5 Methylamine (5) 1.6 T I I l T '— P/ez (P/Oz) 5 E. .C O. 0 § 5 o L l a. o 0.5 )0 E BENZIDINE x (03 (M) a 200— — E O/°—\ O— & \ E T F. . . < 3 — ATP :1 .~ 0 E.T. (+DCMU) CD g ATP(+DCMU) 63.. o—Lfl—_°- o 2::IZJ_1__1_;_4 O 0.2 0.4 0.6 0.8 LO BENZIDINE x (03 (M) FIG. 4. Photooxidation of benzidine with MV as the electron acceptor in chloroplasts pretreated with hydroxylamine plus EDT A. Chloroplasts were pretreated for 9 min at 21 C with 3 mM NH20H plus 1 mM EDTA as described in Fig. 1A. The basic ingredients of the reaction mixture are also given in Fig. 1A. Ascorbate (0.2 mM) was included to keep benzidine in its reduced state, and 0.4 mM KCN to help prevent the autoxidation of ascorbate. Note that the phosphorylation efficiency of this reaction system is the same as that of normal noncyclic photophOSphorylation (P/Ca = 1.1). 144 DISCUSSION Currently, the most widely employed procedures for in— hibiting 02 production are the tris treatment of Yamashita and Butler (31, 32) and the mild heating of chloroplasts (to 50 C for several minutes) (6, 24). However, both of these methods are not quite satisfactory in dealing with critical experiments on photophosphorylation, as briefly mentioned earlier in this paper. Well coupled chloroplasts are rather resistant to Cl'- removal treatment (7) which is also known to inhibit water oxidation (22). Repeated washings with Cl' free media com- bined with room-temperature treatment did severely suppress the Hill reaction, but their secondary effects were quite appre- ciable. In this paper, we have presented a highly effective procedure for abolishing the water-splitting reaction in isolated chloro- plasts which can be extremely useful for photophosphorylation studies utilizing artificial electron donors for photosystem II. This procedure involves the treatment of chloroplast with NH,OH plus EDTA in the presence of Mg“. No signs of un- coupling or inhibition of energy coupling were observed after treated chloroplasts were washed with NI-LOH-free medium. Actually, we found NH20H to be a rather poor uncoupler of photophosphorylation, and therefore it is possible that some uncritical phosphorylation experiments may be carried out in the presence of a few millimolar NH20H which is sufficient to suppress water oxidation. The effectiveness of EDTA in facilitating the complete in- hibition of water oxidation by NH,OH suggests an interesting possibility concerning the mechanism of extraction of Mn from the lamellar membranes by NHQOH. Since EDTA per se has no effect on water oxidation nor is able to release Mn from the membrane (9) (in the absence of Mg‘”, EDTA-uncoupling [23] gives very high rates of the Hill reaction), the observed effect of EDTA on water oxidation is probably indirect. It seems possible that a portion of the Mn extraction by NH20H is reversible, and the binding of released Mn2+ by EDTA does not allow the reversal to occur. The apparent P/ ea ratio of 0.6 found for the ascorbate photo- oxidation in NH20H-EDTA-treated chloroplasts is similar to the value of 0.5 Bohme and Trebst (6) found for heat-treated chloroplasts. They interpreted the data to suggest that the donation of electrons by ascorbate may have occurred after one of two sites of phosphorylation, the site which they sug- gest to be associated with the water-oxidation step. Although this is certainly the simplest and the most attractive interpreta- tion, the validity of the widely used method for computing the electron flux in this donor reaction based on 02 uptake data may be in error (12). A more comprehensive assessment of this complication is presented in a subsequent publication. In this respect the data for benzidine photooxidation may be of more importance. The constant P/e, of 1.0 to 1.1 observed over a wide range of benzidine concentrations confirms and greatly strengthens the brief data of Yamashita and Butler (31). These authors, using tris-washed chloroplasts, found a P/e2 ratio of 0.97 with benzidine (33 pM) as electron donor and NADP’ as acceptor. These P/e. ratios are indeed very close to that of the normal Hill reaction. It seems unlikely, therefore, that there is a phosphorylation site specifically associated with the mecha- nism of water oxidation. However, the possibility still remains that an energy conservation reaction is linked to some step of oxidoreduction reactions on the water-oxidizing side of photo- system II, a step which is involved both in water oxidation and the oxidation of artificial reductants. Research is now in progress testing various donor reactions in NH20H-treated chloroplasts in an attempt to locate more precisely the photo- system II-associated site of phosphorylation. Note. After submission of this manuscript, a paper by J. F. Allen and D. 0. Hall (Biochem. Biophys. Res. Commun. 52: 856—862) has appeared in which the authors demonstrated that the aerobic photooxidation of ascorbate by (untreated) chloro- plasts involves a nonbiological oxidation of ascorbate by super- oxide radicals. Therefore, it is almost certain that the assump- tion 2e‘ : 0, used for computing electron flux and P/e2 in Figure 3 is incorrect. Al'lt'Ilt)?l"t'l!(I"16an'*1‘111‘ authors wish to express thanks to J. M. Gould and N. 1‘}. Good for helpful discussions, and the former for artistic contrilnitions. LITERATURE CITED 1. Anxtsrnoxo, J. 1964. The l'litilt't'llltll‘ extinction coefficient of 2.ti-dichloroplie- nolindophent'tl. Biochim. Biophys. Acta 86: 194 197. . Altxox', 1). l. 1949. (‘opper enzymes in isolated chloroplasts: in Beta ruluaris. Plant Physiol. 24: 1 15. . .-\\'lt()_\'. .\I. 1960. I’lititophosphorylation by Swiss chard cl'Ilort‘IplastS. Biochim. Biophys. Acta. 40: 257 272. . Avnox, .\1. AVD B. CHANCE. 1906. Relation of phosphorylation to electron transport in Isolated chloroplasts. Brookhaven Symp. Biol. 19: 149 160. . Binnie. II. .txn W. A. (human. 1972. Lm-nlization ot a site of energy coupling [0 pt )lyphenolox idase 03 fl in between plastoquinone and cytochrome f in the electron transport chain of spinach chloroplasts. Biochemistry 11: 1155—1100. . Bonnie. 11. AND A. 'l‘luznsr. 1969. On the properties of ascorbate photooxidation in isolated chloroplasts. Evidence for two ATP sites in noncyclic photo- phospltorylation. Biochim. Biophys. Acta 180: 137 7148. . Bork, J. M., C. Boviz. F. R. \an'ruzr. AND I). I. Alixox. 1963. Chloride requirement. for oxygen evolution in photosynthesis. Z. Ni‘IturtorsclI. 18b: 683 1183. 8. (‘ItI-zxixn, G. M. AND I. F. Mxttrix. 1971. Effects of hydroxylamine on photosystem II. I. Plant Physiol. 47: 563 575. 9. ("rinsing G. M. xxn I. F. MARTIN. 1966. Studies on the function of illzlltgzineso in photosynthesis. lit‘tit'Iklitn't'n Sytltp. “inl. 19: 400 417. 10. (,‘ltAMl-LR, W. A. AND W. L. Bl‘TlJ-th. 1969. Potentiometric titration of the fluorescence yield of spinach chloroplasts. Biochim. Biophys. Acta. 172: 5037510. 11. Ers'rxmi, 1*}. F., A. Harem” AND S. VAKIJNUVA. 1970. liber die Ileniniung des pliotosynthetischen Electrononentransports in isolierten (‘hloi'oplasten «lurch l'lydroxylalnin. Z. Pflanzenphysiol. D2: 173 183. 12. ELsTth, E. F., A. IIPZI‘I’EL, xxD S. \'.\KI.I.\'ovx. 1970. fiber die Oxidation \‘oit Hydroxylamin (lurch isolierte (.‘hloroplasten und die Inogliche Funk- 09 ‘3 Factors affecting the decay of 02 evolution. tion einer Peroxidase aus Spinalbliittern bei dcr Oxidation von Ascorbin- satire IIIId Glykolsiiure. Z. Pflanzenphysiol. 62: 184 200. 13. Goon. N. E. 1960. Activation of the Hill reaction by amines. Biochim. Biophys. Acta ~10: 5027517. 14. Gotta). J. M., It. FATHER, AND G. D. WINGET. 1972. Advantages of the use of Cerenkov counting for determination of P32 In photiIphos;thorylation re- search. Anal. BioclIeIII. 50: 540- 548. . Gttl‘Ll), J. M. AND S. IZAWA. 1973. I’httttosyst-em II electron transport and phosithorylation with dibroinotliynioquinone as the electron acceptor. l‘lur. J. Biochem. 37: 135—192. 16. IIIxD, G. AND A. ’1‘. JAGENIKJRF. 1965. Effect of unconplers on the conforma- tional and high energy states in chloroplasts. J. Biol. Chem. 240: 3202 3209. 17. HIxD, G. 11er A. T. JAGENDURI". 1965. Separation of light and dark stages in plnitophospliorylation. Proc. Nat. Acad. Sci. U.S.A. 49: 7157722. 18. “IND, G. AND ('3. 1’. 1963. Reduction of ferricyzmide by chloroplasts: in presence of nitrogenous bases. Biochim. Biophys. Acta 75: H Us W i l l 1"“ so HA M. 194 201. 19. szwx, S. 1970. The relation of post-illuniination ATP formation capacity (XE) to 11* accumulation in chloroplasts. Biochim. Biophys. Acta 223: 165 173. 20. IZAWA, S. T. N. (‘oxxotnq G. I). Wixowr, AN!) N. E. Goon. 1966. Inhibition and uncoupling of photoplnitsphorylation in chloroplasts. Brrmkhaven Symp. Biol. 19: 169487. Imwx, S. J. .\l. Got‘LD, D. R. Our. 1’. FELKHH, AND N. E. 1973. I'llectron transport. and photophos;)iIorylntion in chloroplasts as a function 2 . fl fool). of the electron acceptor III. A dibromotlIanoqIIiinitne-insensitive phos- phorylation reaction associated with I’ln'ttosystein II. Biochim. Biophys. Acta 305: 119—128. 22. IZAWA, S.. R. L. HEATH, AND G. Hist). 1969. The role. of chloride ion in photosynthesis III. The effect of artificial electron donors upon electron transport. Biochim. Biophys. Acta 180: 3.38 398. ‘23. JAGENIHHIF, A. 'I‘. AND M. SMITH. 1962. l'ncoupling Pl]0>plltt1‘)‘lttt1011 in spinach chloroplasts by absence of cations. Plant. I’hySIol. 37: 1357—141. 24. lx'xron, S. AND A. Six Piano). 1967. .-\seorbate supported NADP photo— Iv 0| 26. 27. 28. . KnAAYi-zsiior. R., 145 reduction by heated Euglena chloroplasts. Arch. Biochem. Biophys. 122: 144—152. S. IZAWA, B. CHANCE. 1972. I'se of uncoupling acridine dyes as stoichiometric energy probes in chloroplasts. Plant Physiol. 50: 713718. ()I'iriiAKt‘L, R. AND S. IZAWA. 1973. Electron transport. and photophosphoryla- tion in chloroplasts as a function of the electron acceptor II. Acceptor- specific inhibition by KCN. Biochim. Biophys. Acta 305: 105—118. SAIIA. S‘., R. Ocr'rnAan, S. IZAWA, Axn N. E. Goon. 1971. Electron transport. and photophosphorylation in chloroplasts as a. function of the electron acceptor. J. Biol. Chem. 246: 3204—3209. Tiitznsr, A. AND S. REIMER. 1973. Properties of photoreduction by photo- system 11 in isolated chloroplasts. An energy-conserving step in the photo- ANI) reduction of benzoquinones by photosystem II in the presence of dibromo- tliyrnomiinone. Biochim. Biophys. Acta. 305: 129—139. 29. VAKLINOVA, S.. N. TUMOVA, E. NIKOLOVA, Axn M. Sséisxsxx. 1965. The effect of certain factors in the photooxidation of hydroxylamine in isolated chloroplasts. C. R. Acad. Bulg. Sci. 18: 659—662. 30. Wassrzm, J. S. C. 1958. Studies on photosynthetic phosphorylation II. Photosynthetic phosphorylation under aerobic conditions. Biochim. Biophys. Acta 29: 113—123. 31. YAMASI—IITA. T. An) W. L. BUTLER. 1968. Inhibition of the Hill reaction by tris and restoration by electron donation to photosystem II. Plant Physiol. 44: 435—438. 32. YAMASIIITA. '1‘. AND W. L. BUTLER. 1968. Photoreduction and photo- phosphorylation with tris-washed chloroplasts. Plant Physiol. 43: 1978—1986. APPENDIX V STUDIES ON THE ENERGY COUPLING SITES OF PHOTOPHOSPHORYLATION V. PHOSPHORYLATION EFFICIENCIES (P/ez) ASSOCIATED WITH AEROBIC PHOTOOXIDATION OF ARTIFICIAL ELECTRON DONORS l46 Studies on the Energy-coupling Sites of PhotoPhosphorylation V. PHOSPHORYLATION EFFICIENCIES (P/Cz) ASSOCIATED WITH AEROBIC PHOTOOXIDATION OF ARTIFICIAL ELECTRON DONORS‘ Received for publication August 16, 1973 and in revised form October 12, 1973 DONALD R. ORT AND SEIKICHI IZAWA Department of Botany and Plant Pathology, Michigan State University, East Lansing, Michigan 48824 ABSTRACT The rate of Hill reaction can be measured accurately as 03 uptake (the Mehler reaction) if a rapidly autoxidizable elec- tron acceptor (e.g., methylviologen) is used. However, when an artificial electron donor-ascorbate couple (or ascorbate alone) replaces the natural donor, water, the rate of Os con- sumption is no longer a reliable measure of the electron flux, because superoxide radical reactions contribute to O, uptake. Such radical reactions, however, can be suppressed by adding enough superoxide dismutase to the reaction mixture. Indeed in all of the photosystem I- and photosystem ll-donor reactions tested (except with benzidine which was tested without ascorbate added), the O;- uptake was inhibited by 30 to 50% by the ad- dition of superoxide dismutase. The rate of phosphorylation was totally unafi‘ected by the enzyme. The reassessment of the phosphorylation efficiencies thus made by the use of super- oxide dismutase led us to the following conclusions. The phosphorylation eficiency associated with the transfer of elec- trons from a donor to methlylviologen (than to 0:) through both photosystems II and l is practically independent of the donor used—catechol, benzidine, p-aminopheuol, dicyanohy- droquinone, or water. The P/C: ratio is 1.0 I 0.1. Only ascor- bate gives a slightly lower value (P/ez = 0.9). (NH20H-treated, non-water-splitting chloroplasts were used for reactions with these artificial donors.) The phosphorylation efficiency associ- ated with DCMU-insensitive, photosystem l-mediated transfer of electrons from a donor to methylviologen (then to 02) is again largely independent of the donor used, such as diami- nodurene, diaminotoluene, and reduced 2,6-dichlorphenol- indophenol. The P/ez ratio is 0.6 1' 0.08. Artificial, low potential electron acceptors, represented by viologens and anthraquinone, have been used extensively for the study of chloroplast electron transport and photophos- phorylation involving PS I”. These compounds have advantage 1This work was supported by Grant GB37959x from the Na- tional Science Foundation. The preceding paper of this series, entitled “Studies on the energy coupling sites of photophosphoryla- tion. IV. The relation of proton fluxes to the electron transport and ATP formation associated with Photosystem II," by J. M. Gould and S. Izawa has been submitted to Biochim. Biophys. Acta for pub- lication. "' Abbreviations: PS I and PS 11: photosystems I and II; DAD: diaminodurene; DCIPI-Iz: reduced 2,6-dichlorophenolindophenol; DAT: diaminotoluene; MV: methylviologen; P/eaz the ratio of the number of ATP molecules formed to number of pairs of electrons transported; SOD: superoxide dismutase. over the reconstituted “natural” electron acceptor system ferredoxin-NADP in several respects. They are much less likely to contribute an undesirable rate-limiting step even in very fast PS I reactions and therefore are less likely to permit simultaneous cyclic electron flow around PS I. Also, they are much less susceptible to inactivation by various reagents and conditions and are much more economical to use. The re- duction of these acceptors can be followed most easily as the O. uptake which results from the autoxidation of the reduced acceptors (12). The widely used formulae (e.g., 29) for the transfer of electrons from a donor (Al-1,) to an acceptor (e.g., MV) and the subsequent aerobic reoxidation of the reduced form of the acceptor leading to the formation of H20, may be rewritten as follows, including a new term for the inter- mediate of PLO. formation, the superoxide radical 0.‘(21). chloroplasts “Em A + 2H+ + Z'MV+ (1) AH, + 2MV” spontaneous spontaneous (3) 202‘ + 2H+ -——————* 0: + H202 2e“ transport AH“ + 0’ MV light A + H10: (0: E 26’) (4) In the normal Hill reaction where AH. = H20 (i.e., A : V202). the over-all reaction becomes what is known as the Mehler reaction. “:0 + 1/é02 2c“ transport MV, light H202 (0 E 23’) (5) To date, all the computations of electron fluxes from 02 up- take data have been made on the basis of the above two over- all formulae (equations 4, 5). The basic validity of the mechanism for the Mehler reaction was established by mass spectroscopic studies of the 02 ex- change reactions involved (8). The validity of the formulations for artificial electron donor systems, however, has never been rigorously proven. In fact, the now rapidly increasing knowl- edge of the role of “O...“ in various oxidation reactions makes it more and more diflicult to believe that the events involved in the aerobic photooxidation of artificial reductants are al- ways as straightforward as represented by equations 1 to 4 (11). For instance, if any significant portion of the ‘0; pro- duced by aerobic reoxidation of 'MV‘ (equation 2) directly reacts with the artificial donor AH2, then the same rate of elec- tron transport through the photosynthetic chain would induce a faster rate of O, uptake than predicted by equation 4, and the relation 03 5 2e‘ would no longer be valid. This is because the '02‘, which normally dismutates and regenerates half of the consumed 0, (equation 3) is now simply reduced to H202: AH, + 20.,- _——. A + 2H00- (6) 370 147 However, such reactions and the resulting enhancement of O: uptake should be abolished when the dismutation of '02' is greatly accelerated by added SOD (20). Thus, with suflicient SOD, the aerobic photooxidation of donors should follow ex- actly equations 1 to 4, and the relation 0. e 2e" should hold. This method of detecting or suppressing superoxide radical re- actions has already been in wide use. For instance, Asada and Kiso (2, 3) have successfully adopted this method to detect the involvement of O; in the photooxidation of epinephrine and sulfite by isolated chloroplasts. These considerations prompted us to re-examine a number of chloroplast reactions involving the light-dependent transfer of electrons from artificial donors to MV and then to 0,. The immediate objective was to assess, for each reaction system, exactly what portion of the O, uptake observed should be at- tributed to the true photosynthetic electron transport described by equations 1 to 4, and what portion to the chemical oxida- tion of donors by 02'. As indicated above, this can be achieved by observing the effect of excess SOD added to the reaction mixture which should abolish only the latter portion of 0. consumption. While the importance of such fundamental data seemed obvious in view of the extensive use of these aerobic reaction systems by various workers, our primary concern was to establish the true efficiencies (P/ e. ratios) of the phosphoryla- tion reactions associated with the donor reactions, particularly those involving PS 11. The experiments have not only provided clear answers to the question of the true electron fluxes and true P/e, ratios but have also disclosed an important quantita- tive relationship, in terms of the phosphorylation efficiency, between the reactions involving only PS I and those involving both PS I and PS 11. MATERIALS AND METHODS Chloroplast Isolation and NILOH Treatment. Chloroplasts were isolated from commercial spinach (Spinacia oleracea L.) as described in an earlier paper (26). The hydroxylamine treat- ment was done in the dark at room temperature (21 C) as out- lined below (for original procedure, see Ref. 26). Chloroplasts, at a final Chl concentration of approximately 100 ng/ ml, were suspended in the following medium: 0.2 M sucrose, 5 mM HEPES-NaOH buffer (pH 7.5), 2 mM MgCl,, 1 mM EDTA, and 3 mM NH20H. A stock solution of NH20H (0.1 M) was made by dissolving the hydrochloride salt in 0.1 N HCl and stored at 0 C (fresh solutions were made up every few days). The hydroxylamine was added to the treatment medium imme- diately before the chloroplast treatment, and the pH was ad- justed to 7.5. Unless otherwise noted, the chloroplasts were treated in the above medium for 12 min at 21 C and then washed twice in the same medium (minus NH20H and EDTA) at 0 C to remove the amine. Reagents. Lyophilized bovine blood superoxide dismutase (specific activity, 3000 units/ mg) was purchased from Truett Laboratories (Dallas, Texas). The enzyme was dissolved in 10 mM HEPES-NaOH buffer at pH 7.8 at 2 mg protein/ml and dialyzed against 2 liters of the same buffer solution for 12 hr at 0 C, and then stored at —20 C. DAD, p-phenylenediamine dihydrochloride, and DAT were recrystallized from charcoal- treated aqueous alcohol solution by adding excess HCl at 0 C. Benzidine dihydrochloride, diphenylhydrazine hydrochloride, and p—aminophenol hydrochloride were recrystallized in an identical manner except the recrystallization was from char- coal-treated aqueous solutions. Dicyanohydroquinone(2,3-di- cyano-p-benzohydroquinone) and o-tolidine were recrystal- lized from charcoal-treated aqueous solution by simply lowering the temperature to —25 C. Fresh solutions of these compounds were made up daily in 0.01 N HCl. Measurements. The MV Hill reaction was assayed as the O, uptake resulting from aerobic reoxidation of reduced MV (equation 5). Electron transport from artificial donors to MV was assayed similarly, as 02 uptake. A membrane-covered Clark-type oxygen electrode was used for O. assays. When ar- tificial electron donors were used, the observed rate of 02 up- take in the light was corrected for the slow rate of dark au- toxidation of donors which accompanied some of the reactions and ranged from 5 to 20% of the rate observed in the light. The intensity of orange actinic light (600—700 nm) was ap- proximately 600 Kergs-sec‘1-cm’”. The reaction temperature was 19 C. ATP formation was measured as the residual radio- activity after the extraction of the ”P-labeled orthophosphate as phosphomolybdic acid in butanol-toluene. Radioactivity was determined from the Cerenkov radiation as described by Gould et al. (13). RESULTS Photooxidation of Ascorbate by Normal, 0,-producing Chloroplasts. Aerobic photoxidation of ascorbate by isolated chlorOplasts was a subject of rather intensive studies in 1950’s, but no clearly defined mechanism has emerged (for a review, see Ref. 18). The original observation of Mehler (22) that the ascorbate photooxidation required 0, and was stimulated by catalytic concentrations of quinone, may now be viewed in retrospect as already suggesting the involvement of the super- oxide radical anion in the reaction mechanism. Figure 1 shows the effect of ascorbate addition on the uptake of O, and phosphorylation in normal chloroplasts which are capable of actively transporting electrons from water to MV. As can be seen, the addition of ascorbate doubles the rate of the O. uptake mediated by MV. The phosphorylation rate is T u 3 is O -ASC/ off—0 4 o a @- o/\OASC .9. .C U E’ 1 _l_ .. T’ 300 50 I00 .1: - S.O.D.( lmll E no % < #2 U) j e 5 % 200 / \ - 2% ATP (~Asc) ATP (oAsc) 5 °2 C uptake m '00 (-Asc) _ E 2 O '3. 0 l l l l I O 20 4O 60 80 IOO 5.00. (pg/ml) FIG. 1. Effect of SOD on ascorbate-stimulated O. uptake in normal (untreated) chloroplasts. The 2-ml reaction mixture con- sisted of: 0.1 M sucrose, 50 mM Tricine-NaOH buffer (pH 8.0), 2 mM MgCh, 0.75 mM ADP, 5 mM Nazi-PTO. 0.5 mM methyl- viologen, and chloroplasts containing 40 ug of chlorophyll. When added (+Asc), D-ascorbate was 5 mM. All other pertinent reaction conditions are as described in “Materials and Methods.” Note that O (V202) units are used in this particular figure (not On). I48 scarcely affected (10% inhibition), but, because of the stimu- lated O, uptake, the apparent phosphorylation efficiency P/O (2 P/e. in the absence of ascorbate) drops sharply from 1.2, a value typical of the MV Hill reaction, to 0.5. However, the subsequent addition of SOD abolishes the ascorbate-stimulated portion of O, uptake without influencing the phosphorylation rate at all. Consequently the diminished P/O ratio is restored nearly to the original level of standard noncyclic photophos- phorylation. We may deduce from these observations that, in the normal, Oa-producing chloroplasts employed here, ascorbate photooxidation is supported predominantly by the superoxide radical which is generated by the aerobic reoxidation of re- duced MV, and, therefore, ascorbate does not significantly re- place water as the electron donor. Bohme and Trebst (6) and Epel and Neumann (10) have also noticed the essentially un- changed rate of phosphorylation during the ascorbate-enhanced O, uptake. The above radical reaction mechanism has already been predicted by the latter authors (10). Photosystem ll-mediated Oxidation of Ascorbate in NH,OH- treated Chloroplasts. Ascorbate does donate electrons to the photosynthetic electron transport chain if the water-splitting mechanism is inoperative or destroyed (5, 6, 9, 26, 32), sug- gesting that the inability of water to serve as reductant creates favorable conditions for the strong oxidant produced by PS 11 to oxidize exogenous reductants. In a preceding paper of this series (26) we have documented a new method of inactivating the water-oxidizing mechanism of chloroplasts (NH,OH-treat- ment) without impairing the coupling efficiency of the chloro- plast membrane. Our preliminary data on ascorbate photo- oxidation in these NH,OH-treated chloroplasts indicated a P/ 0. value of 0.5 to 0.6, confirming the observation of Bohme and Trebst for their heat-treated chloroplasts (6). These P/ O, values would have represented the P/ e. values if the ascorbate photooxidation simply followed the mechanism expressed by equations 1 to 4 (O. 5 2e'). Hr wever, it is already clear that these formulae do not apply even in these treated chloroplasts, since the superoxide radical, produced via the univalent reduc- tion of O. by reduced MV, must still react with ascorbate. As Figure 2 shows, the aerobic photooxidation of ascorbate in NH30H-treated chloroplasts indeed contains a large SOD- sensitive component indicative of the involvement of 0;. How- ever, a larger, SOD-insensitive component remains as a well defined plateau. Clearly, it is the latter portion which must be considered as the “true” electron transport expressed by equa- tions 1 to 4. As is clearly seen in Figure 2 (inset) in this plateau region the P/O. (now equivalent to P/e.) reaches 0.9, a level which is no longer greatly different from that of the normal Hill reaction (P/e. = 1.1 to 1.2). The pH dependence of O. uptake and phosphorylation in this ascorbate -> MV -> 0, system is shown in Figure 3. As one would expect, the presence (Fig. 3b) or absence (3a) of SOD only affects the height of the O. uptake versus pH curves with little effect on their shapes. Phosphorylation remains totally unaffected by SOD at all pH levels. The “true” relationship be- tween electron transport and phosphorylation is represented by Figure 3b where the radical-ascorbate interaction is prevented by SOD. The shapes of these activity-pH curves and the marked stimulation of electron transport by Pi (i.e., by concomitant phosphorylation) are remarkably similar to those observed for the standard Hill reaction with MV as acceptor (14). Undoubt- edly both reactions ascorbate -> MV and H20 -* MV are gov- erned by the same rate-limiting phosphorylation step. Photosystem II-mediated Oxidation of Catechol and Other Electron Donors in NH.OH-treated Chloroplasts. In this study we have found that catechol (o-hydroquinone; with 0.5 mM as- corbate as electron reservoir) serves as an excellent electron donor for PS 11. The rates of electron transport and phospho- rylation are comparable to those of the Hill reaction and are highly DCMU-sensitive (Fig. 4). The apparent P/O. ratio of 0.6 in the absence of SOD is close to that observed for ascor- bate photooxidation. The addition of SOD again exerts a marked inhibitory effect (40%) on O. uptake, without affecting the phosphorylation rate at all. The maximum P/O, ratio ob- tained in the presence of SOD (where P/O. = P/e.) now reaches 1.1, which is virtually identical with the ratio asso- ciated with the Hill reaction. Similar results were obtained with another new electron donor, 2,3-dicyanohydroquinone (with I T T: g A l0)“”(3;;:o——”‘O‘ 75 / 2 o‘“ o E 0.50- ------------ 0" E _' L g: (ktnmn O i E..- 50 _ 0 50 100+ <[ A 5.00. (pg/ml) \— :03 A|\‘ 2 ‘3 g a :1. b 25L— ATP ‘ N O m % Residual Hill Reaction (-Asc) E / (Ozuptokd 3 0 o l l l l l 0 20 60 IOO SOD. (pg/ml) FIG. 2. Effect of SOD on Ca uptake associated with photo- oxidation of ascorbate in hydroxylamine-washed chloroplasts. Chloroplasts were pretreated with NH.OI-I as described in “Materials and Methods.” Other conditions are as in Figure 1. '7_ 1 T l t 1 9500 E Q' 0.8 m :3 l20+~ a 0.4 2/\ T -w m /-S.O.D. “e” l. 0 O _ a l J. l _ 1‘: 6.0 7.0 8.0 9.0 . PH ptake 0- 80~ 9 k .p. Can i- .. E Ozupael I) ;pi) b /a¥ O _ 0 \§ 02 uptake .. § / /'\ I 9P.) \ /\ ' / b 40L 0 L ATP— —>- r—a - ON O//; / ATP A / \ \5 in .— o/o/./ _ a?“ 02 uptake J a / /‘ W O E 0 0/ 1 1 1 9 1 t 1 6.0 7.0 8.0 9.0 6.0 7.0 8.0 9.0 FIG. 3. Effect of SOD on O: uptake and phosphorylation asso- ciated with ascorbate photooxidation in NHzOH-treated chloro- plasts at various pH levels. The reaction conditions are identical to those in Figure 2 except for the buffers used: MES-NaOH (pH 6.0 to 6.5), HEPES-NaOH (pH 7.0 to 7.5), and Tricine-NaOH (pH 8.0 to 9.0). The dark oxidation of ascorbate at pH 9 became quite significant (20—30% of the rate in the light); as a consequence, exact determinations of rates became difficult. The concentration of SOD used (B) was 80 pg/ml; no SOD was used for obtaining data in part A. 14.9 /A O -———-o tapers _ - 4’ ...... l o P ..... ./. ‘3 ______ 7 / 95.0.0. /o N N o '— a .< ’”‘£~sou° 9 0.5 F3 ------------- 0- 0,5 ............ O .4 1 O L, 1 O 0.5 LO 0 0.5 1.0 CATECHOLOI'M) S.O.D. (liq/ml) p- 02 V0 t-soo) L an: mom n\ 02 up,“ _ PO ATP / A-——-—_A A\ / IOCH' g/A;\ _ - N O O pmoles 02 or nmoles ATP‘ h". mg chlorophyll" / 02 uptake .-. \9—3 (now up OtuptdtsODClUl _ (scoop) / o I- C l l L I 1 I J O 0.5 LO 0 5O IOO CATECHOL (mM) 5.0.11 tug/ml) FIG. 4. Effect of SOD on O: uptake associated with the PS II-mediated oxidation of catechol in NH20H-treated chloroplasts. All the conditions were as in Figure 2 except for the reaction pH (7.8). When added, DCMU was 1.5 [J.M. Note that the P/Oz ratio (=P/e2) in the presence of SOD approaches 1.1, a value close to that of the Hill reaction. SOD, 80 ug/ ml (left-hand figure). 0.5 mM ascorbate), although the reaction rates with this donor are not nearly as rapid as with catechol (Fig. 5). Yamashita and Butler (32) showed that p-aminophenol acts as a good donor for PS 11 in their tris-washed chloroplasts. They reported a P/e, value of 0.97 measured under anaerobic conditions with NADP as acceptor. We have confirmed this value using the aerobic photoxidation system (P/ 02 = 1.0 with SOD). Benzidine is another compound reported by Yamashita and Butler to give a P/e, ratio approaching 1. As already in- dicated in a previous paper (26), the photooxidation of this compound can be followed for some time (1.5 min), without inducing a cycle, in the absence of added ascorbate. The P/O. ratio obtained was 1.05. As one would expect, the photooxida- tion of this compound (without ascorbate) is quite insensitive to SOD. Photosystem I-mediated Oxidation of Artificial Electron Donors in DCMU-poisoned Chloroplasts. Aerobic reaction systems with autoxidizable electron acceptors are widely used for the studies of PS I-electron transport and phosphorylation, especially those involving very fast rates (e.g., 15, 16, 25). In the following experiments all the reactions were run in the presence of 2.5 mM ascorbate as the electron reservoir. Figure 6 shows the effect of SOD on the photooxidation of DAD, perhaps the most widely used PS I donor today, and DAT, a new donor. Oxygen uptake and phosphorylation in these PS I reactions are faster, by almost an order of magni- tude, than the PS Il-mediated reactions described above. How- ever, a marked suppression of O, uptake by SOD clearly in- dicates that considerable portions (30—40%) of the high rates observed are due to the oxidation of the electron donors and/ or ascorbate by 0;. Once again SOD has no effect on phos- phorylation, and, consequently, apparent P/ 0, ratios of 0.3 to 0.4 in the absence of SOD are elevated to the level of 0.6 to 0.65 as the suppression of 02 uptake by added SOD reaches a maximum. The following electron donors were also tested: DCIPH., o-tolidine, diphenylhydrazine, and p-phenylenediamine. Of these compounds only DCIPH. is well known as a PS I donor. p-Phenylenediamine is better known as a PS II donor (31), but in our hands the photooxidation of this compound contains a significant part (30%) which is DCMU-insensitive and there- fore considered to involve only PS I. All of these compounds are oxidized by PS I at much slower rates than are DAD and LO -/ a a § LO - 6 ° ‘ + 5.0.0. o/ e / °- o.5":’°—-° ‘* 0.5} " S 0.0. 0 1 l 1 m 0 l L I _I_ O 0.5 LO 0 50 IOO Dicyanohydroquinone (11M) S. 0.0. ( uq/ ml) IOO P r ‘ 02 uptdte 980W nmoles 02 or nmoles ATP- h" -mq chlorophyll" " 0 ATP (23.0.11) ‘ ‘ 02 uptake 50 (0500.) —< l 02 we AT p / uocuut wow) - 02 m... - 0,...———-—-—-—/ ...... o/ /(.ocau, .301» o - ‘ ' #- J l l l O 0.5 LO 0 5O IOO Dicyanohydroquinona (mM) 3.0. 0. (pg/ml) FIG. 5. Effect of SOD on O. uptake associated with the PS II-mediated oxidation of dicyanohydroquinone in NH.OH-treated chloroplast. Conditions are as in Figure 4. In the left-hand figure, SOD was 80 ng/ ml. 1 0.75 /DAT 2000 K g E 0502 DAD < 025’— O 1 1 1 I500_ o IOO Zoo 300 \ 3.00. (pg/ml) t O\ n o 02 uptake (DAD) l— .a IOOO 02 uptake (DAT) nmoles 02 or nmoles ATP-h"- mg chlorophyll" \ A A e e \ e 9 ATP (DAD) 500 ~ _ A 5 \ A A ATP (DAT) o I 1 1 O ICC 200 300 3.00. (pg/ml) FIG. 6. Effect of SOD on 02 uptake associated with PS I-de- pendent oxidation of DAD and DAT in the presence of DCMU. The 2-ml reaction mixture contained the following: 0.1 M sucrose, 50 mM Tricine-NaOH buffer (pH 8.0), 2 mM MgC12, 0.75 mM ADP, 5 mM Nag-HmPoh 50 11M methylviologen. 1.5 p.114 DCMU, 2.5 mM D-ascorbate, 0.5 mM diaminotoluene or diaminodurene, and chloro- plasts containing 10 ug of chlorophyll. 150 DAT, their maximum activities as donors supporting only 100 to 200 nmoles O,/hr-mg Chl. Nevertheless, all of these reac- tions support phosphorylation with similar efficiencies (P/Og): 0.3 to 0.4 in the absence of SOD and 0.55 to 0.65 in its pres- ence. Only DCIPH, gives significantly higher P/O2 values—0.5 to 0.55 without SOD and 0.65 to 0.7 with SOD. These data are summarized in Table I together with data for all of the other PS I and PS 11 donors tested in our study. The stoichiometric formation of H20, from the consumed O, was also confirmed for some typical donor reactions, in order to demonstrate that no significant peroxidatic or oxidase reac- tions were occurring which might further complicate the over- all reaction mechanism and invalidate equation 4 (Table 11). DISCUSSION In this work we have placed considerable weight on the characterization of the aerobic photooxidation of ascorbate. Our interest in this ascorbate photooxidation was aroused when Biihme and Trebst (6), using both normal chloroplasts and heated chloroplasts, showed that the P/e, ratio (assuming P/e. = P/O,) associated with ascorbate photooxidation was 0.5, which was close to half of the value for normal noncyclic pho- tophosphorylation. They interpreted this to suggest that one of two sites of noncyclic phosphorylation is associated with the water-splitting step of PS 11, and the entry of electrons from ascorbate occurs after this phosphorylation site. We have con- firmed the P/ 0. ratio of 0.5 to 0.6 using NH,OH-treated chlor- oplasts (26, see also the “Results” of this paper). However, it is now quite clear that this apparently low phosphorylation effi- Table I. Eflect of Superoxide Dismutase on 0; Uptake and Phosphorylation Associated with Various Photosystem II and Photosystem l Donor Reactions Conditions for reactions involving PS 11 are outlined in Figures 1 through 5. The concentration of ascorbate as a direct donor was 5 mM; the other PS 11 donors, 0.5 mM plus ascorbate 0.5 mM ex- cept benzidine which was added alone. Conditions for PS I-donor reactions are similar to those in Figure 6, except that the Chl concentration for reactions with DCIPH}, o-tolidine, diphenyl- hydrazine (DPHZ), and p-phenylenediamine (PD) was 20 pg/ml. The concentration of donors, 0.5 mM plus 2.5 mM ascorbate. For the reaction involving H10 as donor (Hill reaction), untreated chloroplasts were used. NH,OH-treated chloroplasts were used for donor reactions involving PS 11. PS l-donor reactions were run in the presence of DCMU added. DCHQ, 2,3-dicyano-p- hydroquinone. Pb t 25875? 0' . E313: Igviigifid 90"" R0381? ““02,“ $3253.? ‘3'; "$915" System) 0‘ SOD Uptake .. 1:;- maestr- % 11 +1 H30 225 1.22 225 0 1.22 D-Ascorbate 100 0.55 65 35 0.85 Catechol 170 0.70 100 41 1 .15 DCHQ 85 0.55 45 47 1 .00 Benzidine 9O 1 .05 90 0 1.05 p-Aminophenol 145 0.64 85 41 1 .00 l DCIPH, 160 0.50 115 27 0.68 DAD 1750 0.39 1250 30 0.57 DAT 1300 0.35 775 41 0.60 o-Tolidine 100 0.27 50 50 0.53 DPl-IZ 70 0.32 40 45 0.59 PD 220 0.38 135 39 0.62 Table II. Accumulation of ”:0: during the Photooxidation of Various Photosystem II- and Photosystem I Donors For the reaction conditions, see Table 1. Immediately upon turning off the light, 10 ul of catalase (2 x 10‘ units/ml) were injected into the reaction vial, and 0, evolution was monitored to determine H10, accumulation. Note that in all donor reactions the H20, accumulation to O, uptake ratio approaches the theo- retical value of l (or 2 when H20 serves as electron donor). Presence (+) H202 Donor 81¢:fsgnéf) 02 Taken Up Accumulated 11202/02 nmoles H20 — 13.5 27 2.00 o-Ascorbate —— 8 .6 8 .0 0.93 + 6.6 6.0 0.91 p-AminOphenol -— 10.0 10.2 1.02 + 8 .0 7.6 0.95 Catechol — 11.0 9.4 0.86 + 7 .4 6.4 0.86 DAD — 19.2 16.8 0.88 + 13 .4 12.0 0.90 DAT — 15.9 16.4 1.03 + 7 .6 7 .6 l .00 ciency was due to an overestimation of electron flux. Indeed, our data showed that 40% of the total 0, consumption is due to the oxidation of ascorbate by the superoxide radical which is produced by the donor reaction itself: ascorbate -* PS 11 -> PS I -r O,/ 0;. Thus, when this radical reaction is prevented by SOD added, the P/O. ratio is elevated to 0.9. This value, now equivalent to the P/e2 ratio, is no longer low enough to make one suspect bypassing of a phosphorylation site. It is sig- nificantly lower, however, than the average P/e. ratio asso- ciated with the standard Hill reaction—1.0 to 1.2. We have no clear explanation for this at the present time. Ben-Hayyim and Avron (5) did obtain a P/e2 ratio of l for the ascorbate -’ NADP reaction in tris-washed chloroplasts, but their results were not conclusive because of their conditions, which strongly favored a cyclic electron flow. In their original reports on tris- washed chloroplasts Yamashita and Butler (31, 32) listed rather poor P/ e, ratios for the ascorbate -9 NADP system (0.1—0.4). The involvement of superoxide radical ions in ascorbate photooxidation, which leads to an overestimation of electron flow, has already been predicted by Elstner et al. (11). How- ever, their suggestion—that the monodehydroascorbate radical generated by the oxidation of ascorbate by 0; may be the ef- fective electron donor species rather than ascorbate per se— cannot readily explain our results, which showed that the SOD inhibition of O. uptake never exceeded 50% (in any donor re- action). In fact, this 50% inhibition is the theoretical maximal inhibition one could expect, if the whole reaction including O,’-donor interactions is to be essentially described by equa- tions 1 to 6. Before the completion of this manuscript a paper by Allen and Hall (1) was published which deals with the effect of SOD addition on ascorbate photooxidation in normal, O,-producing chloroplasts. Their observations include a marked enhancement of 02 uptake by ascorbate addition (up to 3-fold increase), its complete reversal by SOD addition, and an unchanged rate of phosphorylation. They concluded that in normal chloroplasts ascorbate does not replace water as the electron donor, and the enhancement of O, uptake observed is almost entirely due to the oxidation of ascorbate by 0;. Our results agree with theirs rather well, and we agree with their conclusions, except that the slight inhibition of phosphorylation by ascorbate addition 151 we observed seems to indicate the existence of some interac- tion between the electron transport chain and ascorbate. It is clear that the question of the superoxide radical involve- ment is shared by practically all of the reactions with artificial electron donors assayed as aerobic photooxidation since, in general, an electron donor is administered as a donor/ ascor- bate couple. Besides, many of the commonly used artificial electron donors themselves must be expected to be susceptible to 0;, as exemplified by p-hydroquinone (28) and catechols (23), although we have observed no sign of benzidine, given alone, consuming 0;. Nevertheless, our attempt to isolate true electron fluxes by abolishing donor- 0,“ interactions by SOD addition was apparently successful with all of the donor reactions tested. This is indicated by the fact that the SOD- inhibition curves for O, uptake always reached a well defined plateau and that within the same group of donor reactions (PS I- or PS II-mediated) the P/O2 ratios measured in these plateau regions were remarkably similar (Table I). As men- tioned above, it is also important that the SOD inhibition of 02 uptake never exceeded 50%. It was surprising to find the P/e2 ratios thus computed for all of the PS ll-donor reactions tested fall near 1. Although the ascorbate system showed slightly lower values (0.85 to 0.9), they are still close to the range of normal P/e2 values associated with the Hill reaction. Viewing these data in the light of the recent findings concerning the sites of energy coupling in non- cyclic photophosphorylation (14, 17, 27, 30), there is no doubt that the transport of electrons from these donors to MV uti- lizes both coupling sites I and 11. Here, coupling site I refers to the well known rate-limiting phosphorylation site between plastoquinone and cytochrome / (4, 7). Site 11 is considered to be located very close to PS 11 (17, 27, 30), possibly on its water- oxidizing side, as has been suggested by Bohme and Trebst (6), although the basis on which they made this suggestion is no longer valid. Since the true P/e2 ratio associated with the transport of electrons through both photosystems is near 1 re- gardless of the donor used, it is evident that site 11, if indeed located on the water-oxidizing side of PS 11, does not require water as the obligatory electron donor. However, if one takes the view of chemiosmotic coupling (24), an interesting possibil- ity arises. These artificial donors, with either amino or hy- droxyl groups as the oxidizable groups, might mimic water as the hydrogen donor. That is, if the machinery of water oxida- tion operates in such a way that the protons released by water splitting are discharged to the inside of the thylakoid (which would create a trans-membrane hydrogen ion gradient), then the same directional discharge of protons might occur when the artificial “hydrogen” donors are oxidized by PS 11. Alter- natively, these donors may simply be oxidized inside of the thylakoid. Such a chemiosmotic view is now encouraged by the recent discovery of the involvement of a proton pump in the partial electron transport reaction H20 -* PS 11 -> dibromo- thymoquinone (14). Table I also displays unexpectedly high and uniform P/e, ratios (around 0.6) found for PS I-donor reactions, although the value for the DCIPH, system (0.65 to 0.7) seems apprecia- bly higher than the rest. Significantly, all of these P/e2 values are lower than those supported by the reactions involving both photosystems by about 0.4 to 0.5, a difference which could be explained if one assumes that the PS I reactions do not involve site II. In fact, the characteristics of the phosphorylation reac- tion associated with the DCIPH, -> PS I -> MV system have been well accounted for by the noninvolvement of site II (14). Arknowlcrlomcnia—The authors wish to express their thanks to R. Gee for his kind gift of superoxide dismutase and Valuable advice. Thanks are also due to N. E. Good for discussion. H N uh G N on 19. 2 G 2 . H 2 . N 2 09 24. 25. 26. 27. . (.‘Hi:.\'1.\i-:, G. . lirsrxmt. E. F., A. . 12 um. S.. ..l.u;i:xntnir, .-\. T. LITERATURE CITED . ALLEN, J. F. AND D. C. HALL. 1973. Superoxide reduction as a mechanism of aSCorbate-simulated oxygen uptake by isolated chloroplasts. Biochem. Biophys. Res. Commun. 52: 856862. . ABADA, K. AND K. Ktso. 1973. The photooxidation of epinephrine by spinach chloroplasts and its inhibition by superoxide dismutase: evidence for the formation of superoxide radicals in chloroplasts. Agr. Biol. Chem. 37: 453- 454. . Ason, K. AND K. Krso. 1973. Initiation of aerobic oxidation of sulfite by illuminutcd spinach chloroplasts. Eur. J. Biochem. 33: 253—257. . Avnox, M. AND I3. CHANCE. 1966. Relation of phosphorylation to electron transport Ill isolated chloroplasts. Brookhaven Symp. Biol. 19: 149—160. . Iiux-HAYYIM, G. Axn .\I. Avuox. 1970. Involvement of Photosystem TWO in non-oxygen cyolvmg non-cyclic and in cyclic electron flow processes in chloroplasts. Eur. J. Biochem. 15: 155—160. . 130113115, H. AM) A. TRI-JBST. 1969. On the properties of ascorbate photooxida- tion in isolated chloroplasts. Evidence for two ATP sites in noncyclic pliotoiihosphorylation. Biochim. Biophys. Acta 180: 137-148. . 110113118, 11. .\.\’D W. A. (IRAMER. 1972. Localization of a site of energy coupling between plastoquinone and cytochrome I in the electron transport chain of spinach chloroplasts. Biochemistry 11: 1155—1160. . BROWN, A. H. AND N. E. Goon. 1955. Photochemical reduction of oxygen in chloroplast preparations and in green plant cells. I. The study of oxygen exchanges in NIH) and in riro. Arch. Biochem. Biophys. 57: 3407—354. AM) I. F. .\1Aiiri.\'. 1970. Studies of function of manganese wrtlun photosystem II. Roles Ill 02 evolution and system 11. Biochim. Biophys. Acta 197: 219 239. . EFLL. B. L. .txn J. Nist‘MAxx. 1972. In: VI. Intern. Congr. Photobiol. Rio- chem. Germany. Abstract .\'0. 237 (cited in Refs. 1 and Ill. Illit‘l'la’L, .\\‘l) S. \‘AKIJNLWA. 1970. fiber die Oxidation yon Hydroxylamin «lurch isolierte Choroplasti-n und die moglichc Funktion cincr Peroxidase nus Spinntbliittern bei dcr ()xulation yon .~\scorbin.~‘iiure uud Glykolsiiure. Z. I’llunzcnpIiysiol. 62: 184400. . Goon, N. E. AND R. HILL. 1955. Photochemical rcduction of oxygen in chloroplast preparations. II. Meclnuusms of the reaction with Arch. Blocllt‘lll. Biophys. 57: 355 366. oxygen. . Gottn, J. .\I., R. CATHER, Axn G. D. Wixoiz'r. 1972. Advantages of the use of (‘i-icnkov counting for dctcrmmation of "31’ in lllltllI‘itlllti>]lllill'_\'l1llltlll research. Anul. liioclictu. 50: 540—548. . Got'Ln, J. .\1. AND S. IZAWA. 1973. Studies on the energy coupling sites of photophosphorylntion I. S‘o-pnrntmn of site I and site II by partial reactions of the chloroplast electron transport chain. Biochim. Biophys. Acta 314: 211—223. . Il.~\l'SKA. G. A., R. E. .\I(:(‘.uirr, AND E. HACKER. 1972. The site of phos- phorylation associated with plu'itosystcm II. Biochim. Biophys. Acta 197: 200218. 3. I). Wixoizr. AND N. E. Goon. 1966. Inhibition and uncoupling of plunophosphorylation in chloroplasts. Brookhaven Symp. Biol. 19: 169 187. '1‘. N. ('o}. NULLY. . IZAWA. S.. J. .\1. Colin, D. It. ORT, P. FELKl-Ilt, AND N. E. Goon. 1973. Elec- tron trunsport and photoplios;ihorylation in chloroplasts as a function of the electron acceptor. 111. A dibromothymoquiuonc-insensttivo phosl‘ihoryln- tion I‘Pftt‘lltlll assocmtcd With Photosystem II. Biochim. Biophys. Acta 305: 11!} 123. 1962. Biochemistry of energy transformations during plIOIUSylllllOSIS‘. In: Surycy of Biological Progress, Vol. IV. Academic Press, New York. pp. 181—344. tion by hcnted Euolcua chloroplasts. Arch. Biochem. Biophys. 122: 144-152. K.\ToH, S. AND A. SAX I’irzruo. 1967. Ascot-lrate-supported NADP photoreduc- . .\1c(‘oRn, J. .\I. AND I. anovu'u. 1969. Superoxide dismutase. An enzymic function for erythrocuprein (lu-mocuprcin) . J. liiol. Chem. 244: 6049—6055. Misra, H. P. AND I. Fiunovu'u. 1972. The univalent reduction of oxygen by reduced flavins and quinones. J. Biol. Chem. 247: 188—192. .\li:uLi:it, A. H. 1951. Studies on reactions of illuminated chloroplasts. II. Stuuulntion and inhibition of the reaction with molecular oxygen. Arch. Biochem. Biophys. 34: 339—351. . MILLER, It. W. 1970. Reactions of superoxide anion, catechols and cytochrome c. (.‘an. .1. Biochcm. 48: 93577939. MICHEL, P. 1968. Chemiosmotic Coupling and Energy Transduction. Glynn Research Ltd., Bodinin, London. .\'i:u.\1.tx.\‘, J., C. J. Anxrznx, .\.\'1) R. A. DILLEY. 1971. Two sites of adenosine triphosphate formation in photosynthetic electron transport mediated by phmosystein I. I'lvuh-nce from digitonin subchloroplast particles. Biochem- istry 10: 866 873. Our, D. AND S. IzAWA. 1973. Studies on the energy coupling sites of photo- phosphorylation. II. 'I‘rcatment of chloroplasts with NHcOH plus EDTA to inhibit water-oxidation while maintaining energy coupling efficiencies. Plant. Physiol. 52: 595-600. Ot'lTltAlx'l’L, R. AND S. IZAWA. 1973. Electron transport and photophosphoryla- tion as a function of the electron acceptor. II. Acceptor-specific inhibition by KCN. Biochim. Biophys. Acta. 305: 105—118. 152 AND E. Hn'ox. “1973. Experimental determination of the redox 28. Rao, P. S. Biochem. Biophys. Res. Commun. potential of the superoxide radical ()2‘. 51: 468—473. 29. Tasasr. A., H. Ecx, AND S. WAGNER. 1963. Efi‘ects of quinones and oxygen in the electron transport. system of chloroplasts. Photosynthetic mechanisms of green plants, Nat. Acad. Sen-Nat. Res. Council Publication 1145. pp. l74—194. 30. Tasasr, A. AND S. Reutm. 1973. Properties of photoreduction by photosys- tem II in isolated chloroplasts. An energy-conserving step in the photoreduction of benzoquinones by Photosystem II in the presence of dibrou'iothymoquinone. Biochim. Biophys. Acta 305: 129—139. 31. YAMASHITA, T. AND “C L. BUTLER. 1968. Photoreduction and photophos- phorylation with tris-washed chloroplasts. Plant Physiol. 43: 1978-1986. 32. YAMASHITA, T. Axu W. L. BUTLER. 1969. Inhibition of the Hill reaction by tris and restoration by electron donation to Photosystem II. Plant Physiol. 44: 435 438. APPENDIX VI PHOTOOXIDATION 0F FERROCYANIDE AND IODIDE IONS AND ASSOCIATED PHOSPHORYLATION IN NHZOH-TREATED CHLOROPLASTS 11:11»; arthroi. -m - ”Hum‘f .A :1. ; A . _ llll.ll ! .I l PHOTOOXIDATION 0F FERROCYANIDE AND IODIDE IONS AND ASSOCIATED PHOSPHORYLATION IN NHZOH-TREATED CHLOROPLASTS S. Izawa and Donald R. Ort Department of Botany and Plant Pathology Michigan State University East ansing. Michigan 48824 Summar Hydroxylamine-treated, non-water oxidizing chloroplasts are shown to be capable of oxidizing ferrocyanide and iodide ions via Photosystem II at appreciable rates (:;200 uequiv.h'1.mg chlorophyll'l). Using methylviologen as electron acceptor, ferrocyanide oxidation can be measured as 02 uptake, as ferricyanide formation. or as H+ consump- tion (2Fe2+ + 2H+ + 02 + 2Fe3+ + H202). Iodide oxidation can be measured as methylviologen-mediated 02 uptake, or spectrophotometrically, using ferricyanide as electron acceptor. The oxidation product I2 is re-reduced, as it is formed, by unknown reducing substances in the reaction system. The rate-saturating concentrations of these donors are veny high: 30 mM with ferrocyanide and 15 mM with iodide. Relatively lipo- philic Photosystem II donors such as catechol, benzidine and p-aminophenol saturate the photooxidation rate at much lower concentrations (<0.5mM). It thus seems that the oxidation of hydrophilic reductants such as ferrocyanide and iodide is limited by permeability barriers. Very 153 154 likely the site of Photosystem II oxidation is embedded in the thylakoid membrane or is situated on the inner surface of the membrane. The efficiency of phosphorylation (P/ez) is 0.5 to 0.6 with ferrocyanide and about 0.5 with iodide. In contrast the P/e2 ratio is l.0 to l.2 when water, catechol, p-aminophenol or benzidine serves as electron donor. These differences imply that only one of two phosphory- lation sites operate when ferrocyanide and iodide are oxidized. Ferro- cyanide and iodide are also chemically distinct from other Photostem II donors in that their oxidation does not involve proton release. It is suggested that the mechanism of energy conservation associated with Photosystem II may be only operative when the removal of electrons from the donor results in release of protons (i.e. with water, hydroquinones, phenylamines, etc.). Introduction Recent investigations on partial pathways of electron trans- port have established that there are two sites of energy conservation associated with noncyclic electron transport in chloroplast51'6. One of these sites, which we have called Site I, corresponds to the well- known site of phosphorylation between the plastoquinone pool and cyto- 7’8. The newly recognized site, which we have designed Site II, 2,3,9 chrome f has been placed in the close proximity of Photosystem II This paper deals with the nature of energy transduction at Site II. During the past year we have developed a new method of chloro- plast treatment (NHZOH-washing)10 which is highly effective in inactivat- ing the mechanism of water oxidation yet does not impair the coupling 155 mechanism of the chloroplast membrane. The treated chloroplasts are quite active in oxidizing, via Photosystem II, various exogenous reduct- ants such as catechol, p-aminophenol, benzidine and dicyanohydroquinonell The transport of electrons from these donors to methylviologen supports phosphorylation with a P/e2 ratio of l.0 to l.l. These P/e2 values are virtually the same as the value for noncyclic photophosphorylation involving water oxidation. The P/e2 data for p-aminophenol and benzidine also confirm the data which Yamashita and Butler12 obtained with NADP as acceptor using their tris-washed chloroplasts. Clearly neither of the sites of phosphorylation is destroyed or even significantly impaired when the mechanism of water oxidation is destroyed. Furthermore, Site II seems to function normally when artificial reductants replace water as the electron source. Another important clue to the nature of Site II has recently i been provided by the experiments of Gould and Izawa13’14 , who showed that an energy-linked proton translocation is associated with dibromo- thymoquinone reduction. The reaction is believed to involve only the electron span: H20 +-Photosystem IIa+ plastoquinone poolg‘ They postulated14, in accordance with the original suggestion of Mitchell15 (see also Rumberg gt_§l,16), that the proton translocation consists of two steps: internal discharge of protons through water oxidation fol- lowed by uptake of external protons through reduction of hydrogen car- riers (e.g. plastoquinone). According to the chemiosmotic coupling Abbreviations: DCMU, 3-(3,4-dichlorophenyl)-l,l-dimthylurea; HEPPS. N-Z-hydroxylpiperazine-N'-2-propanesulfonic acid: HEPES, N-Z- hydroxylpiperazine-N'-2-ethanesulfonic acid. 156 15, it is this Photosystem II-driven proton translocating loop theory itself that constitutes Site II. Such a loop, if exists, may well oper- ate when artificial hydrogen donors such as hydroquinones and phenyl- amines replace water, since their oxidation also involves proton release. For technical reasons, however, we have not been able to demonstrate the predicted proton translocation associated with Photosystem II oxidation of artificial hydrogen donors. Nevertheless, a further important test of the feasibility of this chemiosmotic model of Site II is available. That is to test pure "electron" donors the oxidation of which does not involve proton changes. If indeed the inward discharge of protons from donors (including water) represents the primary step of energy conservation at Photosystem II, then no energy conservation should occur at Photosystem II when it oxidizes non-proton producing electron donors. Consequently, the trans- port of electrons from such donors to methylviologen, now effectively by—passing Site II, should support phosphorylation only with the effi- ciency ascribable to Site I alone. The P/e2 expected is approximately 0.5 to 0.6. These are the values found invariably associated with Photosystem I-donor reactions which may be considered to involve only Site I (e.g. diaminodurene +~methylviologen)]]. We have discovered that ferrocyanide and iodide ions can be oxidized by Photosystem II in NHZOH—washed, non-water oxidizing chloro- plasts. The oxidation of these ions at physiological pH's do not involve proton changes. The transport of electrons from ferrocyanide and iodide to methylviologen did support phosphorylation with 157 efficiencies near the predicted value. This paper describes these rather unusual chloroplast reactions in some detail. Materials and Methods Isolation and NH20H-treatment of chlorgplasts--Chloroplasts (unfragmented naked lamellae) were isolated from commercial spinach (Spinacia oleracea L.) as described elsewhere]], and suspended in a medium containing 0.2 M sucrose, 5 mM HEPES-NaOH buffer (pH 7.5) and 2 mM MgCl2. The chlorophyll concentration of this stock suspension was approximately 2 mg/ml. 10 was performed as follows: l Hydroxylamine-treatment of chloroplasts vol. of the chloroplast stock suspension was added to 10 vol. of a freshly prepared medium containing sucrose, buffer and MgClz as above and in addition 5 mM NHZOH plus l mM EDTA. The mixture was allowed to stand at room temperature (2l°C) for 20 minutes, then diluted with cold, NHZOH-free suspending medium, and centrifuged at 4000 x g for 5 minutes at 0°C. The sedimented chloroplasts were washed twice by centrifugation with a large volume of the suspending medium to remove NHZOH and EDTA, and finally suspended in the same medium. Reagents--Potassium ferrocyanide was recrystallized from a warm saturated aqueous solution by slowly cooling it to -l0°C. Colorless crystals of catechol, p-aminophenol hydrochloride and benzidine dihydrochloride were obtained from charcoal-treated aqueous solutions by cooling (catechol) or by adding excess HCl at 0°C (p-aminophenol and benzidine). Liophilized bovine blood superoxide dismutase (specific activity, 3000 units/mg) was purchased from Truett Laboratories (Dallas, Texas). The enzyme was 158 dissolved in l0 mM HEPES (pH 7.5), dialyzed overnight against the same medium, then stored at -20°C. Assaygf-Electron transport from water or artificial electron donors to methylviologen was assayed routinely as the 02 uptake resulting from the reoxidation of reduced methylviologen. A membrane-covered (Clark-type) oxygen electrode was used for these 02 assays. Optical monitoring of reactions (e.g. ferrocyanide oxidation) was performed using a mono- chromator from a Beckman DU spectrophotometer with Gilford electronics. The monochromator was modified to permit illumination of the reaction cuvette (light-path 1 cm) in a thermostated holder. Absorption spectrum determinations were conducted using a Cary 15 recording spectrophotometer. Changes in the pH of the reaction mixture were measured using a Corning semi-micro combination glass electrode connected to a Heath-Schlumberger EU200-300 electrometer equipped with an EU200-02 offset module. The actinic light used was a rate-saturating orange light (600-700 nm: approximately 700 Kergs. cm'z‘s'l). The reaction temperature was l9°C. 32 Phosphorylation was measured as AT P formation by the method detailed elsewhere17. Results Ferrocyanide-supported electron transport and_phosphorylation All of the experiments described in this paper (except those of Figure 13) were carried out using NHZOH—washed, non-02 producing chloroplasts. The electron acceptor employed was usually methylviologen. The chloroplast preparations used were free of catalase activity, and 159 this facilitated the measuring of electron transport as the aerobic reoxidation of reduced methylviologen. The top trace of Figure 1A confirms that the NH20H-washed chloroplasts used were practically totally unable to reduce methylviologen with electrons from water. However, when high concentrations of ferro- cyanide (>lO mM) were added, a rapid uptake of 02 was observed. The fact that exactly half of the 02 taken up in the light was released on addition of catalase indicates that all of the 02 taken up was reduced to the level of H202. The reaction is highly sensitive to DCMU (bottom trace) indicat- ing an obligatory involvement of Photosystem II. The addition of superoxide dismutase had no inhibitory effect on the ferrocyanide-supported 02 uptake even at an enzyme activity of 300 units/ml reaction mixture. This amount of superoxide dismutase is 3 to lO-fold greater than the amount required to completely eliminate the oxidation of ascorbate or of ascorbate-donor couples by superoxide radical anion, the precursor of H202 (see Table I; see also references ll, l8, 19). Clearly there was no radical oxidation of ferrocyanide to contribute to 02 consumption. we may thus safely conclude that the pair of electrons utilized for the reduction of 02 to H202 originates almost exclusively from the light-dependent biological oxidation of ferrocyanide. Uptake of one molecule of 02 therefore corresponds to transport of a pair of electrons from ferrocyanide to methylviologen (hence P/O2 = P/ez; see below). The parallel experiment shown in Figure 18 was performed with normal, water-oxidizing chloroplasts. The active reduction of methyl- viologen with electrons from water was virtually unaffected by the 160 presence of 30 mM ferrocyanide. Apparently ferrocyanide cannot replace water as the electron donor as long as the water-oxidizing mechanism is intact. The spectrophotometric data of Figure 2A,B demonstrate that ferricyanide is formed in approximately the amount predicted from the amount of ferrocyanide-supported 02 uptake. Illumination of reaction mixtures containing ferrocyanide caused irreversible absorbance changes in the 400-450 nm region, and the wavelength dependence of these changes agreed well with the absorption spectrum of ferricyanide. In Figure 23 the reversible change at 480 nm and the reversible portion of the change at 420 nm represent light-scattering changes of chloroplasts. The absorbance changes plotted in Figure 2A have been corrected for these light-scattering changes. In this experiment the absorbance change at 420 nm caused by 60 s illumination was 0.028, which corresponds to a formation of 54 nmoles ferricyanide in the 2 ml reaction mixture. A parallel experiment with a duplicate reaction mixture showed that 24 nmoles of 02 (48 nequiv) were consumed during the same period of illumi- nation. This is in good agreement with the optical data. The "Millipore"- filtration experiment of Figure 20,0 unequivocally demonstrated the light- dependent formation of ferricyanide (for details, see legend for Figure 2C,D). As indicated by the experiment of Figure 2A,B, the oxidation of ferrocyanide could be followed spectrophotometrically. However, the high sensitivity of our optical system to light-scattering changes made this alteriative method rather impractical. The transport of electrons from ferrocyanide to methylviologen supports phosphorylation. In Figure 3, the time course curves for 02 161 uptake and ATP formation were obtained by yield determinations, using a series of identical reaction mixtures illuminated for different periods of time. Both processes exhibited similar, almost biphasic kinetics. The plots revealed an unexpectedly fast initial phase of 02 uptake which continuous 02 tracing failed to detect (cf. Figure lA) obviously because of the slow response of the conventional membrane-covered 02 electrode used. In this experiment the rate of electron transport computed on the basis of the total 02 uptake induced by 5 s illumunation was 220 uequiv (ll0 nmoles 02) h‘1.mg chlorophyll'I. The true initial rate at t = 0 presumably exceeded 300 in the uequiv units. These rates are comparable to the rate of the normal Hill reaction. The cause of the biphasic kinetics will be discussed in a later section. Although the rates of electron transport and phosphorylation decreased quickly with time, their ratio (P/O2 or P/ez; actually the ratio of yields) held level at 0.6 for about 60 s, then gradually rose (Figure 3, inset). This phenomenon can be explained if one assumes that the P/e2 ratio intrinsic to the ferrocyanide oxidation indeed approxi- mates the predicted value of 0.6 (see Introduction). As the reaction proceeds, however, the oxidation product ferricyanide accumulates and begins to replace methylviologen as the electron acceptor, thereby intro- ducing an unmeasurable electron flow from ferrocyanide to ferricyanide. This unmeasurable (cyclic) electron flow still supports phosphorylation at an unchanged rate. Consequently the apparent P/e2 is inflated. Figure 4 shows that such a conversion from noncyclic to cyclic photo- phosphorylation can indeed be induced by exogenously added ferricyanide. A ferricyanide concentration of 30 uM was sufficient to effect a 50% 162 conversion. Close examination of data will show that the apparent rise in the P/ezdepicted in Figure 3 is about the extent one would expect from the data of Figure 4. Very high concentrations of ferrocyanide (330 mM) are required to saturate electron transport and phosphorylation (Figure 5). However, the efficiency with which the electron transport is coupled to phosphory- lation (P/e2 = 0.53 in this experiment) remains constant over a wide range of ferrocyanide concentrations. This latter fact speaks strongly against the possibility that the low P/e2 values (0.5 to 0.6) result from an uncoupling effect of ferrocyanide. In fact, we found no evidence that ferrocyanide could act as an uncoupler in the concentration range tested. The presence of 30 mM ferrocyanide had virtually no effect on the phosphorylation coupled to the Hill reaction in normal chloroplasts (data not shown). The rate of ferrocyanide oxidation peaks at pH 7.7 but exhibits a wide skirt on the acid side (Figure 6). The associated phosphorylation and its efficiency (P/ez) peak at pH 8 and quickly approach zero below pH 7. It should be noted, however, that the rates presented in Figure 6 (also Figures 4 and 5) were computed on the basis of the total 02 uptake yield and the ATP yield resulting from 30 s illumination. Since the kinetics of ferrocyanide oxidation are nonlinear (Figure 3), these are time-averaged values of quickly diminishing rates. As a result, it is quite possible that the pH profiles for the rates of electron transport and ATP formation are somewhat distorted. However, the pH profile for the essentially time-independent term P/e2 is presumably free from such distortion. 163 Iodide-supported electron transport andgphosphorylation Iodide ion serves as an efficient donor of electrons to Photosystem II in NHZOH-washed chloroplasts. Electron transport and phosphorylation both proceed more linearly than in the ferrocyanide- oxidizing system. The P/e2 ratio exceeds 0.5 initially (< l5 s) but falls slightly with the reaction time (Figure 7). Iodide photooxidation and associated phosphorylation become saturated at a KI concentration of about l5 mM (Figure 8). The P/e2 ratio (0.45 as measured after 30 s illumination) was almost completely independent of the concentration of KI. Again this constancy of P/e2 values speaks strongly against the possibility of the low P/e2 resulting from uncoupling by iodide.- Con- centrations of KI up to 30 mM had essentially no uncoupling effect when tested on the phosphorylation associated with the methylviologen Hill reaction in normal chloroplasts. (Iodide does not easily replace water as the donor in normal chloroplasts.) Electron transport and phosphory- lation supported by iodide oxidation share a common pH Optimum at 8 (Figure 9). As in the ferrocyanide-oxidizing system, the phosphorylation, and therefore the P/e2 ratio, steeply approach zero below pH 7. A marked stimulation of electron transport by concomitant phosphorylation was also noted. Attempts to detect the oxidation product of iodide, free iodine (or I3"), were unsuccessful. This is not surprising in view of the high reactivity of 12. The amount of I formed by the end of 30 s 2 illumination (the routine reaction time) would have been of the order of 50 nmoles in the 2 ml reaction mixture, an amount which could have 164 easily been (and indeed was) consumed even by minute impurities of the chemicals present in the mixture such as sucrose and buffer. A test with chloroplasts exhaustively washed and suspended in phosphate buffer showed, however, that chloroplasts themselves can readily consume an amount of exogenous 12 which is nearly equivalent to the amount of chlorophyll present. Since the 12 produced (which would have been able to oxidize ferrocyanide) actually disappeared quickly, it was possible to use ferricyanide as the electron acceptor, replacing methylviologen, and observe the reaction photometrically as ferricyanide reduction. In this study, however, we have not used this optical method for quantita- tive experiments for the technical reasons described for ferrocyanide oxidation. In the methylviologen-reducing system there was no indication of H202 fbrmed being consumed by a reaction with 1' within the reaction time of 30 5. Not surprisingly, neither of the higher potential halogen ions Cl' and Br' was found able to substitute for I' as the electron donor for Photosystem II. The electron transport and phosphorylation data pertaining to the ferrocyanide- and iodide-oxidizing systems are summarized in Table l, and contrasted with the data for several other Photosystem II donor reactions which give P/e2 of l.0 to l.l. For the details of the latter reactions, see ref. ll. Also included are preliminary data for the photooxidation of another metal complex, Mn2+(8-hydroxyquinoline)2, which we found to serve as a good electron donor for Photosystem II. Again a P/e2 value of approximately 0.6 was found. Characterization of this new donor reaction has not been completed yet. 165 Changes in theng of the medium associated with photooxidation of ferrocyanide and’iodide ions The methylviologen-mediated aerobic photooxidation of ferrocyanide should consume protons according to the formula: 2Fe(CN):' + 2H+ + o 26‘ transport 5 2Fe(CN)g' + H202 Eqn. 1 methylviologen Consequently the pH of a weakly-buffered suspending medium should rise irreversibly as the reaction proceeds. One might also expect this pro— cess to support an energy-linked proton translocation since the process is coupled (at Site I, according to our model). Traces a_and b_of Figure l0 verify these predictions. Proton uptake did occur upon illumi- nation, and it did comprise two components: a gramicidin-insensitive, irreversible component (aerobic ferrocyanide oxidation or Eqn. l) and a gramicidin-sensitive, reversible component (energy-linked proton uptake). The initial slope (0 to 20 s) of H+ consumption in trace a_is approxi— mately 230 uequiv-h'1.chlorophyll'], which is diminished to 110 in the presence of gramicidin (trace 9). The latter rate, which should represent the electron transport rate (Eqn. l), is in fair agreement with the rate measured as 02 uptake under comparable conditions (130 uequiv or 65 nmoles 02. h'].mg chlorophyll-1). The methylviologen-mediated photooxidation of I' probably would have exhibited very similar proton changes, if the 12 produced had remained in the medium as did ferricyanide. However, as indicated in the preceding section, the I2 produced is quickly reduced again by ambient oxidizable substances. These chemical reactions apparently involve stoichiometric production of acids (e.g. AH2 + 12 +-A + 2H+ + 166 ZI') to compensate for the proton consumption due to electron transport (215 + 2H+ + 02 +~Iz+ 2H202). Consequently the only detectable proton change is a totally gramicidin-sensitive reversible proton translocation (traces g_and g). Trace g_confirms that NHZOH-washed chloroplasts used in this experiment were totally inactive in the absence of artificial reductants added. The trace also demonstrates an acid production induced by exogenously added 12 (as 13'). Discussion Ferrocyanide (E0' = 0.43 V) and iodide E ' = 0.53 V) can be 0 ranked with benzidine (E0' = 0.65 V at pH 5)20 and dicyanohydroquinone (E0 = 0.97 V; EO' unknown)2] among the weakest artificial reductants oxidized by illuminated chloroplasts. Predictably the oxidation of these substances requires Photosystem II, as indicated by its DCMU- sensitivity. However, ferrocyanide and iodide are unique in that the transport of electrons from these two reductants to methylviologen through the two photosystems supports phosphorylation only with P/e2 values of 0.5 to 0.6. The same pathway mediates phosphorylation with P/e2 values of 1.0 to 1.2 when the electron donor is water, catechol, benzidine or p-aminophenol (Table I; see also ref. ll). The low effi- ciencies of phosphorylation obtained with ferrocyanide and iodide can- not be explained as being due to uncoupling by the donors themselves, since the P/e2 values are independent of donor concentrations over very wide ranges (Figures 5 and 8). Nor is it likely that any kind of pro- duct inhibition is largely responsible for the low phosphorylation efficiencies, since the P/e2 value is essentially independent of the 167 reaction time (<60 5) when ferrocyanide is the donor (Figure 3). The P/e2 does decrease with the reaction time in iodide oxidation, but does so only slightly (0.5 to 0.4 in 60 s; Figure 7). We are thus led to the only remaining simple interpretation: one of the two sites of phosphorylation is inoperative when ferrocyanide or iodide serves as electron donor to Photosystem II. It seems logical to deduce further that the inoperative site is one which is proximal to the point of entry of electrons from the donors, i.e., Site II. Indeed, the characteristics of phosphorylation associated with the oxidation of ferrocyanide and iodide are quite close to what one would expect from 13 have presented the involvement of Site I alone. Gould and Izawa strong evidence that the pathway of electrons from reduced 2,6- dichlorophenolindophenol to methylviologen via Photosystem I includes only Site I. Phosphorylation and its efficiency associated with this pathway peak at pH 8 or slightly above (where P/e2 is 0.65 to 0.6811) and sharply decline toward zero between pH 6.5 and 7. Unpublished data of J. M. Gould for another Photosystem I donor reaction, diaminodurene +-methylviologen, also show an almost identical pH profile of phosphory- lation with a maximal P/e2 of 0.6. Similar maximal P/e2 values of 0.57 to 0.62 have also been found with diaminotoluene and p-phenylenediamine 11. These properties of Photosystem I-mediated phosphoryla- as donors tion, which presumably originate from the mechanism of coupling at Site I, are unmistakably shared in common by the phosphorylation supported by Photosystem II-initiated transport of electrons from ferrocyanide and iodide to methylviologen (Figures 6 and 9). The only perceptible 168 (and unexplained) difference is the slightly higher P/e2 values found for the pathway reduced 2,6-dichlorophenolindophenol-+ methylviologen (see above). It should be added here that the characteristics of phosphoryla- tion attributable to Site II, as observed in the partial electron trans- port systems involving only Photosystem II (H20-+ DBMIB9 and H20-+ DMQ‘3), are quite different. Both electron transport and phosphorylation peak around pH 7.6 and their ratio (P/e2 = 0.3 to 0.4) is remarkably inde- pendent of pH between 6 and 9. As we have already pointed out, the oxidation of ferrocyanide to ferricyanide is a simple valence change at physiological pH's because of the very high stability of both forms of the complex. No detectable H+ or 0H’ change occurs as indicated by the pH-independent E0. value. Neither does the oxidation of iodide to iodine induce H+ changes perhaps except above pH 9 where IOH formation begins. In contrast, all of those PS II donors including water which gave higher P/e2 values (1.0 to 1.2), i.e., catechol, benzidine, p-aminophenol and dicyanohydroquinone, are hydrogen donors, the oxidation of which involves, or presumably involves, nearly stoichiometric release of protons at physiological pH's (Eo'/ pH 2] or for benzidine above pH = 0.06 V; no data for dicyanohydroquinone 6.020). Thus, if one is allowed to ignore the question regarding the fate of the oxidation product of iodide (see below), our results would seem to provide a strong experimental basis for the chemiosmotic model 16'22 or in our terminology, at Site 11. of energy conservation at PS 11 with a strong emphasis on proton relationships rather than on membrane potentials. 169 Our results also offer a possible explanation of the intermediate efficiency of phosphorylation (P/e2 = 0.8 to 0.9) found associated with Photosystem II-dependent oxidation of ascorbate.n That is, the oxidation of ascorbic acid anion (i.e. above pH 4.5) to the neutral molecule dehydroascorbate, the demonstrated end product,23 liberates only one proton for a pair of electrons removed, thus: AH- + A + H+ + 2e'. It may be that Site II is only half-operative with ascor- bate as the electron source. we consider the nominal distinction between one-electron donors (e.g. ferrocyanide and iodide) and two-electron donors (e.g. catechol, benzidine and water) as probably unimportant here. In 24 and of Babcock and fact, the flash experiments of Bennoun and Joliot Sauer25 indicated that the oxidation of hydroxylamine, hydroquinone and p-phenylenediamine by Photosystem II is via a one-quantum (hence pre- sumably via a one-electron process) requiring no charge accumulation. The chemiosmotic model of energy conservation at Photosystem 11 requires, though perhaps not absolutely, that artificial electron donors as well as water be oxidized at or near the inner surface of the thyla- koid membrane. There are several lines of circumstantial evidence for this. The concentrations of ferrocyanide and iodide required to saturate oxidation rates are extremely high: 30 and 15 mM, respectively. These values are in striking contrast with the rate-saturating concentrations (<0.5 mM) of other Photosystem II donors such as catechol, benzidine, p-aminophenol, etc. (Table I; see also refs l0 and l2). Ascorbate, 170 when acts as the direct electron donor for Photosystem II, saturates the oxidation rate at an intermediate but still relatively high concentra- tion of 5 mM‘O. It thus appears that the concentration requirement is primarily a function of the liposolubility of the donor; nonionic, non- polar (lipophilic) donors having the highest accessibility to the oxida- tion site while ionic, highly polar (hydrophilic) donors the lowest. The implication is clearly that the reductants must penetrate the lipid membrane to reach the oxidation site. Such a location of the oxidation site could also explain the biphasic kinetics of oxidation of ferro- cyanide, the least permeant reductant (Figure 3). The initial rapid phase where the oxidation rate approaches the rate of the normal Hill reaction may represent the oxidation of that portion of ferrocyanide which has already permeated the membrane (possibly into the internal space of the thylakoid) while the subsequent much slower phase may represent a duffusion—limited process. The existence of a strong per- meability barrier limiting ferrocyanide oxidation is also suggested by the preliminary experiment of D. R. 0rt (unpublished) which indicated that the rate of oxidation of ferrocyanide can be markedly increased by the addition of sub-uncoupling concentrations of digitonin. Iodide ion seems to permeate the thylakoid membrane somewhat more easily than does ferrocyanide, as suggested by the slightly lower concentration requirement and the more linear kinetics of oxidation. However, the iodide-oxidizing system is not nearly as easily character- ized as is the ferrocyanide-oxidizing system, because the unstable oxida- tion product 12 (or possibly 1') is quickly reduced again by unknown reducing substances in the reaction mixture. (Ferricyanide does not 171 react with these substances.) Moreover, this re-reduction of I2 is accompanied by an acid production. Thus, iodide is indirect1y_a proton- producing electron donor. Nevertheless, the P/e2 data suggest that Site II cannot operate when 1' serves as electron donor. If the chemios- motic model of Site 11 is valid, then the implication is that the protons released by the reduction of I2 are not available for phosphorylation. According to the model, this could only mean that the chemical reduction of 12 and the associated proton production take place predominantly on the outside or the outer surface of the thylakoid membrane. Such a situation does not seem impossible, since I2 is a highly diffusable substance, and since it is conceivable that the region of the membrane where the strong oxidant of Photosystem II is produced may be so defi- cient in oxidizable substances as to allow 12 a brief life-time to diffuse away. The re-reduction of I2 could then take place preferent- ially on the outer surface of the membrane where abundant oxidizable substances are available as impurities of the medium. This interpreta- tion is at least consistent with the fact that no important membrane constituents seem to be destroyed by the I2 produced. As shown in Figure 7 iodide oxidation and associated phosphorylation proceed almost linearly for at least 60 5, during which time the I2 produced and con- sumed reach the amount equivalent to chlorophyll present. Admittedly the interpretation given above for the fate of I2 still leaves ample room for debate. In fact, we feel that the apparent total "uselessness" of protons produced by the nonbiological reduction of 12 may far better be reconciled with the modified chemiosmotic theory 26 of Williams The theory assigns the key role to localized proton 172 concentrations induced withjn_the hydrophobic membrane by the oxidation of hydrogen carriers, rather than to proton gradients subsequently formed across the membrane. Whichever the model, however, our observa- tions described in this paper strongly point to the role of protons in the main sequence of energy transduction at Photosystem II. Acknowledgments The authors wish to thank Dr. N. E. Good for critical reading of the manuscript. This work was supported by a grant (GB37959X) from the National Science Foundation. References l. Ouitrakul, R. and Izawa, S. (l973) Biochim. Biophys. Acta 305, 105- 118 2. Izawa, S., Gould, J. M., Ort, D. R., Felker, P. and Good, N. E. (1973) Biochim. BiOphyS. Acta 305, 119-128 3. Trebst, A. and Reimer, S. (l973) Biochim. Biophys. Acta 305, l29-139 4. Kraayenhof, R., Izawa, S. and Chance, B. (l972) Plant Physiol. 50, 713-718 5. Ort, D. R., Izawa, 3., Good, N. E. and Krogmann, D. N. (l973) FEBS Lett. 31, 119-122 6. Bradeen, D. A., Winget, G. D., Gould, J. M. and Ort, D. R. (l973) Plant Physiol. 52, 680-682 7. Avron, M. and Chance, 8. (1966) Brookhaven Symp. Biol. l9, l49-l60 8. Bahme, H. and Cramer, w. A. (1972) Biochemistry 11, ll55-ll60 9. Gould, J. M. and Izawa, S. (l973) Eur. J. Biochem. 37, 185-192 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24 25. 26. 173 Ort, D. R. and Izawa, S. (1973) Plant Physiol. 52, 595-600 Ort, D. R. and Izawa, S. (1974) Plant Physiol. 53 (in press) Yamashita, T. and Butler, W. L. (1968) Plant Physiol. 44, 435-438 Gould, J. M. and Izawa, S. (1973) Biochim. Biophys. Acta 314, 211- 223 Gould, J. M. and Izawa, S. (1974) Biochim. Biophys. Acta (in press) Mitchell, P. (1966) Biol. Rev. 41, 445-602 Rumberg, B., Reinwald, E., Schrdder, H. and Siggel, U. (1969) in Progress in Photosynthesis Research (Metzner, H., ed.), IUBS, Tubingen, pp. 1374-1382 Saha, S. and Good, N. E. (1970) J. 8101. Chem. 245, 5017-5021 Allen, J. F. and Hall, 0. 0. (1973) Biochem. Biophys. Res. Commun. 52, 856-862 Elstner, E. F. and Kramer, R. (1973) Biochim. Biophys. Acta 314, 340-353 Clark, W. M. (1960) Oxidation—reduction Potentials of Organic Systems, Williams & Wilkins, Baltimore Rideal, E. K. (1925) Trans. Faraday Soc. 21, 143-144 Witt, H. (1971) Quart. Rev. Biophys. 4, 365-477 Bfihme, H. and Trebst, A. (1969) Biochim. Biophys. Acta 180, 137-148 Bennoun, P. and Joliot, A. (1969) Biochim. Biophys. Acta 189, 85-94 Babcock, G. T. and Sauer, K. (1971) Abst. Congress on Primary Photochemistry of Photosynthesis, Argon National Lab. p. 29 Williams, R. J. P. (1969) Current Topics in Bioenergetics 3, 79-156 flq‘d'r’rr‘. #1 wells”? 174 Table I. Electron transport from various artificial donors to methylviologen via Photosystem II and Photosystem I and the associated phosphorylation in NHZOH-washed chloroplasts. Basic reaction conditions were as in Figure 1A except for the variations in the electron donor used. When catechol and p-aminophenol were used the reaction mixture contained 0.5 mM ascorbate as electron reservoir. When used, superoxide dismutase ($00) was 300 units/ml reaction mixture for the first three donors and 100 units/ml for the rest: DCMU, 2 uM._ I' was added as KI. Mn(oxine)2 was a 1:2 mixture (mole/mole) of MnCl2 and 8-hydroxyquinoline. For details of the reactions involving catechol, p-aminophenol and benzidine, see ref. 11. Reaction time, 30 s. firfi 02 uptake ATP formation P/e Donor Conc. (nmoles-h'].mg chlorophyll-1) (= :/0 with (mM) NO addi- +500 +DCMU NO addi- +500 +DCMU SOD) 2 tion tion Fe(CN)g' 30 44* 45* 5 24* 25* o 0.56 I- 20 102 100 8 47 45 <2 0.45** Mn(oxine)2 0.5 75 72 5 43 44 0 0.61 Catechol 0.5 182 102 10 116 112 <2 1.10 p-Amino- phenol 0.5 152 91 10 95 93 <2 1.02 Benzidine 0.5 92 88 5 94 93 0 1.06 * Time-averaged values (0-30 5) of rapidly diminishing rates. Initial rates at t = 0 were presumably 2 to 3-fold faster (cf. Figure 3). **At t = 0 the ratio probably exceeded 0.5 (cf. Figure 7). 175 Legend of Figures Figure 1. A: Electron transport from ferrocyanide to methylviologen (MV) in NH20H-washed chloroplasts as observed by 02 uptake. The reaction mixture (2 m1) contained 0.1 M sucrose, 50 mM HEPPS-NaOH buffer (pH 8.0), 2 mM Mgc12, 0.75 mM ADP, 5 mM Na H32P0 0.5 mM methylviologen, NHZOH- 2 4’ washed chloroplasts equivalent to 40 ug chlorophyll, and the indicated concentrations of potassium ferrocyanide. When used, DCMU was 2 uM. B: Lack of effect of ferrocyanide addition on methylviologen Hill reaction in normal chloroplasts. The conditions were as in A except normal, 02- producing chloroplasts were used. Figure 2. Electron transport from ferrocyanide to methylviologen (MV) in NHZOH-washed chloroplasts as observed by spectrophotometric determina- tion of ferricyanide formation. In A, the curve represents the absorp- tion spectrum of 27 uM ferricyanide: the circles represent light-induced absorbance changes of the reaction mixture at different wavelengths corrected for light-scattering changes as in B. Conditions were as in Figure 1A with 30 mM ferrocyanide and 20 ug chlorophyll/m1. In C, the curve designated "light" indicates the absorption spectrum of a filtrate of the reaction mixture exposed to light for 60 s; "dark", a dark con- trol. D: the light minus dark difference spectrum. In this filtration experiment, the reaction mixture (8 m1) contained 0.1 M sucrose, 50 mM HEPES-NaOH buffer (pH 8.0), 2 mM MgC12, 0.5 mM methylviologen, 30 mM potassium ferrocyanide and NH20H-washed chloroplasts equivalent to 100 ug chlorophyll. Figure 3. Time courses of electron transport from ferrocyanide to methylviologen (MV) and of associated phosphorylation in NHZOH-washed 176 chloroplasts. The time courses were determined using a series of identical reaction mixtures illuminated for indicated period of time. Conditions were as in Figure 1A with 30 mM ferrocyanide and 40 ug/ml chlorophyll. Figure 4. Effect of increasing concentrations of exogenous ferricyanide on the ferrocyanide + methylviologen reaction and associated phosphoryla- tion in NHZOH-washed chloroplasts. The composition of the reaction mixture was as in Figure 1A with 30 mM ferrocyanide, 30 ug/ml chloro- phyll, and the indicated concentrations of ferricyanide added. The reaction time was 30 s. Note that the phosphorylation remains unchanged as the 02 uptake is suppressed by ferricyanide addition, suggesting the conversion of noncyclic photophosphorylation to cyclic photophosphoryla- tion mediated by ferro-/ferricyanide turn over. Figure 5. Electron transport from ferrocyanide to methylviologen (NV) and associated phosphorylation in NHZOH-washed chloroplasts as a function of the ferrocyanide concentration. The composition of the reaction mixture was as in Figure 1A with 40 ug/ml chlorophyll and the indicated concentration of ferrocyanide. The reaction time was 30 s. Figure 6. Electron transport from ferrocyanide to methylviologen (MV) and associated phosphorylation in NHZOH-washed chloroplasts as a function of pH. The composition of the reaction mixture was as in Figure 1A with 30 mM ferrocyanide and 40 ug/ml chlorophyll. The buffers used (all 50 mM) were: MES (2-(N-morpholino)ethanesu1fonic acid)-Na0H (pH 6.0 and 6.5), HEPES-NaOH (pH 7.0 and 7.5) and HEPPS-NaOH buffer (pH 8.0 to 9.0). The reaction time was 30 s. 177 Figure 7. Time courses of electron transport from iodide to methylviologen (MV) and associated phosphorylation in NHZOH-washed chloroplasts. The basic reaction conditions were as in Figure 1A except that potassium ferro- cyanide was replaced by 20 mM KI. Chlorophyll, 40 ug/ml. For procedures, see Figure 3. Figure 8. Electron transport from iodide to methylviologen (MV) and associated phosphorylation as a function of the KI concentration. The basic conditions were as in Figure 1A except that ferrocyanide was replaced by indicated concentrations of KI. When used, DCMU was 1.5 uM. Chlorophyll, 40 ug/ml. The reaction time, 30 s. Figure 9. Electron transport from iodide to methylviologen (MV) and associated phosphorylation in NHZOH-washed chloroplasts as a function of pH. The basic conditions were as in Figure 1A except that ferrocyanide was replaced by 20 mM KI and that the buffers used were as in Figure 6. Chlorophyll, 40 ug/ml. The reaction time, 30 s. Figure 10. Changes in the pH of the medium associated with the transport of electrons from ferrocyanide (traces g_and b), iodide (traces g_and g) and water (trace e) to methylviologen (MV) in NHZOH-washed chloroplasts. The basic constituents of the reaction mixture (2 ml) were 0.1 M sucrose, 0.5 mM HEPPS-NaOH buffer (pH 7.6), 2 mM M9012, 0.1 mM methylviologen and chloroplasts equivalent to 25 pg chlorophyll/m1. When used, potassium ferrocyanide was 30 mM. KI, used alone, was 20 mM. Gramicidin, when added, was 2 ug/ml. In trace g, 12 was added in the form of 13' (final concentrations, 0.5 mM I2 plus 2 mM I'). Trace §_shows the lack of light-induced pH rise in the absence of artificial electron donors 178 added and also an acid production induced by the addition of I2. For explanation, see text. 179 . mudfiaku. . t... . .1...” 1.5!.an .mummpaogo_;u _mssoz cw cowpummm pmw: >2 co coepwuu< muwcmxuoesmu mo uummeu mo xumb m .mgmmFQoso_;u umcmmzizo :z cw >2 op muwcm>UosLmu seem “Loamcmsh cospomFM M2 10d... Am: >2 1&203... :2 180 T 1 l 1 1 1 r] 1 . Off 0.03-A B '39,.” 1 0 ‘ AA: 0 420nm +0028 a: :g(D.C)2!- C) a; .g 480n l 2 . - 0.01 — m'" <1 0 1 L I I l 1 1 i I\O ' 400 450 =30 7\ (nm) CIISN- 0'4 — D Light minus Dark ()13 CLND- 8 ' 4 ii <3 5 005— 8 0.2 - 4: <1 0 l l o" .. 400 450 500 7\(nm) O L I l 1 l 1 I I L ' ' 400 450 500 7\ (nm) Figure 2. Electron Transport from Ferrocyanide to MV in NH20H- washed Chloroplasts as Observed by Spectrophotometric Determination of Ferricyanide Formation. (1132C) males 02 or pmoles ATP/ ml 181 l l 1 l _. Fe(CN)2'—> MV /0 _ O 02 upioke\o .I ATFK\\\\s O /0 iS-o-—q---"""’‘37—”;o I l 60 120 l l I 1 0 30 60 90 120 Figure 3. Time of Illumination (sec) Time Courses of Electron Transport from Ferrocyanide to MV and of Associated Phosphorylation in NHZOH-washed Chloroplasts. i~I.."_.-n-I' , . . .'. - S" W'h .-_ 182 1 l 1 4o- — O FeiCN)§‘ ——> MV 3... '1' FBiCN)6 2’" 30— ._ 8 g /02 "make '5 U" E "1'5: 20— .- e i . . < I ' ' U) \ .92 3 ATP 1 5 10— o .. N o (D 2 O E o () s1 7‘7‘T‘r~——-—__ 1 <3—— 0 so 100 150 Ferricyanide (u M) Effect of Increasing Concentrations of Exogenous Ferri- cyanide on the Ferrocyanide to MV Reaction and Associated Phosphorylation in NHZOH-washed Chlor0plasts. Figure 4. {I‘ll-I'll“ Ill-bl. 183 0.5 a; Fe(CN)4‘ —> MV VOA—O_O§O\o 6 ,3, F Q) \ h 0. N F. O .. \ '7_ 0- r 3%} 15C)h' l 1 1 '- 8- o 20 40 so 5 1'2 O 0\o o \o 40" \ - CJI i k E O 02 UpO e 7.: o. _ s ,_ so <1 IVTF’ .3. / / . E 20- o O \o- 3. L. //////’ o 0 c5“ / + DCMU £33 10 /. 02 upioko ATP\ " g . . . :1. /°/? \ . O 20 4O 6O Ferrocyanide (mM) Figure 5. Electron Transport from Ferrocyanide to NV and Asso- ciated Phosphorylation in NH 0H-washed Chloroplasts as a Function of the Ferrocyani e Concentration. 184 . 1 ' I ' I 50 - 4_ Fe(CN )6 —r MV /02 uptake 3 O"\ .1: a 40* o/ 2 / 2 / 5 o A, E’ 4’ o 1’ so _ P/o2 // 4.: \ I o. [I 2 I g . 15 2!) iAT E :1. t- o N 0 IO .8 - OJ 0 E a. O 1 1 1 1 O 8.0 9.0 pH Figure 6. Electron Transport from Ferrocyanide to MV and Associated Phosphorylation in NHZOH-washed Chloroplasts as a Function of pH. szméy __ J -_,. ._. .9... nun. -run at :1:- - 185 05L ° _ \OxO——o- I —’MV ’7“ 1— 3 O. - N dos—g - CL’ C) E “\ () l l 1 L o. _ 0 30 60 - .3 0.04 o m .9 CD 0 uptake g 0.03— 2 - s. C) C) N O a) 0.02 - .92 (D E :1 0.0l .. 0 l l l 1 0 30 60 Time of illumination (sec) Figure 7. Time Courses of Electron Transport from Iodide to MV and Associated Phosphorylation in NHZOH-washed Chloro- plasts. 186 I’ —> MV '= o 2100— 0/0 o. 02 uptakex.) 8 o. 5‘0 .9 A o\°'—'O o —o 5 a" U, 80P E - E N _ _- o 1.: o R *- .. l I 1 E 6° 00 IO 20 so (D Q) '5 /_———o o :5; 40 -I-/0 /——\ATP " & o N 0 + _o__c1v1u g 20 l./. 0 uptake ATP - — 2 O I g /4 \ A o 1_ 1 :L l 1 \ l 4' 0 IO 20 30 KI (mM) Figure 8. Electron Transport from Iodide to NV and Associated Phosphorylation as a Function of the KI Concentration. 187 I ' 1 CL5- . _ 9” \\b I--———i>iw\/ 1* - \. lCK)-' B: - I. . d“ ' '5" E e . o/OO\ 5- 1 __ l g 80 O 5.0 8.0 Ozupmke 1— A .- c» -‘\‘zs 15 17F 60 h 0 <~ 02 uptake _ '9.- A 1" P; ) <1 3; .A - 41C)“ .,\ - 0 £5 /////// ///,/’\ 1 ' /Z . \ATP " L. O ’0/ N 4 O 20— ° 0) .12 c: E - - :1. O ’9/1 1 1 1 1 1 EEC) '?() ELC) EDC) pH Figure 9. Electron Transport from Iodide to NV and Associated Phosphorylation in NHZOH-washed Chloroplasts as a Function of pH. 188 .mpmmFQosopgu nmgmmzixomzz cw >2 op swam: use onwuoH .mnwcmxuoesmm soc; mcospumPM to peoamcmsp mgp :p_z nmpmvoomm< Ezmum2 mgp mo :2 as» cw mmmcmgu .o_ mszmwm 8:20 >2 3 :_so i $4.1 ii A :o— 2*0 co :1 522880» >2 1% G 75 co v :3 to a22.1% .o 2 to si. e". um..... Om 2522520 :23 21.282283. 3 CO e 2:5 >33: 0. u+I< . EREEEO >21 022823 to .v APPENDIX VII SITE-SPECIFIC INHIBITION OF PHOTOPHOSPHORYLATION IN ISOLATED SPINACH CHLOROPLASTS BY MERCURIC CHLORIDE 189 Site-specific Inhibition of Photophosphorylation in Isolated Spinach Chloroplasts by Mercuric Chloride1 DAVID A. BRADEEN AND G. DOUGLAS WINGET Received for publication June 4, 1973 Department of Biological Sciences, University of Cincinnati, Cincinnati, Ohio 45221 J. MICHAEL GOULD AND DONALD R. ORT Department of Botany and Plant Pathology, Michigan State University, East Lansing, Michigan 48824 ABSTRACT Photophosphorylation associated with noncyclic electron transport in isolated spinach (Spinacia oleracea) chloroplasts is inhibited to approximately 50% by low concentrations of HgCls (less than I pmole Hg’*/mg chlorophyll) when the elec- tron transport pathway includes both sites of energy coupling. Reactions involving only a part of the electron transport system can give a functional isolation of at least two sites coupled to phosphorylation. Only one of these sites, located between the oxidation of plastoquinone and the reduction of cytochrome f, is sensitive to mercuric chloride. The energy conservation site located before plastoquinone and close to photosystem II is unaffected by HgCle concentrations up to lO-fold those re- quired to inhibit phosphorylation by the coupling site after plastoquinone. This site-specific inhibition may reflect a mechanistic difference in the mode of energy coupling at the two coupling sites or a variable accessibility of HgCla to these sites. Concentrations of Hng, which inhibit steady state phos- phorylation, do not inhibit dark phosphorylation after illumi- nation (Xe), suggesting that HgCle affects a step in the cou- pling mechanism prior to the terminal step of ATP formation. Recent data from several laboratories indicate that there are at least two sites of energy coupling associated with noncyclic electron transport in isolated chloroplasts (5-7, 11, 17-20). By utilizing various electron donor-acceptor systems in con- junction with several new inhibitors of electron transport, these energy-coupling sites can be functionally isolated and charac- terized. One site closely associated with photosystem II (5), differs in several respects from a second, well recognized site coupled to electron transport from plastoquinone to cyto- chrome f. (3). The photosystem II-dependent energy coupling site exhibits no control over coupled electron transport and has a P/e,’ ratio of about 0.4 (5-7, 17, 18). Furthermore, this P/e, 1J.M.G. and D.R.0. were supported by Grants GB 22657 and GB 37959X from the National Science Foundation. ’ Abbreviations: P/e. ratio: the ratio of the number of molecules of ATP formed to the number of pairs of electrons transported; DCIPI-Is: reduced 2,6-dichlorophenolindophenol; HEPPS: hydroxy- ratio is practically pH independent (5, 6). On the other hand, the coupling site between plastoquinone and cytochrome I, which is the rate-limiting step for the Hill reaction, exhibits control over electron transport and has a pH-dependent P/e. ratio of about 0.6 (pH 8.0-8.5) (6). Because of the apparent dif- ferences in the characteristics Of the two coupling sites, it was of interest to study the effect of specific inhibitors of the phos- phorylation reaction (energy transfer inhibitors) on each cou- pling site. In this communication, we report that the energy transfer inhibitor HgCl, (11) specifically inhibits ATP forma- tion supported by the coupling site between plastoquinone and cytochrome I while not affecting ATP formation supported by the coupling site close to photosystem II. MATERIALS AND METHODS Spinach (Spinacia oleracea) chloroplasts were prepared as described previously (21), except TES-NaOH buffer replaced Tricine in the isolation media, since Tricine strongly binds Hg. Reactions were run in a thermostatted vessel at 19 C using strong light (>400 kerg/ cm’-sec). In all cases, the chloroplasts were incubated with HgCl, for 30 see before the addition of donors, acceptors, or inhibitors. Electron transport using oxidized p-phenylenediamines, sub- stituted p-benzoquinones, or ferricyanide as the electron ac- ceptor was followed spectrophotometrically, as described else- where (19). MV reduction was measured as oxygen uptake (10) with a Clark-type membrane-covered electrode. ATP forma- tion was assayed using a modified procedure of Avron (2). Ra- dioactivity was measured using the Cerenkov technique of Gould et al. (4). RESULTS AND DISCUSSION Figure 1 shows the effect of low concentrations of HgCl, on electron transport and phosphorylation using three different electron donor-acceptor systems. As previously reported (11), the over-all Hill reaction with ferricyanide (Fig. 1A) or MV (Table I) as the electron acceptor is inhibited in direct propor- tion to the added HgCl, up to approximately 35 to 40 nmoles of HgCl,/ mg Chl. Although very high concentrations of HgCl, (greater than 1 meIe/mg Chl) result in nonspecific electron ethylpiperazinepropanesulfonic acid; FeCy: potassium ferricyanide; DBMIB: dibromothymoquinone; DMQ: 2,5-dimethyl-p-benzoqui- none; PD": oxidized p-phenylenediamine; DAD“: oxidized dia- minodurene; MV: methylviologen. 680 190. AlHZO—rFOCy B)DCIPH2—.MV C) H20 —9 PD” '4 x at. (name) \ C O -.-. 500 1,. ._ _ a A O 1* 400,3 -- 92 xeriuamsi :1- ), xer. 1gp.) - a: he -\ .. 2 .-.- l 3 300 + + I; x E.T. (”11mm ._ er (COMPLETE) I) - —-O—-——O— ° W \ ET (COMPLETE) a 200 "i D—-D — ATP _ a i- 11' ------ . —————— .— o ‘ ET (- P.) E.T. (~P.) h - N "' ‘ ATP ' O 5 100 '-o---—0 —————— o- __ q ‘ --__ ATP E; .' O ————— .0- G 1 1 1 1 1 1 0.2 0.4 0.2 0.4 0.2 0.4 HgClz (uMolos/mg Chlorophyll) FIG. 1. Effect of Hng on electron transport (E.T.) and phos- phorylation associated with various electron transport pathways. The reaction mixture (2 ml) contained 50 mM HEPPS-NaOH buffer (pH 8.2). 2 mM MgC12, 0.1 M sucrose, 1 mM ADP, 5 mM NEhHaaPOt, chloroplasts containing 40 pg Of chlorophyll, and the indicated donor-acceptor system. These systems were: A, 0.4 mM ferricyanide; B, 0.4 mM DCIPHe, 2.5 mM L-ascorbate and 50 pM MV; C, 0.5 mM p-phenylenediamine (PD) plus 1.4 mM ferricyanide. When added, methylamine was 10 mM. In the DCIPH.» -> MV system (B), l 1.1M DCMU was added to block electron transport from photosystem 11. When PD... was the electron acceptor (C), 0.5 uM DBMIB was added to block the photosystem I component of PD... reduction (1). Note that only the H20 -* PD... system, which does not utilize the rate-determining coupling site after plastoquinone, is insensitive to inhibition by HgClo. Rates of electron transport and ATP forma- tion are given in nmoles/hr-mg chlorophyll. Table I. Efl’ect of HgCI, on Phot0p/tosphorylation in Spinach C Itloroplasts with Various Electron Acceptors The reaction mixture (3 ml) contained 20 mM HEPPS-NaOl—l buffer (pH 8.2), 50 mM sucrose, 1 mM MgCl , 1 mM ADP, 5 mM Na, H3’P04, chloroplasts containing 60 pg of Chi, and the indicated electron acceptor system. These systems were: 1: 50 pM MV; 2: 10 uM DBMIB plus 0.4 mM ferricyanide; 3: 0.5 mM DMQ plus 0.4 mM ferricyanide; 4: 0.5 mM DAD plus 1.4 mM ferricyanide. A low concentration of dibromothymoquinone (0.5 pM) was added to reactions 3 and 4 to inhibit the photosystem I component of DAD,“ and DMQ reduction (2). Rates are given in nmoles ATP/hr-mg Chl. Note that only the H30 -» MV system is sensitive to HgClz. ‘ Phosphorylation Rate Experiment Electron Acceptor Micromoles “ng added/mg Chl I ) None 0.05 l 0.25 No i nmoles A TP/hr-mg CH 1 MV i 200 102 105 2 DBMIB I 47 46 42 3 DMQ i 71 65 64 4 DA D.,. i 217 200 195 transport inhibition (11, 13), the low levels of HgCl, used here (less than I meIe/ mg Chl) inhibit ATP formation (and that portion of the electron transport dependent upon phosphoryla- tion) to a plateau of approximately 50%. Contrary to the re- sults of Miles et al. (16), however, we find the same degree of inhibition by HgCl2 when electron transport is measured spec- trophotometrically or as oxygen evolution. Neither basal (—Pi) or uncoupled (+methylamine) electron transport is significantly affected by HgCl,, indicating that HgCl2 does act as an energy- transfer inhibitor rather than an electron-transport inhibitor at these concentrations (11). Chloroplasts which have been un- coupled by EDTA treatment, which removes CF1 (l), are also insensitive to H gCl,. DCIPH, (in the presence of DCMU and ascorbate) donates electrons at a point before the rate-limiting coupling site on the electron transport chain (i.e. before cytochrome f) (6, 9, 15). It has recently been shown that the photosystem I-depend- ent partial reaction DCIPH, -) MV includes only the rate- limiting coupling site after plastoquinone and not the coupling site before plastoquinone (6). Figure 18 shows that HgCl, af- fects the partial reaction DCIPH, -> MV and the over-all reac- tion H20 -> FeCy similarly, indicating that both pathways in- clude the HgCl,-sensitive site. Moreover, since this coupling site constitutes the primary rate-limiting step for both electron transport reactions, a similar 50% sensitivity to HgCl2 should be observed for both systems (compare Fig. 1, A and B). However, when electrons from water reduce lipophilic ac- ceptors (class III acceptors [19]), such as oxidized p—phenyl- enediamine (PD,,), a different effect of HgCl is observed. These lipophilic oxidants (in the presence of the plastocyanin inhibitors KCN or poly-L-lysine (17) or the plastoquinone an- tagonist dibromothymoquinone (12, 20)) accept electrons at a point in the electron transport chain before plastoquinone (5). Thus the partial reaction H20 -9 PDox includes the photosystem II-dependent phosphorylation site, but not the rate-determin- ing site after plastoquinone. As Figure 1C shows, electron transport and phosphorylation associated with the partial reac- tion H20 -> PD... is completely insensitive to HgCl, inhibition. Several other class III acceptors were also tested for HgCl2 inhibition with similar results (Table I). The absence of inhibition by HgCl, with class 111 electron acceptors is probably not a result of fortuitous reaction condi- tions that mask the inhibition. In all cases, chloroplasts were incubated with HgCl, for 30 see before the addition of the ac- ceptor system. Since SH-compounds can reverse HgCl, inhibi- tion (11), it seems likely that Hg“ is reacting with a membrane sulfhydryl group. Thus it is unlikely that binding between Hg”+ Table 11. E fleet of Med. on Posti/Iumination ATP Formation (X g) HgCh, triphenyltin chloride, or methylamine (at the final con- centrations indicated) were present in the dark, phosphorylation stage only. Chloroplasts containing 100 mg OfChl were illuminated with white light (>400 kergs/cm1-sec) for 20 sec in a continuously stirred reaction mixture (2 ml) containing 0.1 M sucrose, 50 mM NaCl, 2 mM MgC12, 10 mM MES-NaOH buffer (pH 6.0), and 5 pM pyocyanine. Immediately after shutting off the light, 0.25 ml of a stock solution containing the test compound was added to the reaction mixture with a syringe. After 5 sec 0.75 ml of a strongly buffered ADP/phosphate mixture (0.15 M HEPPS-NaOH buffer (pH 8.0), 3 mM ADP, 15 mM NazHi’PO.) was quickly injected to initiate ATP formation. After 20 see the dark phosphorylation was terminated by addition of 0.5 ml of l N perchloric acid. All reactions were run in a thermostatted cuvette at 19 C. Note that HgCl, did not significantly affect the yield of X3 at concentrations which strongly inhibit steady state phosphorylation. The tributyl- tin analog triphenyltin, which is a potentenergy transfer inhibitor in chloroplasts (unpublished observations of J.M.G.), does abolish XE, however, as does the uncoupler methylamine. Addition (dark stage) ATP Formed Inhibition nmoles/mg C hl-fi % None 64 HgCI, (20 nmoles/mg Chl) 58 9.5 HgCl, (200 nmoles/mg Chl) 52 18 Triphenyltin chloride (10 MM) 8.8 87 Methylamine hydrochloride (5 mM) 12 81 and the quinonediimides or substituted benzoquinones used here as class III acceptors would be strong enough to reverse the inhibition. Furthermore, UV spectra Of DMQ, DAD“, PD,,, and DBMIB remained virtually unchanged in the pres- ence of high concentrations (33 pM) of HgCh, again suggesting that little or no binding is occurring. This evidence suggests that HgCl, is not interacting chemically with the class III ac- ceptors used here and supports our contention that the energy- coupling site close to photosystem II is insensitive to HgCl,. Table II presents evidence that HgCl, acts on an early step of energy conservation rather than a terminal reaction of ATP formation. Concentrations of HgCl,, which strongly inhibit steady state phosphorylation (e.g. H,O -> MV), were demon- strated to have virtually no effect on the chloroplast’s ability to synthesize ATP in the dark after brief illumination (X3) (8). However, methylamine, an uncoupler, and triphenyltin chlo- ride, an energy transfer inhibitor which acts on a terminal re- action of ATP formation, decrease the yield of XE significantly (8; unpublished observations of J.M.G.). The idea that HgCl, affects an early stage of energy transfer is also supported by the fact that the trypsin activated ATPase activity of CF1 and whole chloroplasts is not affected by the low levels of HgCl, used here (data not shown). The results presented herein lend strong support to the argu- ment that the two known sites of energy coupling associated with noncyclic electron flow in chloroplasts may exhibit mech- anistic differences in their mode of energy transfer (5, 6). It has previously been shown that these coupling sites differ in their response to pH (5, 6), uncouplers, ADP + P. (7), and in phos- phorylation efficiency (P/e. ratio) (6). In addition, we have now demonstrated that the energy-transfer inhibitor HgCl, is spe- cific for the rate-determining coupling site after plastoquinone and before cytochrome f. This selectivity may be due to a vari- able accessibility of HgCl, to the coupling sites, or, altema- tively. to a basic difference in the mechanism of the early steps of energy conservation at the two sites. A discussion of why the HgCl, sensitive coupling site is only partially sensitive (50%) to the inhibitor is beyond the scope of this report. In a subsequent publication, a more detailed study of HgCl, inhibition of chloroplast reactions will be presented, with emphasis on the nature of HgCl, inhibition and the signifi- cance of the 50% inhibition plateau. Acknowledgments—The authors wish to thank S. Izawa for many valuable sug- gest ions. LITERATURE CITED 1. Avnox. M. 1963. A coupling factor in photophosphorylation. Biochim. Biophys. Acta 77: 699-706. 2. AVRON. M. 1960. Photophosphorylation by swiss chard chloroplasts. Biochim. Biophys. Acta 40: 257—285. 191 q 10. 11. 13. I4. 16. 17. 18. 19. 20. 21. . Bhutan, H. AND W. A. CRAMER. 1972. Localization of a site of energy coupling between plastoquinone and cytochrome f in the electron transport chain of spinach chloroplasts. Biochemistry 11: 1155—1160. . GOULD. J. M., R. CATHER, AND G. D. WINGET. 1972. Advantages of the use of Ccrnkov counting for the determination of 33F in photoplnisphorylation re- search. Anal. Biochem. 50: 540—548. . GOULD, J. M. AND S. IZAWA. 1973. Photosystem II electron transport and phos- phorylation with dibromothymoquinone as the electron acceptor. Eur. J. Biochim. 37: 185—192. . Gorw. J. M. AND S. IZAWA. 1973. Studies on the energy coupling sites of photo- phosphorylation 1. Separation of site I and II by partial reactions of the chloroplast electron transport chain. Biochim. Biophys. Acta 314: 211-223. . GorLD. J. M. AND D. R. Our. 1973. Studies on the energy coupling sites Of photophtisphorylation III. The different effects of methylamine and ADP plus phosphate on electron transport through coupling sites I and II in iso- lated chloroplasts. Biochim. Biophys. Acta 325: 157—166. . HIND, G. AND A. JAGENDORF. 1965. Separation of light and dark stages in pho- tophosphorylation. Proc. Nat. Acad. Sci. U.S.A. 49: 715-722. . IZAWA, S. 1968. Effect of Hill reaction inhibitors on photosystem I. In .° K. Shi- bnta. ed.. Comparative Biochemistry and Biophysics of Photosynthesis. Uni- versity Park Press. State College, Pa. pp. 140—147. IZAWA. S.. T. N. CONNOLLY. G. D. WINOET. AND N. E. Goon. 1969. Inhibition and uncoupling of photophosphorylation in chloroplasts. Brookhaven Symp. Biol. 19: 169—187. IZAWA. S. AND N. E. GOOD. 1969. Effect of p-chloromercuribenzoate and mercuric ion on chloroplast photophosphorylation. Progress in Photosyn- thesis Research, Vol. III. pp. 1288—1298. . IZAWA, S.. J. M. GOULD, D. R. Oar, P. Faxes, AND N. E. Goon. 1973. Elec- tron transport and phosphorylation in chloroplasts as a function of the elec- tron acceptor III. A dibromothymoquinone-insensit-ive phosphorylation asso— ciated with photosystem II. Biochim. Biophys. Acta 305: 119-128. KIMIMURA, M. AND S. KATOH. 1972. Studies on electron transport associated with photosystem II. Functional site of plastocyanin: inhibitory effect of IIng on electron transport and plastocyanin in chloroplasts. Biochim. Biophys. Acta 283: 268—278. KRAAvENHOF. R., S. IZAWA, AND B. CHANCE. 1972. Use of uncoupling acridine dyes as stoichiometric probes in chloroplasts. Plant Physiol. 50: 713—718. . LARKUM, A. W. D. AND W. D. BONNER. 1972. The effect of artificial electron donor and acceptor systems in light-induced absorbance responses of cyto- chrome I and other pigments in intact chloroplasts. Biochim. Biophys. Acta 267: 149—159. Mites, D., P. BOLDN, S. FARAG, R. GoomN, J. Lure, A. MOL‘STAFA, B. Ron- BIGUEZ, AND C. WEIL. 1973. Hgtt-A DCMU independent electron acceptor of photosystem II. Biochem. Biophys. Res. Commun. 50: 1113—1119. Our, D. R., S. IzAWA, N. E. GOOD, AND D. W. KnocsuNN. 1973. The effects of the plastocyanin antagonists KCN and poly-L-lysine on partial reactions in isolated chloroplasts. FEBS Lett. 31: 119—122. OUITRAKUL. R. AND S. IZAWA. 1973. Electron transport and phosphorylation in chloroplasts as a function of the electron acceptor II. Acceptor-specific in- hibition by KCN. Biochim. Biophys. Acta 305: 105-118. SARA, S.. R. OUITRAKUL. S. IZAWA AND N. E. Goon. 1971. Electron transport and phosphorylation in chloroplasts as a function of the electron acceptor. J. Biol. Chem. 246: 3204—3209. Tneas‘r, A. AND S. REIMER. 1973. Properties of photoreduction by photosystem II in isolated chloroplasts: an energy-conserving step in the photoreduction of benzoquinone by photosystem II in the presence of dibromothymoquinone. Biochim. Biophys. Acta. 305: 129-139. WrNoa'r, G. D., S. IZAWA, AND N. E. GOOD. 1969. The inhibition of photophos- phorylation by phlorizin and closely related compounds. Biochemistry 8: 2067-2074. APPENDIX VIII QUANTITATIVE RELATIONSHIP BETWEEN PHOTOSYSTEM I ELECTRON TRANSPORT AND ATP FORMATION QUANTITATIVE RELATIONSHIP BETWEEN PHOTOSYSTEM I ELECTRON TRANSPORT AND ATP FORMATION Donald R. Ortz Department of Botany and Plant Pathology Michigan State University East Lansing, Michigan 48824 Abstract 1. The Photosystem I-dependent transport of electrons from diaminodurene (DAD) to methylviologen is linear with reaction time and supports a constant rate of phosphorylation. However. if the DAD is not kept fully reduced by the presence of excess ascorbate. the oxidized DAD accumulates and begins to compete with the methylviologen as the electron acceptor. Thus, although the rate of ATP formation remains unchanged, an increasing proportion of the electron transport becomes cyclic and hence unmeasured. This leads to a rapid increase in the apparent efficiency of phosphorylation which is misleading. 2. In contrast, it is known that the oxidized form of 3,3'- diaminobenzidine (DAB) polymerizes to form an insoluble substance which should not be available to serve as an electron acceptor. However DAB is not a satisfactory donor of electrons in Photosystem I reactions for two reasons: the rate of electron transport quickly falls with reaction time and the oxidized form of DAB seems to be an exceptionally efficient electron acceptor near the beginning of the period of illumination when I92 193 it is presumably not yet polymerized. Thus in the first 2-3 seconds of illumination when the reaction is still rapid much of the electron transport is cyclic and therefore unmeasured. especially in the absence of excess ascorbate. This cycling of electrons, which leads to an inflated apparent efficiency (P/e2 > 2), is particularly pronounced a low donor concentrations. 3. When cyclic electron transport is avoided by the use of ascorbate or by the selection of appropriate reaction times, both DAD and DAB support phosphorylation with an efficiency which is approximately I half of the efficiency exhibited by the overall Hill reaction. The same is true when 2.5-diaminotoluene. tetrachlorohydroquinone. 4,5-dimethyl- o-phenylenediamine, and reduced 2,6-dichlorophenolindophenol serve as electron donors. With these six substances the phosphorylation effi- ciencies were 0.57 i 0.1 molecules of ATP formed for each pair of elec- trons transferred (P/ez). In the same chloroplasts preparations, the transport of electrons from water to methylviologen supported phosphoryla- tion with a P/e2 of 1.2. Introduction Transport of electrons from water to the reducing end of the photosynthetic chain in isolated chloroplasts is coupled to phosphoryla- tion at two sites. one closely linked to Photosystem II (l-5) and one located between plastoquinone and cytochrome f (6.7). We have called the former Site II and the latter Site I. This paper deals with the efficiency of ATP formation at Site I. 194 Gould and Izawa (8) have recently presented evidence that the DCMU-insensitive transfer of electrons from reduced 2,6-dichlorophenol- indophenol +-Photosystem I +~methylviologen appears to utilize Site I. They have suggested that the efficiency of Site I may be about 0.6 ATP molecules per pair of electrons transported. Ort and Izawa (9) have presented preliminary data on a number of Photosystem I reactions using several other exogenous donors. All of the reactions tested supported phosphorylation with a similar efficiency (P/e2 = 0.6 i 0.08). We have pointed out that these P/e2 values could be explained adequately by assuming that not only the DCIPH2 +tMV systems but all Photosystem I reactions utilize one of the two sites of phosphorylation in the main electron transport chain (i.e. Site I).' However. Goffer and Neumann (ll) have recently reported different results. These investigators presented data showing that the photooxidation of 3,3'-diaminobenzidine via Photo- system I supported phosphorylation with an apparent efficiency (P/ez) approaching 1.0. The use of diaminobenzidine should have the major advantage of eliminating unmeasured cyclic electron flow. Cyclic electron trans- port is a complication which is unique to these chloroplast reactions since once the exogenous donor has been oxidized it may begin to function as an electron acceptor. Goffer and Neumann reasoned that oxidized diaminobenzidine (DAB) would not be available as an electron acceptor since it forms an insoluble polymer (ll,l2). Therefore the high P/e2 ratios reported by Goffer and Neumann must be taken seriously. Nevertheless, such high efficiencies of ATP formation asso- ciated with DAB oxidation are difficult to reconcile with the emerging 195 picture of chloroplast phosphorylation. Judging by the method of chloroplast isolation employed by Goffer and Neumann (ll) one would predict an overall non-cyclic (H20 + MV) photophosphorylation efficiency not over 1.0. If DAB-supported phosphorylation approaches this effi- ciency. one must assume either that this particular Photosystem I reac- tion utilizes both Site I and Site II or that still another site of phosphorylation is involved in DAB oxidation. The former seems unlikely since there is evidence that Site II cannot operate in the presence of DCMU (l-3, 13). The latter invokes a special site of phosphorylation outside the main chain of electron transport like the site which has been invoked to account for the phenomenon of "cyclic" phosphorylation (l4). We find no compelling evidence for such an extra site either in the literature or in our own investigations. These considerations prompted us to reinvestigate the phos- phorylation efficiencies of Photosystem I-dependent reactions using DAB and other exogenous electron donors. The apparent high P/e2 values (exceeding 2) observable in a brief illumination with suboptimal con- centrations of the DAB can be attributed to the presence of a short- lived cyclic electron transport that occurs in the first three seconds of illumination. We conclude that the efficiency of phosphorylation (P/ez) during DAB oxidation is actually about 0.5 and is thus no dif- ferent from the efficiencies observed with any of the other donors. Experimental Procedures Chloroplast Isolation. Chloroplasts were isolated according to procedures detailed elsewhere (l6). Spinach (Spinacia oleracea L.). 196 white mustard (Brassica hirta Moench), and Swiss Chard (Beta vulgaris cicla) were purchased from local markets. Tobacco (Nicotiana tabacum cv Xanthi-nc) leaves were selected from greenhouse grown plants. When Swiss Chard or tobacco was used care had to be taken to separate the chloroplast pellet from a white pellet which formed during the first centifuge spin. In the case of Swiss Chard, this pellet is probably calcium oxalate (l7). In tobacco it may be starch granules. Reagents. Superoxide dismutase was prepared from fresh bovine erythrocytes by the procedure of McCord and Fridovich (18). The purified enzyme had a specific activity of 3000 U/mg protein, assayed as the inhibition of cytochrome c reduction by xanthine oxidase. Diaminodurene (DAD), 2,5-diamintoluene (DAT) and 3,3'- diaminobenzidine (DAB) were recrystalized as the pure white dihydro- chloride salts from charcoal-treated aeueous solutions by adding con- centrated HCl at 0°C. Tetrachlorohydroquinone (TCHQ) and 4,5-dimethyl- grphenylenediamine (DMPD) were recrystalized as white leaflets from charcoal-treated aqueous ethanol and ethylene glycol, respectively, by lowering the temperature to -25°C. Fresh solutions of these compounds were made up daily in 0.01 N HCl (DAD, DAT, DAB) or in 1:1 v/v mixture of ethanol-ethylene glycol (TCHQ, DMPD). At all times concentration of organic solvent in final reaction mixture was less than 2%. 0.1 M solu- tions of D-ascorbate adjusted to pH 6.5 with NaOH (at 0°C) were also made up daily. ‘Electron Transport and Phosphorylation Assays. The photo- reduction of methylviologen (MV) was measured as the 02 uptake resulting from aerobic reoxidation of reduced MV. Electron transport with exogenous 197 donors was also measured as 02 uptake. A membrane-covered Clark-type oxygen electrode was employed for these measurements. When exogenous electron donors were used, the observed rate of 02 uptake in the light was corrected for the slow rate of dark aerobic oxidation of donors which accompanied some of the reactions. This rate never exceeded 15% of the rate observed in the light. The intensity of the orange actinic light (600-700 nm) was approximately 600 Kergs-sec 'l-cm'z. Short periods of illumination were timed with an electronically controlled solenoid-shutter arrangement. This device delivered "flashes" which were of essentially uniform inten- sity throughout their duration for illumination times down to 0.1 sec. ATP formation was measured as the residual radioactivity after extraction of the 32P-labeled orthophosphate as phosphomolybdic acid in butanol-toluene. Radioactivity was determined either by Cerenkov radia- tion (l9), or if the ATP containing aqueous phase was colored, with a Gieger-Muller tube. Results Phosphorylation duringkthe oxidation of diaminodurene by Photo- system 1. Reactions involving the donation of electrons to Photosystem I are usually carried out in the presence of excess ascorbate. This is to prevent the accumulation of the oxidized form of the electron donor, since the newly formed oxidant can accept electrons and in so doing contribute an unmeasured component to the electron flow. In the presence of ascorbate the photochemical reduction of electron acceptors such as methylviologen can be measured as the uptake of oxygen since the reduced 198 acceptor is autoxidizable. If excess superoxide dismutase is present to prevent the chemical oxidation of ascorbate or of the donor by the intermediate superoxide radical, the uptake of one molecule of oxygen in the absence of catalase represents the biological transport of pre- cisely one pair of electrons (9): 02 = e2 and P/O2 = P/e2 In all of the experiments described below sufficient amounts of superoxide dismutase (400 ug or 1000 units) were added to the reaction mixtures to ensure the above relationships (cf. Figure 4) and the chloro- plast preparations used were essentially free of catalase activity. The importance of using excess ascorbate is illustrated by Figures 1 and 2. In the absence of ascorbate the apparent value of the P/ez ratio is high and increases with time as oxidized DAD accumulates (Figure 2). When ascorbate is present as an electron pool for the reduction of DAD and the requisite SOD has been added, the P/O2 ratio is concentration-independent and is approximately 0.55; if ascorbate is not included in the reaction mixture the observed P/O2 displays some concentration dependence and is up to 30% higher in value. Figures 1 and 2 show that the increase in P/O2 ratio is not due to an increase in phosphorylation rate but rather to a diminution in the rate of 02 uptake. It is logical to conclude, therefore, that in the absence of ascorbate oxidized DAD is competing for electrons with methylviologen and as a consequence 02 uptake is not a full measure of the electron transport occurring (i.e. P/O2 > P/ez). The time course curves of Figure 2 were obtained using two series of identical reaction mixtures (with or without ascorbate) 199 illuminated for different lengths of time (1 to l4 s.) This procedure is essential to determine the true relationship between electron flow and phosphorylation, especially when the P/Oz ratio is suspected of being a function of time (as in the case of the diaminobenzidine system described later). Both 02 uptake and phosphorylation proceed linearly in the presence of DAD plus ascorbate, again with a P/O2 ratio of 0.55. The same P/O2 ratio can be briefly observed in the absence of ascorbate. However in this case by the third second of illumination oxygen uptake is no longer a full measure of the electron transport occurring. Indeed lO-second illumination produces an amount of oxidized diaminodurene which intercepts more than half of the total electron flux even in the presence of 0.1 mM methylviologen, although the oxidized diaminodurene is pre- sumably no more than 20 uM. The change in P/O2 in the absence of ascor- bate is dramatized by the upper inset of Figure 2 where the change in ATP concentration «SATP) and 02 concentration 0502) are presented for each illumination period. (For details ofASATPAaOz computation, see legend for Figure 2.) Phosphorylation during;the oxidation of 3,3'-diaminobenzidine (DAB) by Photosystem I. Another donor of electrons to Photosystem I which has received attention recently is 3,3'-diaminobenzidine (lO,l2, 20). As has already been pointed out the oxidized form of this com- pound polymerizes to form an insoluble substance which cannot then serve as an electron acceptor and, consequently, one would not expect diamino- benzidine to catalyze any "cyclic" or hidden electron transport (l0). In fact the behavior of diaminobenzidine as an electron donor is quite different from the behavior of diaminodurene, and these 200 differences are not at all to the advantage of DAB. The maximum rate of electron transport with DAB is only about one fourth of the rate with DAD and even this rate is not sustained (compare Figure l, 2 with 3-5). Moreover, the apparent efficiency of phosphorylation is dependent on the concentration of DAB in the presence or absence of ascorbate. This concentration-dependence of the observed P/O2 ratio seems to be the result of an initial cyclic, unmeasured electron transport which is, for some reason, more significant at low donor concentrations. Part of the data presented in Figure 3 closely reproduce the data of Goffer and Neumann (l0). The conditions they selected included a low concentration of DAB (O.l4 mM), no ascorbate and lO-second illumina- tion. They reported a P/O2 ratio of 0.83, almost exactly the value shown for these conditions in our figure. But Goffer and Neumann apparently were not aware of the peculiar effect of the DAB concentration on the apparent phosphorylation efficiency or of the influence of ascorbate thereon. As the increasing concentration of DAB raises the rates of 02 uptake and phosphorylation, the P/e2 ratio quickly falls to a plateau at approximately 0.5 which is essentially the level observed for DAD photooxidation (cf. Figure l). The P/e2 ratio stays relatively constant (0.4-0.5), over the entire range of DAB concentrations, showing no sign of approaching l.O (Figure 3, inset). However, when ascorbate is present, it is mandatory to add large amounts of superoxide dismutase to suppress the superoxide-ascorbate interactions which lead to an exaggerated 02 uptake and consequently to an apparent low P/e2 (see Figure 4). It is interesting to note that the 02 uptake is scarcely affected by superoxide 201 dismutase in the absence of ascorbate. This suggests the reaction of DAB with ~02- is too slow to compete with the spontaneous dismutation of -02'. The cyclic nature of the electron transport which occurs at the beginning of the illumination period when DAB is the donor is clearly illustrated by Figure 5. The cycle is particularly pronounced in the absence of ascorbate (Figure 5a) where the P/O2 shows momentarily an unrealistically high initial value of over 2.0. In this case the high P/OZ results from a temporary suppression of oxygen uptake, indicated by the S-shaped uptake curve. The suppression of oxygen uptake illustrated in Figure 5a seems to result from oxidized but not yet polymerized DAB partially replacing MV as the electron acceptor, since the effect is largely removed if ascorbate is present to reduce the photooxidized DAB (Figure 5b). This interpretation is upheld by the experiment shown in Figure 6. During the initial phase of illumination (< 25), DAB-supported phosphorylation is practically independent of the presence or absence of the electron acceptor MV. No significant oxygen uptake occurs when the DAB concentration is low and no ascorbate is present. Therefore the electron transport which supports phosphorylation must be truly cyclic and not pseudo-cyclic. Phosphorylation duringfithe oxidation of other exogenous elec- tron donors by Photosystem I. We have shown above that the observed efficiencies of phosphorylation associated with the oxidation of DAD and DAB by Photosystem I are actually very similar if precautions are taken to eliminate unmeasured electron flow (about 0.5). The same is 202 true of the photooxidation of a variety of other Photosystem I donors (Table I) and with chloroplasts from a variety of plants (Table II). It should be noted that the conditions used for chloroplasts from various plant species were conditions optimized for spinach chloroplasts and therefore the somewhat lower efficiency with Swiss Chard may reflect suboptimal conditions for chloroplasts from this plant. All of these electron donors (DAD, DAB, DAT, TCHQ, DMPD and DCIPHZ) were photooxidized entirely by System I since the oxidation occurred in the presence of DCMU. Not only is Photosystem II unneces- sary for the oxidation of these electron donors. Plastoquinone, which presumably acts as a carrier of electrons between the two photosystems, seems also to be unnecessary since the oxidations of the donors and the associated phosphorylation reactions are completely insensitive to the presence of the plastoquinone analogue dibromothymoquinone (2l-23), see Table III. Discussion The data presented in this paper emphasize a point long known or suspected, that measurements of the efficiencies of phosphorylation associated with oxidations of electron donors by Photosystem I are often bedeviled by the presence of various amounts of "cyclic" electron trans- port (l4). This is to say, the oxidized donor formed as a consequence of the reaction can serve as an acceptor in a process which has no net oxidation-reduction result. It is necessary to understand the nature of the conditions affecting such cyclic transport if this cyclic trans- port is to be eliminated. 203 Cyclic electron transport in the DAD=oxidizing system is easily understood and can readily be eliminated. As the reaction prog- resses the accumulating oxidized DAD accepts an increasing proportion of the electrons. competing very effectively with MV. Fortunately the level of oxidized DAD can be kept extremely low by having an excess of ascorbate present. In the presence of excess ascorbate there is no change in oxygen uptake or ATP formation orapparent phosphorylation efficiency during the reaction and therefore very probably no "cycle“ whatsoever. Cyclic electron transport in the DAB-oxidizing system is not nearly as easily understood. Cyclic transport seems to be at a maximum near the beginning of the reaction when the concentration of accumulated oxidation products must be very low. This curious behavior of DAB is probably related to the chemical nature of the primary oxidation product of the polyamine, which should be both a fairly strong oxidant and highly unstable substance. Being a strong oxidant and formed in situ it should compete especially favorably with MV for electrons from Photo- system I. Therefore a large cyclic component of the electron transport could develop almost as soon as transport starts. However, as the oxidized product accumulates it begins to polymerize (ll). This poly- merization like many other polymerization reactions is presumably auto- catalytic and therefore, once polymerization has started, the level of oxidized DAB should fall and the cycle diminish. For this reason factors which limit the accumulation of oxidized DAB, such as low DAB concentration and above all short reaction times, might be expected to increase the proportion of cyclic transport by delaying or preventing 204 the onset of polymerization (Figure 5a). However, in the presence of a large excess of ascorbate, the concentration of the primary oxidation product of DAB may at no time reach levels capable of supporting high rates of cyclic transport (Figure 5b). When we have eliminated the contribution of such hidden cyclic transport systems, all the Photosystem I oxidations of exogenous electron donors seem to support phosphorylation with about the same efficiency (P/e2 = 0.57 i 0.1). Moreover this efficiency is approximately one half of the efficiency with which phosphorylation is coupled to the transport of electrons from water through both photosystems and both sites of phosphorylation (P/ez = 1.15 i O.l). It remains to be shown, however, that the site of phosphorylation associated with the oxidation of these exogenous donors is actually one of the sites associated with the Hill reaction. One reason for equating phosphorylation at Site I of the overall Hill reaction to the Photosystem I-dependent, electron donor- supported phosphorylation described herein resides in the principle of economy of hypotheses. As we have suggested, the efficiency of phos- phorylation in the overall Hill reaction can be explained adequately by summing the efficiency of Site II phosphorylation and the efficiency of these electron donor-supported phosphorylations (8). Thus in terms of phosphorylation efficiency there is no real need to postulate a separate site of phosphorylation to accommodate either "donor"-supported or "cyclic" Photosystem I phosphorylations. Phosphorylation supported by the System I photooxidation of exogenous donors has more than efficiency in common with Site I 205 phosphorylation. Gould (24) has demonstrated that the rates of photooxidation of DAD and DAT are nearly doubled when phosphorylation occurs concurrently. We have observed similar stimulations of electron transport by phosphorylation when DAB or DMPD are the donors employed (data not presented). Furthermore, several groups have noted that the photooxidation of DCIPH2 is also enhanced by phosphorylation conditions (8,25,26). Such stimulations, although smaller, bear a conspicuous resemblance to that stimulation of the Hill reaction (H20 + MV) by phosphorylation which has been attributed to the relief of a rate- limitation imposed by the coupling site between plastoquinone and cyto- chrome f at Site I (6). Data presented in Table III of this communica- tion indicate that these electron donors interact with the electron trans- port chain after plastoquinone. In addition McCarty (27) has pub- lished evidence that the photooxidation of DAD involves cytochrome f. The sensitivity which the Hill reaction phosphorylation displays to the energy transfer inhibitor Hg++ (24,28) and the sensitivity of the phos- phorylation efficiency of the Hill reaction to pH (8) do not reside at Site 11 and therefore apparently reside at Site I. Photosystem I "donor" reactions, like Site I reactions, are sensitive to Hg++ and low pH (8, 24,28). When all of the observations are viewed in concert they present a very strong case for the contention that the electron donor reactions we have discussed utilize one of the energy conservation sites (Site I) of the Hill reaction. One feature of the donor-supported phosphorylation which could be interpreted in favor of a separate energy conservation site (e.g. 14) is the exceedingly high rates of electron transport and ATP formation 206 supported by some of these exogenous electron donors. It is not easy to understand how a single electron transport reaction could be rate- determining and subject to similar control by phosphorylation when the rates of reaction differ by nearly an order of magnitude. The reason for this is still uncertain but the explanation may involve such para- meters as the local concentration and potential of the exogenous donor and its reactivity with an endogenous carrier (24). The precise value of the efficiency ratio reported here should not be a primary focus of attention because the value of approximately 0.6 for Site I which we have suggested is probably applicable only to conventionally prepared chloroplasts in which theP/e2 of the overall non-cyclic electron transport is approximately l.2. If the overall P/e2 approached 2.0, as may be the case with the "type D" (29) chloroplasts of Heathcote and Hall (30), we assume that the coupling efficiency at each site (Site I and 11) would approach 1.0, the efficiency believed to represent the theoretical maximum at a single site of energy conserva- tion. Acknowledgments The author wishes to thank Dr. S. Izawa for consultation during the course of the research. Thanks are also due to Dr. N. E. Good for critical reading of the manuscript and for the design and construction of the "flash" device used in this investigation. Footnotes 1. This work was supported by a grant (GBB7959X) from the National Science Foundation to Drs. N. E. Good and S. Izawa. 2. 207 Present Address: Department of Biological Sciences, Purdue University, West Lafayette, Indiana 47905. Literature Cited 1. 10. 11. 12. 13. 14. Ouitrakul, R. and Izawa, S. (1973). Biochim. Biophys. Acta. 305: 105-118. Izawa, S., Gould, J. M., Ort, D. R., Felker, P., and Good, N. E. (l973). Biochim. Biophys. Acta. 305: ll9-l28. Trebst, A. and Reimer, S. (1973). Biochim. Biophys. Acta. 305: ll9- 128. Gimmler, H. (l973). Z. Pflanzenphysical Bd. 68: 289-307. Kraayenhof, R., Izawa, S. and Chance, B. (l972). Plant Physiol. 50: 713-718. Avron, M. and Chance, B. (l966). Brookhaven Symp. Biol. 19: 149- 160. Bdhme, H. and Cramer, W. A. (1972). Biochemistry ll: llSS-ll60. Gould, J. M. and Izawa, S. (1973). Biochim. Biophys. Acta. 3l4: 211-223. Ort, D. R. and Izawa, S. (l974). Plant Physiol. 53: 370-376. Goffer, J. and Neumann, J. (1973). FEBS Letters 36: 6l-64. Seligman, A. M., Karnovsky, M. J., Wasserkrug, H. L. and Hanker, J. S. (1968). J. Ce11 8101. 38: 1-14. Nir, I. and Seligman, A. M. (1970). J. Cell Biol. 46: 6l7-620. Saha, S.. Ouitrakul, R., Izawa, S. and Good, N. E. (l97l). J. Biol. Chem. 246: 3204-3209. Avron, M. and Neumann, J. (1968). Ann. Rev. Plant Physiol. 19: l37-l66. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 208 Ort, D. R. and Izawa, S. (l973). Plant Physiol. 52: 595-600. H111, R. and Walker, 0. A. (1959). P1ant Physio1. 34: 240-245. McCord, J. M. and Fridovich, I. (1972). J. Biol. Chem. 247: 188- 192. Gould, J. M., Cather, R. and Winget, G. D. (1972). Anal. Biochem. 50: 540-548. Chua, N. H. (l972). Biochim. Biophys. Acta. 267: l79-l89. Bfihme, H. and Cramer, W. A. (l97l). FEBS Letters l5: 349-35l. Knaff, D. B. (1972). FEBS Letters 23: l42-l44. Bohme, H., Reimer, S. and Trebst, A. (l97l). Z. Naturforsch. 266: 341-352. Gould, J. M. (1974). Biochim. Biophys. Acta. In Press. Strotmann, H. and von Gosslen, C. (l972). Z. Naturforsch. 276: 445- 455. Arntzen, C. J., Neumann, J. and Dilley, R. A. (1971). J. Bioener- getics 2: 73-83. McCarty, R. (l974). Arch. Biochem. Biophys. l61: 93-99. Bradeen, D. A., Winget, G. D., Gould, J. M. and Ort, D. R. (l973). Plant Physiol. 52: 680-682. Hall, 0. 0. (1972). Nature New Biol. 235: 125-126. Heathcote, P. and Hall, D. O. (l974). Biochem. Biophys. Res. Comm. 56: 767-774. 209 Table I. Stoichiometric relationship between ATP formation and electron transport associated with the PS I-dependent oxidation of various elec- tron donors. Reaction conditions are similar to those in Figure 2, except that the Chl concentration for reactions with reduced 2,6- dichlorophenolindophenol (DCIPHZ) 4,5-dimethyl-9:phenylenediamine (DMPD), or tetrachlorophydroquinone (TCHQ) was 40 pg/2 ml while with 3,3'- diaminobenzidine (DAB) it was 20 ug/Z ml. The concentrations of electron donors used were: DAD (1.0 mM), DAB (l.0 mM), DAT (0.5 mM), TCHQ (1.25 mM), DMPD (1.25 mM), DCIPH2 (0.4 mM). The concentration of ascorbate present in each system was 2.0 mM. Rates for the exogenous donor reac- tion are given in nmoles 02 or ATP per hr per mg Chl. For the reference reaction which used H20 as the electron donor and involved both PS I and PS II chloroplasts containing 40 ug Chl were used and ascorbate, DCMU and 500 were omitted. Note that when water is the donor 0 = e2 but when exogenous donors are employed 02 = e2. System 02 uptake ATP P/ez DAD +-MV 1435 780 0.55 DAB + MV 510 245 0.48 DAT + MV 760 445 0.58 TCHQ +rMV 72 41 0.54 DMPD +-MV 180 91 0.5l DCIPHZ + MV l38 91 0.66 (H20 + MV) (155) (385) (1.24) .muocpmz mom magnumuoca cowum—omw ummpaocopcu Lou .Nm u o Locou :ospumpm «nu ma Lopez :pwz newcocz mm u we .mmm mam mcompummc Locou msocmmoxw .H mg mg» com magma .mgauxms compummc FE m exp 09 umunm mam: Pcu m: oc mew:_mpcoo mummpaoco—cu use umpuwso mew: now use .mumncoumm .zzuo .cocou ecu mm vmxopasm N mm: o I can: .mumacoumm zE o.F saw: 25 o.F mm: mcocou cocpumpm mmwcu mo comm mo cowumcpcmucou m5» .»Fm>wwuwammc m use P mmcamwu cw umcmppzo ecu o l.0 mM) with excess SOD and ascorbate present, where cyclic electron transport is negligible, P/02 = P/ez. Figure 4. Effect of superoxide dismutase on the 02 uptake and phosphoryla- tion associated with diaminobenzidine photooxidation. The concentration of DAB was l.0 mM and the concentration of ascorbate was l.0 mM when present. Other conditions as in Figure 3. Again, in the presence of jimlflmjjlu .9».n. _ . . h— 213 excess ascorbate and excess superoxide dismutase (> 100 ug/ml), P/02 = P/ez. Figure 5. Time courses of 02 uptake and phosphorylation with diamino- benzidine photooxidation in the presence or absence of ascorbate. The concentration of ascorbate was 1.0 mM (when present) and the concentra- tion of DAB was 0.l mM. All other reaction conditions were the same as in Figure 3. In Figure 5a ascorbate was absent. In Figure 5b ascorbate was included in the reaction mixture. Note that after l0-second illumi- nation the rate of the reaction began to decline in both a and b. Note that P/O2 = P/e2 andz_\.ATP/A02 =A ATP/Ae2 in the presence of ascorbate (5b) at illumination times greater than six seconds. In the absence of ascorbate or at illumination times less than six seconds a portion of the electron transport was "cyclic" and unmeasured. Therefore P/O2 > P/e2 and AATPASO2 >AATP/Ae2 under these conditions. Figure 6. Time course of DAB-supported phosphorylation and 02 uptake in the absence of ascorbate and methylviologen. The concentration of DAB was 0.l mM. All other conditions were as in Figure 3. DAB-supported phosphorylation in the presence of methylviologen is transposed from Figure 5a for comparison. 214 | o... _____________________________ illumination * .-A¢c IO sec F E.,/“WA A g- N P / OAOC j C) t 5 \ 3:759----------P.-__-9.- (LND.5 . E a i O - J l l l g?! 2000‘ o 1.0 2.0 .J c) C) \02 uptake (+ Ass) /02 uptake (-Asc) o A A . '\ nmoles 02 or ATP'h"'mg chlorophyll" ATP (3 Ace) O 1 1 1 1 O 0.5 |.O 1.5 2.0 Diaminodurene (m M) Figure l. Photooxidation of Diaminodurene with MV as the Elec- tron Acceptor in the Presence of DCMU. 215 "5” ° Low oao (0.1mm) -Acc 20 slflb A O 2 / q 3 O (I cI> (I> O 02 Uptake +Asc/ (.Agc) \ A 1 O I l l 1 L l l a”: o 4 8 l2 ‘" / u 02 uptake (eASd \ o KAT? (:Amcl “I ‘fl' 5 l nmoles Oz or ATP- ml rx mixture'I 8 10- ° 9 : o+ L F ~A8¢ A 5)" R N b y/ '1 5: 0,5 Lorna to “\ <> 0 a oAsc o. F§:;;::::‘. O l. r C) 2! ‘1 £5 13 l() l2! 144 Time of illumination (Sec) Figure 2. Time Courses of 02 Uptake and Associated Phosphorylation for Diaminodurene Photooxidation in the Presence and Absence of Ascorbate. umoles 0. or ATM"- mg chlorophyll" 600 400 200 216 |.O*- -------------------- ‘\‘/- A” illumunation A , lC) sec: '/ ' ' +Aoc l i l i ()0 LO 2() ON— 02 uptake (oAsc) Ki\”- A 02 uptake (-Asc) ATP (aAcc) / om / 5‘ —1 A A \ / ATP 14») 1 l 1 1 LG l.5 2.0 Diaminobenzidine (mM) Figure 3. Photooxidation of 3,3 -Diaminobenzidine with MV as the Electron Acceptor in the Presence of DCMU. 217 g 7500- llluminatlon l0 sec High DAB (1.0mM)"°f . E. .c o. (D h '2 ( Mel C J: _, o o 500 o o d is” -' A\ A A 11: / - 02 uptake if (A 1 q - 8C L. ATP (OAOQ 0 250r J cu .1 "-————————uo o I . . m .93 \ 2 ATP i-Asc) :. o l I O IOO 200 pg S.O.D./ml Figure 4. Effect of Superoxide Dismutase on 02 Uptake and Phos- phorylation Associated with Diaminobenzidine Photo- oxidation. 218 2.0-'\ Low DAB (O.lmM) L-o ’ no ascorbate N . o l- 5- 3 IO _.. a . - 93 5 - E :5 32' E -2.0 x h E E “l.5 <1 3. C) N N o -I.O o ‘\ g a. 2 C- -O.5 O 1 1 1 1 1 0 0 2 4 6 8 10 Time of illumination (Sec) Figure 5A. Time Courses of 02 Uptake and Associated Phos hor la- tion for Diaminobenzidine Photooxidation in t e A sence of Ascorbate. 219 1.07- b Low DAB (0.! mM) . . plus ascorbate (5 A? _. \\\£L\\\\“‘-‘—_-. £3 ; d \ 0.5 - ' ' E: _ o — q '- 10, 5 .. . ‘ , - L. - 7 a 7 .5. I!" E f 4 1— _ 2.0 E d. '5 3 - - l.5 ‘— 0 ON C37 2 _ - jg lCl it: (3 E c I I— - 0.5 ‘A C) I l l 1 1 C) C) 2: 14 (5 El Ii) Time of illumination (Sec) Figure SB. Time Courses of 0 Uptake and Associated Phos horyla- tion for Diaminobgnzidine Photooxidation in t e Presence of Ascorbate. nmoles ATP (or 02) °ml rx mixture‘l Figure 6. Time Courses of DAB-supported Phosphorylation (and 02 220 3L Low DAB (O.lmM) no ascorbate ATP (+MV) \ BO 1 \ATP (-MV) ll '1 1- / 02 uptake (-MV) 7 /’/O‘_/, .. ’,.O""" o ---- 1 1 1 1 C) 2 4’ 6 8 Time of illumination (Sec) Uptake) in the Absence of Ascorbate and MV. Typed and Printed in the U.S.A. Professional Thesis Preparation Cliff and Paula Haughey 144 Maplewood Drive '7 East Lansing. Michigan 48823 f . Telephone (517) 337-1527